ASCO Plenary – Gilead and Arcus pedal hard with TIGIT

With the enthusiasm surrounding Gilead/Arcus’s anti-TIGIT MAb domvanalimab dwindling fast in lung cancer, the companies are pedalling hard in a second possible indication, stomach cancer. However, early data from the uncontrolled Edge-Gastric study suggest that much more needs to be done to convince investors.

The study will be presented in full at this afternoon’s ASCO Plenary session, so for now all the markets have to go on is the abstract and the companies’ press release. These show an underwhelming result for a TIGIT/PD-1/chemo triplet in front-line disease, leaving the companies clutching at the readout’s most positive aspect – remission rates in patients expressing high levels of PD-L1.

Though Edge-Gastric is a small, mid-stage trial, its results are important because they will be used to handicap Gilead/Arcus’s chances in the phase 3 Star-221 trial, where the same triplet is being compared against Opdivo plus chemo, in a similar front-line setting.

Multiple cohorts

Edge-Gastric in fact has five cohorts, two in front-line and three in second-line or later gastric cancer, and within these several different regimens are being tested.

The ASCO Plenary data, however, relate only to cohort A1, where domvanalimab was given in combination with the anti-PD-1 MAb zimberelimab plus FOLFOX chemo to 41 treatment-naive subjects. No information is provided for cohort A2, which tests just zimberelimab plus chemo in the same setting, and would therefore have shed important light on the relative contribution of domvanalimab to the triplet.

No matter, the headline overall response rate of 59% revealed for cohort A1 looks no better, on a cross-trial basis, than the 58% that Bristol Myers Squibb’s Opdivo plus chemo scored in its registrational Checkmate-649 study, as updated at ASCO-GI in January. (As an aside Keytruda plus chemo has a 16 December PDUFA date in this setting based on Keynote-859, which showed ORR of 51%.)

The good news for Gilead and Arcus comes from Edge-Gastric patients whose tumours expressed PD-L1 at 5% or greater, where the triplet has yielded an 80% ORR – numerically much higher than that in Checkmate-649’s PD-L1 ≥5% population. The clear downside is that for the Gilead/Arcus triplet this cut comprises just 15 patients.

ORRs in first-line gastric cancer: a cross-trial comparison

Note: PD-L1 status determined by tumour activity positivity in Edge-Gastric, and by combined progression score in the other two trials. Source: ASCO Plenary Nov 2023, ESMO Virtual Plenary Feb 2023, ASCO-GI 2023.

Of course Opdivo and Keytruda's registrational datasets are fundamentally based on overall survival, and Edge-Gastric is insufficiently mature to provide these data.

Gilead and Arcus are citing a landmark PFS analysis, saying 77% of patients were progression-free at six months, with medians not reached at 8.1 months’ follow-up. This might compare favourably against mPFS of 7.7 months in Checkmate-649, and 6.9 months in Keynote-859, but with relatively low patient numbers nothing can be said until the Edge-Gastric survival curves are revealed.

Domvanalimab is already facing an uphill battle in first-line NSCLC, where the Arc-7 study has shown a worsening dataset for a zimberelimab-containing doublet. Gastric cancer could offer hope for domvanalimab, for which Gilead paid Arcus $750m to exercise opt-in rights two years ago; this is especially true given the failure of BeiGene’s anti-TIGIT Mab ociperlimab in Advantig-203, though that trial was in second-line gastric cancer.

However, in TIGIT it’s Roche’s tiragolumab that’s generating the most interest, both in liver cancer and – unexpectedly – in NSCLC. Gilead and Arcus need much more than the early promise shown in Edge-Gastric.