Agenus Inc. has reported that 63 of 123 heavily pretreated patients with microsatellite-stable metastatic colorectal cancer without active liver metastases survived for at least 18 months after receiving investigational botensilimab plus balstilimab in the Phase 1b C-800-01 study. Among those 63 patients classified retrospectively as long-term survivors, an estimated 59% were alive at three years, while 15 were alive and no longer receiving any anticancer treatment at their most recent follow-up.
The November 6 analysis, presented at the 2026 Society for Immunotherapy of Cancer Annual Meeting, adds detail to an already mature dataset rather than representing a new prospective efficacy trial. Across the full 123-patient cohort, previously reported peer-reviewed results showed median overall survival of 21.2 months, a 21% objective response rate and estimated three-year overall survival of 33%.
The distinction between those figures is important. The 59% three-year survival estimate applies only to patients who had already survived for at least 18 months and should not be presented as the survival rate for everyone entering the trial. The broader 33% three-year overall survival estimate is the more appropriate measure for understanding the outcome of the complete treated population.
Investors nevertheless responded strongly to the latest update. Agenus shares closed November 6 at $7.15, up 11.0% for the session, giving the clinical-stage biotechnology company a market capitalisation of approximately $322 million. The stock remains below its 52-week high of $10.31, but the move reflects renewed interest as Agenus prepares to take BOT+BAL into the global randomised Phase 3 ROBBIN programme in earlier-stage colon cancer.
Why is long-term survival unusual in this heavily pretreated MSS colorectal cancer population?
Microsatellite-stable colorectal cancer represents the majority of metastatic colorectal cancer and has historically responded poorly to conventional immune checkpoint inhibition. Unlike mismatch repair deficient or microsatellite instability-high disease, MSS tumours generally have biological characteristics that make them less responsive to PD-1 blockade alone, creating one of the more persistent limitations of immunotherapy in gastrointestinal oncology.
Patients enrolled in the Agenus cohort had already received a median of three previous treatment lines, with 67% having received at least three. The trial specifically enrolled patients without active liver metastases, an important selection criterion because metastatic site and tumour biology can influence immunotherapy outcomes and because liver involvement is common in advanced colorectal cancer.
In the full Phase 1b population, median overall survival reached 21.2 months and estimated survival at 36 months was 33%. Objective responses were confirmed in 21% of patients, while median duration of response had not been reached at the mature analysis despite median progression-free survival of only four months.
That apparent separation between relatively short median progression-free survival and prolonged survival in a subset of patients is one reason Agenus continues to focus on the durability of individual responses. It does not establish that BOT+BAL is responsible for every long survival event, however, because C-800-01 was a single-arm study without a randomised control group and overall survival was an exploratory endpoint.
What does the new SITC survivor analysis add beyond the previously reported three-year data?
The latest analysis divides the 123 patients retrospectively according to whether they survived for at least 18 months after beginning BOT+BAL. Sixty-three patients met that definition of long-term survival, while 60 survived for less than 18 months.
Among the 63 long-term survivors, 22 achieved a confirmed tumour response, equivalent to 35%, including three complete responses. Median duration of response had not been reached, responses had lasted at least 37.4 months in some patients, and 10 responses were still ongoing when the dataset was analysed.
Perhaps the most attention-grabbing observation is that 15 of the 63 long-term survivors, approximately 24%, were alive and off all anticancer therapy at their latest follow-up. Thirteen of those 15 had previously achieved a tumour response with BOT+BAL, suggesting that durable tumour control can persist in a subset after active treatment has ended.
The result is clinically intriguing but statistically easy to overinterpret. By definition, the analysis selects patients based on having already achieved a favourable survival outcome, so characteristics identified within that group cannot automatically be considered predictive biomarkers or proof that the treatment produced the long-term survival.
Does the analysis reveal which patients are more likely to benefit from BOT+BAL?
Agenus reported that long-term survivors did not cluster neatly around one obvious favourable clinical profile. Performance status, prior immunotherapy exposure, metastatic disease characteristics and several prognostic biomarkers were broadly similar between long-term and shorter-term survivors.
One difference did emerge around previous treatment exposure. Fifty-six per cent of long-term survivors had received three or more previous therapy lines compared with 78% of patients who survived for less than 18 months, suggesting that treatment earlier in the disease course may be associated with better outcomes.
That observation supports Agenus’ strategic decision to move BOT+BAL into earlier-stage disease, but it remains hypothesis-generating. Patients who receive fewer prior therapies may differ in tumour burden, biology, general health, treatment sensitivity and other variables that themselves influence survival.
