Researchers have published data from a randomized phase I trial of the individualized vaccine TG4050 against head-and-neck cancer. In the study, patients with operable HPV-negative head-and-neck squamous cell carcinoma received the vaccine either immediately after standard treatment or upon later relapse. After a median follow-up of 30 months, none of the 16 evaluable patients in the immediate treatment arm had relapsed, while 3 of 16 patients in the observation arm experienced relapse. An updated analysis at median 41 months confirmed this result, with no additional relapses in either arm. Results were published on August 17, 2026 in Nature Communications.
A personalized vaccine tailored to the tumor
TG4050 is an individualized neoantigen vaccine encoding up to 30 patient-specific predicted tumor neoantigens and delivered via a Modified-Vaccinia-Ankara virus vector. AI-guided algorithms predict neoepitopes; up to 30 target structures are selected based on expression, presentation, and likelihood of recognition by the patient's immune T cells. Successful manufacturing of the individual vaccine was achieved for 53 patients; for 3 otherwise eligible patients it was not successful.
Seventeen patients were assigned to the immediate arm, 16 to the observation arm, which provided TG4050 only upon relapse. The study was open-label, without blinding. The primary endpoint was safety; disease-free survival was recorded as a secondary endpoint. Exclusion followed pre-specified criteria for minimum exposure. The study authors themselves consider the sample size and number of relapses too small to conclude on TG4050 efficacy, pointing to the already-running phase II portion of the study.
In context: the signal at median 16.2 months
In 2024, study authors reported the same difference at a professional conference after median 16.2 months: no relapses in evaluable patients in the immediate arm, three relapses in the observation arm after 6.2, 8.8, and 18.5 months. The updated analysis confirmed this pattern at median 41 months follow-up, with no additional relapses occurring.
What happens in patients' T cells
In 11 of 15 patients from the immediate arm with available samples (73.3%), researchers detected T cell responses specific to tumor characteristics, with a median of 3 responding neoantigens per patient. CD8-positive T cells triggered by the vaccine remained detectable for up to 24 months, one year after the final dose.
Adverse events attributed to the vaccine remained mild to moderate
In the safety evaluation of 19 patients who received at least one dose, no dose-limiting toxicity and no severe adverse events attributed to the vaccine occurred. In 18 of 19 patients (94.7%), mild to moderate reactions occurred, mostly at the injection site.
Who stands behind the study
Transgene sponsored the study, with Transgene and NEC, developers of TG4050, jointly funding it. Bpifrance supported immunological analyses. Immunologist Olivier Lantz is a shareholder of Mnemo Therapeutics and Cereus; his laboratory works as a contractor for Transgene.
In April 2026, pharmaceutical publication PharmaTimes reported that Transgene had completed patient enrollment in the phase II portion of the study with 38 participants. Chief Executive Officer Alessandro Riva announced the primary endpoint should be available by the end of the first quarter of 2028. In the two patients from the observation arm who received TG4050 after relapse, sustained disease control was not achieved. The authors consider combination with checkpoint inhibitors scientifically plausible given the high PD-1 expression in neoantigen-specific CD8-positive T cells.