An exhaustive list of programs surfaced by Sleuth from patents, company websites, news / press releases, conference abstracts and publications.

Plus some thoughts on how this opportunity will evolve in 2025 and beyond:

While the data Summit Therapeutics presented in Oct'24 was stellar, we've seen a ton of examples over the years of IO combo agents initially looking great, only to later fail in a pivotal, global Phase 3 (i.e. Roche's SKYSCRAPER studies measuring PD-(L)1 x TIGIT)

And while there is evidence Asian populations respond better to VEGF therapies, there is good reason to be optimistic that this data will convert:

• The performance of key assets (Summit's ivo and BioNTech's BNT-327) is consistent across tumors and subgroups, and control arm performance lines up with what we'd seen in prior global studies (like KEYNOTE-042).
• These agents have demonstrated monotherapy activity unlike some prior failed IO targets
• The degree and durability of separation in the HARMONi-2 PFS curve does bode well for a stat sig OS result - mOS and mPFS are historically well correlated in 1L IO trials

The market opportunity ahead is huge - arguably one of the most compelling across all of biopharma, behind GLP-1s / obesity - given these and other emerging combos could displace PD-L1s across the board, which are expected to generate >$60B by 2027. The ultra-bull case is they would actually go far beyond that, improving performance in historically poor-performing subgroups such as PD-L1 low and expanding into tumors unresponsive to current IO combos.

Unsurprisingly, we're seeing a gold rush, with leading players rapidly enrolling multiple Ph3 studies, benefitting from the wealth of data Merck, BMS and others generated as to the performance of PD-L1 monotherapies in different tumors and subgroups to guide their clinical development plans.

While the below landscape covers PD-(L)1 x VEGF, we're also seeing an explosion in other PD-(L)1 bispecific approaches, such as with CTLA-4, LAG3 and dozens of other IO targets (AI + 100M biopharma docs to automate CI for you), I compiled landscapes for all of these, including dozens of early-stage companies in China that aren't on many radars yet, mined from proprietary sources. Aggregating this usually takes hours, but with Sleuth I was able to generate landscapes in seconds (+ data benchmarking, BD predictions and more).