Surprisingly, data exists to address lots of these points, and I’ve attempted here to summarize (and link to) a number of prior posts aimed at debunking these myths and sharing a few observations on them.

So here are the top six myths or misperceptions about the biotech venture industry:

1. Returns in Life Science venture investing lag other venture capital sectors.  The data do not support this premise.  Both in an analysis by Bijan Salehizadeh and myself in Nature Biotech (here), as well as in an unpublished analysis from Correlation Ventures, healthcare venture capital actually outperformed all other venture sectors in the past decade, in particular with realized returns.  In the latter study, healthcare services, followed by biopharma, were the two best sub-sectors in venture.  This outperformance exists at the median, top quartile, and even top decile return thresholds for the last decade.  As this chart shows (republished from here), of all the deals with an initial financing between 2000-2010 that have exited, roughly 8% of life science deals vs 4% of tech deals delivered above 5x realized returns.

One part of the disconnect in perception is the impact of unrealized returns on holding values of funds, especially during fundraising cycles.  The “paper” value of write-ups looks good and validate the “market value” of a deal, but these are notoriously poor predictors of realized returns for a fund: as cited in a Kauffman report last spring, “write-ups in value prove seductive to [LP] investors even though they are not consistent predictors of a fund’s ultimate performance” (here).  Further, they significantly bias against Life Sciences due to the “unrealized return problem” (discussed here by Bijan); “Tech unrealized IRR overstates actual tech venture realized returns and Healthcare unrealized IRR understates it.”  One example is Avila Therapeutics: it was held at roughly the same share price for 18 quarters (~cost) and then in six months was written up to its ~6x exit value.  Its hard to argue value wasn’t being created during the preceding 4.5 years.

But a bigger issue underlying this perception is the reality of the “long tail” of the return distribution.  This is best summed up by the question of why aren’t there life science VC funds with >5x-fund level realized returns, like Accel’s Fund IX?  It is most certainly true that the top-performing dozen or so IT-biased venture funds do massively outperform other sectors and have powerful brands as a result.  This is because of an important caveat about the distribution of outcomes: healthcare outperformance described above doesn’t hold at the top <1% part of the return distribution.  Biotech lacks real exposure to those high profile “halo” IT deals that deliver 100x+ returns.  But those deals are 1 out of 700 (here): of 25,000 startups formed in the last decade, 35 exited above $1B in value.  Since there are 400+ active VC funds today, statistics would suggest that the 95%+ of venture funds will never score a halo deal.  With statistics like this, its also hard to maintain persistence of performance: Kleiner Perkin’s recent acknowledgement of weak results was tied to a “lack of home-run internet investments” since its 1999 Google-dominated vintage (here).  This is a big driver of the state of “haves” and “have nots” in the venture world; its very hard to raise a IT-focused fund without at least one halo deal under your belt in your past few fund vintages, or at least the perception of buzz that a halo deal is likely to be forthcoming in the next one.  To be a top 5% venture fund today almost by definition requires some exposure to a halo deal outcome.  This is where hybrid fund models (here) are particularly interesting: in diversified strategies, a good healthcare team offers the potential of stable top quartile-plus returns with the possible outlier of a halo IT-led outcome.   Avalon is a good example of this with Zynga.  These are the nuances of return distributions. As an asset class, it’s hard to claim that only the top 5% of funds matter since much of the capital is allocated elsewhere.  As an industry, we need to be able to deliver attractive returns at the top quartile at the very least, and this is where healthcare can do and does very well.

2. When biotech deals blow-up, they blow-up big.  Said another way, this myth goes “when you lose money in biotech, you lose lots of it”.  The data also do not support this perception.  I discussed this point in a blog post last fall (here).  In an analysis from Adams Street Partners, different venture capital sectors were compared by their capital-weighted loss ratio (i.e., what percentage of dollars flow into money-losing deals by sector, not just what percentage of deals are loss-making).  They found that biotech had a 36% ratio, vs Internet at 59%.  This analysis was further validated by a large LP’s proprietary database: healthcare had a ~40% capital-adjusted loss ratio vs >60% in technology/media/telecom.  This higher loss ratio in tech is presumably driven in part by deals that raise money at high valuations and then come back to reality in subsequent rounds.  These data do suggest, however, that the belief that firms have historically been able to kill their tech losers early on small amounts of capital isn’t well supported: if it were, you’d see lower capital-weighted loss ratios in tech (more deals dying but less dollars invested in them).  More recently, some firms (like Atlas) are exploring seed-led strategies aimed at reducing the post-Series A loss ratios that may change these data in the future.

3. Biotech takes far longer from inception to “exit” than other sectors.  A post earlier this month (here) dealt with the data behind this point.  Biotechs seem to go public “younger” than their tech counterparts, and M&A happens at roughly the same time.  For “interesting” >$100M exits, they appear to occur faster in biopharma than in tech sectors.  I won’t repost the charts, or the lengthy dialogue, but you can read it here.  Lots of caveats to this analysis are included as well.  An important one being that holding period isn’t the same thing as time from founding to exit.  But its fair to say the data do not support the perception that biotech takes longer.

4. The overall biotech venture capital funding environment is drying up.  I heard last week a discussion about “have we hit the bottom yet” – implying that we’ve seen a collapse in funding.  This one makes me scratch my head as the data is easy to assess with NVCA/PwC Moneytree.  Check out the chart below.  This is overall biotech venture capital funding by quarter for the past 13 years (blue lines) as well as total number of biotech companies getting financing (red lines).  Although we have considerable quarterly variation, the trend line is slightly up, not down, over time.  When comparing the last two years to the distribution of fundings in the past 53 quarters (since 1Q 2000), we’ve had three “top-quartile” funding quarters, and three “bottom-quartile” funding quarters.  Those latter “soft” quarters seem to bring out the Chicken Little commentators in droves.  Yes, there’s lots of variability for sure, but this isn’t the sign of a venture capital funding environment that has “collapsed” or “hit bottom”.  With an average of ~100 financings per quarter, these data are inherently very lumpy; however, on a 12-month rolling basis it reflects an active flow of financing.

5. Early stage venture capital is even worse – it’s a “barren wasteland”.  This is typically said as a corollary to #4 above.  And its true that “first-time financings” are indeed down in the NVCA/PwC Moneytree data– compared to the past 53 quarters as before, five of the last eight quarters have been “bottom quartile” in the number of new startups.  The number last quarter hit the 1995 quarterly number, which is indeed disturbing.  But first time financings aren’t the only definition of “early stage”.  A $1-2M seed financing can be a first time financing event, or a $40M Series A.  Those are obviously different.  Same with a “Series B” into a drug discovery platform company whose lead asset is in preclinical – most observers would call that an early stage company.  So I’d caution against high-level conclusions without digging into the underlying mix of data.  In the past I’ve examined different definitions about “What is Early Stage” and how the data around financings stack up to them: check out the post here.  I found that 60% of the biotech deals done in 2011 were into “early stage” deals.  Although I haven’t updated it for 2012 or 1Q 2013, I have no reason to believe that this has changed since 2011 in a material way.  By my guesstimate, in the Boston ecosystem alone, Third Rock, Flagship, Polaris, and Atlas have probably started/seeded ~25 companies in the past 1.5 years.  There are still lots of interesting new and emerging early stage startups being funded.

6.  Life Science VCs can’t raise new funds.  Venture capital fundraising is tough regardless of your sector, and lots of “zombie” venture funds across all industries are out there that aren’t likely to raise new funds.  And it’s particularly tough in LS today to raise funds, especially in a world driven by big venture brands, anecdotal stories, and a limited number of halo deals.  But funds are being raised.  While the media covers Scale Ventures’ move to shutter its healthcare practice, Sofinnova Ventures quietly shuttered its technology investing in its 2011 fund (here).  New healthcare-only funds like Third Rock Ventures’ Fund III, Longitude Fund II, and more recently Lightstone Ventures’ Fund I have been raised.  Index Ventures raised a healthcare only fund to complement its core tech fund.  Hybrid funds like NEA, Polaris, Canaan, Avalon, etc have all raised funds recently.  It’s not easy, but funds are getting raised.  To get a sense of the available capital for startups, those fundraising efforts need to be combined with the huge influx of corporate venture capital that our sector has witnessed.   GSK’s SR One and Novartis Venture Fund are two of the largest healthcare-only funds active in all of venture capital, as I’ve noted before (here).  When commentators claim less is being raised in Life Science VC than being invested, most fail to account for $500M-$1B in Pharma corporate venture capital that gets allocated annually.

I’d love to hear from readers if the data-driven debunking above doesn’t align well with analyses from others, or if there are other myths worth tackling with further analyses.

For those of us in the trenches of venture capital every day, we constantly here the myths above: at meetings with fellow investors or entrepreneurs, at conference panels and discussions, and often at meetings with limited partners.  The latter group is obviously important for the venture world, and one where brands and momentum often matter more than data, especially as the partners in firms evolve.

One of our LPs raised a provocative point in a recent email that part of the issue regarding Life Science’s “ugly step child” status in venture is the perception that there are no great healthcare venture “brands” today – many of the early ones with solid track records in the 90s (Oxford, Healthcare, Prospect, etc…) have been unsuccessful in either raising new funds or generating top quartile returns recently.  Some of those and others have struggled to evolve their firms.  No doubt there are still a number of very good 20+ year old healthcare-focused firms (e.g., OrbiMed, Sofinnova, etc…), but creative marketing and branding isn’t in the DNA of most healthcare firms.  Healthcare teams within diversified firms may be an important part of maintaining their firms’ brands, but its still hard to hold mindshare of LP’s with drug discovery startups when a micro-blogging site they can touch and feel is up for discussion.  Dan Primack’s point earlier this week sums it up: “why are us media folks so obsessed with the acquisition of a low-revenue blogging platform and so dismissive of an $11 billion combined revenue company that tries to cure disease and improve health? For one, it’s easier for most of us to understand what Tumblr makes than what Actavis or Chilcott make.  Second, it’s a tech story and there are far more tech-focused media sites than healthcare-focused media sites”.  As a related point, there aren’t a lot of active biotech VCs in social or digital media; of course LP’s want returns first and foremost, but they also want to see thought leadership and market-level brand awareness.  Newer healthcare-only firms, like Third Rock Ventures and Longitude, certainly have real brand momentum with the market and LP’s, but only time will tell if top-tier returns can be consistently delivered.  Interestingly, this LP also stated that in healthcare lots of LP’s today want to find managers with “novel strategies” and outside-the-box models of life science investing, whereas in IT-investing most LP’s believe many of the top tier venture brands will continue to deliver on their tried-and-true model of tech venture capital.  An interesting dichotomy.

Hopefully in the next decade Life Science venture can break free of these myths and can get its mojo back – we can’t let misperceptions become reality, our collective brands do matter, and so do data around realized returns.  Lets strengthen all of these going forward.

The intent of these relationships is to provide Amgen and Novartis with strategic proximity to Atlas Venture’s startup formation activities around innovative, potentially high impact medicines, and catalyze future collaborations around translational research across Atlas Venture’s early stage portfolio.  Amgen and Novartis bring their deep knowledge in pharmaceutical R&D with Atlas Venture’s focus on creating and building new biomedical startups with experienced entrepreneurs.

Importantly, the CSP relationships function with an “open market” principle.  There are no obligations around deals, no exclusive rights, nor any fixed commitment of Amgen and Novartis to pursue opportunities.  That said, through the mutual commitment of team bandwidth and dedicated liaisons/”EIRs” from our CSPs, we anticipate finding a set of opportunities for working together.

As noted in the press release, the range of potential collaborations could include launching new academic spinouts focused on taking novel science from bench to bedside, creation of drug discovery platform companies to exploit new therapeutic modalities or emerging biology, or the formation of new single-asset entities structured in line with our Atlas Venture Development Corp model, for example.  We expect to start and/or fund more than fifteen biotech companies out of Atlas IX, and we are hopeful that a number of them will be collaborative efforts with our CSPs.

These CSP relationships are part of a macro trend in the life science ecosystem: larger corporate entities creating tighter relationships with venture firms as both direct equity partners in deals, and as LP’s and strategic partners.  Today, over 75% of our deals have corporate venture groups as coinvestment partners; this number was below 5% a decade ago (as discussed here).  The role of Pharma as an LP in venture funds is another element of the overall theme of supporting early stage innovation, and the list of relationships is significant and growing: Index linked up with both J&J and GSK; Flagship with Merck; Healthcare and TVM with Lilly on their dedicated “mirror” funds; MPM with Novartis and J&J; Longwood, Hatteras, Sanderling are in GSK’s venture portfolio.  Each of these strategic LP commitments has their own expectations and agreement structures.  It remains to be seen which model(s) will work best, but our belief is that the truly open market, strategic proximity model envisioned here with our CSPs will be one of the more mutually beneficial and productive approaches.

Looking forward to working with them as we invest our new fund.

Although surprising given the entrenched dogma, biopharmaceutical companies have experienced M&A and IPO events at a younger age than technology companies in the recent exit vintage.  The bias is especially true with M&A events greater than $100M.  Although I’ve not conducted rigorous statistical significance calculations on the datasets, the distribution of the timelines allow some interesting observations to be drawn that are not well appreciated by the venture community.

Data Dump.

1. Over the past four years, Biopharma IPOs have debuted as “younger” companies than their Tech counterparts.  Using Capital IQ data*, and focusing on all the January 2009 – April 2013 IPOs of companies founded after 1995, the median time from company founding to public offering was under 6 years for BioPharma vs. 9 years in Technology, even when looking at the Software and Internet subset of Technology.  As examples of speedy biotech plays look at clinical roll-up deals like Clovis and Tesaro that went public within a couple years of founding, as did preclinical play Verastem.

2. Private BioPharma companies got acquired in a shorter time from their founding than their Tech counterparts.  Looking at all the private venture-backed M&A deals from 2009-2013 that were larger than $25M (there are tons of micro-deals that bias many datasets), and filtering for those companies founded since 1995, we found that the 70+ BioPharma exits had a median time from founding to exit of 7 years vs 8 years or more in Software/Internet as well as the broader Tech sector.  These data resonate with our experience: Avila and Stromedix were both six years from founding to exit.  Lastly, if one filters for the larger and hopefully more interesting exits greater than $100M, the difference becomes more profound: Software and broader Tech extend out to~10 years vs “speedy” Biopharma at 7 years.  See the two graphs below.

3. Founding a company and getting a first financing are clearly different, but biotech holds its own even in the “time from first financing” to an “exit”.  Using a completely different dataset, this one from Dow Jones Venture Source**, the median time from first financing to a >$25M M&A event was roughly the same (6.2 years for BioPharma and 5.5 years for Software/Consumer Internet), and greatly favors BioPharma on the IPO timeline.  Software alone was equivalent to BioPharma at 6.1 years, as the addition of Consumer Internet shortens the timelines by ~6 months.  These two subsectors are clearly big parts of Technology but not the entire sector; its fair to say other Tech subsectors, like semiconductors, clean tech, hardware, etc.., have longer timelines than these.

Observations, Reflections, Caveats. 

First, these data are in striking contrast to the conventional wisdom.  The image of the overnight success in Tech vs the decade-long laggard in Biotech is just not born out in these data.  One has to wonder about the reason why these findings are so different than widely-held perception.  One significant contributor is likely because of the lack of unrealized write-ups in the Biotech venture world.  In Tech, a company can get founded, build a team, move up the revenue ramp, and raise money at higher and higher valuations over time.  If it takes 7-8 years to get to an IPO, like it did with Facebook, that’s ok because we’re able to communicate the value through unrealized gains on paper.  In Biotech, that rarely happens over time.  While Biotech holding periods are marginally better than in Tech according to these data, the valuations of these companies typically stay near cost or at very modest step-ups for years in the private markets (~2x or less).  This is why we form syndicates that can “go all the way” to an exit, and tranche the financings over many years.  The lack of step-ups limits any unrealized “paper” gains, which makes the holding period appear excessively long.  Fortunately, as other analyses have shown, the realized returns in biotech at the end of the holding period are amongst the best sector in venture.

Second, there are credible business model reasons why the timelines look like this.  In Software and many Technology sectors, predictable monthly recurring revenues in the $5-10M range are often needed before contemplating an IPO.  Acquisitions clearly happen earlier as well, but often real revenue ramps are the norm, and building the sales pipeline to support these financials takes time.  On the other hand, most BioPharma exits happen pre-revenue: as a sector, we are able to market compelling biomedical data to a set of buyers (Big Pharma or the equity markets) that are willing to put a value on the company’s equity based on that data.

Third, it’s fair to say that a BioPharma IPO is not an exit for investors.  It’s a financing event for the company, and insiders typically participate and hold well after the lock-up period.  This could explain some of the difference in timing.  In Technology IPOs, insiders rarely participate in purchasing shares in the offering, and in some cases they sell their shares.  More often than not though, insiders begin to sell after the IPO lock-up period – but this usually takes months to years to exit a position completely (both in Tech and Biotech).  Furthermore, as Groupon and Zynga demonstrate, the IPO price and the exit price are vastly different things, and many late stage investors are licking their wounds (although early investors in both still crushed it even at current prices).

Fourth, an investor’s holding period is not the same thing as time from founding to an exit event.  I’ve focused on the start-to-finish timeline here as a proxy for how long it takes to generate returns, but obviously later stage rounds of financing would have shorter holding periods – a point that holds in any venture sector.  In addition, even the time from founding to the close of the first financing shortens the timelines in #1 and #2 datasets above.        

Fifth, the lack of micro-M&A deals in these datasets could bias the findings.  The datasets are filtered to only include those greater than $25M and therefore eliminate micro-deals.  Importantly, the findings above around founding-to-exit timelines aren’t sensitive to a cutoff at $25M vs $10M – the same pattern of differences in exists.  But these micro-deals could be two very different types of outcomes: companies that sell on tiny amounts of equity (bootstrapped, angel-backed, or “acqui-hire” deals), or they could be “garage sales” for failed companies.  Hard to know based on the dataset here.  But the timelines in the sub-$10M exit range interesting: Tech micro M&A exits have a median time from founding to M&A of 6 years (n=828) vs. 9 years in Biotech (n=29).  Neither of those sounds like a quick flip on tiny amounts of angel capital, even less so for biotech.  However, as these sub-$10M exits aren’t of great interest for venture returns, I’ll stop there but certainly lots of possible hypotheses to ponder.

Conclusions.

Venture capital as an asset class may have a holding period problem, but Biotech isn’t the worst offender.  Achieving liquidity in 5-8 years should be considered a successful path from founding to an exit, irrespective of sector; if this is too long for GP’s and their LP’s, than early stage venture investing is probably not for them.  This is why venture funds are 10-12 years in duration.

Founding entrepreneurs, prepare for a long road.  Fast exits are rare in any sector. Building great companies takes time, as does building mediocre ones frankly.  These exit data suggest that you should expect to hammer away at for the better part of a decade before liquidity.

Working on capital velocity – shortening holding periods – is an important element for improving returns.  This post isn’t meant to downplay the importance of holding periods in generating returns, just to share data on the recent 4-year vintage of exits to challenge the conventional wisdom/biases.  Holding periods are a key driver of rates of return.  In early stage investing, returns could undoubtedly be improved if we can shorten the holding period by accelerating the “capital velocity” within a deal and portfolio.  We and others in early stage biotech venture are working closely with Pharma to help address this issue, as earlier liquidity-generating partnerships are a key component of this.  By exploring novel corporate structures like LLC-holding companies (like Nimbus, RaNA, Forma, Viamet, and others), or by structuring buyout timelines at the outset of a new investment (like our AVDC deals Arteaus and Annovaiton), the capital velocity of early stage biotech deals can be greatly improved.  This might also address the lack of unrealized step-ups in value by smoothing out the liquidity and valuation curves in biotech.

Lastly, there are lots of myths in the venture capital and startup world – about returns, losses, timelines – many of which aren’t supported by data (which will be summarized in a blog post coming soon).  While I would love data-driven analyses to change sentiment on their own, that’s not going to happen.  But it will help us dig into the real issues facing both biotech and the broader venture business, rather than wasting time chasing phantoms that don’t exist.

There’s much to be done in biotech today to improve returns, just as there is much to be done in venture capital as a whole.  But these data are clear: with timelines like these, venture capital needs to be patient capital.

Data Footnotes:

Data was assembled from two sources: Capital IQ and Dow Jones Venture Source.  A number of groups helped me wrestle down the right ways of cutting those data appropriately, including Capital IQ’s Andrew Centauro, Dow Jones’ Hind Wildman, E&Y’s Sandra Feldner Vandergriff, and Torreya Partners’ Margaux Babich.  Two analyses were done: time from founding to an “exit” event (Capital IQ), and time from first financing to an exit event (Venture Source).   Both analyses focused on the ~4 year period from Jan 2009 until April 2013. 

* The Capital IQ dataset included all companies with at least one “venture capital” investor where there was a disclosed founding date later than 1995 (to eliminate the old private firm from biasing the data).  For M&A, it was limited to those exits >$25M to eliminate the huge number of micro-exits.  Technology included fifteen primary industries including Internet Software & Services, Application Software, Systems Software, Internet Retail, Communications Equipment, Computer Hardware, Data Processing Services, Semiconductors, Electrical Components, Wireless, etc…   Most of the data points come from the first four sectors.  BioPharma included both Biotech and Pharma.  Capital IQ provides founding dates in years.  To treat all companies the same, the calendar year of their event date (IPO, M&A) was used.  This leads to a rounding off of the numbers of years. 

** Dow Jones Venture Source dataset included all venture-backed companies with an IPO or M&A between Jan 2009 and today, according to their criteria.  M&A was again limited to those deals above $25M in disclosed value.  Timelines were constructed from first financing date to exit date. 

But before doing that, a short history on Atlas: founded in 1980, our first four funds were very successful early stage vehicles – focused on building real connections with entrepreneurs in the local ecosystem.  We invested in Europe and the US.  In biotech, Atlas was an early backer of great companies like Exelixis, Morphosys, DeCode, Crucell, Actelion, etc…  In the dot-com bubble, we like many venture firms strayed from our early stage focus.  Atlas raised too much, too fast, and attempted to scale the venture business.  We had six offices and a couple dozen partners, three layers of them in fact.  It’s clear to us that early stage venture capital doesn’t scale.  Not only because the math doesn’t work well (discussed here about biotech, here more generally), but culturally the organization loses its edge, its entrepreneurial roots, as it tries to scale.  By the time I joined in 2005 we had four offices and 18 partners, and were closing on Fund VII.   Shortly thereafter, the real generational change and firm transformation began.  Fund VIII was closed in the fall of 2008, during the financial crisis and accelerated our evolution.  Fast forward to today and we’ve got one office, seven partners, with a flat and equal partnership around Fund IX.  Hallelujah!  For more on the last five years of “reinvention” at Atlas, see my partner Fred Destin’s excellent blog post (here). 

Today, we are a seed-led firm, committed to early stage venture formation, funding startups across both Life Sciences and Technology sectors.  Although a hybrid sector model, we have a shared philosophy and strategy around getting involved early in a company’s life, helping with strategy and product development, and working closely with entrepreneurs to help build them into successful business.  Many have asked us why we stick together when many hybrid funds are splitting.  First and foremost, we like each other.  Second, it works: some vintages Tech has outperformed, some vintages Life Sciences has outperformed.  Lastly, as a Life Science investor, I think its great discipline to have a Tech counterpoint to keep us honest on things like capital efficiency (and data suggests as much – healthcare deals done by hybrid investors have outperformed, see here).  I suspect the same is true for my Tech partners. 

Geographically, we are focused on New England – Boston is the epicenter of Life Sciences today, and in Tech there’s a real renaissance in the entrepreneurial ecosystem in Boston.  That said, we will invest opportunistically elsewhere if there’s a unique Atlas angle on the deal.  For more about our Tech strategy specifically, follow Fred’s blog (here).

Life Science Investing at Atlas

To boil it down, there are six key elements of our Life Science strategy: 

1. Venture Creation.  We like to encode the DNA of our new startups; we are co-founders of more than 75% of our deals today.  shaping strategy early with a roll-up our sleeves approach to partnering with entrepreneurs.  Like other early stage investors, we take acting operating roles in our new companies.  My partner Peter Barrett was the acting CEO of Zafgen for its first 15 months, I did the same at Nimbus, etc…  This is often critical in helping a fledgling company get off the ground successfully.
2. Seed-led model.  We have evolved from having a seed strategy in Fund VII to being a seed-led firm today.  Our typical initial investment size is $500K – which we use to do early derisking, validating of academic findings, recruiting a team, securing IP, etc…  But we can be full capital partners to our startups as well – reserving upwards of $15M+ into deals that are working well.  We believe strongly that this seed-led “prove” approach will bend the traditional risk curve and lead to overall lower post-Series A loss ratios.  This elimination of more “false positives” during seed rounds  is critical; historically, the straight-to-Series A $10M round has consumed (and destroyed) lots of investor capital.  Now we base our Series A investment decisions on real “signal”: data that we have generated in house, tangible “market” interest from Pharma, recruitment of great talent willing to commit the scarcest resource of all, their time.  These signals around a technology, market, and team are key for our seed strategy (across LS and Tech).
3. High bandwidth, high ownership.  Because we take an active role in our new startups, we seek to achieve a commensurate high ownership stake in our deals.  High ownerships help drive overall fund returns, especially with the math of biotech venture funds; as an early stage investor, our ownership will also never be higher than when we found a company and close a Series A round (ie., syndicating deals over time is dilutive).
4. Capital efficiency.  Lots of firms talk about this, but we focus on it explicitly.  As a small fund, disciplined deployment of capital is critical to our model, as is finding partners with lower cost of capital to join us in helping build and grow our companies.  Capital intensity is a related (but different) theme: some deals can be very efficient but still raise tons of equity dollars.  We also avoid capital intensive deals because the margin for error on returns is much much smaller for early stage investors.
5. Fund real innovation.  We believe the only way to generate outsized returns today is to back “high innovation quotient” programs and platforms – medicines with the potential to unlock new areas of biology, transform clinical care, and improve patient outcomes.  New biology, new modalities, new approaches – built with either asset-centric or platform based business models.  This has always been our bread and butter at Atlas, but like others in the life science venture world, we flirted with a few deals in the spec pharma model during 2003-2006.  Its low tech, low innovation model was believed to be lower risk.  In fact, its not lower risk (see here), and certainly in our experience hasn’t generated outsized returns in general.  In order to access lower cost of capital corporate funding early in the life cycle of a company (e.g., partnerships, structured sales, exits), the products need to be real innovations.
6. Creative structuring.  We have been pioneering new approaches that break out of orthodox models of biotech (“go big, go bust” strategies, capital intensive product “portfolio” roll-ups, etc…).  Our Atlas Venture Development Corp initiative around structured asset-centric investing has closed two deals, Arteaus and Annovation, both of which have structured paths to top 5-10% venture return outcomes.  We’ve broken out of the dogmatic C-corp structures, with Nimbus’ 2009 launch leading the charge for asset-centric LLC-holding company models for platforms that seem to becoming popular.  We have creative partnership with Shire and with Monsanto, and anticipate announcing additional partnerships in the future.  These new models are critical to success in an ecosystem that is becoming tighter and more networked.

To deliver on this strategy, there are “about a dozen” of us on the Atlas life science team.  I say “about” because the Entrepreneur-in-Residence team-based model around starting new companies is very fluid.  While our dedicated investing team consists of three partners (Jean-Francois Formela, Peter Barrett and myself) and an associate (Michael Gladstone), we have 6-10 EIRs or venture partners working with us at any moment.  Dave Grayzel runs two startups and leads up our Atlas Venture Development Corp initiative.  Ankit Mahadevia is a Principal and EIR working in two new seed-stage deals right now.  Nessan Bermingham, Katrine Bosley, Adam Friedman, Marty Jefson, Kevin Pojasek, Josh Resnick, and Tom Schuetz are all working with us in some entrepreneurial capacity – either in new seed-stage projects or helping us evaluate them.  There’s a very permeable boundary between who is an entrepreneur or executive in our startups and who is actively on our investing team.  We also leverage our portfolio executives extensively.  Take Tom Hughes, for example – he’s CEO of Zafgen, but also on the Board of Miragen, and the SAB of both Nimbus and Calorics.  This type of cross-pollination in the portfolio is invaluable.  Fundamentally, early stage venture is a team sport (and a ‘contact sport’ at that), all of us are involved in more than just “our thing” or “our deals”, and that spirit is what helps us start and build great companies.  The same team-based approach holds true for our Tech franchise.

