Background
KEYNOTE-A18 was a phase III randomized double-blind trial (n=1060) evaluating the addition of pembrolizumab to standard concurrent chemoradiotherapy (cisplatin + external beam radiotherapy + brachytherapy) in newly diagnosed, high-risk locally advanced cervical cancer (FIGO 2014 IB2–IIB node-positive or III–IVA). It tested whether PD-1 blockade integrated with definitive curative-intent radiotherapy could improve outcomes.
Interventions and follow up
Arm A (placebo): Cisplatin-based chemoradiotherapy (EBRT + brachytherapy) + placebo ×5 cycles concurrent → placebo maintenance q6w ×15 cycles
Arm B (pembrolizumab): Cisplatin-based chemoradiotherapy + pembrolizumab 200 mg q3w ×5 cycles concurrent → pembrolizumab maintenance 400 mg q6w ×15 cycles
Primary endpoints: Progression-free survival (PFS) and overall survival (OS)
Median follow-up: 17.9 months (PFS); 29.9 months (OS analysis)
Arm B (pembrolizumab): Cisplatin-based chemoradiotherapy + pembrolizumab 200 mg q3w ×5 cycles concurrent → pembrolizumab maintenance 400 mg q6w ×15 cycles
Primary endpoints: Progression-free survival (PFS) and overall survival (OS)
Median follow-up: 17.9 months (PFS); 29.9 months (OS analysis)
Results
24-mo PFS: 68.0% (pembro) vs 57.3% (placebo); HR 0.70, 95% CI 0.55–0.89, P=.0020
36-mo OS: 82.6% vs 74.8%; HR 0.67, 95% CI 0.50–0.90, P=.0040
Subgroups: Benefit consistent across PD-L1 expression; stage III–IVA and stage IB2–IIB node-positive both showed benefit
36-mo OS: 82.6% vs 74.8%; HR 0.67, 95% CI 0.50–0.90, P=.0040
Subgroups: Benefit consistent across PD-L1 expression; stage III–IVA and stage IB2–IIB node-positive both showed benefit
Adverse events
Overall: Grade ≥3 AE 78.4% (pembro) vs 69.7% (placebo)
Immune-mediated: Any grade 35% (pembro) vs 15% (placebo); hypothyroidism most common immune AE
GI/GU: No significant increase in late GI or GU toxicity from combined radiotherapy plus immunotherapy
Immune-mediated: Any grade 35% (pembro) vs 15% (placebo); hypothyroidism most common immune AE
GI/GU: No significant increase in late GI or GU toxicity from combined radiotherapy plus immunotherapy
Conclusions
Adding pembrolizumab to chemoradiotherapy significantly improved both PFS and OS in high-risk locally advanced cervical cancer, establishing the first new standard of care for LACC in decades and representing a major practice-changing result for curative-intent treatment.
Key Limitations
Long-term radiotherapy-related toxicity with maintenance immunotherapy not fully characterized; brachytherapy required (MRI-guided preferred); no biomarker threshold identified; FIGO 2014 staging used (differs from FIGO 2018); the contribution of concurrent vs maintenance pembrolizumab phases cannot be separated by trial design.
Clinical Context
KEYNOTE-A18 represents the most significant advance in LACC treatment since concurrent cisplatin-based CRT was established in the late 1990s. FDA approved pembrolizumab with CRT for LACC (FIGO IB2–IIB node-positive or III–IVA) in January 2024. INTERLACE (induction carboplatin-paclitaxel before CRT) also showed an OS benefit in a similar population; whether these approaches are combinable or complementary is under investigation. ESMO and ASCO guidance now incorporate concurrent pembrolizumab for high-risk LACC.