KEYNOTE-048
In recurrent/metastatic HNSCC, how do pembrolizumab alone or with chemotherapy compare with cetuximab plus platinum/5-FU?
Trial design and population
Registry: undefined.
Key results
CPS ≥20: median OS 14.9 vs 10.7 months (HR 0.61); CPS ≥1: 12.3 vs 10.4 months (HR 0.74). Pembrolizumab-chemotherapy: total-population OS 13.0 vs 10.7 months (HR 0.77, 95% CI 0.63–0.93; p=0.0034).
Outcome visualization
Median OS, CPS ≥20 · primary analysis · 14.9
Bars scale to the larger reported value. Median survival values are time-to-event summaries, not fixed-time absolute event risks.
Absolute Effect
Endpoint/time horizon: Median OS, CPS ≥20 · primary analysis · 14.9.
NNT: Not calculated from median survival or hazard ratios; calculate only from defensible fixed-time absolute risks.
NNH: Not calculated unless a fixed-time excess-harm risk difference is available. Monotherapy interpretation depends on PD-L1 CPS and clinical need for rapid response. Not superior in the unselected total population.
Safety, burden and QoL
Monotherapy interpretation depends on PD-L1 CPS and clinical need for rapid response. Not superior in the unselected total population.
QoL and discontinuation percentages are not asserted where not verified from the full primary report; complete source-table audit remains a QC item.
Subgroups, maturity and limitations
https://pubmed.ncbi.nlm.nih.gov/31679945/
Distinguish prespecified from exploratory subgroups, immature from mature endpoints, and superiority from non-inferiority. Historical comparators may not represent all current treatment sequences.
Current-practice context · 21 October 2026
Verify current guidelines, regulatory status, biomarker requirements, prior treatment, comorbidities and patient goals before applying these historical trial results. Clinical freshness and full QoL/toxicity source review remain open where noted.