As we begin deploying Fund IX, we’ve assembled a set of life science seed projects that we’ll be launching into new companies spanning a broad range of therapeutic categories: oncology, neurology, metabolism, muscle regeneration, cell therapy, antibiotics, in silico chemistry, novel angles on epigenetics, metabolism, unique antibody approaches, immuno-oncology, inherited ataxias, etc…  All of these are “high innovation quotient” startups.  A good number of them won’t graduate out of the seed phase – which is part of the model – but we expect more than a dozen new Life Science startups to grow into our full lifecycle portfolio. 

It’s an exciting time to be in the early stage biotech venture capital business.  The ability to have potentially huge impact on patients while also generating attractive returns for our LPs is a real win-win opportunity.  Onward and upward. 

But I’m far less sanguine about who is going to fund the new innovations required to bring these diagnostics to market.  As I noted in last week’s post-mortem regarding On-Q-ity, venture-backed diagnostics have been a painful sub-sector to invest in and the wreckage is adding up.  An SVB report from last year has it pegged as the most challenging subsector in the med tech space (here).  According to that report, during 2005-2011, over $1.6B was invested in 106 companies focused on diagnostics.  In 2012, according to MoneyTree, another $168M was invested.  I’m skeptical on the return profile of lots of these investments.

Here are the six tough realities of the diagnostic space today:

1. Diagnostic science isn’t any less risky than therapeutics.  The underlying bioscience uncertainty around the relevance and robustness of an assay or test for impacting/predicting outcomes isn’t any different than the uncertainty about whether a drug is pharmacologically active.  Many initial observations don’t get validated.
2. Capital intensity remains very high for many Dx companies, as the exit point typically requires commercialization.  It takes a lot of capital and time to validate a diagnostic and bring it to market.  Large validation sets and clinical trials are expensive.  OncotypeDx, Genomic Health’s and the personalized medicine industry’s flagship product, was developed with samples from 2600 breast cancer patients as well as ~500 clinical trial patients, according to their 2005 S1 filing.  The bar has only gotten higher since then.  There are a number of solid private companies with launched Dx products that have consumed >$100M in equity capital: CardioDx, XDx, Tethys, Crescendo, to name a few.  These will need to be very large exits to generate a venture return.  MDV, a big backer of these plays in the past, has decided to deemphasize diagnostics (here) – it’s a telling sign when one of the leading investors with deep expertise in the field has decided to leave it.
3. Diagnostics have a hard time extracting value-based pricing.  The difficulty of communicating value and then sharing in that value (be it cost savings or better outcomes) is significant, making many diagnostic models more of a high volume, low cost commodity business where scale is essential.  Further, unlike chronic or routine therapeutic use, there are many single use tests on the genetics side, one-offs on acute settings (MRSA), etc that make for challenging price-to-value considerations.
4. Regulatory uncertainly is far higher than in therapeutics.  The FDA may have a high bar for new drugs, but it’s a bar that in most areas is fairly certain.  There are plenty of guidance documents and new, innovative drugs often get very clear paths to approval (SPAs, breakthrough status, accelerated approvals, etc…).  There are occasions in the past where the goal posts have moved (like antibiotics, obesity), but even in those areas the regulatory certainty has recently improved.  In diagnostics, regulatory uncertainty is a systemic risk.  It stems in part from the complexity of CLIA-certified home brew testing vs FDA-approved pathways, and what the future status of those pathways will be.  The requirements for approval need to be much more clear if this concern is to be overcome.
5. Reimbursement remains a mess.  When you have multiplexed tests (like many new personalized medicine approaches) where the diagnostic itself involves multiple activities, reagents, characterizations, etc…, it’s well accepted that the old model of CPT code-stacking just isn’t going to be viable.  But its very unclear what the new model will be.  The process of getting a “new code” is arduous and uncertain, and typically requires mountains of data.  It’s simply not clear how to get paid for expensive but high value tests.
6. Lastly, the exit market for venture-backed diagnostic companies remains weak.  For reasons #1 and #2, it’s very rare to see a big “early stage” M&A transaction (the equivalent of a $300M Phase 1 exit).  But getting companies to revenue stage requires lots of capital (#3) before acquirers are interested.  The big national labs (Quest, LabCorp) prefer accretive deals so don’t make frequent buyers of innovation.  The public markets may open up for diagnostic and life science tools players, but that’s not been a great path for future appreciation in the field.

All the above is not to say that you can’t make money in diagnostics, and a number of smart investors have and will continue to do so.  For instance, HBM made 21.6x or more on the sale of BRAHMS to Thermo a few years ago.  And there are a lot of very interesting emerging diagnostic companies, like Foundation Medicine here in Cambridge, that may achieve the escape velocity required to scale.  But it’s a tough space, or as I wrote a few weeks ago, “not for the faint of heart”.

At Atlas, we’ve reflected on these realities and decided to stick to what works for us in the Life Sciences – which is early stage therapeutics investing.  Looking at our aggregate track record over past 20 years, therapeutics have outperformed diagnostics/tools by a huge margin (5-fold).  Over the past few years, four challenging experiences helped tee up the perspective: OnQity, Aureon, Molecular BioMetrics, and Helicos.

• OnQity.  Our failed tissue biomarker and circulating tumor cell play; see my recent post for the summary of the story and lessons learned (here).
• Aureon Biosciences.  After a decade in the “systems pathology” space, Aureon got two products onto the market via the LDT/CLIA pathway.  But payors were reluctant to reimburse fully for the tests in the absence of “more data” (despite several large trials), CPT code-stacking wasn’t working, and the timeline for getting clarity on reimbursement was uncertain.  Although a good number of urologists/surgeons wanted to use the test, keeping a commercial organization going was prohibitive and so in late 2011 the company was shut down (here) – even after getting two products launched.
• Molecular Biometrics.  MBI had developed a diagnostic in the IVF space for identifying embryos more likely to be viable.  We joined the B-round to fund the launch of what we thought was a ready-to-go test.  But unfortunately the underlying data wasn’t nearly as robust as initially believed: in hindsight, it was fundamentally a failure of over-fitting data during the hypothesis testing and validation stage.  The initial diagnostic signature looked great but was an illusion without the outliers.  The product was voluntarily withdrawn and the company shutdown.
• Helicos.  One of several companies in the Next Generation Sequencing space, a hot technology area plagued by capital intensity.  NGS isn’t necessary a “diagnostic” space, but it’s a new technology area that supports diagnostic applications as commodity solutions.  NGS technology and commercial development has just cost a ton and yet the space is commoditizing rapidly.  Pacific Biosciences has raised over $600M, Complete Genomics over $200M, and Helicos over $200M.  Lots of losses for investors here as these NGS companies struggled to compete with the established competitors, differentiate their offerings, etc…

These four ill-fated deals (spread out over several fund cycles) aren’t representative of all diagnostic companies.  I’m sure there are many good diagnostics companies that will succeed and deliver value to both patients and shareholders.  But for us, at least for the time being, they give us pause about diving into more diagnostics deals, as they frame up lots of the issues and challenges with the field (the six realities above).

It’s certainly true that investors can mitigate some of the above risks by focusing on near-commercial bets or funding through to the revenue-stage.  Roll-ups of smaller diagnostic plays can be had to accrue scale benefits faster (though more of a growth equity than classic venture play).  But these investment strategies preclude the funding of early stage innovative diagnostic opportunities.

Further, this approach raises the question of how does one generate a return from early stage diagnostic investing – the translation of academia-derived insights into new clinically-validated diagnostic tests.  Who is going to fund the next generation of new innovative Dx plays?  I’m hopeful that models of consortia-based funding or pooling of biomarker insights and capabilities will come together.  Qiagen’s CEO Peer Schatz speaks eloquently about these types of ideas, among others.  The Personalized Medicine Coalition has certainly been a positive force for pushing the field around policy changes and these more collaborative models.  Much like the Pharma industry has engaged actively in the early stage therapeutics ecosystem through novel partnership models and active corporate venture investing, the larger players in diagnostics need to do the same in personalized medicine. Much more of this collaborative activity is likely required to attract the much-needed private capital into the field.

Lets start with some background.  On-Q-ity was formed in 2009 through the merger of two predecessor venture-backed diagnostic companies, Cellective Dx and DNAR (DNA Repair Inc).  This Series A round was $26M and had four VCs: Mohr Davidow Ventures led the round (and was the lead investor in the prior entities), and Atlas, Physic, and Bessemer all participated.  I looked back at my December 2009 investment memo, and these were the key reasons to invest:

(a)   “Hot space at nexus of two exciting themes in cancer and personalized diagnostics”: DNA Repair pathway tissue-based biomarkers for drug response prediction in breast and lung cancer; and, a circulating tumor cell (CTC) capture and characterization platform.  These technologies had lots of potential to change the treatment paradigm in a number of cancers, and its fair to say that CTCs have only gotten more interesting over time

(b)  “Strong team with a track record in building diagnostic businesses”, hailing from Genzyme Genetics, Myriad, Genomic Health, Dana Farber, etc… We also had a great board, including a former President of Affymetrix as the Chairman.  At the outset, it looked like a great team and syndicate for Dx startup.

(c)   “Robust financials with attractive recapitalization” at a good valuation (the $31M invested in the prior two entities was recap’d into <$5M pre-money valuation), with clear path to revenues, or so I naively thought;

(d)  “Very strong intellectual property”, especially in CTCs, which actually got stronger over the past few years; and

(e)   Perception of “multiple possible exit paths via either diagnostic or oncology-focused pharma M&A or possible IPO”.  Unfortunately none of these materialized.

I also flagged a few potential risks in the 2009 memo: the biomarker technology needed clinical trial validation, CTCs could be a research-phase money pit before the market matured, execution against the aggressive plan was going to be tough, merging management teams and technologies under one roof is not easy, etc…   At the time, I judged most of those risks to be manageable.  Well, let’s just say I got the risks right but their magnitude wrong – these risks became the big factors that brought down the company.  Here’s the short story:

By mid-2010, only six months after the Series A came together, it was clear that the DNA repair biomarkers were going to be tough, as an early trial failed to reproduce the nice Kaplan-Meyer curves of the original academic work.  By late 2010/early 2011, two more larger trials read out negatively so we decided to terminate that effort.  But unfortunately those trials and the biomarker lab work required to support them consumed 60%+ of the capital in the Series A round.

Not much had gone into the CTC platform in that first year and so early in 2011 the company refocused exclusively on CTCs and streamlined the team, but the clock was ticking.  As we dug in to the status of the CTC platform, it was very clear that lots more work needed to be done – the paper descriptions of what it was supposed to deliver didn’t map to the platform’s actual robustness (or lack thereof) at that time.  Antibodies that were supposedly functional turned out not to work, and several other things like this.  An early LabCorp partnership fizzled because of this realization.  Our new CSO and the R&D team rebuilt much of the platform and reagents from scratch, got it up and running in a reproducible and robust manner, validated it favorably against Veridex’ CTC approach with patient samples, and set us on sound technology footing.  It was an impressive technology turnaround, frankly.  But that took a year, and a very precious year.  We were out of money at the start of 2012.

We also had a revolving door in the management suite during 2010-2011: we lost or transitioned the original CBO, CFO, CSO, CMO, COO, CEO, another CSO, etc… Some of the changes happened fast – within the first 6-9 months – but we didn’t replace them fast enough.  Then as the company’s strategy shifted from near commercial (DNA repair biomarkers) back to R&D (CTC) in early 2011, additional changes to the senior team made sense.  It’s also fair to say the background and experience of some of the team didn’t map well to the high strategic entropy and lack of organizational support common in startup roles.  As you might expect, it’s very hard to stabilize a new technology platform while shifting leadership.

As 2012 started, we had to make some tough decisions.  We were running out of money in a few weeks and had a limited senior management team, but felt we had a technology that had finally gotten its sea legs and a very good R&D team.  MDV, Physic, and Atlas decided to do a small Series B to carry the company to a CTC partnership (Bessemer dropped out, which history now shows that was the smart move).  We recruited an entrepreneurial executive, Mike Stocum, to lead the company, and hired Foundation Ventures to help him run a strategic partnering process.  Together they did more than any board could ask.

But at the end of 2012, despite heroic efforts by the On-Q-ity team to keep the technology maturing and an extensive process for a potential strategic partner/acquirer, we were once again low on cash.  And this time we chose to wind down the company in an orderly fashion.  The final sale of assets is wrapping up this month.

What are some of the lessons I’ve learned:

• Crashing two Fords together doesn’t make a Porsche.  This is the obvious one.  Recap mergers aren’t easy, and bring all sorts of baggage that make the “attractive” valuation an illusion.  Trying to merge two really interesting diagnostic technologies embedded in two struggling, cash-burning businesses into one company and have it smoothly integrate was a very bad assumption.  The first year gave little airtime to the CTC technology, and we paid the price for it by being behind the eight-ball in the second year.
• Tranche the capital.  I failed to follow my instinct here.  I remember pushing hard for tranching back in 2009 as the round came together, but ended up going along with the full raise against my better judgment (in part because the first $21M had already been wired).  I violated the axiom that if a technology can be validated on a smaller financing, it should be.  It gets back to our mantra at Atlas of Prove-Build-Scale.  Instead of raising a “Build-stage” $26M Series A without any tranche or milestone points, we should have broken up the capital to Prove the story first.  Sadly, we knew within months that the DNA Repair story was going to be an uphill battle.  I remember the late 1Q 2010 board meeting well: the initial trial data were far from clear, but clearly far from what we had all hoped for.
• Team, team, team.  If we had in place a stable management team from the start, things might have been different.  There were questions about how to integrate the different management team members right at the start of the deal in fall 2009, and we ended up with the revolving door I mentioned.  This leadership flux kills startups, and certainly was a big part of the mortal wounds we had at On-Q-ity.
• Getting the technology right, but the market-timing wrong, is still wrong, confirming cliche about the challenge of innovating.  The capture and characterization of CTCs will be an important part of oncology care in the next decade.  This I’m fairly sure of, and all the news from AACR earlier this week about “liquid biopsies” supports this premise.  But the reality is it’s a research-stage story right now, and in diagnostics (unlike drugs) it’s hard to get paid for research-stage stories.  We may have been right that CTCs are “hot” and will be important in the future, but we certainly didn’t have enough capital around the table to fund the story until the market caught up.  It will be great in 5-10 years to see CTCs evolve as a routine part of cancer care, though clearly bittersweet for those of us involved with On-Q-ity.
• Diagnostics aren’t for the faint of heart and are a much tougher place to make returns today than other life science subsectors.  Despite the frothy commentary about personalized medicine and the dawn of diagnostics, it’s a very tough business that faces many of the risks and costs of drug R&D but without the upside.  It’s often not less capital intensive than therapeutics, faces similar “academic validation” concerns, is confronted by larger reimbursement and regulatory uncertainties, has commodity-oriented high volume, low price demands, and typically needs to get to commercialization before a material exit outcome.  All these things add up for a challenging investment sector.

My first total write-off in nine years of venture capital.  On-Q-ty, R.I.P.

While I’m sure there’s real value in a big analytics approach to this, I also think there’s a lot to be gained through the appreciation of some simple truisms accumulated by the truly seasoned practitioners of drug R&D.  One of those practitioners is Dr Phillip Needleman.

Phil was the Chief Scientist of Pharmacia when it merged with Pfizer, and before that held the same role at Searle, and before that Monsanto; he’s overseen a bunch of great drug programs, including COX2.  He’s also a National Academy of Sciences and Institute of Medicine member, an accomplished scientist at heart.  I’ve had the pleasure over the past three years of getting to know Phil though our role together on UCB Pharma’s Scientific Advisory Board.

Never one to shy away from sharing his candid feedback about R&D programs, his impressive catalogue of experience gives him a unique perspective.  I don’t always agree with him, nor do others, but his opinions are always grounded in a thoughtfully data-driven, experience-based rationale.  In talking with him last week, he shared his “ten commandments or ten platitudes” with me – and I thought they were worth sharing more generally, as many are still as relevant to Pharma R&D today as they were a decade ago when Phil (in theory) retired and distilled this list.

1. “There’s only one Nobel prize…” and industry researchers aren’t likely to get it.  While great science is an important foundation, the true strength of pharmaceutical industry is in the inventions required in generating clinical proof of concept (PoC) and development of new products.  Those tend not to be what Nobel’s are about, but then again Nobel’s don’t help treat patients.
2. “Phenomenology is very different than pharmacology”.  Phil’s admonition is to only take drugs into patients where you know the mechanism of action and can therefore optimize the drug’s SAR rationally.  Phenotypic screens are fine as starting points, but be sure to deconvolute the mechanism before spending a ton of resources on it.  Pfizer’s “3-pillar analysis” earlier last year supports and extends this view (here), and we certainly share this view at Atlas.
3. “Anything put on hold is dead”.  Projects that get parked or slowed down dramatically in a resource-constrained world are very hard to revive.  If you don’t have the resources inside, and it’s a good program, find a partner on the outside to fund it – patients deserve it.
4. “Define and do the Killer experiments”.  This speaks for itself.  But, in Phil’s words, “you need ice in your veins – if a killer experiment doesn’t work then kill the program”.  This does take a lot of discipline, especially for a team that may have given its heart and soul to a project for 3-4 years in drug discovery.  Defining the right killer experiment is the key skill: too stringent could lead to a false negative, which kills good programs, and vice versa.
5. “If you don’t know where you are going, all roads will get you there”.  I’ve heard Phil use this quote from the Cheshire Cat in Alice in Wonderland, and its very true – strategic direction and very well-defined endpoints are critical to measuring success.  While lots of clinical studies don’t need to focus on registrational endpoints, like early experimental medicine PoC studies, that doesn’t mean having loose endpoints; have a clear sense in advance of what you are after in a project.
6. “Make sure it’s doable in my lifetime”.  If the PoC study or pivotal endpoints will take “many years without a convincing marker, surrogate, or a unique insight opportunity, then don’t bother trying to be patient”: go invest your time and energy someplace else until the field finds more realistic measures.
7. “Find the shortest route to heaven”.  In discovery, don’t spend years doing every preclinical pharmacology model – do a few, not half-a-dozen or more – and then with confidence sprint quickly to the drug’s PoC in patients.  In the clinic, if you need to focus on a limited-use first, go for it and get approved with a narrow label fast, even over the cries of your commercial colleagues.  Get to heaven and you can figure out the rest later.
8. “I’m from Missouri… show me data”.  As he says, “the coinage of the realm is data” so spare the hand-waving talk, wordy powerpoints, and show the data, “gleamed both at the bench but also occasionally in hard-nosed reviews with experts”.  He’s definitely one of those hard-nosed experts now.  I share Phil’s appreciation of the great Edwards Deming quote: “In god we trust, all others must bring data”.
9. “Mechanisms should become lifecycle management platforms”.  Working in a novel area of biology is hard, but once a mechanism has been successfully nailed down, “it should become some peoples life’s work” to find new opportunities and uses for that mechanism.  Without the broader work on newly precedented targets, lots of medicines wouldn’t have reached patients (like beta-blockers in heart failure 15 years after hypertension).
10. “The world belongs to finishers”.  Finishing requires focus: pick a few things and deliver on them.  In drug discovery, it’s always easier to “turn over new rocks than to concentrate your attention and resources to meet expectations and deadlines”.  While it’s exciting to push new discoveries, it takes “discipline, management, people skills, encouragement, toughness to finish” and bring a drug through to PoC and eventually to market.

Nothing on this list of learnings is a blazing or unique insight, but they do help codify a range of themes around how to successfully run an R&D project or portfolio.  Importantly, lots of R&D organizations and leaders have great looking strategic plans that follow these ten themes, but only when this list is executed with “taste, judgement, and urgency” does it have any meaning.  I also applaud its focus on data-driven decision-making and focused execution – a key for any venture-backed company as much as a large Pharma R&D organization.

It’s fair to say, though, that when I share one of these truisms in a board meeting few people listen; somehow when Phil says them they sound much more credible.  It must be his Missouri accent.

With all this optimism sending valuations skyward, one would hope that it is translating upstream into better deal terms for startups to raise venture financings.   The Cooley Venture Financing Report released last month captures the recent 5-year trends for all the venture rounds they’ve been involved with, and there’s definitely some optimism in the industry.  However, the majority of the financings the Cooley report tracks (80%+) are not in biotech, so the trends in there aren’t easy to dissect for our sector.

Fortunately, they were willing to share a life science-only analysis with me, and it’s a fairly large dataset – they have participated as counsel in over 380 life science rounds in the past five years, including nearly 60 in 2012.  Here’s a summary of a few of the interesting deal terms, with a chart below to capture them.

The good news is that on most dimensions the deal terms are improving for startups and reverting to pre-2009 financial crisis levels.

Two-thirds of all life science financings were “up-rounds” last year: This is a healthy, high percentage of rounds with step-ups in valuation, and is the highest level in five years, way above the meager 24% in 2009.  Further, it’s only slightly lower than the rate across all of venture, including frothier Tech, which was 75% in 2012.

Recapitalizations as a share of financings are down: Recaps are obviously signs that companies (and the sector) are struggling, so its good to see the rate of recaps is down by 30% or so, with only 8.6% of rounds in 2012.  This is down from a 10-12% share of financings in the past three years.

Punitive “full ratchet” anti-dilution provisions are less common: Last year only 1 out of 20 financings had full-ratchet ownership protections for new investors; in 2010 it was as high as 1 out of 6 financings.  This term may seem esoteric, but its presence is a real sign that new investors have more leverage and are raising the effective cost-of-capital.  It essentially preserves the ownership percentage of a new investor in a future downround by extracting ownership from others in the cap table – it can lead to particularly painful outcomes if future rounds are challenging to price.  The term really only existed in a few life science financings last year and was absent from deals in other sectors.

Tranching of rounds is less frequent: The share of financings that include the tranching of capital around specific milestones hit a high in 2011 of 55%, and in 2012 that dropped to 39% or near its 2008 level.  While its fairly common in life sciences, tranching remains out of fashion in most other venture sectors – less than 15% of venture financings in other sectors were tranched.

 So while those deal terms trended well in 2012, at least two deal term components haven’t improved in 2012, or at the very least haven’t reverted back to their 2008 levels: pre-money valuations and the use of fully “double dip” preferred securities.

Median pre-money valuations are flat or down across most of the financing series that Cooley has tracked for LS.  Series A pre-money valuations are off from 2008 by 20-30%, around $5-6M in 2012, the lowest level in the past five years.  Later Series are flat or down in overall valuation.  Its hard to square these data with the up-round statistic mentioned earlier, but it might suggest that the companies that have been raising new money had previously priced lower rounds in the past, giving them headroom on valuations to support the higher frequency of step-up rounds of 2012 that Cooley witnessed.

Fully participating preferred stock issuance has been on a steady climb since 2008 in Life Sciences.  This term peaked last year at 47% of all financings in our sector, compared to 30% in 2008.  This full “double-dip” preferred (vs straight preferred, or capped participation) is clearly investor-friendly: in essence, it enables investors to be paid as both a debt-holder first, and an equity-holder second – and clearly raises the cost-of-capital for startups.  The trend runs counter to the more entrepreneur-friendly changes cited above.  It also is counter to the trend in other non-LS sectors of venture, as I’ve mentioned on this blog before, where full participation feature is now present is less than 1 our 4 financings, about half the rate in Life Science deals.

All in all, the general trends captured in the Cooley dataset are positive, and bode well for venture and for startups trying to raise money at a reasonable cost-of-capital.  As an early stage investor focused on company creation, I’m obviously quite pleased about the frequency of up-rounds, a lower recap rate, less full ratchet anti-dilution clauses, and attractive pre-money valuation ranges.  That said, the lack of tranching in the majority of deals isn’t something that I’m excited about, nor something that I’m actually seeing in practice.  I’d guess that close to 100% of the deals we see and do are tranched, but that may be because we only occupy the earliest end of the venture spectrum.  As I’ve mentioned before on this blog, tranching is a great tool for derisking while limiting capital exposure and has an important place in early stage biotech financings.

Looking forward to the continued positive trajectory of these deal term trends in 2013.

The scale of value creation from immunokinases in the last few years is really quite staggering.  By my estimate, north of $10-12B in value has flowed to companies (and their shareholders) that possess hot immunokinase programs.  Here’s a brief tally and I’m sure I will have missed some:

• Private M&A deals have racked up over $1B in upfronts and potentially up to $2.6B including milestones across multiple targets, on roughly $300M in equity capital. In 2010, TargeGen was snapped up by Sanofi for its JAK2, and CGI Pharma by Genentech for its Syk and related programs.  In 2011, Gilead jumped for Calistoga’s PI3K-delta inhibitor CAL-101, and Takeda for Intellikine’s PI3K-delta and related chemistries.  And in 2012, Celgene bought Avila, in large part for its BTK program.
• Publicly-traded mid-cap stocks have appreciated by some $8B+ over the last few years.  Pharmacyclics has the lead BTK inhibitor (that it shares with J&J), and its value has risen more than 15x from ~$400M in early 2011 to over $6B today.  Infinity Pharma has appreciated 8x since early 2011, triggered in part by dropping its hedgehog program in favor of its PI3K-delta program (that it licensed from Intellikine); its now worth over $1.8B.  Incyte has gone up 10x since mid-2009 on back of Jakafi (JAK2) data, launch, and now sales; its now a $3B company.  Rigel Pharma’s stock hasn’t faired as well in recent years (largely flat), in part because of prior super high expectations for its Syk inhibitor.
• Big Pharma has seen its share of excitement as well, though harder to dissect out the value on their market cap’s: Pfizer’s JAK1/3 inhibtor Xeljanz is the first immunokinase approved outside of oncology (where it is indicated for Rheumatoid Arthritis).  Furthermore, Pharma has been very active on the licensing front: J&J, Novartis, and AZ are active partners on a number of programs listed above (where they paid considerable sums).  A slew of other interesting deals have been done, including Biogen’s deal for Portola’s Syk inhibitor, among others.

So its fair to say the past few years has been a fantastic period of value creation for those lucky or smart enough to have “hot” immunokinase assets around a small set of targets: JAK2, JAK1/3, PI3K-delta, BTK, and Syk to name the top five.

But should companies keep attacking those targets?  For high cost-of-capital venture-backed biotech companies, I’d definitely suggest against it: leave the war of attrition around differentiating in these crowded target spaces to the big R&D budgets of Pharma.

Our quick review of Thomson Pharma’s database says that there are 125 programs in the industry against those five targets above.  Nearly 80 of those are still in drug discovery and preclinical testing.  That’s a lot of early stage programs, and a lot of resources.  A good example would be JAK2, where there are dozens of programs, including ruxolitinib, SB1518, TG101348, XL019, CYT387, AZD-1480, R723, LY2784544, among many others.  PI3K-delta has 16 programs in drug discovery and preclinical today, despite half-a-dozen in the clinic already.

So if not these five, what’s next?

We think there are plenty of targets to get excited about in the field of immunokinase modulation, many with strong preclinical rationale and compelling human genetic linkages despite lacking clinical-stage pharmacologic validation.  The list of potentially “hot” immunokinases is considerable, with targets like Tyk2, ZAP-70, NIK, Fyn, Lck, among many others, and I’m sure there are many companies advocating for these and others.

One immunokinase that we believe is quite interesting is IL1-Receptor Associated Kinase 4 or IRAK4, and we’ve invested a significant portion of the resources at Atlas’ portfolio company Nimbus Discovery to design and develop inhibitors against it.  I thought sharing briefly our rationale for why we like it could be more generally instructive about how we view the field.

Like most of these next generation immunokinase targets, IRAK4 is not new to the industry: it was cloned over a decade ago at Millenium and Tularik.  Its role in lymphocyte signaling was characterized and explored through lots of cell-based and in vivo models, including RNAi and other target validation approaches.  Efficacy has been shown in arthritis, psoriasis, and gout models, among others (here).  Furthermore, a human mutation leading to IRAK4-deficient individuals has been well characterized.  IRAK4 is a critical signaling protein downstream of IL1R, IL18R, IL33R, and a set of Toll-Like Receptors.  Upon ligand engagement, these receptors in turn engage MyD88 which signals through IRAK4 to turn on a variety of pathways, including NF-kB and the secretion of pro-inflammatory cytokines.

While IRAK4 is of great interest in autoimmune diseases (e.g., lupus, psoriasis, IBD, RA, etc…) as a key node for regulating lymphocyte activity, recent data in hematologic malignancy has really put the spotlight on IRAK4: the key upstream signaling protein, MyD88, has activating mutations in 35+% of B-cell lymphomas that make those cancer cells dependent on IRAK4 signaling and therefore very sensitive to IRAK4 inhibitors (here).  Furthermore, recent data suggests that IRAK4 inhibition synergizes with inhibition of other B-cell receptor signaling pathways, like BTK (here).

