NEW YORK – Arcus Biosciences is pushing forward with a Phase III trial of domvanalimab, an anti-TIGIT antibody, and zimberelimab, an anti-PD-1 drug, with chemotherapy in the frontline setting for patients with PD-L1-positive non-small cell lung cancer after seeing promising overall survival outcomes and safety in a Phase II study.
At the Society for Immunotherapy of Cancer's (SITC) annual meeting this week, Arcus presented data from the ARC-10 trial, a three-arm randomized study comparing frontline treatment with domvanalimab-zimberelimab against zimberelimab monotherapy against platinum–doublet chemotherapy alone. The trial included 95 patients who had advanced or metastatic NSCLC with a PD-L1 tumor proportion score (TPS) of 50 percent or higher.
At the meeting, researchers reported that median overall survival was not reached at more than two years' follow-up for patients on domvanalimab-zimberelimab, besting the comparator arms. This has given Arcus confidence in the outcome of an ongoing Phase III STAR-121 study, in which the firm is pitting domvanalimab-zimberelimab-chemo against an arm testing Merck's checkpoint inhibitor Keytruda (pembrolizumab) plus chemo and another arm assessing zimberelimab plus chemo.
Earlier this year, Arcus and Gilead, which are codeveloping the anti-TIGIT program, decided to refocus domvanalimab's development to go after a potentially larger patient population than the one in ARC-10. Arcus CEO Terry Rosen noted on a call to discuss the Phase II trial data with investors on Wednesday that the firm hopes to further expand the treatment population for the combination in the STAR-121 trial by using a more generous PD-L1 positivity threshold — a TPS score of greater than 1 percent — versus the more limited PD-L1-high group enrolled in the ARC-10 trial.
"We stopped ARC-10 for strategic reasons based on an evolving market where physicians are moving from just using Keytruda to using Keytruda plus chemo even in high PD-L1 patients," Rosen said during the call. He added that early data from another Phase III trial of the domvanalimab-zimberelimab-chemo combination in gastric cancer has shown that the chemo-containing regimen is well tolerated by patients.
Rosen also said the company has "optimized [the domvanalimab-zimberelimab studies] for probability of success, also while addressing the largest number of patients." The firm is aiming to expand its potential market for the domvanalimab-zimberelimab combo in first-line NSCLC. With the wider PD-L1 criteria in STAR-121, Arcus expects the drug could be an option for more than 300,000 patients worldwide, with a market potential of about $10 billion, if approved.
In terms of efficacy, in ARC-10, the addition of the anti-TIGIT antibody domvanalimab to the checkpoint inhibitor zimberelimab improved progression-free survival, overall survival, and objective response rates in patients compared to those on just zimberelimab. A smaller number of patients in the trial received chemo alone, as it was used as a control arm in countries where checkpoint inhibitor monotherapy is not yet standard of care.
At a median follow-up of 24.5 months, median overall survival was not reached in the domvanalimab-zimberelimab arm. Median overall survival was 24.4 months in the zimberelimab monotherapy arm and 11.9 months in the chemo arm. Twelve-month overall survival rates were 68 percent with domvanalimab-zimberelimab, 57 percent for zimberelimab, and 50 percent for chemo.
Median progression-free survival was 11.5 months on domvanalimab-zimberelimab versus 6.2 months on just zimberelimab and 9.6 months on chemo. In ARC-10, 44.7 percent of those on domvanalimab-zimberelimab saw their tumors shrink, compared to 35 percent on zimberelimab and 35 percent on chemo.
"The results are supportive of domvanalimab's ability to meaningfully extend survival in patients and the zimberelimab control arm performed right in line with benchmark studies of pembrolizumab in this setting, with a median overall survival of 24.4 months," said Arcus Chief Medical Officer Dimitry Nuyten, adding that the other endpoints reported from ARC-10 were also comparable to benchmark studies of Keytruda, the current standard of care in this setting.
Arcus also highlighted the safety and tolerability of the domvanalimab-zimberelimab combo. Rosen noted that other anti-TIGIT molecules have reported serious safety concerns but domvanalimab does not appear to have these toxicities.
The difference, explained Nuyten, is that domvanalimab is an Fc-silent anti-TIGIT antibody, while other anti-TIGIT antibodies in development are Fc-enabled drugs. Fc-silent drugs are designed to reduce their interaction with the immune system by not binding to Fc receptors, but Fc-enabled drugs do interact with Fc receptors, leading to activation of the immune system and immune-related adverse events.
Rosen added there are only two Fc-silent anti-TIGIT antibodies currently in development, domvanalimab and AstraZeneca's anti-TIGIT and anti-PD-1 bispecific antibody rilvegostomig. AstraZeneca is also evaluating rilvegostomig in first-line NSCLC and first-line gastric cancer. In Rosen's view, based on recent data on rilvegostomig, its efficacy and safety profile is comparable to the domvanalimab-zimberelimab combination.
In ARC-10, the combination arm did not show any safety concerns compared to the zimberelimab arm or the chemo arm. Immune-mediated adverse events were similar among those on the combination, 24 percent, and those on just zimberelimab, 20 percent. Immune-related events were not reported for the chemo arm. Grade 3 treatment-related adverse events occurred in 21 percent of patients on domvanalimab-zimberelimab, compared to 15 percent on zimberelimab and 47 percent on chemo. The two investigational arms also had similar rates of treatment discontinuation due to adverse events, 10.5 percent in the domvanalimab-zimberelimab arm and 7.5 percent in the zimberelimab arm.
One patient died due to an adverse event in the domvanalimab-zimberelimab arm, four patients died in the zimberelimab arm, and two patients died in the chemo arm.
"These [safety] results are substantially different from data reported from studies with Fc-enabled anti-TIGIT antibodies, where immune-mediated adverse events and treatment interruptions and discontinuations have been meaningfully higher than the anti-PD-1 or anti-PD-L1 alone arm," Nuyten said. "In fact, [safety] has been cited as the cause of trial failures [for Fc-enabled anti-TIGIT antibodies]."
This year alone, sponsors stopped two late-stage anti-TIGIT antibody trials due to adverse events or lack of efficacy. Merck, for example, stopped the Phase III KeyVibe-008 trial of its anti-TIGIT antibody vibostolimab, Keytruda, and chemo in August due to lack of survival benefit and an increased rate of adverse events on the combo compared to Roche’s checkpoint inhibitor Tecentriq (atezolizumab) plus chemo. Genentech also ended a Phase III study of its anti-TIGIT drug tiragolumab with Tecentriq and chemo in July due to lack of survival benefit compared to Keytruda and chemo.
Nuyten added that the efficacy and safety profile observed in ARC-10 "increases our confidence that combining domvanalimab and zimberelimab with chemotherapy in STAR-121 will further improve outcomes and provide lung cancer patients with the best chance of success in their first-line treatment."
Arcus does not yet have an estimate for when results will read out from the STAR-121 trial. Rosen said the company will provide more details on that timeline in 2025.
Beyond the STAR-121 trial, Arcus is also evaluating domvanalimab combinations with zimberelimab and chemo in an all-comer gastric cancer patient population and with AstraZeneca's immunotherapy Imfinzi (durvalumab) in NSCLC.