ESMO Congress 2026

The broader route of educating the immune system with antigens, to which personalized mRNA cancer vaccines belong, already has a multi-layered evidence base in the literature: studies of cancer immunoediting show the immune system can eliminate and shape tumour immunogenicity, checkpoint immunotherapy can produce durable responses in some previously refractory patients, and both spontaneous and therapy-induced anti-tumour responses require tumour-antigen-specific CD8+ and CD4+ T cells, with MHC-II-restricted neoantigens needing to be expressed at sites of successful rejection ^[2] — the core rationale for neoantigen vaccine design; in melanoma, the intensity of tumour-infiltrating lymphocyte infiltrate is believed to correlate with outcome, and insight into TILs has driven immune strategies including blocking antibodies ^[6]; on the engineering side, nanoparticles carrying the full array of cancer cell membrane antigens coupled with an adjuvant can induce tumour-specific immune responses in preclinical settings ^[1], while delivery systems such as polymeric nanoparticles, despite fine control over size, charge and surface functionality, still see minimal broad clinical application ^[3].

The only numbered readout on the ground at ESMO26 (Madrid): Merck and Moderna reported that in a phase 3 melanoma trial, adding an experimental mRNA vaccine to immunotherapy reduced the risk of melanoma recurrence or death by 42% compared with immunotherapy alone, with Memorial Sloan Kettering Cancer Center (MSK) among the participating institutions. Another post from @sevincballi gave no numbers, only the remark “Maybe we found new questions. But we’re closer than before.”

This readout pushes the CD8+/CD4+ dual, neoantigen-driven immune logic depicted in ^[2] to efficacy numbers at the phase 3 trial level, and continues melanoma’s role as the validating tumour type for immunotherapy ^[6]. The same post also mentions a second, earlier-stage thread: David Hoyos, a researcher in the Benjamin Greenbaum Lab at MSK and a 2024 Hertz Fellow, outlined at the Hertz Summer Workshop his approach of teaching the immune system to recognize mutant p53, focused on Li-Fraumeni syndrome — an inherited condition that greatly increases the risk of several cancers including breast cancer — exploring whether mRNA vaccines could one day help prevent or delay breast cancer in these women, and hoping to extend mutant p53 targeting to a much broader range of cancers. In antigen strategy this is a shared tumour-antigen route, akin to the Tn antigen found in a majority of human carcinomas ^[5], a different track from the personalized recurrence-prevention vaccine reported at the meeting.

The limitations are equally clear: the post gives only the 42% relative risk figure; sample size, follow-up duration, absolute benefit, overall survival and safety are undisclosed, and the specific drug names and trial identifiers remain to be confirmed; Hoyos’s p53 direction is an early-stage could-one-day exploration with no clinical data. Delivery platforms such as polymeric nanoparticles still face a long road from bench to broad clinical use ^[3], a real constraint for subsequent translation in this field.