These data and others have caused several companies to initiate programs against IRAK4, but few have succeeded in drugging its shallow and unique pocket.  Its been challenging to make selective, potent, and in vivo active inhibitors of the protein, and most of the early efforts failed – but it is a challenge that’s being overcome, and Nimbus Discovery is one of a small handful of companies making real progress on the target.

Five active programs exist, according to Thomson Pharma, and they are all in drug discovery or preclinical studies.  A number of the earlier attempts against the target failed to achieve cell-permeability, selectivity against IRAK4, or sufficiently clean toxicology profile.  We believe we’re on track to addressing all of those issues at Nimbus with a first-in-class program, so it is an exciting time for the company.

But stepping back from IRAK4 specifically, the “next wave” of immunokinase inhibitors are likely to remain a rich substrate to go after in terms of regulating different types of lymphocyte biology.  The key opportunity and challenge will be articulating (and demonstrating) precisely how these immunokinase signaling interventions can differentiate from the existing, precedented approaches.  While its tough to predict exactly how much value remains to be created around these more novel targets, its fair to say that if history is any guide it should make for fertile ground.  Enormous unmet medical need still exists in these pathways.

For those more interested in exploring the space, I’d suggest checking out an interesting data analytics and visualization approach from Relay Technology Management (check them out more generally, very cool interface).  The chart below is a small sampling of what they can do – it captures the “Established Target Index” and the “Therapeutic Targeting Index”.  The former reflects the steady state nature of the literature noise about a target (e.g., publications, grants, patent disclosures that mention it every year).  Targets on the left side are more emerging/novel, those on the right side are more “established” as targets of interest.  The y-axis reflects the “Therapeutic Targeting Index”, a score from 0 to 1 that reflects how often these targets appear in the literature next to words like inhibitor or drug or antagonist.  The higher the score, the more likely there are drugs being pursued against them.  This analysis reaffirms the emerging nature of IRAK4, NIK, Tyk2, ZAP-70 as intriguing targets.  A more extensive report can be found at this link, and includes a range of other analyses including publications, patents, deals, and R&D pipelines over time.

Here are the quick takes on each:

• Resolve Therapeutics struck a deal with Takeda (here).  The deal included an $8M upfront payment to fund the company’s work through a Phase 1b lupus study, and a predefined option to acquire in 2 years of ~$60M or so (from what I can guesstimate from the disclosures), and up to $247M if you include the two downstream development milestones.  Resolve raised about $8M from its investors over the past 2-3 years.  The lead program is in preclinical IND-enabling work.
• Zacharon was bought by BioMarin for its rare orphan genetic disease programs (here).   The company raised about $4M with a post-money of $7.5M, according to VentureSource.  BioMarin paid $10M upfront and a set of undisclosed milestones.  The most advanced programs were in the lead optimization phase of drug discovery.
• RQx Pharma struck a unique deal with Genentech (here).   On the surface it looks like a normal R&D collaboration, but from what I understand Genentech is funding the R&D from this point forward and so the upfront will be distributed.  The company drew down a little more than half of its $7M Series A round across a number of micro tranches of capital, and when the upfront is distributed it should net about ~1x the invested capital.  The total payments could go up to $111M.  The lead gram-negative bug program was in drug discovery stage. 
• Lotus Tissue Repair was acquired by Shire (here).  The company was in preclinical development for a very rare orphan disease, dystrophic epidermolysis bullosa.  They had raised about half of the announced $26M Series A, all of which came from Third Rock.  The upfront payment was $49M, or roughly a 3x return, with potentially $275M in milestones.    

These deals have lots of similarities worth highlighting:

• Innovation.  All were high risk, high science plays.  Two were orphan disease focused, two in areas of larger markets but very high unmet need conditions (lupus, gram negative bugs).  More evidence that innovative approaches are what it takes to get Pharma to the table early.
• Capability-sharing and risk-sharing deals.   They all chose to bring on a strong partner’s expertise, capabilities, and resources to increase the odds of success in the future, and by doing so offset some downstream risk while deferring the real returns into the future through creative structures.
• Capital efficient.  These companies only raised between $4M and $12M in equity capital prior to these deals, and none of them should have to raise additional funding going forward as they leverage their partners’ larger balance sheet.
• Lean, asset-centric plays.  To my knowledge, all of these companies really only had one asset or tightly related set of assets, and the acquirer/partner was clearly interested in that program.  Even were there was a “pipeline” they were tightly focused around the same target or MoA and were more legitimately “backup” compounds than standalone #2 programs.   In addition, I’m pretty sure all of them had very lean teams: all had only single digit numbers of employees, and in the case of Lotus, I think only one.
• Downstream return potential.  All four appear to have the potential, however small the probability, of achieving outsized 10-20x returns should the deal hit all the downstream milestones.  The “all-in” value of these deals could be between $100-300M inclusive of upfronts and milestones.  Its worth noting these exit valuation ranges aren’t extreme: they are at or above the median for all biotech M&A deals, but not in the top decile of exit values (>$400M).  But because they were so equity capital efficient, the return multiples could easily be in the top 1-5% of all venture deals in the vintage should the milestones deliver.  Even assuming only a third of the milestones are met, which appears to be the recent historic average, then all of these are likely to deliver 3-6x returns, certainly in the top quartile of outcomes in venture.

There are, however, a few key differences worth noting though, and the upfront return is one of them.  Only Lotus really delivered an attractive multiple and particularly IRR off the upfront payment (~3x in 18 months); congrats to the team at Third Rock Ventures who owned 75% of the deal.  RQx Pharma and Zacharon took their money off the table with these deals (~1x upfront) so are protected the downside while looking to their partners’ diligent efforts to drive the real returns.  Resolve is more of an option structure, so while attractive (i.e., no equity dilution, capable and committed partner, control over execution), the shareholders there are unlikely to see a return until Takeda (hopefully) exercises their option to acquire.

These are all interesting deals, and each board made the decision to sell early rather than go longer.  The tradeoff is a tough one.  Several have given up control entirely, others maintain it through the achievement of meaningful milestones.  The important calculus was whether raising more equity and “going it alone” would increase or decrease the odds of getting to that $100M-300M exit range and attractive multiples.  They all clearly came down in the camp of partnering early as the preferred route.

Broader questions are raised by deals like this:

• Are these types of deals going to replace the typical Series B round?  B-rounds were often what got drug discovery startups into the clinic; to a big extent, these deals were done in lieu of that Series B raise.  Asked another way, will early stage partnering deals fill the perceived gap in early stage funding?
• If more deals get taken out early in preclinical development, will we have enough substrate as an industry to create the next $500-1B exits, or the next $5B Pharmacyclics?  I don’t think this will be the issue, but its a question often asked.  I think there’s plenty of companies that are still choosing to “go long” into clinical development.  Most of those companies end up regretting it unfortunately.
• Will the dearth of interesting unpartnered Phase 2-3 programs that exists today begin to shift backward in development as earlier deals are being done?  This issue faces both the capital market buyside and the strategic Pharma buyer eager to find new clinical stage innovation.  I doubt there will be a mad rush to consume all the hot preclinical programs, but even an slight uptick will affect the pool of available clinical stage assets in 2-4 years.
• Are these types of deals likely to shorten the time to liquidity for drug discovery deals, or once in the arms of a big bureaucracy will the pace of the projects slow and actually extend the timelines?  This is especially important for the IRRs of deals where the upfront doesn’t provide a healthy return on capital.
• Will these deals deliver big enough $ returns to support some of the large Life Science funds out there?  Returning only $40-80M to a fund is a solid win for a $250M fund, but not much for a $500M fund.  These early deals clearly raise this issue, though owning 75% of them may address it.
• With more and more of the returns in these early deals embedded in milestones, it poses a bunch of interesting questions.  How should VCs value their milestones with their LP’s?  We currently build discounting models to create a fair market value for them, but there is no standard.  Also, will we finally see the emergence of a market for milestone payments like that which has emerged in the royalty space?  I continue to believe this market liquidity is likely to get created.

Many more questions come to mind, but those are a few.  Interesting deals though, and fully anticipate that we will be seeing more creative early stage partnering going forward.

It’s hard to judge how much we really learned from the 2007 process and applied to our deals since then, but it’s fair to say we’ve evolved our strategy around a number of these observations.  Like all learning loops, it’s been a process of building the institutional memory around what’s generating returns, and appreciating the mistakes of the past.  As my friend and coinvestor Kyle Lefkoff of Boulder Ventures often says: “let us make some original mistakes” – trying not to fall into the same troubles that hurt returns in the past.

Below is a very distilled and rather sanitized summary of our “Lessons Learned”; at the macro level, many of these are rather prosaic and not very insightful, but practical reality of each deal is where insight becomes actionable (and inaction causes issues).  To spare the innocent, I’ve dropped most of the company names, but all these points were largely informed by experience in the trenches not abstract thinking.

Here are five buckets of reflections:

1. Management, management, management.  Its not news to anyone that the success of a startup biotech depends in large part on the management team.  This is an axiom in venture: getting the right group of early entrepreneurs and executives around the table is critical.  But this is often not easy in early stage biotech companies.

• Different management teams are often required at different stages of a biotech, and the reality is that many seed- and early-stage deals don’t need a CEO.  They are science-driven companies that need great Chief Scientific Officers to build the fundamentals of the story, and a BD executive to help build the broader vision.  It’s upon that progress with which a company can recruit a great CEO.  Putting a founding CSO in as the CEO early on often creates unnecessary conflict in the future: having the conversation about a perceived “demotion” to CSO when hiring the future CEO is uncomfortable and avoidable.  Same goes for putting the lead BD entrepreneur in as CEO early on to “fill the role”.  Keeping the role vacant in the beginning prevents future discomfort, or at least minimizes it
• With weak management, Boards often begin to run companies.  And then it’s a vicious cycle: At the 1Q Board meeting, the management thinks they are responding to the Board so they chase after XYZ; then at the 2Q meeting, the Board says they think chasing ABC is a better idea, which the management does thinking its being responsive; and then at the 3Q meeting, the Board wonders why the company has no direction and chaos ensues.  Never a good cycle, but of course its not as simple as this.  A good Board is able to provide direction, governance, and input, and a good management is able to distill that feedback and integrate it into the strategic direction of the company.  It’s a healthy balance and tension.  But keeping the Board away from whiplashing the “day to day” program choices of what to “chase” is key.  An important nuance is worth mentioning here though: an active Board Chairman or single Lead investor playing the role of an Acting CEO is very typical in an early stage startup, and is a good thing (especially given the point above about not having a formal CEO in a science-led startup at the beginning).
• Making management changes quickly is almost always the right answer.  We have historically not moved fast enough to make senior management changes even when we knew it wasn’t working.  Trusting one’s instincts is important: if it feels like its not working, it probably isn’t.  And the team working in the company probably sees the same thing from their view of the executives.  Further, if real management questions are present at the closing of a new investment its unlikely to improve: its often worth being explicit about this with the existing team before the closing to lay out expectations and possible action plans.  Closing the deal and then firing the CEO immediately after doesn’t feel like the high road.
• Sometimes we’ve delayed for apparently good reasons (e.g., “we were close to a BD deal and didn’t want to rock the boat during the process”), which more often than not turns out to be poor reasons (e.g., “the deal never got done”).  Ripping the bandaid off quickly is almost always the right answer.
• Don’t hire for the resume, hire for real talent.  Lots of folks in biotech have good resumes, were part of stories with great drugs or great exits but didn’t actually shape them (or at least no where near as much as they think they shaped them).  We’ve certainly hired our fair share of great paper-resume CEOs that didn’t translate into excellent leaders and operators in our startups.  Diligencing the specifics of their actual contributions in the past is an important part of reference checking and recruiting.  Some of the disconnect between paper and practice is that the transition from large to small companies is hard despite great past roles; others is that their past success was more luck than skill and repeating luck is a challenge.

2. You can’t pick your family, but you can pick your co-investors.  The vast majority of biotech deals require >$15M in venture capital to get to an exit, and the average is close to $60M.  This means that most deals require syndicates of at least three VCs and often more.  This creates a lot of potential for entropy.  Our observations:

• Big syndicates are often dysfunctional, and we try to avoid them. The bigger the syndicate, the more cooks in the kitchen, the more distracting the differing priorities become: different exit preferences (“go long” vs “sell early”), different capital appetites (capital-sparing vs pushing money into a deal), different views of a management team (change vs no change), different disease area interests (one made money on eye-deals, another lost money on eye-deals).  Unless there’s a real lead investor and well-vetted group of like-minded partners, the big venture syndicates are frequently a struggle to manage.  This is not always the case (e.g., Avila had five VCs and was a well-behaved Board), but it’s frequently an issue and becomes increasingly problematic as a company raises larger amounts of equity capital from broader sets of investors.  One of the reasons we get involved at the seed-stage is so that we can shape the co-investors we work with.
• Misaligned investors can cause as much harm as weak management.  This is undoubtedly true, and, for the reasons stated above, big syndicates increase the odds of misalignment.  The best management teams work hard to keep their investor syndicates aligned and maintain a continual open Board dialogue about that alignment on strategy, vision, financing, exit, etc…
• Experience beyond the venture skillset can add real value on early stage boards.  We are big believers that bringing great independent directors into deals early can be a huge help for startups.  For example, getting former Vertex President and head of R&D Vicki Sato involved at Nimbus, even when the board was larger than the company, was the right thing to do and a great addition.

3. Diligence would benefit from more realistic crystal balls.  The best way to not lose money in biotech is to not invest in biotech.  It’s not a simple business, and deep due diligence around the science, programs, team, plan, patents, etc is critically important.

• Most drug discovery companies fail to deliver on their overly-optimistic initial Series A plan, so factor that into the financing plan.  Almost every new discovery-stage startup comes in with a pitch that says “in 30 months we’ll have an IND”.  Unless you have your lead scaffold in hand now, you’re not likely to get there in 30 months: novel modalities or new scaffold identification and optimization just take time.  In our experience, it takes between 36-48 months to get to an IND around novel chemistry when the plan hits reality.  Even a superbly executed plan can fall short: Avila’s BTK program filed its IND 42 months from the start of the company, 12 months later than forecast in the Series A plan.  Factoring in the almost certain delay and slippage is important when thinking through capital and time requirements.  The Catch-22 here is that if an entrepreneur pitches a plan for 48 months to an IND, even to an early stage investor like Atlas, its likely to receive a lukewarm reception; this leads to overly optimistic plans from everyone that 30 months can do it.  It’s very healthy to have a frank conversation about this paradox and the issues around it early in a diligence process.
• Do your own primary diligence.  Often when syndicates form around a new deal, the proverbial summer pool effect happens: since every parent thinks other parents are diligently watching the kids, no one does and bad things happen.  Just because great VC firm XYZ is committing to do the deal doesn’t mean they did great diligence.  We’ve learned the hard way that different firms, and different partners in the same firm, often do varying degrees of diligence.  Sharing consultants and experts is fine, but having a direct dialogue with them is important: taking a fellow VC’s word for it that “former head of R&D John Doe thinks this is the hottest program ever” is simply not sufficient.

4. Super models and investment models: beauty is in the eye of the beholder.  Lots of different investment models work in biotech, and most firms have their favorite.  Some firms have done well with PIPES and later stage assets, like Longitude and Sofinnova, but it’s just not a space we’ve engaged in given our early stage innovation bias and the desire to keep doing what works for generating returns for us.  Our experience with different models over the past decade has led to a few lessons for us:

• Going in early allows us to shape the DNA of the company.  Whether it turns out to be the bleeding edge or leading edge, some of the best deals in our portfolio over time have come from the roll-up-your-sleeves model of venture creation.  This is a recurrent theme on this blog, so I’ll leave it at that.
• If you can tranche the capital into a deal around important milestones, you should.  By metering in the money, an investor can monitor not only the derisking of the science, but equally important we can observe how the team executes and delivers what they claim.  We’ve been burnt repeatedly in the past on the big raises that weren’t tranched.  These can’t be anorexic high frequency tranches of funding, but appropriately designed 9-18 month tranching is optimal.
• Later-stage ‘opportunistic’ or spec pharma deals are often much riskier than meets the eye.  I’ve riffed on this theme before more generally (here), but we know this first hand because we’ve previously been seduced by their siren song: we’ve not done well with low-risk repositioned late stage assets like Ivrea, ARCA, Shogoo, Newron, Xytis, Nitec, etc from the 2003-2005 period.  Our big takeaway lesson was to avoid these type of plays, and you won’t find them in any new deals since early 2007.
• Quick flips in biotech don’t happen often, unless it’s a flip to the trashcan. Promises of quick FDA approvals or near term M&A interest are most often empty claims.  We’ve all heard the “this is the last financing before an exit” pitch, or “the ongoing Phase 2 is going well” in order to close a financing before opening the envelope.  This latter pitch burnt us twice in the mid-2000s: Ivrea and Proprius both had inherited ongoing Phase 2s, both of which readout as completely uninterpretable due to protocol issues among other confounding variables.  So much for the quick path to Phase 3.  We’d prefer to focus on sustainable value creation as the thesis is much more credible.
• Company diversification by adding assets across risk or clinical stage often destroys value (another theme I’ve mentioned before).  Forming a company with a Phase 2 in-licensed asset and a preclinical program isn’t smart due to the stage and value asymmetry: the Phase 2 program consumes the capital and dictates the future of the company, dragging the preclinical asset either up or down.  Furthermore, even if the assets are of similar stage, they typically aren’t of similar merit.  Adding a second asset to a story that’s not quite as compelling to the first just to make use of a management team feels like the wrong use of capital; figure out other ways of leveraging a team (e.g., part-time roles at other portfolio companies) and keep the bar for asset dilution very high.  These observations 5+ years ago led to our asset-centric investment model around single program entities, like Stromedix’s STX-100 focus.  Our AVDC initiative was born out of this observation in conjunction with Pharma’s externalization trend.
• Ultra-lean virtual models are great, but they have their limitations.  For instance, big biology plays are hard to do without your own wet lab to do the bespoke work (though its fair to say we haven’t really gotten that wrong yet).  Where we’ve felt some pain is around the constraints of the virtual model on other operational elements: when you have only a few key entrepreneurs in a startup, its very hard to go on the road for fundraising or BD without stalling or slowing the work on the R&D front.  It can be incredibly distracting for a lean team to enter full due diligence with an army of experts from multiple pharmas or potential investors.  Figuring out how to focus the BD campaign quickly is essential, and if one can avoid fundraising by having a strong insider syndicate that’s incredibly helpful.  This is certainly something we are mindful of in our more virtual plays.

5. Lawyers are a necessary evil.  We also looked at our legal, IP, and deal-making experiences to see if there were lessons here.  Getting good legal advice is clearly very important to successfully navigating deals, especially in preventing bad outcomes from becoming worse.

• Patents are critical to biotech, so don’t skimp on doing deep IP diligence before investing.  This has helped prevent us from doing some bad deals.  We’ve only really had one big IP blow-up that changed a company’s trajectory: within months of the Dynogen $50M Series B closing in 2004 (no tranching!), a unexpected patent appeared that impaired the lead DDP200 program.  So in some ways we’ve been lucky.  But IP diligence goes beyond doing work before an investment; we’ve found that biotechs with best practice IP strategies often have an IP audit by an outside firm every couple years to ensure they are on strong ground
• For critical deals, investors should read the contract details.  Similar to the diligence point above, trusting that others in the Boardroom have read the critical deal documents isn’t always a great strategy.  Having a surprise appear after the fact can be a painful realization.  We had this happen 5+ years ago a couple times with uncomfortable outcomes.
• Structured deals with Pharma have a long gestation time, no matter how many lawyers and accountants are in the room.  Don’t assume you can speed them up too much.  When Big Pharma XYZ tells you they are spinning out some assets and they show you the initial list, assume it’s going to take something in between a human and a pachyderm pregnancy to get to the closing, especially if there’s a structured buyback with complex consolidation details included.  This gestation just comes with the territory of large bureaucratic organizations and complicated deal structures.

While we’ve had no major “Ah-ha” moments in reviewing our Lessons Learned (wish it were that easy), we do find that revisiting and updating them every few years is helpful.  And during our champion and challenge process for new deals, we do bring up these lessons to gently bludgeon each other once in a while.

The quick summary from a venture-backed biotech perspective: it was a good year although not a great one.  Upfront and total deal values for venture-backed M&A of $3.5B and $8.5B, respectively in 2012.  Those total upfronts are smaller than 2011, but larger than 2008-2010 annual numbers.  Total annual deal numbers for VC-backed private acquisitions have been relatively steady since 2005 around ~30 deals, except for 2009 when only 19 occurred.  About two-thirds of those deals have financials disclosed.

Here’s my quick take on the Good, Bad, and Ugly:

The Good: median investor multiples on upfront have risen from 1-2x during 2005-2010 up to 2.5x in 2011 and an attractive 3.5x in 2012.  This trend bodes well for returns in the LS venture sector.

The Bad: European biotech exits have lagged in this data.  The median exit value to invested capital multiple in European deals in 2012 was 1.3x vs. 3.5x in the US.  For all but one of the last five years the US multiple has been higher by 50% or more.  While small deal numbers start to test the significance of this, it’s certainly in line with sentiment that it’s been hard to drive great returns in Europe.  Certainly a number of outliers to this point, and a few top venture firms have been able to consistently deliver, but as a sector its been even more challenging than in the US.

The Ugly: The trend on time from founding to trade sale looks painful.  This appears to have gone up considerably from 2005 (~5 years) to 2012 (~9 yrs) in this dataset.  I don’t have the raw numbers per deal around this, so don’t know how much these numbers are skewed by a few ‘older’ companies in each year (e.g., in 2012 DeCode and Enobia were founded 16 and 15 years before, respectively; both had recaps more recently).  But its fair to say the trend line makes the signal hard to argue with and clearly of concern.  Fortunately, time from founding to exit isn’t all that related to product stage and its possible to shorten the investment cycle time with different investment models: of 2012 exits, Skinmedica was 13 years from founding to exit with a marketed product, whereas Avila and Stromedix were less than 6 years from founding to exit with Phase 1 assets.

Lots of food for thought, as usual, in the HBM M&A report.  Thanks to Uli and team for putting it out.

For years we’ve complained about a triple threat of worries: FDA obstacles, poor public markets, and a perceived lack of funding in biotech.  Some of this we’ve done to ourselves (and deserved), as I’ve noted before (here), but 2012 in particular has challenged these concerns with positive data which should manifest into real improvements in the early stage climate.  I’ve been accused of being a Pollyanna about early stage by some, but I think the data support a more optimistic view than is widely held by many stakeholders and observers.  Unfortunately even with good data it will take time to turn bearish sentiments into bullish ones, but I think the fundamentals are pointing in the right direction.  Let me take on each of the big three “worries” that has plagued our sector of late.

FDA and the regulatory landscape.  Last year we witnessed 39 new molecular entity drug approvals by the FDA, one of the three best years in the past half-a-century (yes, 50 years!).  2012 was only bested by the outlier year of 1996, which at 53 included a big “drug lag” cohort pushed through in the wake of PDUFA I (passed in 1992).  Coupled with a big year in 2011, this is a great sign that the decline in approvals may be reversing.  According to FierceBiotech, McKinsey forecasts 36 or so approvals each year through 2016.  This is great news, and would make for one of the best five years in the history of the Agency.

Lots of others have already written about the positive and negative attributes of these 39 approvals (here, here, here).  It’s worth noting that most of the critique is that many of the drugs last year were either approved for orphan diseases or were me-too incremental innovations.  These criticisms may well be true, but this isn’t much different than others years.  Taking the outlier 1996 cohort of approvals as an example, how many readers of heard of blockbusters like Albenza, Cerebyx, ProAmatine, Elmeron, and Zanaflex?  All of those were orphan drug approvals.  And what about the incremental me-too innovations approved in 1996: Lipitor, Crixivan, Humalog, Zyprexa, and Diovan.  They all did reasonably well.  I have no reason to believe the Class of 2012 drug approvals won’t have some strong performers.

Irrespective of whether these drugs help drive big economic gains for Pharma, they certainly reflect an FDA that is increasingly open to approving new medicines, especially those that are real, innovative products.  Many of the first-in-class agents approved last year were reviewed faster then their full PDUFA-directed review time.  This and other efforts, like the Breakthrough medicine concept, reflect an FDA that is trying, and succeeding at least in part, to adopt a more innovation-friendly posture, and this bodes well for early stage biotech as we need clear paths to regulatory approval.

IPOs and the public market for biotech.  While I’m often a critic of the “go-public” path for aspiring biotechs, its fair to say that post-IPO performance of the class of 2011-2012 has done exceptionally well.  Of the 20 biotechs that priced IPOs in that period, 75% are above their trading price, with an average and median gain of 27% and 17% respectively.  Intercept and Pacira have more than doubled, Kythera is up 80%, and even preclinical stories like Regulus and Verastem have held their own.  This is much better than in many prior biotech vintages of the past decade where two-thirds traded down in their first 12-24 months under the pressure of future fundraising expectations.  However, it remains a real issue that the pricing of IPOs is a massive challenge (and I still won’t advocate for the IPO path for most biotechs), but for much of the 2000s both pricing and after-market performance were issues.  At least we’re trending well on the latter and hopefully the former will follow suite.  We obviously need both to create a healthy public market for new biotech offerings. 

A related point to the improving IPO landscape is that many of the buysiders that would buy into these offerings are recognizing that great companies are getting plucked off by Pharma before they get to the public markets.  This has led to a influx of buyside interest in late stage stories again.  Take Agios and Foundation Medicine – both have raised considerable amounts of capital from buyside investors at very high valuations.  In fact, Agios may be the highest valued preclinical stage company I can think of, reflecting the confidence the market has in David Schenkein, his team, and the cancer metabolism field.  Seeing some of these emerging success stories come to fruition with solid IPOs or M&A deals will be further catalysts for addressing the public market worry that many early stage investors currently have. 

Funding.  The quarterly vacillations in venture funding usually create lots of fanfare, especially amongst the Chicken Little’s of the sector, but the aggregate annual funding numbers for the past few years are surprisingly strong.  Based on BioCentury data, $5.5 Billion was invested in venture-backed biotechs in 2012, up from 2008-2011, and it includes ~$1 Billion for Seed/Series A investments.  Given the outcry about funding drying up, this is pretty robust.

At the New Paradigms meeting (a satellite of JPM Conference), a panel I was involved with discussed the perceived funding gap and whether great companies were still getting financing.  The unanimous view was that innovative new startups were continuing to attract capital.  I do share the concern of many about the number of new startups being formed in biotech; its down from its historic range of 80-100 per year, and is worth watching closely.  These new startups will be what feeds the later stage funding cycle and future IPOs.  That said, a flip side to this is that as an early stage investor involved in creating these new startups, the supply-and-demand dynamic reinforces my belief that this a great time for early stage venture creation.  Many entrepreneurs I know feel the same way.

Although the aggregate funding numbers remain strong, its very clear that the mix of investors in these financings has changed.  As noted above, the buyside and in particular hedge fund investors are supporting some of the later rounds.  In addition, the early stage rounds have seen a significant increase in corporate venture capital involvement (discussed here, here).  As I’ve commented previously, some 80%+ of our new deals are now syndicated with Pharma VCs.  These corporate players are also actively filing the coffers of standalone venture firms through an increasing number of LP relationships.  Lastly, a new crop of life science VCs have emerged.  Beyond big new players, like Third Rock Ventures and VenBio, there are a number of smaller and/or regionally focused LS investors including Remeditex Ventures (coinvestor with me at Miragen), Allied Bridge Gap Ventures in Boston, City Hill Ventures, NaviMed Capital in the device space, and a number of others.

I’d be remiss if I didn’t acknowledge that there’s a huge bias against Life Sciences in the current LP environment; LS is clearly a contrarian bet today, and its not easy to raise funds from traditional LPs for either hybrid funds (Tech + LS) or LS-only vehicles.  This negativity is a major concern, but hopefully continued solid data on LS returns, coupled with an embrace of data-driven decision-making by LP’s, will change this sentiment for the better.

Beyond the directional improvement around these big three worries, I am encouraged to see that innovation is continuing to be rewarded, through both earlier stage M&A deals with great venture returns (like Phase 1-2 exits at Calistoga, Amira, Avila, Stromedix, Enobia), as well as in the public markets, as I’ve written about recently (here).  Further, Pharma’s posture has changed for the better towards early stage innovaiton, and it appears keen on helping facilitate but not suffocate translational efforts beyond their walls (here).   These trends also bode well for early stage biotech.

We live in an era of big biology: we’ve taken ten years, but the real fruits of genomics are finally ripening and exciting new medicines are being developed.  This is a great time to be in early stage translational research, and I’m hopeful the FDA, the public markets, and the funding environment continue to trend favorably.  As a sector, we’re scaling the wall of worry and I look forward to a great run of successful investing in innovation over the next few years.

