Development

Agenus emphasizes positives after Phase II miss in PD-1 refractory gastroesophageal cancer

Edited by Robert McTiernan··3 min read

Agenus Inc. (Nasdaq: AGEN) reported Phase II data from an investigator-initiated trial combining botensilimab, balstilimab, and agenT-797 in PD-1 refractory gastroesophageal cancer, showing a 77% disease control rate and 43% survival at 18 months in patients who received an induction strategy — though the trial missed its primary endpoint of objective response rate.

Trial specifics

The NCT06251973 trial is a Phase II, investigator-initiated study conducted at Memorial Sloan Kettering Cancer Center, enrolling patients with advanced PD-1 refractory gastroesophageal adenocarcinoma who had progressed after frontline therapy.

Across 17 enrolled patients, the induction arm — in which patients received agenT-797 alone or with botensilimab and balstilimab before the full combination — produced a median PFS of 6.9 months versus 3.5 months without induction (HR 0.19; p=0.015). Median OS was 9.5 months in the induction cohort versus 5.2 months in the non-induction cohort, with 43% of induction-treated patients alive at both 12 and 18 months compared with 20% and 0%, respectively. A subset of patients survived beyond 20 months. The trial did not meet its primary endpoint of objective response rate. The safety profile was consistent with the component agents.

The data carry substantial caveats. Seventeen patients across two unequal arms is a small sample from which to draw efficacy conclusions, and the trial's failure to meet its pre-specified primary endpoint tempers the survival observations. Cross-trial comparisons are limited by differences in patient populations, prior treatment histories, and study designs. Still, the survival curve separation between the induction and non-induction arms — and the mechanistic correlates accompanying it — gives the data a degree of biological coherence that purely numerical results would not.

Research context

Botensilimab is an Fc-enhanced anti-CTLA-4 antibody designed to engage both innate and adaptive immune responses more broadly than conventional CTLA-4 inhibitors. Its Fc engineering amplifies FcγR-mediated effector functions, enabling Treg depletion within the tumor microenvironment and myeloid cell activation alongside T-cell priming. Balstilimab is a standard IgG4 anti-PD-1 antibody. AgenT-797 is an allogeneic iNKT cell therapy derived from donor-derived invariant natural killer T cells, intended to provide innate-like antitumor immune activity. The combination pairs three distinct immunological mechanisms with the established second-line backbone of ramucirumab and paclitaxel.

The AllSci BriefSystematic R&D and deal news. Daily.

The post-PD-1 refractory gastroesophageal setting is one of the more constrained areas of oncology. Once patients progress on a PD-1-containing regimen — now standard in the first-line — the approved options revert to anti-angiogenic and cytotoxic strategies. Ramucirumab plus paclitaxel, approved by the FDA in 2014, remains the standard second-line treatment for unselected patients. Fam-trastuzumab deruxtecan (Enhertu) is available for the HER2-positive minority.

Gastroesophageal adenocarcinomas are predominantly microsatellite stable and immunologically cold — characteristics that have historically predicted poor responses to checkpoint inhibition. Botensilimab's Fc-enhanced design is specifically intended to overcome this biology by engaging myeloid cells and inducing tertiary lymphoid structure formation, which the correlative data from this trial documented in at least one patient with prolonged benefit. Intratumoral T-cell and dendritic cell infiltration and peripheral CD4 and CD8 T-cell activation were also observed. These findings provide a mechanistic rationale for the survival differences between arms, though they require validation in larger cohorts before they can be treated as predictive biomarkers.

Agenus has reported clinical activity for botensilimab and balstilimab across nine metastatic, late-line cancer types, with approximately 1,200 patients treated across Phase I and Phase II trials. The gastroesophageal program is one of several indications under evaluation, and the company has indicated that ongoing biospecimen analysis from this trial is expected to yield further insight into optimal sequencing and potential patient selection markers. Additional analysis of the full dataset is ongoing.

Spot something wrong? Report an issue with this article