Background
Phase III, double-blind RCT (INTerpath-001 / V940-001). N=1137 (2:1). Completely resected high-risk cutaneous melanoma, stage IIB–IV. Intismeran autogene = individualized mRNA neoantigen therapy (up to 34 patient-specific neoantigens) added to adjuvant pembrolizumab.
Results
Interventions and follow up: Arm A: Intismeran 1 mg IM q3w × 9 doses + pembrolizumab 400 mg IV q6w × 9 cycles (n=757)
Arm B: Placebo IM q3w × 9 + pembrolizumab 400 mg IV q6w × 9 (n=380)
Primary endpoint: Investigator-assessed RFS
mFollow up: 24.6 mo
Results: 2yr RFS: 82.2% vs 70.0%, HR 0.58, 95% CI 0.45–0.75, P<.001.
RFS piecewise HR: 0.99 (0–6 mo) vs 0.42, 95% CI 0.30–0.58 (>6 mo).
2yr DMFS: 88.5% vs 80.2%, HR 0.62, 95% CI 0.45–0.85, P=.002.
OS: immature; HR 0.77, 95% CI 0.45–1.31, P=.33 (NS).
Arm B: Placebo IM q3w × 9 + pembrolizumab 400 mg IV q6w × 9 (n=380)
Primary endpoint: Investigator-assessed RFS
mFollow up: 24.6 mo
Results: 2yr RFS: 82.2% vs 70.0%, HR 0.58, 95% CI 0.45–0.75, P<.001.
RFS piecewise HR: 0.99 (0–6 mo) vs 0.42, 95% CI 0.30–0.58 (>6 mo).
2yr DMFS: 88.5% vs 80.2%, HR 0.62, 95% CI 0.45–0.85, P=.002.
OS: immature; HR 0.77, 95% CI 0.45–1.31, P=.33 (NS).
Adverse events
Grade ≥3 (any): 23.6% vs 20.3%
Grade ≥3 treatment-related: 14.2% vs 13.4%
Reactogenicity: injection-site pain 51% vs 10%, pyrexia 52% vs 8%, flu-like illness 25% vs 9%, fatigue 39% vs 30% (mostly low grade, transient)
Immune-mediated: 36.0% vs 34.7% (no potentiation of pembro irAEs)
Discontinuation (TRAE): 14.0% vs 8.7%
Deaths from AE: 8 (1.1%) in intismeran arm, 5 before intismeran given
Grade ≥3 treatment-related: 14.2% vs 13.4%
Reactogenicity: injection-site pain 51% vs 10%, pyrexia 52% vs 8%, flu-like illness 25% vs 9%, fatigue 39% vs 30% (mostly low grade, transient)
Immune-mediated: 36.0% vs 34.7% (no potentiation of pembro irAEs)
Discontinuation (TRAE): 14.0% vs 8.7%
Deaths from AE: 8 (1.1%) in intismeran arm, 5 before intismeran given
Conclusions
Adding individualized mRNA neoantigen therapy to adjuvant pembrolizumab reduced recurrence/death by 42% and distant metastasis/death by 38% in resected stage IIB–IV melanoma. First positive phase III trial of a personalized cancer vaccine.
Key Limitations
Benefit emerges only after ~6 mo (HR ~1.0 early); OS immature, NS. Control limited to pembrolizumab (no nivolumab, no BRAF/MEK comparator). Requires tumor tissue + sequencing; ~17% of screened pts excluded for inadequate samples; manufacturing turnaround constrains real-world use. Wide CIs in some subgroups (e.g., US DMFS). Sponsor-funded (Merck/Moderna).
Clinical Context
Builds on phase IIb KEYNOTE-942 (RFS HR 0.56). Prior standard: adjuvant anti-PD-1 alone (KEYNOTE-054, KEYNOTE-716, CheckMate 238) or dabrafenib/trametinib for BRAF V600. Presented ESMO 2026 (LBA); not yet FDA/EMA approved — sponsors engaging regulators. Not yet in NCCN/ESMO guidelines; likely new adjuvant option for high-risk resected melanoma if approved.