The resounding theme we heard from Pharma was about their desire to not only identify, track, and potentially partner around innovative early stage assets, but their interest in helping build new startups with us prospectively.  Continuing a trend over the last few years, Pharma clearly wants to get closer to venture creation around new startups.

Unlike past JPMs, I didn’t repeatedly hear the “this is too early, keep us posted when you get your Phase 2 data” message from prospective partners.  What I did hear a lot of was “what new startups are you working on…”, “there aren’t enough innovative assets in Phase 2 or 3″, and “we’re looking for pre-IND stage innovative programs”.  Their collective interest has definitely moved upstream into earlier stage R&D, and includes late discovery and development candidate programs in IND-enabling work as well as new platforms for exploring novel biology.  Whether or not this will translate into more deals remains to be seen, but it certainly is a change in the posture of Pharma’s external engagement relative to only a few years ago.  Innovative preclinical programs like Nimbus’ IRAK4 and Acetyl-coA Carboxylase inhibitors, miRagen’s cardiac miR-208 inhibitor, and RaNA’s gain-of-function approach were of great interest to many prospective partners.  Lotus’ acquisition this week by Shire highlights this as well – its lead program was a preclinical orphan disease protein replacement therapy.  And multiple early stage platform collaborations were announced this week with GSK-Priaxon, Gilead-Macrogenics, and Cubist-Heptares.  

Here are a few sub-points of this theme worth highlighting:

• Pharma wants to creatively engage in earlier BD around high innovation-quotient new medicines.  One exec mentioned that all of their recent deals but one were for pre-IND stage candidates, and that all of those were based on relationships that had been built over 12-24 months of tracking, monitoring, and engaging on the biotech’s scientific plan.  Building trust around the robustness of the science and integrity around collaborative potential is critically important.
• Seed-stage discovery projects were of particular interest to Pharma.  The opportunity to work with us closely around some brand new startups, primarily for shaping and proximity and not for exclusivity, was of clear interest.  Its also not about primarily about providing access to seed-stage capital: Pharma recognizes (as do we) that they have a deep bench of talent in areas like lead optimization, pharmacology, preclinical, and early development that could greatly aid startups’ early plans, and doing so would put them in the pole position for future deal-making.  J&J’s new Innovation Centers concept is aimed at this type of interaction.
• Accessing “open market” input as a way of shaking up their internal R&D culture was also a recurring message: getting objective, external market views on their early stage programs, leveraging the currency of entrepreneurship, engaging externalization partners as a means of building a prospective pipeline outside of the confines of the internal R&D organization.
• Lastly, Pharma is clearly very interested in tighter linkages with VCs, and vice versa.  These relationships go way beyond just information sharing around “VC-days” and the like.  They increasingly involve collaborative arrangements (like Atlas-Shire), as well as more significant LP arrangements (e.g., GSK just announced another LP commitment into Sanderling; Merck invested in Flagship last year; etc…).  I suspect this type of interaction will increase in frequency in 2013.  Importantly though, these deeper links are likely to happen primarily between early stage LS funds and Pharma; late stage venture funds and spec pharma focused investors just don’t cater to the high innovation-quotient interests that most Pharma seem to be expressing.

As a side note, its worth highlighting that as Pharma goes upstream, risk-sharing is going to continue to accelerate, and the structured deals we’re working through now are likely to continue to be in favor.  As early stage investors we’re very open to these types of models: securing a return for early innovative risk-taking with an opportunity to leverage a larger R&D budget to achieve future success milestones.  The alternative is to take on significant dilution from later stage investors to carry these projects forward without a partner in the hopes of a bigger exit or IPO.  The calculus around this is clear – if the immediate returns are interesting enough, and the capital required to finance a development program large enough, the clear case for early stage deal-making can be made.  Further, Pharma typically has a strong perspective around how to conduct clinical development, and for them there’s a real premium for them in taking on responsibility for projects early in clinical development.  These are win–win deal structures from my perspective, combining risk-sharing and reward-sharing: venture–backed biotech take risk around new biology, modalities, and programs, while Pharma engages as those projects mature and integrate them into their global development models.  The emerging data around milestone-payouts support this premise: about a third of milestones are getting paid, a third are still hoped for, and a third have been missed.  This is a reasonable hit rate compared to historic R&D attrition rates.  Risk-sharing can be structured to be good for all parties.

To close out, sitting on the JetBlue flight home from JPM yesterday with 150 other Boston biotech folks, the post-conference mood seemed very positive, albeit exhausted.  We’re all hopeful 2013 delivers on the excitement of this year’s kickoff circus – and that’s going to require pharma, biotech, and venture to work constructively to support the ecosystem.

The basic premise we hold is that high innovation quotient opportunities have the best chance for making real impact on patients and hence generating attractive returns.  Furthermore, only these types of deals have a shot at early stage exits (e.g., Phase 2 Proof of Concept or earlier), as Pharma is only willing to engage earlier on transformative medicines vs incremental innovations.

In 2011, we saw a bunch of great high innovation quotient wins: B-raf at Plexxikon, PI3Kd at Calistoga, LPA at Amira, dual PI3K/mTor at Intellikine, oncolytic viruses at BioVex, and Astofase Alfa at Enobia.

Real innovation delivered great results again this year.  Three big notable wins:

• BTK.  Bruton’s Tyrosine Kinase was the belle of the ball at ASH again this year, with fantastic data from Pharmacyclics’ ibrutinib.  The excitement around that program propelling their stock price up over 300% in twelve months.  Avila obviously benefited from the excitement around this hot target with its sale to Celgene earlier this year.
• CARs.  The engineering of T-cells with Chimeric Antigen Receptor (CAR) technology has delivered some astounding, breakthrough data in B-cell cancers.  We’re not talking about killing millions of cancer cells, we’re talking about removing kilograms of tumor load (trillions of cells).  The early and limited patient response data are staggering.  The results from Carl Jung’s lab at Penn prompted a very exciting early deal with Novartis, and I expect this field to accelerate rapidly in 2013.
• DMD.  Sarepta’s surprisingly positive data this year with its exon-skipping technology in Duchenne’s Muscular Dystrophy brought hope to patients and their families, and a 450% increase in their stock price since earlier in 2012.  Its been a long road for Sarepta (f.k.a. AVI BioPharm) and for RNA therapeutics, but this is a truly remarkable demonstration of this new modalities promise.

Technology platform innovation also closed the year with momentum.  Envoy’s cutting-edge neurology target platform was acquired by Takeda for $140M (including a nearly 4x return upfront), and the saga of DeCode Genetics completed a nearly two-decade run with Amgen’s $415M acquisition (a 6x return).  Both of these deals were fundamentally high innovation quotient platform biology acquisitions.

It wouldn’t be fair though to just highlight the victories on the side of innovation this year.  It’s also tough to innovate.  There’s real risk in pushing new biology into clinical development, especially in diseases where gold standards already exits.  2012 was marked with a number of challenging events for novel programs in later stages of development: the failure of AZ’s Syk inhibitor to differentiate from Humira in Phase 2b,  Lilly’s anti-BAFF mAb in Phase 3 was terminated due to insufficient efficacy, Pfizer and Lilly’s anti-amyloid mAbs missed their Phase 3 endpoints,  Oncothyreon’s immunotherapy for lung cancer blew up this week as did Allon’s cognition program…

The big question on several of those late stage blow-ups is whether a more critical view of their early development data would have prevented those Type I errors (false positives) where the teams pushed them into big expensive Phase 2b/3 studies.  I have no idea, but certainly lots of pundit speculation seem to think so.

As we close out the year and think ahead to 2013, it’s nice to celebrate some of the high innovation quotient deals in our space.  I know there are many more than the above.  I hope readers can share their thoughts on companies or concepts that hit high marks on innovation from their perspective and why.

Lastly, its worth noting that most breakthroughs come from new places – RNA-based drugs (Sarepta), genetic engineering of T-cells (CARs), unprecedented targets (BTK), to name a few.  And upon demonstrating promise, these breakthroughs reward the risk-takers behind them – the scientists, entrepreneurs, founders, and investors.  But most importantly, they have huge impact on patients and their families.  It’s this wonderful combination of doing well by doing good that sums up why we come to work everyday.

The same holds true in early stage venture capital; distributing realizations from M&A deals to Limited Partners enables the raising of future funds, which keeps the innovation cycle going and well funded.  We obviously need more of this in the venture arena: as has been reported by others, life science venture funds have struggled to convince LP’s in today’s environment and the sector is consolidating around a new mix of players.

But beyond the importance of recycling capital, there’s an equally important dimension to consider: M&A deals help to recycle talented teams so they can spend their time on new emerging startups.  Seasoned biotech executives who have been a part of prior M&A deals are highly sought after.  These are the “been-there, done-that” management teams who’ve been through the challenges of building biotechs, but importantly have also seen what success looks like.  Knowing the hallmarks of success – product, culture, team, Board, process – is a big factor in helping guide younger companies down the right path.

A quick review of our biotech portfolio suggests that roughly 45% of the CEOs were prior executives in successful M&A exits.  Adding in the senior leadership teams, I’m sure much of our portfolio has exposure to these veterans.  This is obviously of great value to our companies, and we actively recruit and cultivate relationships with serial executives to help us build successful companies.

More importantly, this recycling of talent is of huge value to the ecosystem.  This diaspora of biotech executives into the local environment helps disseminate the learnings and nuances around what success looks like, how to run a BD process, how to avoid pitfalls, etc…  This creates the virtuous cycle that reinforces the importance of biotech clusters like Boston, San Francisco, or San Diego.

Take a few case examples.

Avila was a 2012 exit for us, and some of leadership team is moving on to new things.  CBO Nagesh Mahanthappa is the founder and CEO of a new startup called Scholar Rock, CFO Andrew Hirsch is now at BIND, and Executive Chair Dan Lynch is now in a similar role at both Nimbus Discovery and RaNA Therapeutics, among others.  And I’d guess that CEO Katrine Bosley won’t be unemployed for long.  In aggregate, that leadership team will be involved in 6-8 biotechs in Boston within a year of the closing of the acquisition by Celgene.

A less recent example highlights the post-M&A effect: Adnexus’ leadership team went all over. Katrine to Avila, former CSO Eric Furfine is now CSO at Eleven Biotherapeutics, former COO John Edwards is Executive Chair of our bispecifics play F-star as well as Permeon Biologics, former CMO Marty Freed is founder and CMO of Civitas Therapeutics,  and former CEO John Mendlein has served as Executive Chair of aTyr Pharma and Fate Therapeutics.

Other biotech clusters are also good at recycling their successful teams: Amira’s leadership has now set up several startups: Peppi Prasit set up Inception, and Jilly Evans and her colleagues have founded PharmAria with Celgene.  Anadys’ former CEO Steve Worland is starting a new company eFFECTOR, and Calistoga’s former CEO Carol Gallagher is the Chairman.   Intellikine’s former CEO Troy Wilson has started a new entity called Wellspring Biosciences.

So if each successful exit sends out a diaspora of leadership team veterans to settle into half a dozen or more other biotechs, including both new startups and existing companies, the impact on a local ecosystem can be profound.  These veterans become the mentors of the next wave of biotech leadership teams, helping coach and cultivate the real insights around success.  Importantly, they help first-time C-level executives navigate the trials and tribulations of those roles.  It is this flow of entrepreneurs and executives, and the learning loops around them, becomes the lifeblood of a tight biotech community.

In less established biotech clusters, the lack of this “success diaspora” is a serious limitation.  Not only does it impact the current biotechs, but it prevents the inter-company transfer of the hallmarks of success that help shape the winners of tomorrow.  Figuring out how to access these insights is an imperative for clusters that don’t have a steady flow of deal-enabled entrepreneurial movement.

More M&A does far more than just generate returns: it liberates talent to keep the early stage biotech ecosystem vibrant.

There are many possible modes of failure in this process, and all of us involved in this part of the ecosystem spend a lot of time thinking through how to derisk projects, reduce those costly false positives, and channel capital towards the right innovations.  The ingredients of success are multiple – sound science, titration of capital, clear early development path, the right exploratory markers, etc…

Form follows function, and to integrate all these things many “forms” of early stage translational R&D have emerged.  Academics have tried to develop projects on their own, through new translational R&D centers; the jury is still out on how these will do.  Pharma has tried to integrate projects directly out of academic labs through licensing arrangements, though once inside Pharma’s research behemoths these projects often lose momentum.  To avoid this, Pharma has more recently also tried to incubate projects within academia through large scale direct-to-academia deals.  It will be interesting to see how those play out, and if they will be different from the lackluster relationships of the past.  The NIH is also taking a shot at translational work through a shift in its emphasis and the NCATS effort.  These are all important efforts.

But they are likely to fail to deliver.

Fundamentally, these “forms” don’t embrace at their core the quintessential nature of this early R&D: it requires real risk-taking.  And risk-taking requires entrepreneurship.  Venture capitalists take risks with their capital, but ultimately the real risk-takers are those who dedicate their time, the scarcest resource of all, to these projects.  These are the entrepreneurs who help create new biotechs, throw themselves at them 24-7, and relentlessly drive them forward.  They don’t want to waste their time on the false positives – those drug projects that advance, consume resources, but fail to prove out in the clinic.  It’s not the existence of a job that they are worried about.  They are just as aligned as investors to want to kill projects early that don’t work.  Great entrepreneurs are totally aligned with other shareholders around wanting the capital efficient discharge of risk from their project or portfolio.  And lastly, all of these entrepreneurs tend to be motivated with the concept of “doing well by doing good” – create great medicines and the financial rewards will follow.

So why can’t the other “forms” of translational R&D work as well?  As my partner Jean-Francois Formela likes to say, it’s because they lack the fundamental currency of entrepreneurship: equity ownership in the startup.  This is the unit that rewards the successful entrepreneur and binds them together with the academic founders, key advisors, and the broader shareholders.  It’s this currency that creates the mutual ownership that drives teams and success.  Those small, focused startups that we call “biotech” are the only “form” of translational R&D enterprise that harnesses this currency effectively.

It’s the very nature of this equity currency that enables impact.  A great chemist in a drug discovery startup can personally change the trajectory of the equity of the startup with a new lead series.  Working over the weekend actually might change the outcome of a startup.  When a superbly crafted development plan, developed by a few individuals in a startup, achieves clinical PoC, it can create huge equity appreciation potential.  It’s also how scientific founders are rewarded: they are real owners of the translation of their work through their startups, and can play an active role in shaping it over time.

Unfortunately, the other “forms” of translational R&D don’t have this level of alignment or impact.  A great drug discovery researcher at Merck has no real ability to alter Merck’s stock price.  An academic investigator that excited partners his work with Pharma for some lab funding often regrets it later when they see their project get buried in a portfolio of 100s of discovery-stage efforts.

As the entrepreneur’s currency, equity provides a clarity of focus and transparency around value that few other measures can.  Equity is the only currency that rewards risk-takers.  Like other currencies, there’s a market out there that prices it on a daily, monthly, or quarterly basis.  Early stage investors help price the equity initially, but importantly so do downstream pharma partners, later stage investors, and eventually traders on the more liquid capital markets.

In early stage biotech, it’s easy to argue pricing of this currency isn’t very efficient: there are too few buyers of equity to enable a marketplace.  This is undoubtedly true in part.  But it’s also true that if the pricing was systemically too low, then the asset class’ returns would be far higher and it would equilibrate.  The sad reality is that venture returns, across all sectors, haven’t been great for a decade, so at a macro level its fair to say that on average the pricing of equity, or risk, hasn’t been too low.

Equity is also priced by entrepreneurs, not just by those with cash.  Entrepreneurs effectively value their equity as the price for their scarce time.  If there’s not enough equity at Biotech ABC, I’ll spend my time elsewhere.  This gets at the importance of biotech hubs.  Boston excels in large part becauase the pricing of startup equity is more efficient here than elsewhere.  More choices for entrepreneurs, more choices for capital providers – creates a more efficient (not fully efficient) market for the value of biotech startup equity.

Importantly, all currencies are based on some underlying value.  As a currency for entrepreneurship in early stage biotech, it’s fundamentally based on data: has the entrepreneur and her startup been able to generate a compelling data package around a platform or product that supports the appreciation of the underlying value of the equity.  This data package, coupled with the intellectual property that enables appropriate time-delimited monopoly pricing to extract returns over time, is what creates the futures market for their currency.

As with most marketable currencies, equity in a biotech can suffer from deviations from this underlying data-supported value, which can result in over-pricing of assets that lack the intrinsic quality or robustness.  Entrepreneurs often try to maximize the value of their currency by pushing their biotech’s stories beyond where the data supports.  But this is self-defeating in the end.  Over-promotion leads to reputational consequences and in the long-term bubbles in an entrepreneur’s equity get efficiently repriced.  More often, great entrepreneurs recognize that their time is the ultimate asset, and spending it on opportunities with better promise is the way to maximize one’s potential.

If other “forms” of translational R&D are going to work, they need to figure out how to inject the currency of entrepreneurship into their models.  Encourage and reward real risk-taking.  Cultivate a culture of rule-breakers not rule-takers.  Move the career-ist phenotype out of management role to liberate the entrepreneur within.  This can’t just be “fiat currency” either; it needs to be meaningfully grounded in external market value and provide the significant startup-equity-like financial incentives for those that succeed. This will be a huge challenge culturally, and may cause atrial fibrillation in most HR departments or academic hierarchies, but it’s the best way to unleash the entrepreneurial spirit required to for successful translational research.

The other alternative, of course, is figuring out how to leverage the existing currency of biotech equity to facilitate early stage translational efforts, and this is happening with increasing frequency: Pharma-Venture collaborations, open source externalization of R&D portfolios, pre-competitive equity-based collaborations like Enlight, Academic-Pharma alliances that involve creating startups or partnering with biotechs, etc…  At their core, all of these involve liberating the entrepreneurial culture that is so adept at challenging the status quo, advancing new innovation, and testing new translational hypotheses.

Lastly, markets function best when rewards manifest over appropriate periods of time to free up existing resources.  To do this effectively, a viable futures market for this currency of entrepreneurship must exist.  Whether through M&A or IPOs, equity realizations in this market are critical to unleashing this entrepreneurial currency and recycle the time, talent, and capital into the next wave of new ideas.

A thesis we believe in at Atlas is that one needs to be able to make a reasonable return at a median exit valuation – i.e., the math needs to work for a deal outcome at the 50^th percentile.  Of course, most venture investors live in the “top quartile only” world of Lake Wobegon, so medians don’t tend to engage our sector’s thinking.  But as an asset class, we should focus on them more.

So what are median value exit outcomes in biotech?  The data are not readily available, so to put some meat around that number I’ve pulled together a few datasets.  Using BioCentury M&A data along with some banker-provided IPO statistics, I assembled a list of the 194 biotech deals that either were sold as private companies for more than $10M upfront, or went public from 2007-2011.  Keep in mind these data don’t include the ~20% or so of biotechs that go bankrupt or liquidate for zero value, or those that sell for marginal values of <$10M upfront.  For IPOs, their pre-money valuation is used as a proxy for exit value.

The median valuation in this five year dataset is ~$105M, and the top decile is ~$430M.  With this dataset, that means about 20 exits were at or above $430M during this period.  The average exit value was $185M, skewed above the median as expected due to some of the larger deals.

Here’s the broader distribution:

There are multiple implications of a distribution like this.  If you believe this historic range of returns set up a reasonably stable constant like gravity, then building companies with these metrics in mind is critical.

One observation is around the equity capital intensity that supports venture returns: $25-40M appears to be the sweet spot for the amount to raise before a major value inflection or “exit” via M&A or IPO.  The simple math is hard to argue with: by keeping to this range, if a biotech is unfortunately only able to hit a median valuation ($104M), it is still feasible to return a modest 2-3x on invested capital.  But if a company can achieve a top decile valuation (>$430M), the exit delivers a 10x+ return.  Lots of room in between for great outcomes.  But if a company raises $60M or more in equity, the median outcome starts to look less interesting, and above $80M there’s a good chance of a loss on the invested capital.  This math is undeniable.

As discussed previously (here), the probabilities of getting to top decile exit valuations aren’t different for companies that have raised $25M vs $100M – but the differential returns are considerable.  The corollary to this point is that high capital intensity late stage deals, with ingoing assumptions requiring $100M+ in fundraising, need to bet they’ll be well above top decile outcomes or they just don’t make sense to back.  That’s arguing against the gods of chance.

This $25-40M target sweet spot of equity is entirely feasible today given the dynamics of the ecosystem for both single asset plays or drug discovery platforms.  Both of these models of biotech can and have be done in a lean, capital efficient manner (as discussed here).  Stromedix only raised $32M before Biogen acquired them; Amira was in a similar ballpark before BMS picked them up.  Leveraging large Pharma balance sheets and their lower cost-of-capital is an important part of this equity capital efficiency through either less-dilutive R&D collaborations or risk-sharing earnout deals.

We think about this lifetime “equity capital” bogey before we start a new company or invest in an early round; knowing that the initial plans won’t survive contact with reality, spending and equity requirements are almost always much higher than a priori expectations.  An entrepreneur who initially thinks he will need to raise $60-80M to get to an exit is likely to require $120-160M if he’s lucky enough to get that far.  I’d rather back the lean entrepreneur who initially thinks $20M will do it, knowing we’ll be closer to $40M once time has passed.

Another implication is about capital titration and allocation.  By tranching your capital into a deal over time, as risk is discharged, you can assess where on the distribution curve a deal is tracking.  This is where dynamic allocation of reserves across deals can play a big role in enhancing the capital-adjusted returns of a fund.  If it looks like a deal is going sideways, then facilitating a median value exit might be appropriate Board and investor strategy.  If things are going well, this may be the time to inject more equity capital at higher valuations to support the growth of the story.  With a supportive public capital market, this is where they’d in theory step in.  We’re still waiting to see that happen with any real cadence, and in their absence focusing on the sweet spot for equity returns is more likely the winning strategy.

Importantly, we’re not in the business of shooting for medians.  We don’t push our biotechs for $100M exits at the start.  That isn’t in our investment memos.  We aim and aspire to create and fund innovative companies with top decile valuations supported by high impact medicines.  The median isn’t the goal.  But it can and often will be the endgame, so being conscious of the gravity of valuations while we’re building new companies is a key to overall portfolio returns.

In the past five years, more than a dozen companies have been funded under the premise that they can rationally design or screen for “cyclosporin-like” macrocyclic drugs to find uniquely differentiated therapeutics, including Aileron, Bicycle (where for full disclosure Atlas has invested and I’m on the Board), Ensemble, Peptidream, and Ra, among others.

In September, first SciBX Summit on Innovation in Drug Discovery & Development was convened to tackle the challenges that face the development of these macrocyclic drugs and “puzzling out the basic science underlying these molecules”.  They also wrote up an outstanding review of the Summit and space in last week’s SciBX (free pdf download here); for those interested in the space, this is a must-read.

Since they’ve already done a great job summarizing the four major challenges, I thought I’d just reiterate them and then move to the biggest question of all.  Here were the four major challenges that need to be overcome for the next wave of macrocycle R&D efforts to be successful, which were cited by the Summit and articulated in the SciBX piece: improving pharmacokinetics (serum half-life, tissue exposure); understanding cell permeability; achieving oral bioavailability; and understanding their binding modes.

These are serious challenges.  So why have (and why should) venture investors bother channeling capital towards macrocycle plays?  Because this modality could be big.  Macrocycles offer a number of advantages over bigger biologics, including better ligand efficiency, tissue penetration, delivery, and cost.  And versus conventional small molecules, they offer improved selectivity and larger binding modes (useful for PPIs), among others.  But this new wave of programs will only be big – and individual deals will only be successful – if their team’s figure out how to channel scarce resources into pursuing the right differentiated applications.  Like most drugs, the key will be to differentiate them from known modalities (and other macrocycles), and this will fundamentally depend on a target-by-target, disease-by-disease analysis.

This strategic target selection question is paramount for all new modalities: if you pick targets that are “well validated” in order to stay with “low risk” opportunities, you may win the battle but not the war: you prove the modality works but poured funding into a program that lacks lack significant differentiation or at the very least faces crowded classes.  If you pick targets that are too hard or too risky, then you could end up never proving that your platform is worth anything.  This target choice tension is evident in every early stage platform company that I’ve been a part of, and can be a huge value driver or value destroyer.

At Atlas, back in 2008 when we were initially evaluating the macrocycle class, we felt the “differentiable” target space was best approached with this matrix:

Obviously the most exciting applications would be for macrocycles that are orally active, cell penetrant drugs that open up totally novel, untapped targets.  But its also the most high risk, and the challenges, as highlighted by the Summit, make this unlikely to be low hanging fruit any time soon.  However, if it could be done, this quadrant would certainly deliver on the high cost of capital demanded by venture investments in the space: this is high risk, high return domain.

At the opposite corner are the injectable macrocycles that have to compete in the extracellular space.  It’s a high bar here, and while the risk is low, the opportunity set is constrained.  Just another injectable approach to the same set of well known serum proteases or cytokine targets is simply not that interesting, nor are most locally administered versions.  The only realistic way to achieve a venture return on these approaches is creatively demonstrating what is likely to be a narrower differentiation profile.  Here are a few: agonist approaches are tough to do selectively with other modalities and might work well here especially with shorter PK profiles; dual-acting macrocycles, engineered to hit two targets selectively, could open up more cost-effective and tunable bispecific functionality; lastly, conjugating toxic payloads onto macrocycles may prove more interesting than ADCs if they provide better tissue and tumor penetration, as well as clear stoichiometry, than bulky biologic approaches.  All these are theoretically differentiable profiles, but the killer experiments around these and other angles are being done now by many macrocycle startups.

The other two corners of the matrix are somewhere in between with regards to their degree of both risk and opportunity.  The top left quadrant offers access to truly novel target biology with an injectable drug, which is an interesting mix.  The reliance on an injection for delivery shouldn’t impair its product profile if the biology is truly distinctive.  In some ways, this is one of the quadrants with the best balance of risk and opportunity given the obstacle of high oral activity.

At the bottom right, orally-active compounds against more conventional extracellular targets is a tricky one.  Is simply making an orally available anti-TNF or anti-IL17 interesting enough?  If its as efficacious as a marketed mAb, then of course as it will also have big cost-of-goods advantages.  But is that a bet whose risk is dischargable early in development?  Probably not.  This is a big challenge for venture investors to wrestle through.  Oral active versions of the narrower differentiation profiles mentioned above (agonists, dual pharmacology,  conjugates) could be interesting, but gastric stability and adsorption may preclude those from working optimally and may conflate the risk profile.

In the long run, solving the challenges that the Summit laid out (PK, permeability, oral bioavailability, and binding modes) will be key to this class delivering the big therapeutic impact we all hope to witness.  But in the short-run, getting some early points on the board with other more modestly differentiated approaches will be important markers of success for the next wave of molecules around the macrocycle modality.  Getting the balance right between these short- and long-run product aspirations, and the investments required to get the field there, will be of critical importance.   Hopefully at least of few of the current crop of emerging startups in the space will successfully address these challenges.

This combination – clinically meaningful advance coupled with sound economics – is what those of us in the life science R&D funding ecosystem focus on every day.  On the venture side, figuring out how to generate returns from medical advances requires us to innovate both with regards to products and business models.

Pfizer’s recent acquisition of NextWave Pharmaceuticals sparked my interest in exploring this topic of innovation.  As many of you probably saw in October, Pfizer announced it was purchasing NextWave for $255M upfront and up to $425M in earnouts if sales milestones are met.

Founded in 2004 by Mahendra Shah in concert with Vivo Ventures and Sofinnova Ventures, NextWave was initially focused on developing more than 25 products leveraging various liquid and combination delivery technologies.  They raised a $10M Series A in 2005, a $40M Series B in 2007 led by Aisling Capital and Fidelity (which we looked at), a bridge round in 2010, and a $45M Series C in early 2011 led by Panorama and Bay City along with Kearny Venture Partners.  This recent financing also focused the company onto just a pair of lead programs rather than a large basket.

Kudos to all of those involved.  After ~$105M was invested over the past 8 years, the venture investors will likely see a 2-3x return upfront and a path to significantly more over time.  If some of those milestones pay out, this is certainly a top quartile deal outcome.  Beyond the interesting financials, I see three other pieces of good news worth highlighting from this deal.

First, it represents the successful exercise of an option-to-buy deal structure.  Pfizer put in a $20M option earlier in 2012, and upon approval exercised their pre-negotiated right to purchase the company.  Always good to see those deals play through the way we hope.

