Direct identification of specific tumour-derived HLA-bound peptide targets is enabling the development of personalised immunotherapy strategies
Initially promising research in cancer vaccination has failed time and again to translate into an approved cancer treatment. Immunopeptidome analysis enables peptides to be effectively ‘washed’ directly from the tumour cell surface to provide specific information about the antigens presented by a tumour to the immune system. The latest tools enable precise selection of key antigens for personalised vaccine development, and three main strategies are being explored for the clinical development and application of peptide-based cancer vaccines: ‘off-the-shelf’ vaccines based on common human leukocyte antigen (HLA) allotypes or highly frequent tumour antigens; pre-manufactured immunopeptidome-defined ‘warehouse’ antigens that may be used to assemble multi-peptide, patient-tailored products; and personalised vaccines produced in real time based on a patient’s tumour mutational profile and tumour-specific peptides.
An ongoing challenge in cancer vaccine development is determining the optimal timing of therapy. Many of the early vaccine studies were conducted in patients with advanced disease and a significant tumour burden, and outcomes were not improved substantially. Subsequent trials have demonstrated that cancer vaccines are most impactful when given in the adjuvant setting where they can induce continuous immune stimulation and durable T-cell responses (Signal Transduct Target Ther. 2025;10:107). In later-line settings, patients are most likely to benefit from a combinatorial approach, not only with immunotherapies, but also small-molecule targeted therapies that can work synergistically with cancer vaccines by inducing antigens – a previously unknown mechanism of these agents.
Data from recent research in the field are encouraging. Fusion-VAC-XS15 is an off-the-shelf, peptide-based vaccine developed for patients with fibrolamellar hepatocellular carcinoma (FL-HCC), a rare form of liver cancer that typically affects children and young adults and is associated with a poor prognosis (Nat Commun. 2022;13:6401). The vaccine targets a tumour-specific HLA presented peptides from the FL-HCC oncogenic driver, DNAJB1-PRKACA fusion transcript and its immunotherapeutic application in a single FL-HCC patient resulted in persistent DNAJB1-PRKACA-specific T-cell responses and durable relapse-free survival (RFS) >5 years post vaccination (Nat Commun. 2022;13:6401). Clinical trials of the fusion-based vaccine in combination with immune checkpoint inhibitors are ongoing in the locally advanced/metastatic setting (NCT05937295) and in the adjuvant setting as a single agent for the prevention of relapse after surgery (NCT06789198).
Using HLA allotyping and immunopeptidome analysis of individual patients with chronic lymphocytic leukaemia (CLL), our group created iTAC-XS15-CLL01, a personalised warehouse-based multipeptide T-cell activator. A recently completed phase I trial of iTAC-XS15-CLL01 provided proof of principle that this approach can be successfully applied in CLL; T-cell responses targeting multiple peptides were induced in 19/20 patients (95%) and persisted in 16 patients (84%) at 6 months of follow-up without any significant toxicity (Lancet Haematol. 2026;13:e74–e85). Several trials are now underway following this warehouse principle in other tumour types, including pancreatic cancer.
Personalised neoantigens are unique mutated peptides predicted based on a patient’s tumour genome mutation profile and they vary between patients and tumours, making them a significant source for developing personalised neoantigen-based cancer vaccines, such as autogene cevumeran. The vaccine was produced in real time from surgically resected pancreatic ductal adenocarcinoma tumours, and a maximum of 20 neoantigens were administered per patient as adjuvant treatment in combination with atezolizumab and chemotherapy in a phase I trial (Nature. 2023;618:144–150). A significantly prolonged RFS was demonstrated at 18 months of follow-up in patients who showed evidence of vaccine-expanded T-cell activity compared with those without T-cell activity (median RFS not reached versus 13.4 months, respectively; p=0.003). A programme at the University of Tübingen exploring the activity of personalised vaccines based on a combination of immunopeptidome analysis and tumour genome sequencing is also showing promise across multiple solid tumours (Association for Cancer Immunotherapy Annual Meeting 2026;Abstract 58).
