Rojas 2023: a personalised mRNA vaccine trained T cells against each patient's pancreatic cancer, and those who responded stayed cancer-free longer
Individualised mRNA vaccines encoding up to 20 of each patient's tumour mutations generated strong T-cell responses in half of pancreatic cancer patients after surgery, and those responders had far fewer relapses.
In this phase 1 study at Memorial Sloan Kettering, 16 patients with resected pancreatic ductal adenocarcinoma received atezolizumab followed by autogene cevumeran, an individualised uridine mRNA-lipoplex vaccine encoding up to 20 predicted neoantigens manufactured within about nine weeks of surgery, and then modified FOLFIRINOX chemotherapy. The vaccine expanded high-magnitude neoantigen-specific T cells in 8 of 16 patients. At a median follow-up of 18 months, responders had not reached median recurrence-free survival whereas non-responders had a median of 13.4 months (hazard ratio 0.08). Vaccine-induced T-cell clones were shown to be newly generated and, in a 2025 follow-up, persisted for years with continued separation of recurrence curves. A randomised phase 2 trial (IMCODE003) in the same setting is under way.
- 16 patients with resected pancreatic cancer; atezolizumab, then individualised mRNA neoantigen vaccine (up to 20 neoantigens), then mFOLFIRINOX.
- Vaccine manufactured and delivered within about 9 weeks of surgery in most patients.
- High-magnitude neoantigen-specific T-cell responses in 8 of 16 (50%).
- Median recurrence-free survival not reached in responders vs 13.4 months in non-responders; hazard ratio 0.08.
- Responding T-cell clones were de novo and persisted for up to several years in the 2025 follow-up.
Pancreatic cancer was thought to be immunologically inert because of its low mutation burden; this study showed that with the right vaccine platform its few neoantigens can still be targeted. It is the strongest human evidence so far that personalised cancer vaccines can generate durable, tumour-specific immunity, and it justifies the randomised trials now running in pancreatic cancer and melanoma. Benefit is not yet proven, because responders may simply have had more immunogenic tumours.
- Tiny, non-randomised study; the association between immune response and recurrence could reflect confounding.
- Half the patients did not mount a response, for reasons that are only partly understood (for example, splenectomy).
- Manufacturing an individual vaccine within weeks of surgery is expensive and logistically complex.
- Clinical benefit awaits the randomised IMCODE003 result.
Pages like this
not linked directly; found by shared links- IdeaPersonalised cancer vaccines at commodity cost through fully automated manufacturing
Shares BioNTech, Autogene cevumeran, Intismeran autogene, Neoantigen.
- IdeaPersonalised vaccines given only when the blood test turns positive
Shares BioNTech, Autogene cevumeran, Intismeran autogene, Neoantigen.
- Key paperKEYNOTE-942: a personalised mRNA cancer vaccine plus pembrolizumab after melanoma surgery
Shares INTerpath-001 (V940-001), Intismeran autogene, Tumour mutational burden (TMB), Neoantigen.
- InstitutionUniversity Cancer Center Mainz (UCT Mainz) / Universitätsmedizin Mainz
Shares BioNTech, Autogene cevumeran, Neoantigen, Personalised neoantigen (mRNA) vaccines.
- IdeaVaccines aimed only at mutations shared by every tumour cell
Shares BioNTech, INTerpath-001 (V940-001), Neoantigen, Personalised neoantigen (mRNA) vaccines.
- IdeaWhat makes a neoantigen actually immunogenic?
Shares Autogene cevumeran, Intismeran autogene, Neoantigen, Personalised neoantigen (mRNA) vaccines.
- PersonTimothy A. Chan
Shares Tumour mutational burden (TMB), Neoantigen, Melanoma.
- IdeaDe-acidify the tumour so T cells can work in it
Shares Cold tumours and the immunosuppressive microenvironment, No one can predict who responds to immunotherapy, Melanoma, Pancreatic ductal adenocarcinoma.