Second, and a related point, during the option period Pfizer collaborated extensively with NextWave to ensure a good outcome (approval with the right label) and an appropriate manufacturing plan was in place.  Witnessing successful big-small company partnerships is also a good thing, especially given how challenging joint efforts can be.

And third, this deal was done by Pfizer’s Established Products business unit.  Unlike the Primary Care and Specialty business units at Pfizer, this business is supposed to be where old products go to die on the tail of their lifecycle.  Yet this group has been remarkably creative in how they structured the deal with NextWave and worked to a great outcome.  Unexpectedly innovative business units and their strategies should be celebrated and rewarded.

But let me now share my broader concern and where my enthusiasm for the deal dims: why is Pharma spending hundreds of millions for a reformulated 50-year old molecule, methyphenidate, in a world where healthcare costs are spiraling out of control?  At a time when Pfizer is cutting its internal R&D budget and missing out on innovative BD deals (i.e., they aren’t one of the top buyers of innovation these days, or known as a preferred strategic partner in the biotech world), why are they going to fork out up to $700M for a liquid version of Ritalin?

To counter the criticism that I might be shortchanging the story, lets look at the market and the product, Quillivant XR.  Methylphenidate (Ritalin) went generic ages ago and is widely used to treat Attention Deficit Hyperactivity Disorder or ADHD in children and adults.  ADHD is a serious medical condition and I’m all in favor of diagnosing and treating it to help those who suffer with it.  In the Ritalin-related armamentarium for ADHD, there are pill forms for once-daily and twice-daily use.  There are chewable tablet forms for those kids that can’t swallow pills.  There are even liquid forms for those kids who prefer drinking fluids, albeit prior to Nextwave these were only twice-daily versions as far as I’m aware.

This is where Quillivant XR comes to play.  It’s a once-daily liquid formulation, based on innovative liquid delivery technology from Nextwave’s partner Tris Pharma, NextWave.  Its interesting technology for sure – attractive extended release profile, stable across food effects, less nighttime disruption when given in the morning, etc – and offers parents with kids who don’t like pills a nice incrementally improved alternative.

The big question is what should we – and we all are paying for it in our insurance premiums – pay for this type of product advance?  Is it worth a significant price premium over all the generic alternatives?  For a parent with a child who doesn’t like pills and wants convenience with effectiveness, a once-daily liquid would be great – especially if health insurance pays for the bulk of it.  But should this narrowly differentiated improvement demand a big premium over other ADHD alternatives, especially generics?

Apparently Pfizer must think so.  To justify the price tag Pfizer paid, they must have a forecast above $500M at peak sales.  And to get to these forecast numbers amidst a sea of generics and competing ADHD franchises (like Adderall), the price has to be an order of magnitude, or more likely two orders, higher than generic or even branded generic versions.

That’s hard to stomach at a time when the fiscal cliff of healthcare costs looms large.  It’s obviously great for Pfizer to get an approved product with $500M+ in top line revenues that they can drop into their sales rep’s bags, and their option-to-buy structure derisked the bet enormously for them.  So for each individual Pharma company, acquiring this type of incremental, marginal product may make sense to their P&L in the near term.  Wall Street will likely reward them.  But for the healthcare ecosystem, is this really worth it?  I’m doubtful.

The question I keep coming back to: where’s the commitment to real biomedical innovation – focusing on high impact, clinically-meaningful advances with unquestionably positive health economics?

A second order consequence of deals like this is that, reflecting the herd mentality of many venture investors, these deals channel more risk capital towards these less innovative, more incremental medical advances.  I’ve already heard VCs talking about “NextWave 2.0”.  I wish them luck, and hope they make money off the deals.  Returns will help the asset class as a whole.  But I’ll remain on the sideline on this “2.0” thesis, and focus on real innovation solving big medical challenges.

The NextWave flavor of specialty pharma reformulation play was in vogue back in 2005-2007, where lots of similar reformulation plays were funded (e.g., Zogenix, Alexxa, Alimera, Cogentus, Horizon, etc).   Atlas did some as well unfortunately.  But given the challenging returns in that space, I had hoped the model had run its course.  The number of truly early stage deals (see blog here) suggests that the recent tide is moving towards more meaningful medical innovation.  And fortunately, funding the latter type of play still pays dividends when done well: great returns have been achieved in deals like Avila, Stromedix, Amira, Plexxikon, BioVex, Intellikine, and many others focused on cutting-edge innovation.

In the capital markets, where companies and their projects compete for funding, I’m hopeful the winners in the long run will stand on the side of real innovation.

Fortunately, Adams Street Partners recently helped shed some light on this data gap with some interesting industry statistics, and I thought I’d call out a few points from their presentation.  As context, they assembled a robust dataset examining over 5600 fully realized exits from 1979-2011, including over 2000 since 1999.

Some observations:

• The best performing deals are not creating as much relative value as they did in the past.  The Top 10% of invested capital yielded 60% of the overall value created by venture between 1979-2011. In the most recent decade, these data suggest that the Top decile has only delivered about half as much of the value (about a third).  The top three deciles in this recent period generated about 70% of the total value, also well shy of the 90% or so over the longer time period.  However, despite differences in relative value created, the proportion of venture dollars yielding above 2x has been remarkably constant (~30%).  These data suggest that it’s not the lower frequency of winners in general, but the lower frequency of outsized winners, that has dampened returns in the asset class.
• VCs haven’t been able to channel disproportionate capital into winners: the top venture-backed performers on a dollar-weighted basis are similar to those on a deal-weighted basis.  The Top 3 best-returning deciles (Top 30%) on a dollar basis represent exits with return multiples above 2x.  This is true over three decades (1979-2011) as well as just the past decade (1999-2011).   The Top 10% is roughly above 4-5x returns, in line with data posted here previously that reflect deal-weighted metrics (here, via our Nature Biotech piece) and dollar-weighted metrics (here, Correlation Ventures).  The lack of difference between dollar- and deal-weighted metrics implies that the winners don’t consume a dramatically different amount of capital (smaller or larger) than the typical deal.

• VCs aren’t any better or worse in past decade than VCs were in the distant past at avoiding losers: the dollar-weighted loss rate hasn’t changed much.  Over both 10- and 30-year periods, share of dollars invested that go to losing deals has been roughly 55%.  The capital-weighted loss rate has been 45-46% over these periods.  This suggests that separating signal from noise (knowing what will work and what won’t) is hard when applying the traditional venture model of capitalizing startups: $10M Series A, $20M Series B, etc….  It also suggests that despite talk over the past decade about a “fail fast” model, few venture firms have done this successfully.    At Atlas, across both Technology and Life Science franchises, we believe a seed-led, measured approach to investing can help us channel capital towards winners more effectively: we provide startups with small amounts of seed capital (<<$1M) to derisk their theses, and we watch for early ‘signal’ around team, product/technology, and market traction.  We’re expecting our post-A-round capital-weighted loss rate to be far less than historic industry averages. Only time will tell.
• Life Sciences have a much lower loss ratio than most other sectors.  These new Adam’s Street Partners data (no chart unfortunately) also show that while over 30-years the aggregate biotech sector returns are similar to internet deals (~2x), the dollar-weighted loss ratio in biotech is significantly less: ~36% vs ~59%, respectively, over the past 30 years.  In English: biotech is far less lumpy than internet venture investing, far less dependent on the lottery ticket outlier.  Despite a perception of “binary” bets in biotech, these data suggest that losing money in other sectors happens at a much higher rate than biotech.  In practice this means that biotech investing is less random, where learnings can actually be applied to evolution of the investment model over time.  We apply this thesis in our early stage LS strategy: taking disciplined approaches to capital titration, shaping the DNA of these companies through active startup roles, focusing on repeatable models for how to create new companies, find value-added corporate partners, etc…  We aren’t just scratching off numbers on the MegaMillions ticket.  We think we can repeatedly build companies with >5x return potential, and improve our odds of getting to those outcomes.

Some might ask why I focus this blog so much on the aggregate data.  For me, these distribution curves define the asset class and help to frame the reality – or gravity – we need to work within.  The venture capital field is full of great anecdotes and sadly these shape lots of the media and LP interest in the space: Groupon and Zynga at 100x+ were great anecdotes.  I hope for their venture backers that they sold their shares a while ago otherwise much of those paper returns have evaporated.  Biotech doesn’t make for good anecdotes despite lots of interesting winners.  But one of my favorite quotes applies in spades to venture: “the plural of anecdote isn’t data.”

Informed by these historic curves, we can then ask ourselves the key question at the heart of venture capital going forward: how do we bend this distribution curve? We need a higher frequency of 5x+ winners.   We need to avoid this high rate of dollar-weighted losses.  As the business cliché goes, hope is not a strategy.  Hoping reality changes without changing the way we invest isn’t smart.  We need to push the venture model.

Using a reaction equilibrium metaphor, Fear and Greed are two reasonably stable, low-energy states for a Board, and which one is the preferred resting state depends on the makeup of the group.  But there’s typically just one, and it can obviously change over time.  Maintaining a Board where its balanced actually requires more free energy – like holding reaction in a transition state – and is rather unusual to do.

Fear is all about loss aversion – and since losses happen to about 40% of biotech investments, there’s plenty of fear in most boardrooms.  In biotech, the #1 reason biotech investors lose money is for their company to run out of it.  And running out of it can be the consequence of a drug candidate’s failure or a financing failure; the former leads to the latter, but the latter can occur even without bad data.  Fear of these losing outcomes often leads to a number of suboptimal choices, including over-capitalizing a company (trying to protect against the downside), cutting the burn so much as to cut corners and impair a program (doing an under-powered clinical study), or doing less-than-optimal business development deals.  A example of a “bad” deal: partnering a lead program just to get the cash to “feed” a biotech’s burn rate even though it was the best asset in the company.  That may be non-dilutive on the cap table, but its ultimately very dilutive.

Nothing pushes a pendulum faster toward fear than troublesome clinical data or unexplainable findings.  Often down-rounds of financing (raising money at lower prices) take advantage of fear from the existing investors – if they were bullish, why wouldn’t they do an insider round in the face of a challenging external financing.  Down-rounds also further contribute to the fear factor in a boardroom.

Greed in the boardroom typically manifests as an irrational optimism about the value of a program.  Strong clinical data clearly drives this kind of bullishness, but greed often pushes optimism beyond rational basis.  Does the data really support such a rosy view of value?  Often it doesn’t: Boardrooms where the pendulum has pushed to greed often ignore the deficiencies or challenges of a program and push forward without due caution. Greed is often about ‘keeping up the Jonses’ on the deal front: I’m sure the phrase “if Sirtris is worth $720M, aren’t we?” has been uttered in dozens of biotech boardrooms.  Of course, I love it when greed wins in the boardroom and great deals get done – its outliers like this define the best returns in venture.

But beyond the helpful optimism, greed can certainly lead to counterproductive decision-making.  For instance, Boards frequently debate the “walkaway” partnering or M&A deal value; nothing signals that greed has taken over the board more than a walkaway value a log-order higher than what the market is bidding on a program.  Maybe there’s a rational basis for ignoring the market and pushing ahead, as inefficiencies in the market can be opportunities for making outsized returns.  But too often greed’s bullishness becomes bullheadedness in the face of reality – leading to lost deals and impaired relationships.

Greed can often lead to similar suboptimal decision-making outcomes to fear, but for different reasons.  In fundraising, the same over-capitalized outcome can occur except the motivation is different.  Instead of worrying about loss aversion and raising more than you need, greed-induced overcapitalization is often driven wanting bigger warchests in the face of lots of interest.  Why not raise more?  Bigger funds eager to put more to work – so why not?  The answer is because raising more is excessive dilution, and dilution is the bane of returns.  Same goes for conducting the right clinical trial: greed (or more aptly over confidence) can lead to an irrational belief that our drug is so good its likely to turn an underpowered study into a big exit.  Raising exactly what you expect to need to do the right set of value-creating steps (no more, no less) is a tricky balance between fear and greed.

As states of nature, fear and greed themselves aren’t that interesting; they are just what you’d expect from human interactions.  But the most interesting moments occur when the pendulum swings between fear and greed: the down-round in a tough market followed a year or two later by a great M&A offer; getting into the clinic for the first time only to see a safety signal appear; receiving an FDA complete response letter in the midst of deal discussions; or the successful trial (or so we were told it was successful) that fails to generate any partner or financing interest.  It is moments like these when balance seems hard for many boards to maintain, and often you can feel the board’s emotions shift between greed and fear and vice versa in real time.  The tenor of the discussion just shifts and its palpable.

Great CEOs know how to manage their boards’ manic pendulum swings in good times and bad.  Sort of like enzymes, they stabilize the unfavorable transition state in between fear and greed – making it energetically less unstable and bringing some balance to a board’s emotional state.  This sort of maturity can be incredibly helpful in biotech.  “Do you fancy yourself as an enzyme?” might be added to my CEO interview list.

But are these recent IPO’s really signaling a change in public market buyside appetites?  It’s true that the biotech sector has performed very well this year in the public markets, and the IPO classes of late have outperformed in the aftermarket vs IPOs of other sectors (here).  All of us are hopeful that a true IPO window opens at attractive costs of capital – this would enable lots of companies to raise the significant financing required to grow into late development and commercialization.  But I’m not sure that’s happening – the early 90s model of a 2x up-round each step of the way from the Series A to the IPO is still not back in fashion.

That said, it’s clearly a good thing to see companies like Kythera and Intercept do well in their offerings – a combination of great assets, strong teams, and smart IPO pricing.  And I’m sure there will be more offerings like these over the next few quarters.

Unfortunately though, I think it would be an over-statement to conclude that these IPOs are “lifelines” to venture investors, as my friend Luke Timmerman has suggested (here).  IPOs in recent years have traded well in the aftermarket, but in aggregate they have not generated a great return for their VCs (due in part to challenging pricing).  And LP’s are wary of celebrating paper values of IPO stories only to see them trade down as illiquid stocks before the lock-up expires.

To better understand the relative attractiveness of the IPO market, and how their venture backers have done so far, I thought it would be illustrative to look at the biotech IPOs that have made it public since early July in greater depth: Durata, Hyperion, Regulus, Kythera, and Intercept.

At present, none of these five offerings reflect fantastic “homerun” returns, but there’s room for optimism around a few of them.  In aggregate, they’ve raised nearly $750M privately and in their IPO offerings, and are worth a combined market cap of $1.2B today. The average return for investors who bought into private rounds in these companies is 1.2x or so after 3-5 years (some lower, some higher).  By my estimate, none of these on paper are above a 3x for any of their venture backers at this point – so all remain below the top quartile from a current “exit value” perspective (though none of the VCs have sold shares or are likely to in the near term).  But these IPOs have provided significant funding for these companies to push their products forward, and hopefully see significant stock appreciate from here.

To highlight the nuances around the pricing and financing of these stories more clearly, I’ve taken a look at their share value over time.  These data are gleaned from their S1’s, so may be wrong (not always easy to calculate) and likely miss the potentially large impact of warrants and other conversions that provide lower implied cost-basis for some investments.

But the snapshot below shows the trends:

Few notes on each:

Durata: This was a deal with a heavy-weight syndicate from the start to support the company through its public offering.  They knew they’d need a lot of money to get dalbavancin approved and the syndicate funded it each step of the way at a pre-defined $8.00 a share for $78M.  The IPO was at a slight step-up, and involved insider participation as expected.  VCs in the deal are now sitting at a 1.2x (20% up) on their investment from 2009.  Despite the lack of paper returns, they are now well-financed to get dalba across the goal line and see the stock appreciate over time.

Hyperion: A tougher story than others in this IPO lineup, with over $110M raised prior to the IPO and real FDA challenges.  The original $22M into the Series A and B rounds of 2007-2008 were washed out to zero essentially in the $60M recap of the large Series C.  Then a series of bridges in 2011-2012, convertible into shares at the IPO price, were put in place for $32M.  The extra burn was almost entirely due to the FDA issues around the SPA on the lead program (so much for late stage deals being less risky).  This makes the implied private share cost basis for the VCs in the deal from the start above $12+ per share, even with the warrant coverage of the bridges.  They were able to dollar-cost average downward with their insider IPO purchases.  So at today’s share price, this one feels like a paper loss right now; but like Durata, its now well financed to be able to turn the card and create value from the program.  Onward and upward.

Regulus:  This is a unique one.  Not only is it the sole preclinical stage company in the mix, it also is the only IPO (in a long time) to have never been priced by institutional investors as all the shares sold privately went to its parents, Isis and Alnylam, or its strategic partners.  In fact, the participation of its partners in the IPO was considerable and certainly helped in getting this early stage company out into the public markets (kudos to them on using smart BD to anchor the IPO).  On pricing, it was tough: the IPO valuation was at the same share price as the Series A round back in 2008, and Sanofi’s shares are underwater.  I’m a big fan of microRNA biology so am hopeful that they (and others in the field, like Miragen and Mirna) help deliver great therapies and real value from the approach.

Intercept: Another relatively unusual deal in that Intercept raised most of its capital from the Genextra SpA group: both its Series A and B rounds (~$50M) were committed by them at ~$10 per share, according to the S1.  Their position is now sitting close to a 2x.  The recent Series C was at ~$11.50 in August 2012, where Orbimed joined the syndicate.  Based on the current price, Orbimed is already sitting at 1.7x (up 70%) in a few months.  New Leaf looks like they also bought into the IPO according to the S1; already sitting at 30% uptick in a handful of days, so looking very smart right now.

Kythera: This aesthetic medicine company is probably the most interesting IPO of the group from a venture perspective.  It began life as AestheRx in 2005 with a $900K seed investment from Versant, backing CEO Keith Leonard and the founding team.  It raised a Series B less than year later at ~$7 per share with ARCH and Prospect.  Both a Series C and D followed, at ~$13 per share, where the existing investors and others committed the bulk of the private financing for the company – they’ve raised nearly $110M privately to drive their lead dermal fat reduction therapy.  Based on the ownerships in the S1, the average cost basis for Versant is probably around $8 or so.  At the IPO price of $16, and the first day close near $19, this looks like a solid 2.5x return for Versant at the current valuation.  Congrats to them for a solid investment here.   If the team continues to deliver, this could be a significant winner over time.

Time will tell if these five IPO’s shares will appreciate in value in the public markets, which is obviously the hoped for plan for the venture investors.  As of right now the jury is out on whether these are really worth celebrating with LP’s as homeruns or even “lifelines” for raising funds.  But several have lots of promise.  The reality is the top decile of venture exits in today’s environment, or north of 5x returns on invested capital, continue to be had at a higher frequency from M&A deals with Pharma than from going public.  It will be nice if this changes, but for structural reasons this is unlikely.

Its not that the public markets are tough places to make money: its fair to say that the public markets, and PIPES into beaten up stories in particular, have generated sizeable returns of late (e.g., witness Sarepto’s 12x swing upwards in 90 days, Jazz’s 50x return from the spring of 2009); unfortunately, these aren’t typically coming from recent IPO stories.

But maybe they will soon.  I’m hopeful these and other recent IPOs will bring life back into this part of the market.  But hope isn’t a strategy, so I’ll keep focusing on finding acquirers for the portfolio in the meantime.

Instead of taking a “birth” vintage approach to returns, which is the way venture fund vintages and most benchmarking analyses tend to cut the data, Correlation Ventures’ team has chosen to look at all the realized exits and their returns in any given year.  This data is based on financings (e.g., Series A and B rounds in the same company are two data points), and includes all prior financings into companies that went out of business, got sold, or got public (IPO) in that year. They are also dollar-weighted (e.g., for the top decile, or 90^th percentile, 10% of all the dollar’s invested in financings of companies that exited in that year were above the realized multiple). Since 2003, each year represents 500-700 exits and over 2000 financings – so it’s a very robust dataset.

Three observations worth calling out from this – though all are obvious from the chart:

First, top decile venture financings yield attractive return multiples.  A top 10% outcome has delivered above ~4x over the last 5 years, aggregating all sectors of venture capital. For a sense of proportion, this represents the top 200 or so financings into companies that exited that year (ranging between 50-100 exited companies per year). This is very much in line with data Bijan Salehizadeh and I published in Nature Biotech last year: top decile outcome was roughly 4x for exited companies first invested in during the last decade (with a slight edge to Life Science vs Technology top decile deals). Although not in the chart above, Correlation’s database estimates Top 1% of financings (5-10 top exits per year) have been in the 15x range for the past five years; these are much more likely to be Technology deals. This also implies that the ‘halo’ lottery-ticket deals in the 100x+ range are in the Top 0.25% or less (top 1-2 deals out of 700+ exits in any given year). Its hard to base an investing strategy on being in the top 0.25% of anything. This is similar to data flying around Twitter earlier this month (here).

Second, venture-backed exits have been getting better over the decade.  This is obviously good news for those of us in venture and in the startup ecosystem.  Returns have rather steadily moved upward across the 2000s, with a minor correction in the 2008-2009 timeframe.  Top decile returns nearly reached 5x in 2011, and the average exit is now above 2x.  This trend clearly bodes well for venture capital.

Third, the big albatross on returns has been the post-bubble crisis.  The horrific average and top decile returns in the exit vintages of 2001-2003 represent lots of the companies that got flushed out when the bubble burst.  Many of these raised capital at the high equity valuations and when the lofty markets collapsed, they fell to the ground hard.  But lots of companies from that vintage survived, albeit wounded, and many entrepreneurs and their venture backers spent much of the decade trying to recoup their equity values for their LP’s.  Part of the positive trend in the chart above is that companies born in those challenging bubble years have been exiting and now represent a smaller proportion of the exit universe.

Given how little transparency there is on venture capital returns, its nice to see some additional data on the topic – so thanks to Correlation Ventures for letting me share this material.

For those that don’t know them, Correlation has a unique data-rich investment approach: applying predictive analytics to the likelihood of a deal being successful, and using that knowledge to inform rapid co-investment decisions around new financings.  They’ve built a robust database to support these analytics, and have made ~30 investments to date across a range of sectors.  Definitely worth checking them out.  I’m working with them on some additional analyses that should shed more light on sector return profiles and other topics.  More to come.

One of the earliest blogs here at LifeSciVC was on the dirty secret that more than half of academic work couldn’t be replicated in an industrial setting, and how it shaped the way we view starting new companies as venture investors.  It got the attention of BioCentury/SciBX as well as the Wall Street Journal.

Later in 2011, some real data was added to strengthen the case: a Bayer Healthcare team published work showing that only 25% of the academic studies they examined could be replicated (Prinz et al. Nat. Rev. Drug Discov. 10, 712, 2011).  And then earlier this year, Glenn Begley (formerly Amgen) and Lee Ellis (MDACC) showed that of 53  “landmark” oncology studies from 2001-2011, each highlighting big new apparent advances in the field, only 11% (only 6!) could be robustly replicated in work done at Amgen (Begley & Ellis Nature 483, 531–533, 2012).  Adding insult to injury, the number of citations for the unreproducible findings actually outpaced those with reproducible findings according to the Amgen work: averaging 248 vs 231 citations, respectively, for papers in high impact factor journals and an even more astonishing 169 vs 13 citations for papers from other journals.

These are frightening statistics for an industry predicated upon building on the prior work of others and the integrity of peer review for sorting the good from the bad.

As we think about starting new companies, initiating drug discovery campaigns, or even launching clinical studies, how do we deal with this issue?

First and foremost, we need to get better at assessing the scientific literature, and all of us involved in translational medicine need to hold ourselves to a higher standard – including investigators, their institutions, journal editors, grant-funding bodies, VCs, Pharma, etc…  From an industrial perspective, we should also do better diligence – assessing science with better filters about what’s robust and what’s not.

To that end, Glenn Begley has some up with a great list of rules for what makes a robust, high quality paper with the hallmarks of reproducibilty, based on their review of scores of papers.  Like Lipinski’s Rule of Five for predicting oral activity of small molecule drugs, I’d like to propose calling these Begley’s Six Rules for Reproducibility:

1) Were studies blinded?

2) Were all results shown?

3) Were experiments repeated?

4) Were positive and negative controls shown?

5) Were reagents validated?

6) Were the statistical tests appropriate?

Lets take each one of these in turn.

1) Most studies aren’t blinded with experimental and control arms.  Furthermore, by my estimate, less than 20% of Methods sections even mention whether the work was blinded to prevent experimenter bias, and in most cases the blinding methodology isn’t included.

2) Results from multiple studies are rarely shown in the same paper, as its usually only the “representative” example figure (read = best single result).  Outliers often disappear from figures  (e.g., telltale sign are n’s differing randomly between arms).  Many western and northern blots show only a computer-generated slice of the gel, without size markers.  Its also often unclear if the exposures were in the linear range of the staining.

3) N-of-1 experiments are sadly fairly commonplace in the literature.  Assays often don’t have replicate values included.  Nor are aggregate n’s often used.  Its true that some long term animal models are a chore to do multiple times, and often critical reagents are expensive – but repeating studies before publishing should be the bar.

4) The use of both positive and negative controls to benchmark an experimental system is frequently not done.  In fairness, with a novel model, there might not be a positive control.  But if there is, it should be included and described.  Selection of the right controls is also an issue: e.g., when studying the role of a single kinase in a disease, a promiscuous dirty kinase inhibitor that happens to hit the target of interest is probably not a great control.

5) Validated reagents are essential to draw robust conclusions.  Unfortunately, Begley and his colleagues found this to be frequently overlooked, especially the strength of immunohistochemistry probes and western antibodies (e.g., species cross-reactivity).  Authors should highlight where the validated reagents were obtained.

6) Statistics is a big gap for most papers.  Proper powering of animal studies with a pre-agreed stat plan is a rarity.  Showing n’s and SEM bars in figures is important.  Also, what’s the right p-value to use; for instance, p-values of 0.05 aren’t relevant to post hoc analysis hunting for signals on a chip.

These Six Rules are good guidelines for those of us in the business of finding and commercializing the next cutting edge science.  Thinking about these during diligence around an investigator’s work will undoubtedly improve the outcome of academic-to-industry translational efforts.  Furthermore, Tech Transfer Offices should hold these Rules up when they are working on invention disclosures and external outreach for the work.  Lastly, more CROs should track the literature and propose to do reproducibility work in line with these Rules for high impact science out of top tier academic centers; my guess is many academic institutions would support those studies.

Importantly, adherence to these Rules won’t make reproducible translation work 100% of the time.  Fundamentally, there are “language” differences between academic and industrial work.  The often used phrase “safe and well-tolerated” in an academic animal study means the animals didn’t look sick nor did they die.  But it doesn’t mean that even gross organ pathology was ruled out, much less full histopathology, chemistry and blood counts, liver enzyme levels, etc…  This language difference is an important factor in translation, but is much more nuanced than Begley’s Six Rules and needs to be considered in any academic-to-industry transfer.

The entire ‘system’ of biomedical research needs to change in order to address these issues and raise the bar.  Grant funding bodies need to demand it, as do journals.  Investigators should want to do it and be rewarded for their work’s robustness.  As one effort aimed at addressing this systematic issue, the Science Exchange’s Reproducibility Initiative launched earlier last month and I’m honored to be on their Advisory Board.  The Initiative aims to provide “both a mechanism for scientists to independently replicate their findings and a reward for doing so”.  It’s received lots of press in Science, Nature Biotech, Slate, BioCentury, and others.  I’m hopeful that it will help create the momentum needed to address this troubling issue.

But, as a parting remark, let’s not forget that this is all about cutting-edge science.  There will always be studies that can’t get repeated – that’s part of the iterative nature of the scientific method of articulating and challenging hypotheses.  But as a system we can’t continue to tolerate ‘hit rates’ of reproducibility below 50% from academic scientific literature, especially from top tier journals and biomedical institutions.

New models come up all the time because we’re an industry in the midst of dynamic change and adaption: changing (and shrinking) mix of venture funds, increasing role of corporate venture efforts, largely inaccessible IPO markets, LP (mis-)perceptions about Life Science returns vs Tech affecting asset class allocations, fund size math pressures, the capital efficiency imperative, etc…  Many of these topics have been discussed before in this blog, but I thought I’d revisit the “new model” subject to provide some practical points and examples.

But before that – some context: generally speaking, there are two conventional models for biopharma companies: either drug discovery platforms or product-focused plays.  Broadly, these two approaches were reviewed in a prior post on “worldviews” of biotech that framed up some of the key success factors.

To recap some of the challenges with those, it’s clear that both have struggled with industry-wide issues around capital intensity (burn rates) amidst stubbornly high costs of capital.  This can only be managed through tighter titration of capital, focused on derisking the story, while seeking non- or less-dilutive sources of funding to take it forward (like accessing big balance sheets of larger partners).

