Today we’re thrilled to announce the acquisition of Nimbus’ allosteric Tyk2 autoimmune program by Takeda for up to $6 billion in an upfront and future milestones.
As a potential best-in-class asset in the significant new therapeutic category of selective Tyk2 inhibition, our recent Phase 2b data set the stage for a broad Phase 3 campaign. Partnering with Takeda will accelerate this potential therapeutic reaching its full potential.
Kudos to CEO Jeb Keiper and the Nimbus team for continuing to innovate around new medicines, and for driving this partnership to closure. We wouldn’t be here today if it weren’t for our exceptional science-driven organization and culture at Nimbus.
Like all major programs, the history of our Tyk2 program highlights the long and non-linear enterprise of drug R&D: a decade-long overnight success.
Starting with some background, Nimbus isn’t a new startup – in fact, we just celebrated our 13th anniversary. During the spring of 2009, when the markets hit bottom, Schrödinger’s Ramy Farid and I co-founded the company, then called Project Troubled Water, and we incubated it in the Atlas offices. I was fortunate to be the acting CEO in the early years, and have served as Chairman since. In the spring of 2011, we came out of stealth with a blog “Discovering Nimbus”, which highlighted the three key elements of the “experiment”: a cutting edge in silico discovery engine, with a truly virtual and globally distributed operating model, and a unique asset-centric LLC-based corporate structure (also described here). Nimbus CEO Jeb Keiper revisited these founding corporate hypotheses in his 2019 post “A Decade of Discovery” as part of celebrating Nimbus’ first 10 years.
Like many startups, during the early years we spent countless cycles on the selection of targets to work on – probably the most important choice any platform company makes. We knew that if we chose the “wrong” projects, we weren’t likely to succeed. If the targets were too easy, anyone could do it… and if they were too hard, perhaps no one could. Both of these poles are common failure modes for new discovery companies.
Along with CSO Rosana Kapeller (now CEO of Rome) and CBO Jonathan Montagu (now CEO of HotSpot), among other key team members, we “scored” hundreds of possible targets on many of the typical attributes (biological validation, genetics, unmet need, clinical path, competition), but also and importantly where we felt our privileged access to cutting-edge computational insights would lead to a discovery advantage. The key was to identify what targets would best be cracked open by our approach. In hindsight, it’s clear we were fortunate to have chosen some great targets: two of the initial programs were IRAK4, which we partnered with Genentech in 2015), and acetyl-CoA carboxylase (ACC), which we partnered with Gilead in NASH in 2016.
Our third program was Tyk2, and it’s a great case study on the patience and pivoting required across the long timeline of drug R&D – because our Tyk2 inhibitor program is just now finishing its 11th year.
In the fall of 2011, our “New Targets” team triaged Tyk2 and got excited at its prospects in both cancer and autoimmune applications, based on the emerging understanding of human genetics. We kicked off our first Virtual Screen with our partner Schrödinger in late 2011 and pushed a number of series into lead optimization over the next few years. All of these initial campaigns were focused on the ATP-competitive kinase catalytic site, where we were attempting to drive selectivity vs the other JAK family members given safety concerns associated with JAK1, -2, and -3. We leveraged a huge amount of processing power for free energy perturbation modeling (here). During these early days, increasingly compelling genetic data further strengthened the linkage of Tyk2 with autoimmune disease, increasing our enthusiasm for the target across a wide range of diseases.
Given the high similarity between JAK1/2/3 and Tyk2, only modest selectivity was achieved via the catalytic site; we weren’t convinced it was enough to be clinically profound. But as we understood the target more, as well as the protein’s structure-function relationship, the allosteric JH2/pseudokinase site caught our attention, as it possesses enough structural differences to drive real selectivity. After multiple years of working on the target, we chose to fully pivot our discovery campaign to drugging this site, and fully accelerated our efforts there. As this progressed in discovery, we also partnered with Celgene, which was subsequently acquired by BMS, as an example of our LLC deal-making model.
The team’s discovery efforts successfully culminated in NDI-034858, an exquisitely selective allosteric inhibitor of Tyk2. We filed our IND in November 2019, and had 28-day Phase 1b data in patients in our hands by spring 2021 (thanks to Covid delays). In August 2021, we launched our Phase 2b clinical trial and recently announced the top line data for ‘4858 from that trial in early November 2022. It’s now poised, in Takeda’s capable hands, of starting Phase 3 studies in 2023.
As with all emerging new medicines with huge potential, there’s been a fantastic drug R&D team behind its success. Keiper, who has led the company as CEO since October 2018, deserves enormous credit for building the current team and advancing our strategic priorities. At the risk of not mentioning many of the incredible contributors, I’ll call out both our CSO Peter Tummino and our Tyk2 clinical lead Bhaskar Srivastava. Peter and Bhaskar were and are a spectacular discovery-development pairing. And Abbas Kazimi was a fantastic CBO working with JPM on this Tyk2 deal process. Many, many others at Nimbus have helped deliver us to this important milestone.
Golden Goose
Back in 2016, when Gilead acquired our ACC program for NASH by securing “Nimbus Apollo Inc”, we shared the view that our unique structure allowed Nimbus to be like the Golden Goose. With the sale of our “Nimbus Lakshmi Inc” subsidiary, Nimbus has once again demonstrated it can lay golden eggs.
Nimbus 3.0 is already well underway: HPK1 is in the clinic for immuno-oncology, and it’s followed by a broad and rich discovery pipeline of WRN, CBL-b, CTPS1, and AMPK – and that’s just the publicly-disclosed set of programs. Importantly, we’ve continued to augment and refresh the talent pool at Nimbus as part of building towards Nimbus 3.0 – most recently, we added Nathalie Franchimont as CMO to the team. We’re also very proud of our Nimbus alumni: every great company creates a diaspora of veterans growing into different roles, and Nimbus is no different.
For Atlas, Nimbus is one of the most successful biotech investments we’ve ever made on a number of metrics – including the fact that we’ve invested across four different fund vintages (2009, 2015, 2018, 2021) with long-term conviction in the story. And we have continued conviction going forward.
As we like to say at Atlas – “Onward and Upward” – congrats to the team and our new partners at Takeda on advancing ‘4858 as a potential best-in-class Tyk2 franchise. Looking forward to Nimbus 3.0!
