TL;DR: Key Takeaways
- Intismeran autogene (V940 / mRNA-4157) is the more advanced programme by trial stage
- EVX-01 has the clearest published link between AI-predicted targets and T-cell responses
- Different populations and designs — no head-to-head trial exists
- Key figures on both sides are company disclosures, not peer-reviewed detail
- Both remain investigational; neither is commercially approved
Side by side
The table below reflects public information as of 22 August 2026. It is a summary of what each programme has disclosed, not a comparison of effectiveness.
| Dimension | Intismeran autogene (V940 / mRNA-4157) | EVX-01 |
|---|---|---|
| Developer | Moderna and Merck | Evaxion |
| Format | Personalised mRNA vaccine | Personalised peptide vaccine |
| Population studied | Melanoma, adjuvant setting after surgical resection | Advanced melanoma |
| Computational selection | Developer describes integrated AI algorithms processing tumour and blood sequencing, predicting up to 34 patient-specific neoantigens | PIONEER / AI-Immunology platform selects patient-specific neoantigens; a peer-reviewed Phase 1 documents the AI-based selection |
| Clinical phase | Phase 3 (INTerpath-001); randomised Phase 2b KEYNOTE-942 completed | Phase 2 (NCT05309421), single-arm |
| Strongest published evidence | Randomised Phase 2b published, plus a company announcement on 19 August 2026 that Phase 3 recurrence-free and distant-metastasis-free survival endpoints were met | The clearest published link between AI-predicted targets and measured T-cell responses |
| Main evidence gap | Phase 3 results are topline company statements — detailed effect sizes, subgroups and full safety data were not public as of 22 August 2026 | Small, single-arm study; response and biomarker figures are largely company disclosures and conference material |
| Regulatory status | Investigational, not approved | Investigational, not approved |
Developer
- Intismeran autogene (V940 / mRNA-4157)
- Moderna and Merck
- EVX-01
- Evaxion
Format
- Intismeran autogene (V940 / mRNA-4157)
- Personalised mRNA vaccine
- EVX-01
- Personalised peptide vaccine
Population studied
- Intismeran autogene (V940 / mRNA-4157)
- Melanoma, adjuvant setting after surgical resection
- EVX-01
- Advanced melanoma
Computational selection
- Intismeran autogene (V940 / mRNA-4157)
- Developer describes integrated AI algorithms processing tumour and blood sequencing, predicting up to 34 patient-specific neoantigens
- EVX-01
- PIONEER / AI-Immunology platform selects patient-specific neoantigens; a peer-reviewed Phase 1 documents the AI-based selection
Clinical phase
- Intismeran autogene (V940 / mRNA-4157)
- Phase 3 (INTerpath-001); randomised Phase 2b KEYNOTE-942 completed
- EVX-01
- Phase 2 (NCT05309421), single-arm
Strongest published evidence
- Intismeran autogene (V940 / mRNA-4157)
- Randomised Phase 2b published, plus a company announcement on 19 August 2026 that Phase 3 recurrence-free and distant-metastasis-free survival endpoints were met
- EVX-01
- The clearest published link between AI-predicted targets and measured T-cell responses
Main evidence gap
- Intismeran autogene (V940 / mRNA-4157)
- Phase 3 results are topline company statements — detailed effect sizes, subgroups and full safety data were not public as of 22 August 2026
- EVX-01
- Small, single-arm study; response and biomarker figures are largely company disclosures and conference material
Regulatory status
- Intismeran autogene (V940 / mRNA-4157)
- Investigational, not approved
- EVX-01
- Investigational, not approved
Reading the difference honestly
These two programmes answer different questions well. Intismeran autogene has the stronger clinical outcome evidence: a randomised Phase 2b plus a Phase 3 announcement on its recurrence-free and distant-metastasis-free survival endpoints. EVX-01 has the stronger published mechanistic evidence that computationally selected targets provoke the intended immune response.
Why you should not rank them
- The populations differ — adjuvant post-resection disease versus advanced melanoma.
- One design is randomised; the other is single-arm, so selection bias cannot be removed.
- Response rates across separate small trials are not comparable numbers.
- Neither developer discloses its algorithm in enough technical detail for independent comparison.
What would change the picture
Full peer-reviewed Phase 3 publication with effect sizes, subgroups and complete safety data would settle much of the intismeran autogene question. For EVX-01, randomised data would be needed before its response figures can carry comparative weight. Until then, both belong in the "promising and unfinished" column of the evidence overview.
See every programme in one table
Five personalised neoantigen vaccine programmes with format, phase, evidence strength and stated limitations.
What this means for someone with melanoma today
Trial access is decided by an oncology team based on stage, prior treatment and eligibility criteria — not by anything on this site. What remains within reach for everyone is earlier detection and consistent follow-up: our lesion monitoring guides and what to do after a concerning result cover that part.
Frequently Asked Questions
By trial stage, intismeran autogene (previously V940 / mRNA-4157) is further ahead: it has a completed randomised Phase 2b and an ongoing Phase 3, whereas EVX-01 is in a single-arm Phase 2. Trial stage is a measure of development maturity, not of superiority.
No. They have never been studied against each other, the populations differ (adjuvant after resection versus advanced disease), and one is randomised while the other is single-arm. Response rates from a small single-arm study cannot be placed beside randomised outcomes.
EVX-01 provides the clearest published link between AI-predicted targets and measured T-cell responses. Intismeran autogene provides the strongest clinical outcome evidence. Those are different questions and both remain partly unresolved.
Neither is commercially approved. Access is through clinical trials, and eligibility is decided by an oncology team, not by a screening tool or an online article.
Not on its own. A successful trial validates the entire chain — sequencing, target selection, construct, formulation, manufacturing and the companion therapy given alongside it. The algorithm's specific contribution is not isolated by that design.