But there are also some very specific challenges to each model.  Drug discovery platforms require significant money and time to build, but are frequently only valued for the lead products that they produce and only after they’ve made it into the clinic; in a world where exits are only M&A driven, this greatly handicaps the value of the platform itself, delays time to liquidity, and hence diminish returns.  Product plays have typically been roll-ups or spec pharma stories, where portfolios are accumulated under the pretense of product diversification and efficient use of a “management’s bandwidth” – but the crispness of a tight product thesis is often eroded, suboptimal assets consume resources (lemons being outlicensed by Pharma), and the eventual value of the story focuses on a single lead program over time.

These challenges are in large part due to a singularly company-centric view of investing, building, and realizing value that remains commonplace today.  But by taking an asset-centric view of both types of models, a number of these concerns can be alleviated.

Here are a few practical thoughts on two asset-centric corporate structure models to platform and product plays.

Drug Discovery Platforms: LLC Holding Company Model.

Instead of building drug discovery plays as singular entities (simple C-corp), we believe that taking a more modular approach to discovery engines through the creation of LLC holding company structures can optimize value.  The figure below captures the set up:

Here’re the key elements to the structure: The LLC itself is passive and non-operating.  This makes for a more tax efficient model for institutional investors (a straight LLC can work well if individual investors are involved rather than VCs).  The team and the technology platform itself are housed in a subsidiary C-corp (“Mgmt Corp”), which sets up all the Master Services Agreements and such.  Each drug program, right before generating valuable IP, should be moved into its own C-corp subsidiary of the LLC – the data packages and IP related to a program are all housed in their own entities.  A variant of this subsidiary model is to put several related programs into a single subsidiary – assuming that the “exit path” for that basket of programs is the same then this makes sense.

Liquidity can be driven through all the traditional manners (M&A of the whole thing, or an IPO), but the real virtue of the model is that single assets can be monetized to create returns without losing the platform: as a program matures and garners Pharma interest, that single subsidiary can be acquired by the partner in a risk-sharing, earnout-based structure.  The payments flow into the LLC as capital gains (since it was the purchase of equity), and can be passed through to the shareholders of the LLC.  The Board of the LLC can determine how much gets distributed vs recycled into new drug discovery operations.

Compare this LLC return profile to a traditional drug discovery company with the illustrative chart below.  The dotted line below tracks the relatively flat valuation typically given to a drug discovery biotech over its Series A/B/C rounds; then, like a hockey stick, if things go well it exits at a great multiple.  This illustration is realistic: both Avila and Adnexus had flat/slightly down Series B rounds only a couple years prior to their big exits.  Now take the green line, illustrative of the LLC model: each asset sale incrementally improves the ROI, and “smooths” the curve over time as deals around Dev Can or IND mature more quickly than waiting until mid-clinic results to catalyze a big exit.  So the aggregate return curve and flow of capital looks far more interesting.

More importantly, this model also enables the potential for an ‘evergreen’ biotech model – since you don’t have to sell the entire company to generate returns, and you don’t need to take a company public to get liquid, this LLC holding company model enables you to invest, build, and harvest returns out of a drug discovery engine over much longer time frames.  It’s easy to argue for “going longer” when you’re sitting on healthy return already.  This feature certainly has real appeal to many management teams that want to build a lasting company in the current ecosystem.

It’s worth noting that nothing is free, so this structure costs something.  Care must be taken to track the flow of funds for accounting reasons across these entities – which requires more sophisticated finance expertise.  You also need more lawyer time to get this structure in place and maintain the appropriate governance.  But these costs are dwarfed by the potential benefits.

This model is more than just a good thing to talk about at conferences: at this point, all of our recent discovery plays at Atlas have adopted this model.  We pioneered the structure with Nimbus Discovery LLC in 2009, launched RaNA Therapeutics LLC in 2011, and stealth-stage Boreal Therapeutics LLC in 2012.  Other drug discovery plays like Forma and Viamet (non-Atlas deals) have been convinced of the value of this LLC approach, and have reorganized from C-corps into LLC’s due to the clear advantages of the structure.  I suspect that most drug discovery plays will move towards this model over time.

Product Plays: Single Asset Project Financings with Structured Buyouts. 

Single asset project financings aren’t new news; we and others have been investing in these types of deals for years (e.g., Stromedix, Zafgen, Infacare, and others).  By creating a portfolio of these approaches we are able to balance the binary risk profile of these companies.

A more recent trend shaping our view of project financing models has been Pharma’s increasing interest in finding creative P&L-sparing means to fund more R&D efforts externally, especially in the context of shrinking internal budgets.  This dynamic, coupled with the increasing sophistication and specialization of the CRO ecosystem, has enabled the creation of single-asset deals with defined liquidity paths via structured buyouts.

Here’s the “simple” description of how these deals get done: Pharma has a handful of assets it believes are interesting but can’t be funded due to budget constraints.  These assets are typically not on the out-licensing list per se, because the Pharma is interested enough to want them back if they work; most pure out-licensing plays are sour lemons (though many biotechs have made lemonade).  The Pharma then opens up its kimono and shares the profiles of these interesting assets and their development plans with us.  We do our diligence on those, share an “open market” view of the strengths/weaknesses (which most Pharma’s have found valuable), and then identify one or two we’d like to push forward on.  We typically bring in a team of experts to rework the development plan with a lean, fast-to-PoC project-financing mindset.  At that point, assuming we like the way its coming together, we negotiate a pre-defined exit path for a possible investment: upon successful completion of XYZ clinical study, the Pharma will buyout the asset-centric entity in an upfront plus milestones type transaction.  For Atlas to consider a deal, the upfront plus near-term milestones needs to generate the potential for a top decile outcome (>5x); longer term milestones and royalties help secure outlier potential if the drug makes it to market.  We also need to customize the deal’s more nuanced attributes: the accounting and tax structure of the deal, the nature of the IP transfer (or not) to the new entity, what to do in different “grey” outcomes, timing of specific financings, etc…   Then the final proposal with all the juicy details needs to be approved by the right internal decision-making committees at the Pharma partner.  All this takes a lot of time, lots of time, and with lots of potential for frictional loss.  But, once completed, we think they make for very interesting deals for both Pharma and for Atlas.

We set up our Atlas Venture Development Corp (AVDC) initiative to push these types of deals forward back in 2010, and our Shire Alliance in 2012 is a variant of this theme.  Several other groups have recently announced efforts to also seek deals of this type, including CMEA’s Velocity initiative.  I’m not sure how many have been done in total across the industry, but to date we’ve closed two such transactions.

Arteaus is working with Eli Lilly on an anti-CGRP antibody for migraine prevention, and is in the midst of its clinical proof-of-concept study.  Upon successful completion of that study, Lilly has the right to acquire Arteaus in order to pursue the antibody’s further development.  This project raised an $18M Series A financing to complete Phase 1a, 1b and 2a; OrbiMed joined us in the financing and things have gone well so far.

More recently, Annovation struck a similar structured deal with Medicine’s Company (MDCO) around a novel anesthesia asset.  In this case though, we actually sourced the asset from Partners and during the seed funding secured the relationship with MDCO as a natural buyer.

Since starting AVDC, we’ve reviewed well over 200 opportunities and only closed two.  Our Shire Alliance has reviewed over 100 assets, and is actively evaluating a few maturing opportunities; we’re hopeful we’ll have more to announce by yearend.  Reality is that it’s not easy to get these deals done – patience and creative problem solving are some of the most important ingredients.

But when done well we think they address a number of the issues that plague product-focused biotech plays – they offer a pre-defined and tightly-titrated capital intensity, a virtual development efficiency, and a clear path to liquidity in a reasonable timeframe.

Looking forward to discussing these and other “new models” at BioPharm America on Wednesday.

SRS has assembled an intriguing dataset of Life Science deals which offers insight into the metrics around M&A-related earnouts, especially as a comparison to an older dataset I’ve posted about in the past (here).  This data involves 47 undisclosed private-target pharma and medtech deals, including 39 with earnout payments (representing some 200+ discrete milestones).  These include nearly $9B upfront payments and $7B in potential milestones.  All the deals in this set are rather young, having closed from 3Q 2008-2Q 2012, with a bias towards more recent closings (in line with the growth of SRS’ business).  It’s a dataset that broadly mirrors the sector’s value distribution: 27% of deals had upfronts worth $250M or more, which is in line with other statistics for deals of this valuation range (here).

Although many of these aren’t new insights, here are some of the key observations gleaned from the data:

• LS deals have much higher frequency of earnouts than other sectors SRS tracks: 83% vs 15% of M&A deals.  This isn’t “new news” given the well discussed trend toward risk-sharing deals in LS, but the quantification of the stark contrast to other sectors is still eye-opening.
• Unsurprising to readers of this blog, the capital invested doesn’t correlate with bigger exit returns; in fact the 7 largest deals on an upfront multiple basis all raised a below average amount of equity capital vs the rest of the dataset (<$60M).  Further, the range in upfront return multiples is huge and renders averages useless: deals went from 0.1x up to 16x upfront.
• LS investors and boards are far more generous with regard to vesting for their employees at deal closing than boards in others sectors.  Boards voted to fully accelerate the vesting of options for 82% of LS deals vs only 24% in other sectors.  I’m not sure why this is the case; one speculation is that the more generous acceleration practice may be a reflection of the fact that most tech entrepreneurs have far higher equity ownership to begin with (here).
• Biopharma deals were faster to exit than the Medtech deals in this dataset: median of 6.5 years vs 8.0 years.  Unclear whether the Medtech future milestones are of shorter or longer duration than the BioPharma payments.
• Post-acquisition diligence requirements vary across the board, but the majority of development-stage milestones require only the “commercially reasonable efforts” standard.  Further, only 8% of deals managed to craft language around clawing back a program if the buyer didn’t advance it.  I know this gets complicated, but it’s a shame not to see more of this given portfolio shuffling that goes on in Pharma (where good assets can die through neglect).
• Lastly, the achievement and payout of near-term milestones is reasonably high to date.  In this rather young dataset, only 25 milestones had come “due” in SRS’s assessment.  Of those, 36% were met and paid (reflecting 30% of the dollars in this subset), and another 16% were delayed but likely to be paid.  About third had missed, and the remainder were delayed and unlikely to be met.  This distribution implies that nearly 50% of near-term milestones have or are likely to be paid.  SRS also quantifies the value of near term milestones; it appears that roughly 50% of milestone values are within the first four years, and for most deals they can add another 2-3x multiple with those near term milestones.  Since upfronts are often in the 2x range, this additional return in the near term payments can propel a deal into a top decile outcome (>5x is top decile for any venture deal, regardless of sector).

Earlier this year, this blog (here) and Alex Lash at Startup (here) published on earnouts with a more mature set of seals (Jan 2005-Dec 2009 deals in this blog’s set), and concluded that 25% of milestone values had already been paid out.  While those analyses are a different cut and sample set than SRS’ data (which looks only at milestones that should have been met by now), the SRS findings in the last bullet above are reasonably consistent with our prior findings that a meaningful percentage of milestones are getting paid.

Now that more of these earnout deals are maturing, it will be interesting to see how SRS’ database evolves; it’s likely to become one of the best data-rich sources of this milestone payment information.

Two further comments more generally about milestones and implications for our sector:

• VC firms need to figure out how to standardize the accounting of these earnout payments in our LP reporting.  Since a reasonable percentage of the early milestones are statistically likely to pay out, these can reflect a significant amount of value.  In the past, most firms reported the future value as zero, which clearly undervalues them.  We now build expected value models based on industry attrition rates and the time value of money, and include the parent value of the future payments on our books.  It’s never very much, but it’s only appropriate to determine the fair value of these financial instruments.  At Avila, we’ve valued the future earnouts at 10% of their nominal value – which is about a full 1x on the deal.  Its clear that venture firms are all over the map here in how they do this; I think our industry and sector needs to set some standards for these approaches.
• It’s only a matter of time before dedicated biotech milestone-buying funds get raised to monetize these payment streams, especially since defined earnouts have a certain probability of being met and can be modeled off both attrition data and historic patterns.  Royalty funds emerged after the R&D vehicles of the 80s and 90s created a lot of royalty streams; there are now a bunch of these funds buying actual and synthetic royalties.  With ~$12B+ in earnouts floating out there in deals since 2005, there’s plenty of substrate for new dedicated funds to try to monetize.  The pressure to create a market for these financial instruments is only going to increase as the 10-12 year venture fund life approaches with significant milestones remaining out in the future.  If not new dedicated funds, it seems to me that secondary funds which buyout a venture find’s stakes may be in a good position to pick these future streams up.

Milestones aren’t going away any time soon, and are likely to contribute in a meaningful way to the returns of our sector over the next decade.  Thanks to SRS for sharing this useful data, and look forward to updates over time.  As an industry, we’re going to need to build the right processes, standards, and markets for taking these future milestones into consideration.

To help the startup entrepreneur think through their SAB, I’ve polled a few of our portfolio exec’s* about what works, and combined it with my own experiences.  This post is the synthesis of that – highlighting the Pros, Cons, and Best Practices for early stage biotech SABs.

Pros – Reasons for having an SAB

• Good SABs can provide a young biotech with objective, external perspective and quality authentication.  This can involve constructive criticism and helpful problem-solving, and typically helps provide an “open market” view of where the biotech should focus going forward.
• They can raise the profile and awareness of a startup with thought leaders in the scientific community.  For raw startups, this is sometimes just “window-dressing” for a story, especially with “Hollywood-star” SAB members, but a great lineup can provide a meaningful stamp of approval to the external world, including current or potential investors.
• Quality SABs can involve real data-sharing with collaborators, and provide a conduit for exchange of cutting-edge ideas.  They can also be a good forum for sharing public, non-confidential information about competitive programs and how one’s efforts stack up relative to others
• They can be fun meetings – appropriately crafted agenda’s leave time for far-reaching scientific questions, more than just the next 3-6-9 months of work.

Cons – The real costs of a poorly used SAB

• SABs don’t come for free.  Good ones are a lot of work.  And they are also costly for a startup in terms of dollars and equity.  A typical SAB member may charge $2.5-5K per day and receive 0.1-0.3% equity.  Hollywood-star advisors can be much, much more expensive than that on the equity front.
• SABs, especially larger ones, have a high degree of entropy pent up in them.  Lots of potential opinions can lead to a herding of cats during the meeting.  Furthermore, a strong negative point of view from an influential SAB member can derail programs, especially with investors or Directors listening in.
• Pet projects of SAB can create challenges.  If your SAB is composed of collaborators, where sponsored research may be a part of the relationship, it can become very difficult to kill “pet” projects affiliated with that SAB member.  This can lead to unpleasant meeting dynamics.
• Promiscuous SAB members may be too involved in other roles to be focused and constructive, and often aren’t as tight-lipped about confidential information as one would hope.

All in all, the Pros clearly outweigh the Cons for most companies.  But enlisting the help of an SAB isn’t simple and requires a good deal of forethought about the why, what, and how of putting them together.

Best Practices for a Productive SAB. 

Here are a few areas where our experience suggests some general guidelines.

• Why does your biotech need an SAB?
  • It’s important to define the reason you want to have an SAB and build it for that purpose.  This obviously changes over time from pre-funding days to drug discovery to clinical development, and its important that your SAB mature along with the company.  Is the SAB primarily for external validation (raw startup), or is it for specific guidance on a portfolio or project (maturing into development)?  Is it something where discussion in a broader group is going to be helpful to getting to the right answer (big science questions), or can you do it via one-on-one dialogues (programmatic input)?  In my experience and that of many of our portfolio executives, much of the value from an SAB comes during the informal, regular contacts in ad hoc discussions, not during the SAB meeting itself.
• What should your SAB look like?
  • In general, its good advice to keep it small.   I think 3-5 core members are best, and for each SAB meeting invite ad hoc members with particular expertise to bear on the questions being asked.  To keep it fresh, I’d advocate 1-2 year renewable “terms of service” of your core members, so you can sign them up again if they are adding value.
  • Avoid the “Hollywood SAB” of Nobel Prize-winning scientists who haven’t done much in years or don’t have the time to engage thoughtfully.  They look good on paper, but often don’t add much with regard to shaping the story, but certainly do add on the costs.  They may help attract investors, but unless the Advisor is fully up to speed on your work it could also dampen interest upon deeper diligence: having a disconnected “Hollywood SAB-member-in-name-only” can hurt rather than help.  Nothing is worse than hearing a purported SAB member say during a diligence call “I haven’t heard from them in months” or “I haven’t actually reviewed the data”.
  • Avoid members who consume too much oxygen in the room.  Its more than just annoying when you have a blowhard who talks too much, often about subjects they actually don’t know anything about, and who doesn’t play well with others.  We’ve all seen this happen.
  • Try to mix up the backgrounds and experiences of the group.  The best sessions I’ve attended tend to have a mix of cutting-edge academics and industry veterans, in order to bring orthogonal perspectives.  Most startups default to a string of academics and miss out on the veteran drug hunter or development perspective.  Have your Founders help with the selection, but not so much as to limit diversity of thought.  Also, get out of the local bias – e.g., Boston startups should think about SAB members from other parts of the country if not abroad.  With all of the career transitions out of Big Pharma lately, there are lots of talented folks available who are willing to advise.
• How to conduct an SAB meeting:
  • It sounds obvious, but send the materials in advance.  Most companies always say they will, and then they still pile on the slides a day or so beforehand.  I’d suggest sending papers/pre-reading at least a week if not 3 weeks in advance.  Importantly, the materials should include the key questions with which you want input from the SAB.
  • Actively managing the meeting is a must.  As mentioned above, these are tests of your ability to herd cats.  Managing the time and cadence of the dialogue, pulling out of the inevitable rabbit holes in the dialogue, and getting the right balance of presentation and discussion is key.
  • Require your SAB to work for their compensation.  They shouldn’t just plan on sitting down, listening, and occasionally pontificaitng.  Make them actively contribute by putting them on the agenda to give presentations or 10+ minute perspectives on an aspect of the field.  Facilitate the gathering of specific feedback to the key questions during the meeting, and the sharing what your SAB sees out there in the market on the competitive front. Lastly, make them review the pre-read material by jumping right into their feedback on some aspect of that material.  An engaged, up-to-speed SAB is critical to having a great dialogue.
  • Include your team in the SAB!  Nothing motivates and fires up the juices of the scientific team than seeing SAB luminaries and industry veterans debating the merits (or demerits) of their scientific programs.  And its helpful to have folks with real data on the tips of their tongues in case the tough question gets asked.

SABs can be quite valuable to startup biotechs if constructed and conducted in a thoughtful manner.  After the SAB, its important to follow-up with minutes and a set of management conclusions – why you did or did not accept the consensus view of the SAB, what changes were implemented, etc…  Maintaining frequent, open dialogue with an SAB via email or informal one-on-one’s is also helpful.

Beyond just these fledgling biotechs, bigger Pharma companies can benefit from having an external SAB aimed at bringing much-needed open market, objective input into the quality and robustness of internal programs.  I’ve been serving on the UCB Pharma Research SAB for the past few years, and find that the quality of the meeting and dialogue is exceptionally good.  Its certainly one of the most productive SABs I’ve witnessed (and they follow most of the Best Practices above), and I’d like to believe that UCB finds the external, objective input into their internal Research programs helpful.

Seems to me that more organizations, big and small, would benefit from having greater external “open market” objectivity injected into their portfolios through constructive SABs.

* Mike Gilman of Stromedix (Biogen), Tom Hughes of Zafgen, and Bill Marshall of Miragen, along with my Atlas colleagues, helped contribute ideas for this post.

Unfortunately, as this blog has sounded off about frequently, these beliefs are in direct contrast to reality: the bane of returns in biotech has been its equity capital intensity.

Raising lots of money is not indicative at all of successfully generating a return for shareholders, but as a sector we tend to celebrate it.  I’ve vented on this subject before, highlighting our tendency to cheer for trophy financings (here); the first company I highlighted as potentially challenged in that post was Pacific Biosciences.  About 1.5 years later, its the poster child of “Go Big, Go Bust” stories, having raised ~$600M over the past five years and now has an enterprise value of -33M.

And raising a lot of money is not always indicative of a management team that’s doing what it’s supposed to: working on behalf of their shareholders.  When your option pool and ownership are always reloaded to the latest Radford compensation survey benchmark, you’re indifferent to dilution and of course want to raise more money.  BD candidates that brag to me about having raised $100M for XYZ Biotech, without big returns to back it up, instantly lose points in my book (as my brain is thinking, “how come they didn’t do more with less equity capital?”).

Two important and related observations on this theme of raising capital are worth emphasizing.

• The amount of capital raised by a venture-backed private biotech has no clear correlation with its exit value.  The scatter plot of Figure 6 from this Kevin Lalande analysis is an excellent visualization of this point. The odds of a $300M exit are seemingly just as likely for a company that has raised $40M as it is for those that have raised $150M.  Obviously capital starvation doesn’t give a new company a chance, but above a reasonable threshold this observation appears to hold true.
• For “successful exits” (IPOs, M&A), the amount of capital raised by a venture-backed private biotech tends to be inversely proportional to returns (here).  Because your odds of getting to a top decile outcome don’t change with raising more capital, then raising more must by definition inversely related to the returns (you can only play with numerators and denominators).

To exemplify these rather abstract observations with some specific companies, I thought it would be instructive to look at the Top Deals of 2007, the annual list that is pulled together and annotated well by FierceBiotech.  I’ve looked at the top 10 largest financing rounds of the year, all of which were greater than $50M.  In general, these are companies that have raised $100M+ in equity capital over their “lifetimes”.  Since five years have passed, it gives us a chance for more of a post-mortem, to see what they’ve done with all that capital.  Here’s the rundown:

1. Zogenix - 2007 financing was $78.8M, the biggest of the year.  The company struggled to IPO but got public in late 2010 with a major valuation haircut (fraction of cash invested) as the only way to feed a very high burn rate supporting its formulation improvement for a generic migraine drug.  Stock is off nearly 50% since its 2010 IPO.  Venture investors lost significant portion of their capital.
2. Ception Therapeutics - $77.7M.  After progressing their lead antibody into the clinic, Ception was bought in an earn-out deal by Cephalon and ended up getting paid both the upfront the milestones ($350M in total).  Positive exit in the 3-4x range.
3. Targanta Therapeutics - $70M.  Company got public after the 2007 pre-IPO round, and continued its effort to get a late stage antibiotic program approved; the FDA rejected it with a Complete Response in late 2008 in the midst of the financial crisis.  Company lost most of its value and was sold to Medicine’s Company for $15M in 2009.  Venture investors lost most of their capital.
4. Portola Pharmaceuticals - $70M.  Company has been developing several interesting drugs in partnership with Novartis, Biogen, and Merck, having raised $307M in venture capital to date.  From what I’ve heard, things are going well, but with Merck handing back the Phase 3 platelet drug the burn is likely to increase.  Valuation is very high today, which could handicap returns unless it becomes a rare outlier (a possibility given the story).  Fair to say from a returns perspective, it’s too early to tell but a positive work in progress.
5. AVEO Pharmaceuticals - $58.5M.  Company got public in 2010; share price is now 10-20% below its Series C (2005), Series D (2007) and Series E (2009) rounds.  So despite having a drug in front of the FDA, AVEO is currently a negative return for its venture investors (the ones who have held large portions to date) or at most a marginal win (1.3x) for those that exited in 2011 (looks like Flagship has exited, whereas MPM, Highland have been holding at least some of their positions).  Time will tell if they can recover and generate a return, but the odds are against it at this point.
6. Cogentus Pharmaceuticals - $55M.  Company was focused on a combo pill of Plavix with Prilosec, and never finished its Phase 3 COGENT trial (got 3800 of 5000 patients) before running out of money and declaring Chapter 7 in 2009.  The company had raised $85M over its lifetime, but bankruptcy led to a full loss for its venture investors within two years of this 2007 financing.
7. Phenomix Corporation - $55M.  After having raised nearly $165M for its clinical program (DPP-IV) and platform, this San Deigo based biotech shutdown after a Forest Labs bailed out on their Phase 3 DPP-IV partnership.  Full or near full loss to its investors.
8. Sagent Pharmaceuticals - $53M.  Backed by Vivo Ventures, this Spec Pharma story got public in April 2011 and shot to a valuation over $500M.  This was a >5x homerun for Vivo, assuming they exited.  Its now valued at closer to $230M, and has raised $250M to date.  So it was a win for Vivo and the other venture investors, though a more modest one today than in 2011.
9. Microbia - $50M.  Now called Ironwood, it raised its 2007 round at $6.25 a share.  Ironwood’s currently trading near ~$13, so its 2007 investors are sitting at approximately 2x after 5 years.  Similar to the company’s aggregate investor returns: IRWD has raised nearly $600M and has a market capitalization of $1.3B.  It’s been a modest ~2x positive return for its private investors.
10. Sangart - $50M.  This two-decade old biotech has now raised over $280M for its blood substitute products, including another $115M since this 2007 financing.  Its lead program remains in Phase 2 studies, as it was in 2007, and the company has been through a restructuring in 2009.  It’s hard to believe this one will generate returns for its pre-2009 shareholders, but who knows.  Clearly still a work in progress.

So of the 10 biggest financings of 2007, here’s the scorecard: 40% are significant losses (Zogenix, Targanta, Cogentus, Pheomix), 10% sitting at or near cost (AVEO), 30% are “winners” above 2x in returns (Ception, Sagent, Ironwood), and 20% are still works in progress (Portola, Sangart – if we’re generous on the latter).  I’ve not done the analysis, but my guess is that most of the Top Financings of the 2005-2008 timeframe have a similar distribution.

What’s fascinating is that this distribution of returns is actually not very different from the aggregate biotech venture distribution over the past 30 years (here): 50% of biotechs have lost money or returned only their cost, and 35% made more than 2x invested capital.  But this historic industry distribution includes not only later stage plays like the above ten deals, but also all of the raw, high-risk startups and earlier stage companies that never make it to the perceived “big leagues” of Top 10 financings.

Conventional wisdom would think that these more “mature biotech companies” capable of raising piles of capital would have improved outcome distributions vs less well-financed biotechs both in terms of invested capital loss ratios (risk) and upside return potential (return) – sadly, they seem to have a distribution no better (and perhaps worse) than the historical venture-backed biotech return curve.  Given the later stage nature of the above Top 10, it reinforces a prior blog post about the misperception of risk in later stage biotech investing (here).  Biotechs that move downstream into full Phase 3 development, FDA registration, and commercial stages face considerable and often hard-to-manage investment risks, but fundamentally lack the margin of safety that these activities have within Big Pharma’s budgets and organization – and therefore pay huge penalties when the inevitable delay or rejection occurs.

Two conclusions from these observations: (1) big biotech financings haven’t delivered lower risk or better return profiles, highlighting yet again that capital intensity is a biotech investor’s bugbear; and (2) as a sector, I think we need to reframe the types of deals and managers that we celebrate: equity capital efficiency, not chart-topping financings, is a virtue we should be extolling in biotech.

In recent years, plowing it back into internal R&D hasn’t been the preferred option given pipeline productivity questions.  Returning capital to shareholders via dividends has certainly been high on the list.  Another, albeit indirect, way of paying shareholders is through share repurchases (stock buybacks), and it has also been quite popular.  The expectation (or hope) with these indirect stock buybacks is that the stock will move upwards because the shares oustanding goes down (or at least the buybacks offset the dilution from the exercise of options).

But buybacks have a more mixed assessment in practice (here, here) and are typically only a smart if a company is (a) under-valued and (b) has no better uses of capital.  This latter point is where they draw my ire, especially given their scale in our industry and the many strategic alternatives.

By my estimates (assembled via SEC filings and Reuters, so take it as directionally correct), over $75B in stock repurchases have been or will be completed between January 2011 and December 2012 by 15 of the top Big BioPharma companies.  BioCentury’s lead article “Pharma Bull Run” over the weekend suggested $34B over a twelve month period, so in line with my estimates.  The figure below captures the range across a set of Big Pharma/Biotech:

To appreciate the magnitude of these buybacks, it’s worth comparing them to other important financial values in the biopharma ecosystem.  It’s bigger than the NIH budget for both 2011-2012 by nearly 25%.  It’s 4.5x bigger than all of the private venture-backed M&A that occurred in the past 18 months – and that involved over 70 biotech companies.  It’s 12x bigger than the sum total of venture dollars invested in biotech in that period.  And its nearly 80x bigger than all the capital raised by fifteen biotech IPOs during that period.  This is a huge amount of capital washing into stock repurchases.

