← Back to Home

Preoperative Dual Immunotherapy Shows Pathological Responses in Colon Cancer; Phase 2 UNICORN Findings Await Validation in Larger Trials

Short-course dual therapy showed signs of substantial tumor regression in pMMR colon cancer, which has limited responses to immunotherapy. However, with only 14 patients per group, the study cannot yet establish whether this translates into a lower risk of recurrence.

By SURL BioNews

When colon cancer can be surgically removed, treatment aims not only to remove the existing tumor but also to reduce the chance of future recurrence. Could giving immunotherapy before surgery benefit cancers that typically respond poorly? The phase 2 UNICORN study, published in Nature Medicine on November 2, offers preliminary clues: after short-course dual therapy, some patients had substantially fewer viable cancer cells in their resected tissue, but the long-term benefits remain unconfirmed.

This multicenter, open-label study enrolled 56 patients with resectable, locally advanced colon cancer. Patients were divided into four groups of 14 according to proficient (pMMR) or deficient (dMMR) DNA mismatch repair status and treatment with monotherapy or combination therapy, without random assignment. Botensilimab, used in the study, is an Fc-engineered anti-CTLA-4 antibody; the combination groups also received the anti-PD-1 antibody balstilimab. The dual-therapy groups received one dose of botensilimab and two doses of balstilimab before surgery, which was originally scheduled for approximately day 35.

The most notable signal came from the pMMR population, which generally has greater difficulty benefiting from conventional immunotherapy. In the dual-therapy group, 5 patients, approximately 36%, achieved a major pathological response, defined as no more than 10% residual viable tumor in the resected tissue. Of these, 4 patients, approximately 29%, achieved a pathological complete response, with no viable cancer cells found in either the primary tumor or the lymph nodes. However, the 95% confidence interval for the 36% response rate ranged from 12.8% to 64.9%, indicating substantial uncertainty in the estimate because of the small sample.

Results also differed by molecular type: all 14 patients in the dMMR dual-therapy group achieved a major pathological response, including 13 who achieved a pathological complete response. By comparison, no patients in the pMMR monotherapy group achieved either endpoint. These findings support further investigation of the dual-therapy strategy, but because the groups were not randomly assigned, they cannot establish an efficacy advantage from adding balstilimab. A pathological complete response also does not mean that a patient has been cured.

Preoperative therapy must also clear another hurdle: side effects must not interfere with surgery. All 56 patients in the study ultimately underwent resection, with only one experiencing a 10-day delay because of grade 3 hyperthyroidism. Immune-related adverse events occurred in 29% of monotherapy patients and 54% of dual-therapy patients, while the overall rate of grade 3–4 immune-related adverse events was 2%. This supports the feasibility of surgery following short-course treatment, while also underscoring the risk of immune toxicity with dual therapy.

As of the February 9, 2026, data cutoff, neither dual-therapy group had reported recurrence, and both had a 12-month event-free survival rate of 100%. Median follow-up was 19.2 months in the pMMR dual-therapy group and 16.7 months in the dMMR dual-therapy group. Blood analyses showed that circulating tumor DNA was undetectable before surgery in all 5 pMMR patients who achieved a major pathological response and had evaluable samples. This exploratory finding can be followed in subsequent research, but it is not yet sufficient to demonstrate that the blood marker can reliably predict long-term recurrence.

The limitations of the current evidence remain clear: small group sizes, differing follow-up durations, and an almost entirely White study population limit the precision and generalizability of the findings. Agenus said the planned randomized phase 3 ROBBIN trial will compare preoperative dual therapy followed by standard treatment with standard treatment alone, using event-free survival as the primary endpoint. Whether deep responses seen in pathological specimens can translate into more durable disease control remains a question for large comparative studies of this kind.

References

  1. Agenus via BioSpace
  2. Agenus
  3. Targeted Oncology
  4. CancerNetwork