Does NCT06788990 (FORTIFI-HN01) meet its overall-survival co-primary endpoint — ficerafusp alfa + pembrolizumab vs placebo + pembrolizumab in first-line HPV-negative recurrent/metastatic HNSCC, PD-L1 CPS≥1 — at primary completion?

60% likely to miss · raimo

target EGFR + TGFB1 · moa checkpoint-inhibition · failure mode target-not-rate-limiting

1. premise Pembrolizumab acts only at the PD-1 synapse checkpoint; KEYNOTE-048 is the 1L recurrent/metastatic HNSCC CPS>=1 benchmark that ficerafusp alfa + pembrolizumab must beat on overall survival. pmid:31679945
2. premise HPV-negative HNSCC is frequently immune-excluded and re-suppressed via TGF-beta at points upstream (dendritic-cell maturation, tumour-bed exclusion) and downstream (regulatory-T-cell suppression) of the synapse; a high TGF-beta signature predicts checkpoint resistance. These are failure modes a synapse brake-release alone cannot reach. pmid:34921236
3. premise Ficerafusp alfa adds an EGFR-localized TGF-beta trap plus direct EGFR blockade; its Phase 1b in HPV-negative HNSCC produced an objective-response rate markedly above pembrolizumab monotherapy, with durable responses on extended follow-up. pmid:40643947
4. premise The closest precedent, bintrafusp alfa (a TGF-beta-trap + checkpoint bifunctional), produced single-arm signals across tumours and converted none into a randomized Phase 3 win. pmid:38485188
5. inference Ficerafusp's cis-localized trap and EGFR activity genuinely differentiate it from bintrafusp, and the HPV-negative depth/durability signal is real, so this is not a dead concept. But single-arm ORR rarely converts to a randomized OS win over pembrolizumab in 1L HNSCC, the TGF-beta-trap+checkpoint Phase 3 base rate is a run of failures, and the in-platform checkpoint-combination record is a coin-flip across two resolved cases. Setting the clean multi-position coverage rationale against that precedent holds the call below even rather than above it.
6. prediction The overall-survival co-primary is more likely than not NOT met versus pembrolizumab in the CPS-admitted ITT (p ~ 0.40). Any benefit concentrates in the TGF-beta-exclusion-high subgroup, which PD-L1 CPS does not capture - directly checkable now by re-stratifying the Phase 1b paired biopsies by a baseline TGF-beta exclusion signature.

Evidence: pmid:31679945, pmid:34921236, pmid:40643947, pmid:38485188