So in light of their relative scale, what have these multi-billion dollar buybacks done for these companies’ stock performance?  It’s not clear.  While all of their share prices are up since January 2011, but there’s no correlation in this small sample set between stock movements and the size of the buyback (figure below).  And they’ve not risen faster than related indices: the median of this group is up 21%, the SPDR S&P Pharmaceuticals XPH ETF is up 26%, and the NASDAQ Biotech Index IBB is up 41%, and the latter two track lots of non-buyback Pharma/Biotech stocks.  But who knows what would have happened had they not done these massive repurchases.

Importantly, at a macro level, this huge quantity of buybacks sends a clear signal that Boards and management teams of these big companies believe these repurchases are the best use of capital.  So their sentiment must be that there’s few if any attractive investment opportunities in the life sciences at this point that are likely to create returns better than the hard-to-measure and questionable impact of their buybacks.  That’s discouraging, and cynically suggests to me that many of these companies must be over-valued if that’s how they are thinking.

Big BioPharma clearly believes the Street rewards them for buybacks, and the Street’s short-term nature probably encourages them to keep doing it.  And given how much cash they have on their balance sheets, and generate each year, I certainly believe a healthy amount of share repurchases per company (on top of 1.5x more that gets distributed via cash dividends) is the right thing to do.  But $76B seems like an egregiously large amount for the biggest players in the industry to be pouring into an inefficient vehicle for long-term return generation.

I firmly believe that there are better long-term uses of capital for these companies, with better ROI prospects, than these buybacks.  In an environment where we have an apparent dearth of early stage innovation and a challenging path to advance it from bench to bedside to pharmacy, I can think of a few ideas.

Lets do a thought experiment.  Imagine if only a few percentage points of these buybacks were used for longer-term investments.  It would be billions of dollars per year.  Here are few uses:

• More creative early stage deal-making: I’m clearly conflicted here, but think that Big BioPharma could do a lot more early stage deal-making through risk-sharing, company-building “open R&D” collaborations.  Genentech-Roche type models come to mind – ones that could help build the next great companies.  Further, one could envision more partnerships with structured relationships like Agios-Celgene, WARP Drive-Sanofi or Arteaus-Lilly. Taking a couple percent of the annual buyback number and investing it through creative deals would significantly increase the deal pool and greatly reduce the cost-of-capital for startups.  The issue will be the P&L exposure these deals may cause, so creativity will be important.
• Supporting new public offerings of emerging biotech firms: Stelios Papadopoulos, John Maraganore & Moncef Slaoui wrote a BioCentury editorial in January about the need for Pharma to step up with IPO funding to build confidence for other investors.  It’s an interesting concept.  Without access to longer-term growth capital, our sector will struggle to build the next Gilead or Celgene.  If these Big BioPharma’s channeled just 1% of the annual buyback value into an IPO funding vehicle, it would double the size the biotech IPO funding environment.
• Helping anchor new or existing early stage venture funds with LP commitments: As has been widely reported, fewer venture funds are focused on truly early stage biotech innovation, so helping launch, support and sustain these funds would help engender a more vibrant biotech ecosystem while providing a better prospect for returns.  There’s some of this going on (e.g., Merck and GSK both have a few LP relationships), but it could be more significant.  The paucity of syndicate partners is a real concern for those of us active in early stage, so we’d welcome the launch of more funds.  Further, like the IPO fund, these investments would be balance sheet items and should essentially treated like a repurchased share (an asset).
• Accelerating the funding around pre-competitive work, like that of SAGE Bionetworks or the Structural Genomics Consortium (SGC).  These nascent efforts are being funded piecemeal today, but could have huge longer-term impact if they were resourced to help validate new targets efficiently.  Even a fraction of a percentage of the annual buyback would massively increase these funding efforts.

These four ideas could be very well funded with 5% or less of the aggregate share buybacks being done today.  And I’m sure others have lots of better ideas.

The critical Board question these Big BioPharma’s need to answer is whether these ideas – doing more creative deals, becoming an equity holder in emerging biotech IPOs or an LP in biotech venture funds, or fostering these precompetitive advances – will lead to more value creation for shareholders of these Big BioPharma’s in the long run than that extra ~5% they tossed into buybacks.

For me, the answer is clearly Yes.

According to the National Venture Capital Association, 83 venture-backed IPOs occurred since January 2011 through end of 2Q 2012: 15 Life Science deals and 68 “Technology” deals (which include social media, retail, software, etc…).  I’ve taken a crack at examining the relative performance of these Biotech vs Tech IPOs, and here are the takeaways.

Biotech IPOs have massively outperformed Tech in their post-offering trading.  Biotech median and average percent change since IPO are +19% and +26% versus Tech performance of -9% and +3%, respectively.  This is ~2000+ basis points of outperformance – an impressive metric.  To put that in perspective, asset managers often talk about beating benchmarks by 300-500 bps as a success, so this type of outperformance is remarkable.   Below is a graph comparison across a few dimensions.

But all deals aren’t created equal.  Big deals need to be weighted as much more impactful than tiny ones.   Adjusting performance by the market capitalization immediately post-IPO, the Biotech basket’s weighted average is up 21%.  Tech IPOs are down 31%.  Excluding Facebook, which accounts for close to 50% of the aggregate Tech market cap, the Tech performance is still poor: -20%.  So on a dollar-weighted basis, Biotech has outperformed Tech by 4000 basis points.

Lastly, when comparing performance of IPOs to the market, Biotechs have also done well.  By taking the S&P500 performance since the day of their offerings and subtracting it from their change since IPO, stock performance adjusted by market returns can be calculated.  As shown in green bars above, Biotech IPOs have, on average, maintained a market-adjusted average stock performance of +19% (implying S&P500 is, on average, up 7% during this period).  Tech has underperformed with a market-adjusted stock change of -4%.  So Biotech is not only doing well against Tech, but also very well against the overall stock market.

The outliers don’t even breakout in favor of Tech.  The top decile performers in biotech beat their tech counterparts: biotech’s 90^th percentile saw stock appreciation of 98% (top ~2 biotech deals) vs 84% in Tech (top  ~7 deals).  With regard to the big Tech IPOs that have become household names, there were 20 offerings with >1B valuations in this dataset.  Only four of them are positive today (LinkedIn being the most impressive, up 126%).  Sadly, not one of the biotechs is valued at >$1B, at its offering or today.

The aggregate market capitalization statistics are also interesting.  The total market cap of all fifteen Biotech’s immediately following their IPOs was $3.6B.  Tech dwarfed that with $172B, or excluding Facebook at $91B.  Today the Biotech market cap’s have gone up to $5B, or up 39% (slightly higher than the dollar-weighted number above because of additional share offerings).  Tech has dropped to $119B or $72B without Facebook, which is the destruction of some $54B (or $19B) in value since they went pubic.  Lets put that in perspective: The 68 Tech IPOs have lost more market capitalization in aggregate than 10x the entire value of the 15 Biotech IPOs that squeezed onto the public markets.  It’s also more than 10x the annual pace of venture capital funding into private biotechs.

So since January 2011, the underdogs of Biotech have surprisingly done pretty well.

Reflections on the Data.

It’s clear that on the Tech side, and in social media in particular, there has been a massive wealth transfer from the public investor (retail and IPO-buying funds) to the entrepreneur and private investor.  A handful of these big deals have driven massive (100x) returns to their venture investors (assuming they sold a chunk at the IPOs).  For those that have held their shares with hope of trading out after a post-IPO runup, I’m sure the paper write-ups and write-downs have been a rollercoaster.  But in the end, this type of massive wealth transfer isn’t helpful to the venture-backed ecosystem: this pump-and-dump IPO model isn’t sustainable at these levels, nor will it make going public in the future easier for Tech companies.

On the Biotech side, the stellar post-IPO performance is little consolation to the venture investors in those deals.  Most went out at or near their invested capital (~1.5x) so even 26% share increase doesn’t transform the multiple into something to get excited about.  But it’s better than the poor performance of prior IPO classes of the 2000s (described here), and hopefully will continue.

Generalist public market investors remain on the sideline from Biotech IPOs, which has limited demand and created a cartel of hedge funds that control access to the IPO market.  One would hope that with relative performance numbers like these we’d see some greater interest in Biotech IPOs from the broader investment community.

Many long-term Specialist Biotech buysiders are frustrated that the great Biotech’s of tomorrow are being cherry-picked pre-IPO by Pharma buyers.  This is certainly the case for companies like Avila Therapeutics.  But the challenge has been pricing: Pharma is paying more than these buysiders are willing to.   With strong continued post-IPO performance, perhaps the Biotech classes of 2011-2012  will help support a more attractive IPO window in the near future.

The JOBS Act was signed into law in April 2012, and a part of the legislation was aimed at increasing the number of IPOs.  Yet the data doesn’t suggest a lack of IPOs was the problem., as most of the 150+ IPOs of the past 18 months came before that.  Post-IPO performance may be the bigger problem (hopefully Congress will tackle that next).  On a more serious note, one of the provisions of the JOBS Act will enable confidential S1s – and greatly facilitate the pre-marketing of potential IPOs stories before the formal roadshow.  This could greatly help with marketing more complicated stories, like Biotech, where potential investors can’t just go try out a new photo-sharing website.

I’m hopeful that the confluence of better stock performance, increased interest in innovation from the buyside, and new legislative changes to the IPO process will combine to make for a more interesting IPO market in Biotech over the next couple years.   But I won’t hold my breath – I’ll just keep rooting for the Biotech underdogs.

It all comes down to what do we mean by “early stage”.  To a biotech hedge fund trader, it might be pre-Phase 2.  To those of us in the venture world, early stage could simply mean a new company, or a preclinical program, or a new drug discovery initiative.  Unfortunately, without common and easily measured definitions, generalizations about early stage biotech venture data are challenging and often misleading.

Most data sources, like VentureSource, CB Insight, or NVCA/Moneytree, track financings by round or sequence (1^st, 2^nd, etc…), which doesn’t reflect much about the technological stage of the underlying assets or platforms.  They also track “by product status” but that’s predominantly a Tech sector nomenclature.

So what is the right definition of an “early stage” biotech company?  I don’t know, but a very simple framework around two dimensions – company stage and asset stage – can be useful here (btw, as a former consultant, long live the 2×2):

Most people instinctively think of the Quadrant 1 (upper left) as “early stage” – and it is the quintessential biotech NewCo: a combination of a brand new company and a new discovery stage platform or program.

But Quadrant 2 (upper right) is doing early stage R&D as well: a company raising its 2^nd or 3^rd or later round of financing to support its emerging discovery and preclinical pipeline.  But this latter group is tracked as “later stage” by most data providers (i.e., its not a “First Financing”).

Quadrant 3 – startups formed around clinical stage assets – essentially represents the in-licensing model in biotech, often called the Spec Pharm, No Research Development Only (NRDO), or Search & Develop (S&D) strategy.  These are brand new companies raising their first financings, so are considered early stage by most data providers.

Quadrant 4 captures later stage biotechs: companies with their assets maturing that have raised multiple rounds of funding.

The tranching of financing rounds makes the difference between Quadrant 1 vs 2, and 3 vs 4, challenging to parse from the data in a meaningful way.  Take Biotech A, which raised a $5M Series A, $5M Series B, and $10M Series C.  Its a later stage deal in most of the data.  On the other hand, Biotech B, with a $20M 3-tranche Series A,  is tracked as an early stage “first financing” deal.  Obviously there’s not any meaningful difference between them on the surface, but these are tracked very differently in most datasets.

Unfortunately, no major data provider actually annotates the funding data by these or similar buckets in a systematic manner across the hundreds of rounds and companies that get funded per year.  But I’ve taken a crack at estimating the breakdown for 2011.

• VentureSource data reveals that ~40% of the Biotech rounds closed in 2011 were Seed/First Rounds.  This is up over 2009-2010, but in line with 2007-2008.
• BioWorld (through their Snapshot) and OnBioVC have tracked what percent of rounds are for preclinical stage companies: in 2011, roughly 70% of Series As, and 33% of later rounds (Series B+) were in pre-clinical stage companies.  This rolls up to about 50% of the overall number of companies funded.  This is an uptick from 2007-2009.
• Combining these statistics, one can estimate what proportion of biotechs fall into these different buckets:

So these data clearly support the idea that in 2011 the relative proportion of financings done around “early stage” deals is reasonably high.  And the data I’ve seen isn’t dramatically different for 2007-2010 as a whole.  Early science companies (Quadrants 1,2) represent ~ 50% of the financings (by number of deals), and Spec Pharm like models (Quadrants 3) add on another ~10%.  So upwards of ~60% of deals that get funded have some claim to being called “early stage” investments.  Note this is by deal numbers, not dollars: Quadrant 4 eats up a much greater share of the capital than 40%, probably closer to 60%+.  When the data on 2012 is collected, it will be interesting to see where it comes out given the abysmal top line numbers from the first half.

Frameworks like this are more than just definitional: they can help dissect what it takes to make a successful investment.  In particular, when considering new investments into companies in these three “early stage” quadrants, there are clearly very different substrates and skills required.

Quadrant 1 is a classic biotech startup – typically roll-up-your-sleeves venture creation around academic science or a whiteboard concept.  Both platforms and asset plays can be found in this Quadrant, as can bricks-and-mortar biotechs along with virtual, distributed R&D models. They require formative company-building skills.  At Atlas, we spend much of our time working in deals of this type.  Quadrant 1 startups obviously mature into Quadrant 2 or 4 over time as they raise more capital and their pipelines progress.

New investments in Quadrant 2 deals (or “Other People’s Deals”, OPDs, as we call them) are typically “special situations” for venture creation-focused firms like Atlas – occasionally we’ll invest in them if we feel we can add real value, the rounds’ pricing still offers attractive opportunities for large ownership stakes, etc…   The capitalization structure can often be an albatross on these deals, and syndicate management typically becomes more challenging as the number of investors across different rounds (and preferred stock terms) increases in these boardrooms.

Creating and investing in Quadrant 3 startups requires significant negotiation skills and deep Pharma connections, as the process of spinning out an asset or portfolio from a larger company isn’t simple.  Its also requires lots of patience.  These Quadrant 3 plays can be product roll-up strategies, like Clovis and Tesaro, or they can be single asset spin-outs, like Stromedix or Arteaus.  Our Atlas Venture Development Concept has been focused on this type of deal: helping a Pharma externalize the development of a program through some form of a strategic out-licensing mechanism.

Importantly, there is no right Quadrant to be in – new investments into any can make attractive returns.  The fundamentals of team, product/platform, ownership and deal dynamics, and an efficient use of equity capital determine that outcome.

To finish up – “early stage” biotech means a lot of things to a lot of people.  But definitions matter, and lots of rather important nuance gets lost in generalities – like the fact that “early stage” deals still represent a good proportion of the biotechs getting funded today.

However, although concerning, it’s not the reduced capital flows into biotechs that I find so worrying – quarterly variations happen, these things are cyclic, etc…  It is certainly as hard as it’s ever been for bioentrepreneurs to raise money, but deals are getting done.  There was still $1.5B that flowed into ~200 venture-backed biotechs in the US in the first half of 2012; this volume represents more companies than any of the 16 other sectors tracked by the NVCA/PWC Moneytree venture capital survey outside of software (541), including media and entertainment (186) and IT services (165).  So we’re still a sizeable venture sector.  And assuming the economic and political uncertainties don’t create havoc, I expect venture-backed biotech’s numbers to rebound somewhat in the 2^nd half with the typical lumpy variability of quarterly data.

But what I do find really disturbing is the longer term macro trend around the shrinking number of venture firms active in biotech – and the number of partners in those firms.  Seems like every month there’s another diversified firm leaving healthcare (like Scale Ventures), or healthcare VC firms winding down altogether (like Prospect).  See recent tweets alluding to the subject here, here.

Even in a capital constrained world – and clearly the data suggest biotech venture is getting leaner – one of the biggest constraints on the ecosystem is the bandwidth of individual, active investors: what new startups are you working on forming, how helpful you are to the entrepreneurs you’re backing, what existing portfolio companies do you spend time on, how effectively are you leveraging your network for the portfolio, how many frogs (new opportunities) are you kissing, etc…  And as an active investor looking for syndicate partners in our portfolio, the dwindling “available bandwidth” of the biotech VC universe is a real concern.  Many are too busy working on their portfolios to do more than 1-2 deals a year (including me).  This bandwidth issue is as or more important of a constraint on the number of biotechs that get funded than the amount of capital itself.

To put some statistics around this concern, here are some data from the Dow Jones VentureSource database.  In the chart below, I’ve plotted the number of investment firms participating in at least one biotech financing by year since 1998, for both “first financings” (startups) and all venture financings (first and later rounds).

This reveals a 40% drop from 2007 in the number of participating investors overall, but also startup (“first financing”) investors.  For context, the number of “active” VC firms according to the NVCA only dropped from ~10% from 2007-2011.

These data help quantify the massive contraction in the number of firms investing in biotech that we all know and feel.

A surprisingly positive finding embedded in the data is that there are so many investment firms on the list to begin with.  Since 2008, 169 venture firms have invested in at least one “first financing” in biotech – I skimmed the list and it seems credible.  Hard to know exactly how many of these actually did more than a few investments, and undoubtedly a version of the Pareto principle is in effect: 60-80% of the deal volume is probably done by the most active 20%.  But with over 2000 rounds of funding in biotech since 2008, the ranks of the less active 80% are still doing a few rounds a year on average.

But even with those players, the data in the chart above would suggest we’ve lost nearly half of the investment firms willing and able to do deals in LS over the past five years.  A reasonable number of those are dead or are zombies unlike to ever raise another fund.  A good number are probably distracted by lengthy fundraisings, and praying for an exit to catalyze a close.  Whatever the reason, they are out of the picture today and possibly forever.

In line with this data, Luke Timmerman of Xconomy recently wrote a very good piece on who are the last remaining active early stage VCs in biotech, and shared a bit of data from Thomson Reuters.  Its clear that the shortlist of active biotech VCs is indeed shorter than its been since the mid-1990s.

Is this going to change anytime soon?  Not likely.  Jon Norris of SVB recently published a great pair of reports (here, here) worth reading.  Among many insights that I’ll bring up in the future, there were two great analyses with regard to venture investors:

• Big Exits in the Life Sciences are up from 2005-2011 – but they are concentrated in the portfolios of relatively few firms.  Of the 170 Big Exits since 2005, the Top 10 venture funds (in terms of numbers of these hits) were involved with 52% of all the big exits.  The Top 20 expands that to 66%.   It’s a world of Haves and Have-nots.  For those funds without big exits, fundraising won’t likely happen.
• Investing has outpaced Fundraising.  Since 2005, SVB estimates that there’s been 25% more money flowing out from VCs into biotech comapnies than into new VC funds earmarked for biotech.  This deficit in dry powder has to catch up to the sector, and it appears the numbers are beginning to reveal it.

The silver lining in the SVB data is that Life Science liquidity (distributions) in 2011 has outpaced both of investing and funding (as I also pointed out back in January) – and over time these improved returns would undoubtedly increase LP interest in Life Science investing, which should help to stem the tide of ever-shrinking biotech financings.  Looking forward to the day!

One might ask why and how a company that won’t have profits for a decade can raise any money through the issuance of debt, but it happens frequently, and the “venture lending” business is actually very robust.  Players like Silicon Valley Bank (SVB), Oxford Finance, Hercules Technology Growth Capital, and Horizon Technology Finance (and many others) are all very active supporters of emerging life science companies.

But venture debt is also a very opaque business, with little public information out there around the overall market size, the typical returns to venture debt providers (or its flipside: the cost-of-capital to those who access it), and what are the “market rate terms” and conditions of the debt.  This is yet another area where we as a sector would benefit from greater transparency.  And entrepreneurs seeking to build their companies with alternative forms of capital should certainly spend the time to learn more about it.

Here’s a quick Q&A with some data points on venture debt in biotech.

How big is the venture debt market in the life sciences?  It’s bigger than most people think.  Industry insiders have suggested to me that the debt market in the life sciences is about 10% of the equity market: since $7.7B was raised for Life Sciences (Biotech plus Med Tech) in the US alone in 2011, according to PWC Moneytree, that implies the venture debt market is about $800M per year in life sciences alone. 

It’s hard to know the exact number of companies that raise debt but it’s presumably at least 200 per year (US only).  SVB is the market leader by volume, and has done approximately 150 venture loans since January 2011.  Oxford Finance has done nearly 50 deals since then according to its website, though my understanding is these are probably at slightly higher average deal size than SVB.  Hercules, a pure-play specialty finance company, does a handful of deals per year, and has some 50 LS portfolio companies on its website (though many were done prior to 2011).  A similar specialty finance player, Horizon, has 34.  There are another dozen or so venture debt providers for life science companies, so it’s clear that a lot of firms raise some of their capital this way.

To give a more specific example, in our maturing Fund VII portfolio of early stage life science companies (all started between 2005-2007), >40% of them have taken on debt of some type over the past five years.  For example, Avila and Miragen both have had a $2M line from SVB, and put them in place when their lead programs were in Phase 1 and drug discovery, respectively.  It’s worth noting that Miragen only had $4M in equity at the time, so it provided a material extension of our early stage runway.

Why would a biotech want to take on debt?  There are lots of reasons, but the single most important is that a company wants a longer cash runway and chooses debt financing because raising additional equity financing is more costly and more dilutive in the near term.  Companies use this debt financing for a variety of things: extend visibility to reach key value inflection (e.g., completion of a clinical study); purchase expense capital equipment; strengthen their balance sheet prior to deal negotiations or IPO; or, expand the pipeline by purchasing new assets or advancing secondary programs, among other things.

What are the typical financials and terms?  For an individual company, the size of a debt issuance varies widely and caters to the needs of the issuer: only $1-2M for equipment or capital expenditure loans, and up to $15-30M for later stage growth or working capital.  It obviously depends on the needs and balance sheet of the company.  Typically they are done in concert with and to augment an existing equity financing.  As SVB likes to say, venture debt requires a symbiotic relationship between the debt provider and equity-provider (venture capital).

The “market rate” for terms also varies greatly across these different types of loans, but typically has three important financial components: interest rate, warrant coverage, and fees.  In the US market, fully loaded interest rates range today 10-13% (far higher than in 2007-2008 when it was half that rate).  Warrant coverage is typically 2-8% of the offering in the recent preferred stock security.  Fees are 40-100 basis points of the loan.  Other important terms to consider include whether there are liens on the IP (a negative), Material Adverse Change clauses, cash-collateral elements (where a bank can sweep the account), and other particular covenants.  When all of these are considered, the cost-of-capital is probably in the 15% range.  This is in general much less dilutive than the cost of capital of venture equity.

These deals can generate good returns for the debt providers.  Two specific examples, pulled from Hercules’ public disclosures: Aegerion Pharmaceuticals’ debt yielded them a fully realized IRR of 17%; Sirtris provided a 21% IRR.  These both obviously benefited from the run-up in equity value from the warrants in those deals.  On the flip side, their $25M debt offering with Anthera is now close to the latter’s entire market cap.  Obviously there are winners and losers in any portfolio.

Comparing terms is often a complicated analysis and the practical cost-of-capital of a given debt term sheet is sometimes hard to calculate.  Firms like the Capital Advisors Group specialize in helping biotechs frame up their options and seek competitive offers.  As a general rule, having multiple debt term sheets (much like equity) is critical to understanding what the true “market rate” is for any given company’s debt.

What are the downsides to venture debt?  As they say, debt makes the good times better but the bad times worse.  When lead programs fail or stall, debt can kill a company – force an ugly sale or liquidation, limit future financings, impair the intellectual property, etc…  This isn’t theoretical, it happens.  At Dynogen, having $5M in venture debt when the clinical programs didn’t work as planned significantly hampered our strategic options.  I won’t list additional companies in order to protect the innocent, but across the industry it’s not a short list where debt has become a bugbear in the boardroom.

Fortunately, good venture debt providers often work with the equity holders to come up with a solution that works for rescuing the most value in these scenarios.  But it’s certainly the case that having debt on the balance sheet significantly changes the dialogue during tough times.  Boards and management teams should think through scenarios before raising venture debt to ensure the covenants aren’t likely to become onerous, the core IP isn’t pledged, the Material Adverse Change clauses aren’t easily triggered, etc…

Does debt hurt or help BD or IPO prospects?  If it can meaningfully impact a balance sheet and provided for an extended corporate development runway, than its probably helpful on the margin.  But should the BD process drag on, or S1 get pulled, debt can adversely impact an outcome.

Its worth noting though that it can however be used creatively in deals.  For example, last fall Myriad did a strategic debt investment into Crescendo Bioscience (here), where the former loaned the biotech $25M for 6 years at 6% interest in exchange for an option-to-acquire the whole business at pre-defined revenue multiples.  I don’t know the specifics of this deal, but it appears a creative way to provide P&L-sparing access for Myriad to an interesting platform, while sufficiently funding Crescendo to build an interesting business.

I’ve not heard of Big Pharma doing these types of deals, but given the perceived dearth of equity capital supporting innovative early stage biotechs, it seems to me that they could use their massive balance sheets to offer venture loans to complement their corporate venture capital businesses.

What are some further examples of biotechs that have done debt deals?  Here’s a list of publicly disclosed transactions from SVB, Hercules, and Oxford.  Interesting to see Aegerion, Vitae, Ceptaris, Pacira with multiple listings from different venture debt providers: often these debt providers work together and syndicate their loans much like VCs work together on equity rounds.

• SVB: Aegerion Pharmaceuticals, Affymax Inc, Alimera Sciences, Avila Therapeutics, CardioFocus Inc, Ceptaris Therapeutics Inc, CodeRyte Inc, Enobia Pharma, Furiex Pharmaceuticals Inc, NuPathe Inc, Ocular Therapeutix Inc, Sunesis Pharmaceutucals Inc, TearScience Inc, Tetraphase Pharmaceuticals Inc, TRIA Beauty Inc, Vitae Pharmaceuticals Inc
• Hercules: Aveo Pharmaceuticals, Portola Pharmaceuticals, Anthera Pharmaceuticals, Sirtris Pharmaceuticals, Acceleron Pharma, BarRx, Novasys Medical, Dicerna Pharma, Adiana, Aegerion Pharma, Pacira Pharmaceuticals, and Cempra Pharmaceuticals.
• Oxford: Avanir Pharmaceuticals, Pacira Pharmaceuticals, Affymax, Ligand, Transzyme Pharma, Cadence Pharma, Protox Therapeutics, Anacor, Synta, Supernus Pharmaceuticals, Zalicus, Zogenix, Nanostring, Cerus, Ceptaris, Vitae Pharmaceuticals, and Achoaogen.

While not back to its free-flowing pre-2008-crisis form, the venture debt markets are open for business in biotech.  It should be considered strategically by entrepreneurs, management teams, and boards as a useful financing tool to build biotechs in our capital-constrained world.

The similarities of these three, and in particular between Tesaro and Clovis, are obvious: great and experienced management teams, Phase 3 oncology programs, well capitalized balance sheets, etc…   The one major difference is that Merrimack is a full-fledged R&D shop, whereas the other two are primarily “search & development” organizations today.  One could fill up a blog post discussing the nature of their late stage programs (i.e., none are first-in-class agents against novel targets, but all have potential in varying degrees to be beneficial additions to the armamentarium in cancer), but I’ll leave commentary on that to others.

As an investor, I’m particularly intrigued at understanding the capitalization history, valuations, and returns from these deals.  Attached below are the graphs of their cumulative capital raised (left axis, blue lines) and price per share (right axis, red lines).  Caveat emptor: all these numbers are derived by me from their S1s, so while directionally correct may suffer poor calculus on my part.

Here are a few observations:

• All raised a significant amount of capital from the private markets: $110M for Clovis, $120M for Tesaro, and $268M for Merrmack.  Including their public offerings, they’ve raised nearly $900M between the three of them: $316M, $202M, and $368M respectively.  This late stage oncology game is only for those very deep pockets.
• As private companies, all achieved some sort of step-up in value from their earliest rounds into their late stage rounds.  Series A to IPO multiples of 2.2-3.9x across the group.  But unfortunately for returns, most of the capital came in the later rounds so the participating insider shareholders’ dollar cost average per share went up considerably over time.
• Their valuations today (and at their IPOs) are considerable – collectively worth $1.6B: Clovis is valued at $550M, Tesaro at $360M, and Merrimack at $650M.  These are three of the largest post-IPO biotechs of the past five years.
• At current prices, the returns for the average private investor in all three are in the modest 1.4-1.8x range.  Clovis’ investors had a weighted average private price per share near $12 and is now $20, Tesaro was near $8 and IPO’d at $13.50, and Merrimack was near $5 and is now $7 (if my calculations from the SEC filings are right).  These current return multiples are remarkably comparable, although the IRR on the two new companies, Tesaro and Clovis, is obviously much more attractive than 12-year old Merrimack.