The survivor analysis therefore does not yet provide a validated patient-selection tool. Its stronger contribution is showing that durable outcomes were not confined exclusively to a small group defined by one conventional clinical feature, creating justification for testing the combination prospectively in a broader, carefully controlled population.
Why is the exclusion of active liver metastases critical when interpreting the results?
The C-800-01 metastatic colorectal cancer cohort was restricted to patients without active liver metastases. That detail materially narrows the population to which the findings can reasonably be applied because the liver is one of the most common sites of metastatic spread in colorectal cancer.
Liver metastases have also been associated with a distinct immune environment that can make systemic immunotherapy less effective in several tumour types. Agenus is therefore developing BOT+BAL in a biologically selected metastatic population rather than claiming that the current results apply to all patients with refractory MSS colorectal cancer.
The restriction could be clinically useful if it identifies a subgroup in which immune-based treatment has greater potential. It also means the survival results cannot simply be extrapolated to patients with active hepatic disease, who may represent a substantial proportion of the broader metastatic population.
Any comparison with other later-line colorectal cancer treatments must take this selection into account. Differences in liver involvement, performance status, previous therapies and subsequent treatments can strongly affect survival, making historical comparisons informative for context but insufficient to establish superiority.
How does botensilimab differ from conventional CTLA-4 checkpoint inhibitors?
Botensilimab is an investigational Fc-enhanced anti-CTLA-4 antibody designed to modify several parts of the anticancer immune response. Agenus has developed the molecule with the aim of enhancing T-cell priming, reducing regulatory T-cell activity within tumours and activating components of the innate immune system.
Balstilimab is an anti-PD-1 antibody designed to block the interaction between PD-1 and its ligands, thereby reducing an inhibitory signal that can restrain T-cell activity. Combining the molecules is intended to attack immune resistance through complementary mechanisms rather than relying on PD-1 blockade alone.
The biological strategy is particularly relevant in MSS colorectal cancer because conventional checkpoint therapy has generally produced limited benefit in this immunologically resistant setting. The 21% response rate and durable responses observed in C-800-01 provide clinical evidence that the combination has antitumour activity in selected patients.
Mechanistic differentiation does not establish clinical superiority, however. Botensilimab’s Fc engineering may help explain observed immune effects, but only appropriately controlled clinical trials can determine whether those properties translate into better patient outcomes than currently available treatment strategies.
What does the safety profile show after longer BOT+BAL follow-up?
In the full 123-patient Phase 1b cohort, the most common treatment-related adverse event was diarrhoea, occurring in 39% of patients, with Grade 3 or higher diarrhoea reported in 8%. Fatigue occurred in 37%, including Grade 3 or higher events in 2%.
The SITC survivor analysis found that immune-mediated adverse events were more common among long-term survivors than in the overall cohort. Immune-mediated diarrhoea or colitis was the most frequently reported event in that group, and 33 of 34 affected long-term survivors experienced resolution, with a median resolution time of 13 days.
Management included corticosteroids and tumour necrosis factor-directed therapy when considered appropriate. Agenus reported no treatment-related deaths in the analysis, while patients in both the overall cohort and the long-term survivor group received a median of only two botensilimab doses.
The association between immune toxicity and longer survival should not be interpreted as showing that toxicity causes benefit. Immune-related adverse events can correlate with treatment exposure or immune activation in some checkpoint inhibitor studies, but the current post-hoc analysis was not designed to establish such a causal relationship.
Why is Agenus moving BOT+BAL from metastatic disease into treatment before surgery?
Rather than conducting its next pivotal programme in the same heavily pretreated metastatic population, Agenus is shifting the central registrational strategy toward high-risk Stage II and Stage III MSS colon cancer before surgery. The planned ROBBIN trial is designed to test whether short-course neoadjuvant BOT+BAL can reduce recurrence risk when the immune system, primary tumour and tumour-draining lymph nodes remain intact.
The strategy is supported by separate Phase 2 studies rather than the metastatic survivor analysis alone. Agenus has reported that the NEST and UNICORN studies each produced pathological complete responses in approximately 30% of MSS colon cancer patients receiving short-course BOT+BAL before surgery.
Those neoadjuvant studies also reported that surgery generally remained feasible on schedule. Agenus said no colon cancer recurrences had been observed in the BOT+BAL cohorts at the respective data cut-offs, although follow-up remains relatively limited and absence of recurrence in small early-stage cohorts cannot substitute for a randomised event-free survival result.
The central development thesis is therefore shifting from rescuing heavily pretreated metastatic disease toward preventing recurrence in patients treated with curative intent. That is potentially a much larger clinical opportunity, but it also creates a considerably higher evidentiary bar because patients with resectable disease may already have favourable outcomes with surgery and established perioperative treatment.
