BioNTech, Roche End Phase 2 Trial of Personalized mRNA Cancer Vaccine; Shares Fall 9%

On August 28, BioNTech announced that it had formally terminated a key clinical trial of an autogene antigen vaccine therapy developed in collaboration with Roche, citing a lack of significant treatment efficacy. Following the announcement, BioNTech shares fell more than 9% at one point, while related stocks such as Roche ADRs, Moderna, and Merck also weakened.

The therapy, known as autogene cevumeran (BNT122/RO7198457), is a personalized mRNA cancer vaccine co-developed by BioNTech and Roche's Genentech. It works by identifying neoantigens based on specific mutations in each patient's tumor and customizing an mRNA vaccine accordingly. After vaccination, T cells are trained to recognize and attack cancer cells carrying these mutations.

The terminated Phase 2 trial, BNT122-01, enrolled patients with high-risk stage II or III colorectal cancer who had undergone surgical tumor removal but still had detectable circulating tumor DNA (ctDNA) in their blood. The drug had faced efficacy concerns as early as 2020: in an earlier study combining BNT122 with the PD-L1 inhibitor Tecentriq, only 9 of 108 evaluable patients achieved an objective response, an objective response rate of approximately 8%. In a small monotherapy cohort, the response rate was about 4%. Testing showed that approximately 77% of patients generated neoantigen-specific T-cell responses, but the actual rate of tumor shrinkage was far lower.

BioNTech disclosed that the study had crossed a futility boundary as early as October 2025. At that time, the independent Data Safety Monitoring Board (DSMB) determined that data maturity was insufficient to make a reliable judgment on efficacy, that follow-up time was inadequate to assess the primary endpoint, and that no safety concerns had emerged. BioNTech therefore decided to continue the trial according to the original protocol.

BioNTech noted that colorectal cancer is a biologically complex disease and is considered an immunologically "cold tumor," which generally responds poorly to immunotherapy and carries a high risk of metastatic recurrence. The company also stated that this failure does not affect the entire autogene cevumeran pipeline. Another Phase 2 trial, called IMcode003, will continue, studying autogene cevumeran in combination with immune checkpoint inhibitors and chemotherapy as an adjuvant treatment for pancreatic ductal adenocarcinoma (PDAC).