What’s interesting and sobering to note is that we’ve got three very successful, public late stage oncology companies with aggregate valuations of $1.6B that have raised $900M – this is less than a 1.8x multiple on invested capital in aggregate, despite the lofty valuation number.  That’s a snapshot of how suboptimal the markets are today: Phase 3 programs led by great teams should be able to generate better returns at their IPOs than modest 1.4-.1.8x-ish metrics.

Its fair to say that all of these oncology plays are clearly longer term growth stories attempting to break out over the next 3-5 years, and its doubtful any of their investors have sold any shares to date.  Their IPOs were geared to access larger pools of capital to build bigger businesses.  Many of their private investors were and are probably buyers in current and future offerings– betting these management teams can create magic like they have in the past.

But what will it take for these stories to become high-returning deals?  Over the past few decades, a 3x venture return is typically a top quartile deal, so lets assume that’s the threshold.  In light of their valuations today and the amount they’ve raised, hitting a 3x will require them all to break out to much higher values.  Assuming they don’t raise any more equity money (which is purely theoretical given their burn rates), here’s what it would take in terms of market capitalization: Clovis at $950M, Tesaro at $600M, and Merrimack at $1.4B.  Hitting these heights would require around 100% appreciations in those stocks.  All three would need to rank amongst the handful of post-IPO performers of the past decade for this to happen.  While this is quite possible with these management teams, its certainly not easy.  And statistically its unlikely to happen to all three.

Looking forward to watching them over the next few years.

Time and again, life sciences VC deals have been outscoring their better-capitalized, super-hyped and oft-praised colleagues who do deals in the tech side of the asset class.

- Jonathan Marino in a PE Hub Post titled “Quietly, VCs in Life Sci Made Biggest Hits“

It was nice to see Jonathan calling out some of the big 2011 Life Science exits.  We may not have social media Insta-gratification deals like our Tech brethren, but 2011 was a solid year for Life Science M&A, as described in this Wilmer report and my January post on the subject.

This prompted me to look at how we’re doing in 2012.  By my count, there have been at least nine venture-backed therapeutic biotech M&A exits this year, and I’ve assembled them in a table below.  If I’ve missed any, let me know.

With $2B in upfronts alone for the first five months of 2012, we’re on pace for a year like 2010 ($4-4.5B in upfront payments from M&A).  Unfortunately this is lower than the very strong pace of 2011, but its still real points on the board.  And this $2B number is probably higher than the amount invested into new venture rounds in biotech for the first half of 2012 – getting paid more than you are dispersing is clearly a good thing (as discussed here).

A few other observations:

• The median upfront deal size of $150M is right in line with the 10-year dataset in CapIQ of ~500 or so life science exits (see also Lalande’s analysis with a trimmed mean of $156M); a few of the bigger earnouts being met would obviously change this analysis favorably.
• Although a big distribution, theses exits have taken about $60M of equity capital to get to the goal-line (median).  This is in line with the 10-year dataset described above (~$56M).  Depending on your vantage point, that’s either a lot or a little.  As an early stage investor, I tend to think the former, but its all a matter of perspective.  History suggests that it’s hard to deliver healthy returns above this amount of capital.  Boston Biomedical’s limited equity infusion and use of partner capital to fund its efforts certainly paid off handsomely in their returns.
• Seven of the nine deals were founded around early stage biotech research with a diverse mix of models (in line with lots of biotech worldviews working).  Ferrokin and Stromedix were virtual asset-centric plays, with a half-a-dozen FTEs  and a network of external CRO and consultant partners.  Avila, Cellzome, Boston Biomed, Kai, and Aldagen were drug discovery platforms of various sorts (small molecules for first three, peptides for Kai, and cell therapies for Aldagen).  The last two, EUSA and EKR, were spec pharma asset development and commercialization plays.

Who knows what the rest of the year will bring.  With the U.S. elections, European dysfunction, and global economic concerns, sentiment may be biased towards taking risk off the table which could make buyers more cautious.  But my guess is we’ll continue to see a steady pace of M&A even with those macro headwinds – and beating the 2010 M&A numbers is not an unlikely outcome.  And lets hope for more than that.

Its worth noting that we’re very pleased to have been a part of three of these 2012 deals – Avila, Stromedix, and Cellzome were all Atlas deals.  Both Avila and Stromedix were solid returns, and will be great deals should (when!) even a few of the earnouts be delivered.  And while Cellzome wasn’t a great return, we’re very appreciative of the team there as they worked very hard to return capital: it was a genomics-bubble 2000-vintage biotech that managed to raise no additional equity since 2004 while building a solid epigenetics platform.  A long road, but this successful exit certainly beats many less desirable outcomes.

But the lemming behavior revealed by this list is frightening: a significant percentage of these programs are chasing the same targets.

We used a different database than the PhRMA report (Thomson Pipeline vs ADIS Insights) but count roughly the same number of active programs in oncology (990 vs ~981)*.  Of those, the concentration of clinical development effort around a handful of targets is staggering: 8 targets are addressed by >20% of the projects, each of which has more than 24 projects in clinical development.  Clinical projects that target VEGF lead the charge at an amazing 70 – and this only includes the one’s addressing oncology.  Here’s the stack below:

Does the industry really need 25+ clinical or commercial stage programs against each of these targets to exploit the full anti-cancer potential of those mechanisms?  Even more puzzling than the clinical development congestion on these targets, what about the huge numbers of preclinical projects addressing these targets (in red above)?  Some may be unique angles (e.g., receptor mutant-selective inhibitors that spare wildtype), but most are probably not.

A quick back of the envelope suggest there’s a ton of wasted industry resource across these programs.  Assume every clinical program has cumulatively spent $20M, and the preclinical ones at $5M (both conservative estimates), this implies well north of $5B+ in R&D dollars has been tossed at this set of programs over the past few years.

How many winners are likely to emerge from this?  Maybe a small handful per target, at most.  That implies lots of zombie programs being funded by budgets and investors that will never be of value.  This is of course always the case in R&D: programs more often than not fail.  But the concentration of industry activity on these privileged targets feels way outside the norm, and in aggregate represents a lot of wasted energy.  Beyond money and time, it also is a waste for patients.  Clinical trial recruitment is hard enough for exciting programs, but doing work on all these unlikely-to-matter programs on crowded targets seems borderline unethical frankly.

All that said, there is some merit to the principle that we should aggressively attack mechanisms that work in order to fully exploit them.  The arguments for this type of pipeline/program concentration are that: (a) smarter follow-on molecules may dial out some of the liabilities of the pioneer molecules through better selectivity, pharmacokinetics, etc… that make for better patient outcomes (after all, Lipitor was 5^th to market after all, Humira was 3^rd, etc…) – and this may be particularly relevant in unique “mixed” or dirty cancer agents; (b) different chemotypes could be terminated for tox liabilities so its good to have multiple shots on goal for high value targets; (c) different chemotypes will lead to differential responses by different patients for genetic/epigenetic reasons – and so personalized medicine requires a bigger arsenal of programs.  And much of the above logic is probably true to an extent.  But do they warrant, for instance, having 29 clinical and 31 preclinical programs against EGFR, many years after Tarceva and Erbitux were approved?

The real reason for this concentration isn’t about the arguments above – it’s fundamentally a reflection of our industry’s collective risk avoidance, as well as a misperception of aggregate risk.   

Portfolio decision-making in large and small companies leads to an overwhelming bias towards precedented mechanisms as a means to reduce biologic risk.  No head of discovery ever got fired for producing too many Development Candidates, and the lowest risk way of doing that is through “fast follower” (and even slow “fast follower”) incremental improvements.  In the shot on goal mentality of R&D, more of these shots are better.  Most R&D portfolio prioritizations punish novel target programs as low “confidence in mechanism”, and therefore riskier than precedented targets.  The math from these models is hard to challenge, having made some of those models in a prior life.

Sadly, the same biology risk avoidance holds for many venture capital investment decisions (“this is a risky target if no one else is working on it”…  “we don’t do drug discovery unless its close to IND”).  There’s a reason only a few VC firms do early stage innovation and venture creation like we do – many investors myopically focus on biology risk when other risk drivers are equally if not more important (which I’ve written on here).

In the end, all this leads to an industry pipeline – big and small companies alike – full of groupthink programs that follow the “hot” target trends like lemmings.

Beyond just the biology, I’d bet the industry pipeline’s chemistry concentration on these privileged targets is also high.  With compound structures from the patent literature, simple analoguing strategies and scaffold hopping approaches can provide for a patentable chemistry around even widely-mined chemotypes.  There’s a reason most kinase inhibitors are derivatives of only a handful of scaffolds.  These chemistries are also perceived to be lower risk (“this series has been in the clinic” etc…), and maybe they are.  But they don’t provide a lot of room for novel discoveries.

But while biologic and chemical risk is often lower for projects in these crowded classes, the differentiation risks skyrocket - including the downstream Phase 3 development risk (e.g., active comparator in your drug class; refractory patients), regulatory risks (e.g., higher bar for approval with other approved agents ahead), reimbursement risks (e.g., why should payors pay for your drug), and marketing risks (e.g., how do you get share of voice).  How does one differentiate the 10^th PI3K inhibitor to enter its Phase 2a program?  Very challenging.  Furthermore, if small biotechs are hoping to be acquired by Big Pharma, being in the first wave of innovative projects against a novel target is important – or certainly facilitates earlier interest.  If a biotech has to do a Phase 3 program to show differentiation before it gets acquired, hope they have deep pockets.  I can think of a few small cap examples of this today.  And, lastly, the returns from spending all this may not be there in oncology especially: at what point does society stop paying $50-100K for a “doubling of survival” that adds only 3 more months of life?   For all these reasons, the next decade will not be kind to incremental innovations against well-known, well-drugged targets, and the companies that have them.    

Importantly, I’m not advocating for big programs against totally unvalidated targets that pop out of the Human Genome (the “Fruits of Genomics” described over a decade ago).  That would be a huge waste of time and money.  But the risks can be mitigated around novel first-in-class targets through many “confidence” building approaches and the tight titration of capital into them over time: (a) early validation of an intervention’s effectiveness in the best available animal models of disease (not predictive, but better than nothing); (b) identify the possible linkages to human genetic variation and if nature has validated the target already; (c) understand if pharmacology can phenocopy the transgenic knock-out or knock-down models; (d) show that multiple chemical series achieve the same outcome in preclinical models; (e) create translational strategies with appropriate PD markers of target engagement to bridge from animals to the clinic; and many, many other strategies.  None of these are a panacea for risk – but they do help strengthen the spine for making a commitment against novel biology to move forward.  Careful deployment of resources, and the weeding out of false positives with a fast-to-fail mentality, can help manage risk in these novel approaches (btw, its hard to fast-fail a precedented program because differentiation is largely untestable for oncology in preclinical models).  And while the biology risk may remain higher than for more established targets, the downstream risks around differentiation are clearly reduced. I suspect the aggregate “net present risk profile” of a preclinically-validated but not yet to human Proof-of-Concept program is probably far less than a similar stage inhibitor of a well known target like EGFR, PI3K, or VEGF today.

I’m also all for advancements in oncology.  Many cancers still have huge unmet medical needs that we should be aspiring to tackle.  But in the mid-1990s, oncology was a far smaller percentage than the ~40% or so of the pipeline it occupies today.  Is this the right allocation of resources?  What about all those other fields (like the contrarian ones I highlighted last week)?  It’s worth questioning at R&D strategy meetings around the industry.

When we see this kind of cancer pipeline concentration, its certainly hard to argue that we as a sector don’t need to make smarter bets about what diseases, what targets, and what approaches we put our scarce industry R&D capital into – both in Big Pharma/Big Biotech (90% of the spending) and in smaller biotech (90% of the companies).

We need to stop being target lemmings, or our group-think and risk avoidance will run our industry off a cliff.

* Analysis thanks to Michael Gladstone at Atlas Venture

But perhaps not always in the way you might think: investing in areas that Pharma is fleeing or deemphasizing can yield great returns.  While there are clearly many examples that don’t fit this generalization, it certainly has held true in lots of areas.  What’s deprioritized today by some Pharma, will likely be re-prioritized by others in a few years.

One of the best examples is anti-bacterials.  In the 1990s and early 2000s, many Pharma companies pulled out or scaled back their anti-bacterial research, as has been widely covered (here, here, here).  From 2000-2005, companies like Wyeth, Aventis, Eli Lilly, GlaxoSmithKline, Bristol-Myers Squibb, Abbott Laboratories, and P&G all deprioritized anti-bacterial R&D.  Many of these companies spun out either their programs or whole divisions.  And yet, over the next five years, some of the best returns in biotech venture capital from M&A exits in the anti-bacterial space: Calixa, Cerexa, Novexel, Neutec, Paratek, Pennisula, Protez, Vicuron, and a number of others.  These deals represent multiple billions in exit value.  Several of these began their lives as startups formed around out-licensed compounds discovered by Pharma, or even entire sites/portfolios (like Novexel).  Others were de novo science around novel targets, especially for MRSA and other resistant bugs.  By having a contrarian view, and taking advantage of a strategic depriorization in Pharma, biotech entrepreneurs and their venture backers were able to create considerable value in the past decade out of the anti-bacterial theme.

I think the a similar opportunity exists today across several areas:

Neuroscience.  In past couple years, there’s been plenty of press about various Pharma cutting back their neuroscience efforts in Alzheimer’s, psychiatry, pain, etc… (here, here, here).  GSK, AstraZeneca, Novartis, Pfizer have all been reworking what’s in or out of their CNS activities.  Many believe classic neurotransmitter approaches around serotonin et al have been over-worked areas of psychiatric drug discovery, and that the clinical challenges around neurodegenerative disease modification, patient variability, and complex background therapy make CNS an unattractive R&D opportunity today.  Recent failures of several gamma-secretases in AD don’t help, and I suspect that if the Phase 3 beta amyloid antibody programs fail to show cognitive improvements in Alzheimer’s this year we’ll see a further exodus from the field.

But I’m incredibly bullish on the neurology field as a whole: I think the convergence of our emerging understanding of neurobiology, genetics (and importantly epigenetics), new models of disease, and neuroimaging/diagnostics provides an exciting foundation for new efforts.  Both symptomatic and disease modifying therapies address real unmet needs in neuroscience, and both are benefiting from the accelerating pace of scientific progress.  Patient selection and stratification approaches will reduce the noise in trials (like Genentech’s recent initiation of a landmark AD prevention trial in a Columbia subset with the NIH).  Its worth noting that most of the high profile struggling Phase 3 programs in the field were discovered at least a decade ago – around the time of the Human Genome Project – and lots has changed since then (some, like AZ/Targacept’s TC-5214, were discovered 50 years ago; Bapineuzumab’s murine parent mAb 3D6 was from late 1990s).  So I think it’s a great time for early stage translational research in neuroscience around novel approaches to these big medical problems.  We recently helped externalize Lilly’s anti-CGRP antibody for migraine prevention through our Arteaus investment – a good example of venture stepping up where Pharma was hesitating.  We’re currently evaluating a number of exciting neuro projects out of both academia and Pharma with the aim of creating and launching some high risk, high reward startups.

Heart failure.  Despite its massive cost on the healthcare system, a number of Pharma with legacy franchises in heart failure have deprioritized their efforts over the past few years, at least temporarily, including long-time heart failure leader Pfizer (here).  Lots of Pharma’s just won’t touch the area.  Earlier in the last decade, with a more limited understanding of molecular basis of heart failure, few predictive markers, and enormous Phase 3 mortality study requirements, one could sympathize with the change in emphasis.  But the world is changing: new disease models, appreciation of the different subtypes of heart failure, better surrogate endpoints to predict Phase 3 outcomes, and prognostic markers of a patient’s risk profile, etc.

Despite having a survival curve as steep as many cancers, heart failure hasn’t seen a new mechanism of action approved since the 1990s; I suspect this will change over the next decade.  New modalities like microRNAs (e.g., Miragen’s programs in CHF), gene therapy (e.g., Celladon’s SERCA AAV), stem cells (e.g., Juventis’ approach), and a plethora of other novel small and large molecule targets (e.g., Cardioxyl, Cytokinetics) offer real promise to change the heart failure landscape.  And there’s a huge number of targets being pre-clinically validated through academic work that will need to get picked up by translationally focused enterprises (after being reproduced, of course)- opening up more venture funding opportunities.

Obesity.  Many Pharma’s have fled obesity due to the challenges at the FDA (like Accomplia’s collapse, and the CRLs of the three weight loss drugs now in front of the Agency), as well as the treatment failures in the clinic throughout the 2000s (e.g., leptin, MK-0557, CCK1 agonist, PYY3-36, MC4R agonists).  No new anti-obesity therapies have been approved since Xenical in 1999.  But this is a huge medical problem, and therapeutics will be part of the solution.

But while Pharma has been generally hesitant over the past five years amidst this uncertainty, smaller biotechs have stepped up.  It’s not an accident that the three drugs in front of the agency today are from emerging mid-/small-cap companies.  Earlier in development, a number of innovative younger companies, like Zafgen, are tackling novel mechanisms to treat obesity that go beyond centrally-acting agents that suppress appetite.  Novel, peripheral targets that attack adipose, liver, and muscle abnormalities present in obese patients are critical, and blending this understanding of disease biology with targeted clinical strategies makes the early development landscape exciting.  Beyond Zafgen, we’re also excited about several early programs in the obesity/metabolic space in our portfolio today (including Nimbus’ acetyl-CoA carboxylase program, and miRagen’s miR-208 program), and certainly wouldn’t shy away from additional bets in the field.

Picking great areas like neurology, cardiology, and metabolism to invest in isn’t enough though.  In domains like these, where the biology is moving fast and R&D expenses can add up quickly, entrepreneurs and their venture backers can lose their shirts if they aren’t careful – so the keys to success are to take a thoughtful, titrated funding approach to derisking these programs, pushing for killer experiments early to terminate false positives and avoiding the omnipresent risk of over-capitalizing them.  Lastly, finding the right partners to power up these therapeutic projects as they move through development is critical because the late stage registration studies will be far bigger than a venture wallet can handle.  This is tough in fields where Pharma is pulling back.  But it can be done.  A good example of this is our partnership at miRagen with Servier, where they secured ex-US rights and will significantly help fund our global heart failure programs.  Deals like this will be critical to offsetting the burden of late stage development of these agents.

So the above areas – neuroscience, heart failure/cardiology, and obesity/metabolism – are just three of the areas I think are incredibly fertile ground for venture capital investing today (and have been over the past five years).  I think they’ve largely been underfunded in recent years by Pharma and VCs alike, while oncology, anti-virals, and orphan diseases have seen relative funding spikes.  As evidence of the latter, just look at the preponderance of EGFR inhibitors, PI3K alpha blockers, HCV programs, and orphan projects within industry pipelines (more on that in a future blog post).

There are also several modalities that I think are great contrarian bets for VCs today – RNA-based therapeutics being a prime example.  With Roche, Novartis, Pfizer, and Merck de-prioritizing their big prior RNA investments, I think they may be throwing out the proverbial baby with the bathwater.  RNA therapeutics have huge clinical potential, especially single-stranded approaches as I’ve mentioned before (here).  We’ve recently doubled-down in the field with the launch of RaNA Therapeutics.

My enthusiasm for the therapeutic areas above runs directly counter to the sentiments captured in the NVCA/MedIC’s Vital Signs report from last year: “survey respondents expect to see significant investment decreases in companies fighting serious and highly prevalent conditions including cardiovascular disease, diabetes, obesity, cancer, and neurological diseases.”  Although I’m enthusiastic, there may indeed be significant drops in funding for these areas over time, especially if venture capitalists follow Pharma’s lead.  But a drop in funding doesn’t mean a drop in returns: those companies that get funded in these areas, and deliver compelling packages in early development, will have a great opportunity to make outsized returns by selling scarce novel therapeutics to innovation-hungry partners.

While a similar takeway has been published before by Windhover’s StartUp about a year ago, these data suggests a fairly robust effect from a large dataset.  The analysis includes 2907 therapeutics companies that raised venture capital dollars between 2000-2010 across 5100 rounds of financing.  Corporate VCs were investors in about 10% of companies, and this pool of 286 companies had what appears to be a markedly higher hit rate: a ~60% higher rate of licensing deals, M&As and IPOs.  The figure from the article is pasted below:

As an aside, its interesting to note that the “CVC participation rate” in biotechs of ~10% over the decade reported in this Burrill article supports the perceived uptick in corporate venture activity today when compared to more recent cohorts of data.  A recent report published by NVCA/PWC in their MoneyTree earlier this year suggested that 18% of all biotech deals had CVC involvement in 2010-2011.  My guess is this rate of 1-out-of-6 biotechs with CVC involvement today probably holds for the aggregate biotech funding rate across all company stages.  However, while I don’t have the data breakdown by stage, from what I can tell the rate of involvement in Series A rounds is probably closer to one-third, at least in the Boston biotech cluster.  As I’ve written about before, Atlas’ recent portfolio of new Series A-stage startups has a participation rate with pharma corporate VCs of 70% or more.

So these new data from Burrill are entirely all consistent with pharma corporate VC playing an important role in the ecosystem, and that they seem to get involved in good companies that possess better than average probabilities of success.

But the killer question is whether corporate VCs are just good company pickers because they look for things their Pharma  R&D organizations may want, or if they actually help shape a better outcome once they get involved either through the value-added “stamp” of Pharma validation or through active deal management contributions.  I’ve certainly seen lots of evidence of the latter, so am biased to that answer – but it’s a chicken and egg type of question.

A big caveat to all of these generalizations is worth mentioning: if there are different styles among independent early stage venture capital firms, then its fair to say there are different species amongst corporate venture capital firms.  The two ends of the spectrum:

• Standalone CVCs:  These firms behave as quasi-independent venture groups, driven foremost by financial metrics but relevant strategically – like SR One, Novartis Ventures, Lilly Ventures, Medimmune.  These firms typically function more like traditional VCs: they lead/co-lead rounds, drive a pricing process, and typically take on Board seats.  They also tend to maintain high walls of confidentiality from their corporate R&D groups.  Lastly, and importantly, they only recuse themselves from Board discussions in a portfolio company if it involves a strategic process with their parent company; otherwise, they are active and often vocal contributors to a BD debate.
• Purely Strategic CVCs.  Other firms have a more explicitly strategic mandate, including firms like Amgen, Shire, Abbott, BI, Baxter, etc…   These firms tend to join existing syndicates in companies focused on areas of deep franchise interest.  They rarely lead the negotiation to price a new round, and most often only take on Observer roles on Boards.  During BD discussions, the representatives of these firms typically leave the Boardroom.

The Burrill dataset may be large enough to explore which of these CVC species has contributed to the outperformance suggested by the data.  But it would be complicated, as many corporate VCs have morphed their strategies over time.  For instance, in the early 2000s, SR One and Novartis didn’t typically lead rounds, they were followers in most syndicates; today, my guess is they lead or co-lead everything they do.

Even if we don’t have performance data broken down distinctly, it is fair to say that the contributions these divergent CVC strategies can bring to an early stage biotech company are very different.  Strategic CVCs may plug a company into their R&D organization much more deliberately and bring the capabilities of the parent to bear more explicitly.  On the flip side, standalone CVCs bring Pharma insight, smart company-building capital, and a strong governance role – in some ways filling the vacancy left by the shrinkage of the traditional venture firm.

We enjoy working with both, and will continue to.  But for biotech’s thinking about their next financing, knowing what you need from your syndicate and how it maps to these differences is certainly important as you choose your next investor.

Its received lots of coverage in the media (here, here) and blogosphere, including an excellent pair of posts from my partner Fred Destin (here, here), and should be required reading for those interested in the venture ecosystem.

It’s fair to say, though, that many of the themes highlighted by the report are not “new” news, but they’ve put quality analytics behind a number of these points.  Insiders have long known about most of these issues, and I’ve discussed many here in this blog.  Some reflections on a few key points:

• The “venture capital math problem” places real constraints on the returns achievable by large funds.  The simple math, and the unimpressive data for >$500M funds, have been covered elsewhere (here, here).  More funds in the $250M range have delivered handsome returns than any other size.  It’s particularly relevant in biotech where Facebook-sized exits aren’t even a remote possibility, as I covered in a recent post looking at the assumptions required to deliver “venture-like” 3x gross returns.  At Atlas, we’re true believers in the “small is beautiful” mantra: our most recent fund is ~$280M and anticipate similar sizes going forward.  We know the consequences of big funds because we’ve made those mistakes at Atlas.  In the bubble, Atlas raised far too much, far too fast, and invested like drunken sailors (much like everyone else).  For the past decade, we’ve been working hard to get LPs their capital and then some.  As most in the industry know, the bubble vintage funds of 1999-2001 will probably go down as the worst years in the history of venture capital.  Lots of lessons learned, but its clear to us that in the long run smaller funds are better for LPs, GPs, and entrepreneurs.
• Partnerships are complex and often blackbox, and understanding what being a “partner” really means should be important to LPs.  Partner titles can mean lots of things: GPs, key men, junior partners, managing directors, operating partners, venture partners – the industry uses a bunch of terms for different roles and there is no consistent definition.  And the economic fund flows within partnerships often require armies of accountants.  These oddities certainly create confusion.  I can’t speak for other partnerships, but I think we’ve got this right (finally): Atlas has undergone an evolution since I joined in 2005, and is a much healthy, leaner partnership today.  We’re now a flat, equal partnership with six GPs and are exclusively focused on early stage innovation in life sciences and technology.  The cohesion of the team and clarity of the strategy is part of why I love coming to work in the morning.
• The median venture fund underperformed the public markets in the Kauffman portfolio.  This is certainly true, and affirms what insiders have long known about returns.  And no LP invests in venture hoping for a median performer.  What’s also true is that the median managed mutual funds underperform the passive equity markets (e.g., according to Vanguard, 84% of actively-managed U.S. large blend funds underperformed their index for the decade prior to 2007).  Actively managed asset classes aren’t about the median performer: they appeal because top quartile performers can vastly outperform.

So in general, I’m in complete agreement with the big themes in the report.

That said, there are a few nuances worth mentioning:

• The “vaunted J-curve” was challenged as a myth in venture.  The J-curve has been described elsewhere, but in summary its that value of a VC fund may post negative IRRs in the early years as money is spent on deals and expenses, but no exits or upticks in value have been achieved.  The Kauffman report does a good job of showing that many VC funds (Tech-biased) didn’t actually have a J-curve: that is, they achieved rapid paper write-ups in the value of their portfolio (and then use those write-ups to raise their next fund).  But I don’t think it’s a myth in early stage life sciences: biotech deals don’t typically achieve big write-ups early in their lives.  Bijan Salehizadeh has written on the “unrealized” problem in biotech, which is the product of the lack of these write-ups.  Many remain flat for 4-5 years and then, if successful, hockey stick upward with a favorable M&A exit, like Avila Therapeutics prior to its deal with Celgene, for instance.  The reason VCs like Atlas like to get into startups early that are likely to be held at flat valuations for years is to shape the DNA and direction of a company while securing out-sized ownership stakes.  Many great deals, like Plexxikon, didn’t raise money for the 8 years prior to their exit – so getting in early was the only way to get into those deals.  So, at least from my limited experience, early stage biotech venture portfolios more often than not exhibit a J-curve early in the life of fund. But its fair to say that “the plural of anecdotes is not data” and I’d love to see an aggregate dataset on this.
• There’s a reference to liquidation preferences and how VCs apply them to their portfolio but not themselves.  Before the bubble of 2000-2001, it was more common to pay carried interest on a deal-by-deal basis (which led to the clawback problems when the funds were underwater), but these terms are less common today.  Our LPA, like many others, effectively has a liquidation preference in place: VCs are only paid a share of the profits when all the committed capital and fees have been paid back.  This is the equivalent of a 1x liquidation preference feature in a typical startup, except that its over the entire fund.
• Kauffman recommends that LPs invest directly into a portfolio of startups.  I’m generally supportive of LPs investing directly, but making sure it’s alongside “trusted” investors is key.  I’m a firm believer that startups benefit a great deal from having value-add, high quality venture capitalists involved on their boards.  Good VCs also bring the best of their entire firm to bear on a portfolio company’s issues across a spectrum of company-building activities: corporate strategy, business development, resource allocation, etc…  So if LPs are going to “go direct” to avoid the fees and profit-sharing of a VC investment, its definitely worth teaming up with some trusted co-investors.

As I’ve said before, the venture model is clearly in flux – hopefully better returns from some of the experiments we’ve been doing in biotech venture capital will bring back some of the enthusiasm for the asset class.

Post was updated after a good discussion with Diane Mulcahy of the Kauffman Foundation, who among other good comments had correctly pointed out that a number of funds are paid carried interest once invested capital has been returned, rather than total committed capital.