What will the 850-patient ROBBIN Phase 3 trial need to prove?
Agenus expects ROBBIN to enrol approximately 850 patients with high-risk Stage II or Stage III MSS colon cancer across multiple international regions. The trial is designed as a randomised registrational study, with event-free survival as the primary endpoint.
First patient enrolment is targeted for the first quarter of 2027. Agenus expects an interim pathological response assessment involving the first 100 enrolled patients during the second half of 2027, but pathological response is not the same as the trial’s ultimate event-free survival objective.
This distinction is particularly important because the Phase 1b metastatic data and Phase 2 neoadjuvant studies have generated several attractive response and survival signals without yet establishing the randomised evidence required for a broad U.S. approval. ROBBIN is intended to make that transition.
The trial will also need to demonstrate that adding immunotherapy before surgery does not create unacceptable immune toxicity, surgical delays or complications that offset any reduction in recurrence risk. In curative-intent disease, the tolerance for treatment-related harm is different from that in patients with refractory metastatic cancer.
How do Armenia’s BOT+BAL authorisations fit with the U.S. regulatory position?
Agenus achieved an unusual milestone in October when Armenia granted the first marketing authorisations anywhere in the world for botensilimab and balstilimab used together in previously treated adults with MSS metastatic colorectal cancer without active liver metastases. Those approvals provide a commercial pathway in Armenia but do not constitute approval by the U.S. Food and Drug Administration or other major regulators.
In the United States, both botensilimab and balstilimab remain investigational. The company’s primary U.S. registration strategy is now centred on ROBBIN rather than treating Armenia’s authorisation as evidence that broader regulatory approval is assured.
Agenus has also generated $17 million of pre-commercial BOT+BAL product revenue during the first nine months of 2026 through authorised patient-access programmes in several countries. Those revenues show that some physicians and patients are seeking access to the combination, but compassionate or named-patient programmes are not equivalent to broad commercial adoption.
The geographic distinction will be particularly important as investor attention grows. An authorised treatment pathway in one jurisdiction can support revenue and real-world experience while the evidence standards, indications and review processes in other countries remain separate.
Can Agenus finance ROBBIN through the point where Phase 3 evidence begins to emerge?
Agenus ended September with $86.9 million in cash, cash equivalents and short-term investments, up from $18.7 million at the end of June after completing an oversubscribed private placement. The July transaction generated approximately $85 million of upfront gross proceeds and included milestone-linked warrants capable of providing up to another approximately $255 million if fully exercised.
Management expects the upfront financing to support ROBBIN initiation, regulatory activities and operations through the third quarter of 2027. The company has said full exercise of the warrants could extend funding for the planned ROBBIN programme and corporate operations through the end of 2031, although warrant proceeds remain contingent rather than guaranteed capital.
Agenus reported third-quarter total revenue of $31.5 million and net income of $2.8 million, while operating income was positive for a third consecutive quarter. Most revenue still includes non-cash royalty income, so headline profitability should not be interpreted as evidence that the company is funding late-stage development entirely through recurring commercial cash flow.
Financing therefore remains closely connected to clinical execution. Reaching enrolment and interim milestones could unlock additional capital through existing warrants, while delays or disappointing data could make that funding structure less dependable.
Why did AGEN stock rise 11%, and what is the next result that really matters?
Agenus closed November 6 at $7.15, up 11.0% from $6.44, after trading as high as $7.20. Approximately 874,000 shares changed hands during the session, while the company’s market capitalisation was about $322 million and its 52-week trading range stood at $2.71 to $10.31.
The latest rise follows a year in which investor sentiment has improved substantially as BOT+BAL generated longer follow-up, peer-reviewed publications, its first national marketing authorisations and a clearer Phase 3 strategy. The stock nevertheless remains volatile because the company’s valuation is still heavily dependent on successful development of a relatively small number of oncology programmes.
The SITC analysis adds evidence that some patients receiving BOT+BAL can experience unusually durable outcomes, but it does not settle the central investment or clinical question. The strongest survival figures come from a retrospectively defined group of patients who had already survived for at least 18 months, while the underlying trial was single-arm and was not designed to prove a survival advantage against standard treatment.
ROBBIN is therefore the more consequential test of the Agenus thesis. If a large randomised trial shows that short-course BOT+BAL before surgery meaningfully reduces recurrence in high-risk MSS colon cancer without compromising surgical care or creating disproportionate immune toxicity, the programme would move from intriguing durable-response evidence into a much stronger regulatory and clinical position. Until that result exists, the latest long-term survivor data are best viewed as a compelling signal that justifies the Phase 3 experiment rather than proof that the experiment has already succeeded.
