Crunch time for Bristol’s TIGIT bet

As if upcoming results of key trials of Roche’s tiragolumab and GSK/iTeos’s belrestotug weren’t enough, the second half of this year could also see the first human data from another anti-TIGIT project, Bristol Myers Squibb’s BMS-986442.

It’s surprising that, despite BMS-986442 having completed one phase 1 trial and started a mid-stage study, no human data on it have yet been made available. As such the first readout will provide the first clues about the validity of the scientific approach used, as well as suggesting whether Bristol was right about doubling down on this asset rather than its own anti-TIGIT MAb.

Before Bristol licensed BMS-986442, then known as AGEN1777, from Agenus in 2021 it had been working on an anti-TIGIT project known as renvistobart. But that was scrapped last September as the company went all in on BMS-986442, a MAb with two distinct features: it has an enhanced Fc region, and it hits the CD96 protein as well as TIGIT.

Opdivo combo

Now investors are awaiting the first data from a phase 1/2 trial, in which BMS-986442 is being combined with Opdivo, with or without chemotherapy. The initial tumours of interest are NSCLC and gastric cancer, and Bristol has guided to a 2024 readout.

This uncontrolled study actually has a somewhat complex design, with multiple cohorts testing different chemo regimens in addition to the MAb doublet. However, it’s unclear what cancers each cohort will test, and the only other tumour type known to be included in it is head and neck.

Summary of BMS-986442's phase 1/2 trial

Note: efficacy endpoints include 6mth & 12mth ORR & PFS. Source: OncologyPipeline.

It’s noteworthy that when Bristol licensed AGEN1777 from Agenus the project was still preclinical, making the $200m up-front fee seem generous. Still, back then the TIGIT mechanism was rather hotter than it is now that the sector has experienced the disappointments of Roche’s tiragolumab and Gilead/Arcus’s domvanalimab. Only last week, Roche discontinued the Skyscraper-15 and Skyscraper-05 tiragolumab trials following the recent failure of Skyscraper-06.

For Agenus, which has just been hit by an FDA refusal to endorse an accelerated filing plan for its lead project, the botensilimab/balstilimab combo, the prospect of success with BMS-986442 offers little solace; in May the group mortgaged away part of its royalty stream due from Bristol, signing a financing deal with Ligand Pharmaceuticals that left Agenus mostly focused on bot-bal.

Before that BMS-986442 completed phase 1 dose escalation in solid tumours (dosing the first patient triggered a $20m milestone to Agenus) and started the phase 1/2 Opdivo combo study that is now due to read out, whose first patient dosing triggered a $25m milestone. 

Despite all this, no actual human data appear to have been disclosed for BMS-986442. Agenus did present a poster at this year’s AACR, but that focused on preclinical backing for the molecule’s mechanistic approach. Whether such backing can be repeated in humans will be one focus of Bristol’s first clinical data.

The logic behind BMS-986442 is that CD96 (also known as “Tactile”) and TIGIT share CD155 as a ligand, and are thought to be complementary targets in the CD226 axis that negatively regulate T and NK cell function in the tumour microenvironment. Bristol claims that BMS-986442 could enhance the “quality and magnitude” of T-cell responses through dual inhibition in antigen-presenting as well as tumour cells.

Fc enhanced or silenced?

But another feature of BMS-986442 is its use of an enhanced Fc region, on the grounds that interaction with the Fcγ receptor is critical for tumour control, and that, it’s argued, anti-TIGIT MAbs with a conventional Fc region lack single-agent activity.

Things haven’t been simplified by the fact that at least three other anti-TIGIT MAbs are pursuing the opposite approach, employing an Fc-silent design. One is domvanalimab; Arcus/Gilead claim that eliminating Fc activity could prevent unwanted killing of TIGIT-expressing cancer-killing cells, and of T regulatory cells that control immune responses against healthy tissue.

The other is Compugen's Fc-silent anti-TIGIT project COM902, which is relevant as it’s the basis of AstraZeneca’s anti-TIGIT x PD-1 MAb rilvegostomig. The Astra bispecific is now in three phase 3 studies, and as more data are generated they could at last shed light on the best biological approach.