Background
EMPOWER-Cervical 1 (GOG-3016/ENGOT-cx9) was a phase III randomized open-label trial (n=608) evaluating cemiplimab (anti-PD-1) versus single-agent investigator's choice chemotherapy in women with recurrent cervical cancer who had progressed after first-line platinum-containing chemotherapy. It enrolled patients regardless of PD-L1 status and prior bevacizumab exposure, making it broadly applicable to the second-line unselected population.
Interventions and follow up
Arm A (chemotherapy): Investigator's choice single-agent chemotherapy (pemetrexed, topotecan, irinotecan, gemcitabine, or vinorelbine)
Arm B (cemiplimab): Cemiplimab 350 mg IV q3w until progression or toxicity
Primary endpoint: Overall survival
Median follow-up: 18.2 months
Arm B (cemiplimab): Cemiplimab 350 mg IV q3w until progression or toxicity
Primary endpoint: Overall survival
Median follow-up: 18.2 months
Results
mOS (overall): 12.0 (cemiplimab) vs 8.5 months; HR 0.69, 95% CI 0.56–0.84, P<.001
mPFS: 2.8 vs 2.9 months; HR 0.75, 95% CI 0.63–0.89, P<.001
PD-L1 ≥1%: mOS 12.4 vs 8.4 months (HR 0.68); PD-L1 <1%: mOS 10.0 vs 7.8 months (HR 0.75)
Histology: OS benefit consistent across squamous and adenocarcinoma subgroups
mPFS: 2.8 vs 2.9 months; HR 0.75, 95% CI 0.63–0.89, P<.001
PD-L1 ≥1%: mOS 12.4 vs 8.4 months (HR 0.68); PD-L1 <1%: mOS 10.0 vs 7.8 months (HR 0.75)
Histology: OS benefit consistent across squamous and adenocarcinoma subgroups
Adverse events
Overall: Grade ≥3 AE 45.0% (cemiplimab) vs 53.4% (chemotherapy); discontinuation due to AEs 5.5% vs 4.0%
Immune-mediated: Any 10.3% vs 1.4%; most common hypothyroidism 12%, rash 10%, hepatitis 3%
Immune-mediated: Any 10.3% vs 1.4%; most common hypothyroidism 12%, rash 10%, hepatitis 3%
Conclusions
Cemiplimab significantly improved OS compared with single-agent chemotherapy in platinum-refractory recurrent cervical cancer, with lower rates of Grade ≥3 AEs than chemotherapy, establishing cemiplimab as a second-line standard of care regardless of PD-L1 expression.
Key Limitations
Open-label design; mPFS only marginally improved despite OS benefit (likely due to pseudoprogression / immune-responder dynamics); no crossover arm; chemotherapy choice varied (5 options) creating control-arm heterogeneity; PD-L1 unselected population makes biomarker subgroup analysis exploratory; increasingly, patients receiving PD-1/PD-L1 therapy first-line (KEYNOTE-826, BEATcc) are not candidates for second-line IO.
Clinical Context
FDA approved cemiplimab for recurrent or metastatic cervical cancer with progression on or after platinum-containing chemotherapy in 2021, one of the first IO approvals in cervical cancer (alongside the pembrolizumab CPS ≥1 indication). EMPOWER-Cervical 1 is the pivotal evidence. With pembrolizumab plus chemo (KEYNOTE-826) and atezolizumab plus bev plus chemo (BEATcc) now standard first-line, second-line IO rechallenge is less applicable; cemiplimab remains relevant for IO-naive second-line patients, particularly where pembrolizumab was not used first-line.