Background
PORTEC-3 was a phase III randomized trial (n=660) evaluating whether adjuvant concurrent chemoradiotherapy (CRT) improves survival over pelvic radiotherapy alone in women with high-risk endometrial cancer. The 10-year follow-up analysis with post-hoc molecular classification (Post et al, Lancet Oncology 2025) provides long-term confirmation of the OS benefit and integrates the TCGA-based molecular subgroups (p53abn, MMRd, POLEmut, NSMP) into prognostic and predictive interpretation.
Interventions and follow up
Arm A (RT alone): Adjuvant pelvic radiotherapy alone (48.6 Gy/27 fx)
Arm B (CRT): Concurrent cisplatin 50 mg/m² weeks 1 and 4 with pelvic radiotherapy → 4 cycles adjuvant carboplatin AUC 5 + paclitaxel 175 mg/m² q21d
Primary endpoints: Overall survival and failure-free/recurrence-free survival
Median follow-up: ~10 years
Arm B (CRT): Concurrent cisplatin 50 mg/m² weeks 1 and 4 with pelvic radiotherapy → 4 cycles adjuvant carboplatin AUC 5 + paclitaxel 175 mg/m² q21d
Primary endpoints: Overall survival and failure-free/recurrence-free survival
Median follow-up: ~10 years
Results
10-yr OS: 74.4% (CRT) vs 67.3% (RT); adjusted HR 0.73, 95% CI 0.54–0.97, P=.032
10-yr RFS: 72.8% vs 67.4%; adjusted HR 0.74, 95% CI 0.56–0.98, P=.034
p53abn subgroup: 10-yr OS 52.7% vs 36.6% (HR 0.52, P=.021); 10-yr RFS 52.6% vs 37.0% (HR 0.42, P=.0027)
MMRd / POLEmut / NSMP: MMRd and POLEmut tumors did not benefit from CRT vs RT; NSMP benefit modulated by ER status
10-yr RFS: 72.8% vs 67.4%; adjusted HR 0.74, 95% CI 0.56–0.98, P=.034
p53abn subgroup: 10-yr OS 52.7% vs 36.6% (HR 0.52, P=.021); 10-yr RFS 52.6% vs 37.0% (HR 0.42, P=.0027)
MMRd / POLEmut / NSMP: MMRd and POLEmut tumors did not benefit from CRT vs RT; NSMP benefit modulated by ER status
Adverse events
Neurologic: Persistent Grade ≥2 sensory neuropathy 6% (CRT) vs 0% (RT alone) at 10 years
GI/GU: Grade ≥2 late GI and GU toxicity higher in CRT arm at 5 years but not significantly different at 10 years; no treatment-related deaths in either arm during follow-up
GI/GU: Grade ≥2 late GI and GU toxicity higher in CRT arm at 5 years but not significantly different at 10 years; no treatment-related deaths in either arm during follow-up
Conclusions
At 10 years, adjuvant CRT significantly improves both OS and RFS over RT alone in high-risk endometrial cancer, confirming the durability of the benefit first observed at 5 years. Molecular classification identifies p53abn tumors as deriving the most substantial benefit from CRT, while MMRd and POLEmut patients appear not to benefit — a finding with major implications for biomarker-driven adjuvant treatment selection.
Key Limitations
Post-hoc molecular analysis available in only ~62% of patients (n=411/660), with potential selection bias; molecular subgroups defined retrospectively; small subgroup sizes limit precision (p53abn CRT n~55); PORTEC-4a addresses biomarker selection prospectively; chemotherapy schedule differs from modern standards (no IO); the benefit applies to RT-eligible high-risk patients, not all endometrial cancer.
Clinical Context
PORTEC-3 is the definitive trial establishing adjuvant CRT in high-risk endometrial cancer (stage III; stage I–II with deep myometrial invasion/LVSI/G3; serous/clear cell). The 10-yr molecular data strongly suggest p53abn tumors should preferentially receive CRT, while POLEmut patients (excellent prognosis) may be candidates for de-escalation (tested in PORTEC-4a). This molecular-guided approach is being integrated into ESMO guidance.
References
Post CC et al, Lancet Oncol, 2025; PMID: 40921169
de Boer SM et al, Lancet Oncol, 2019 (6-yr follow-up); PMID: 31345626
de Boer SM et al, Lancet Oncol, 2018 (primary report); PMID: 29449193
de Boer SM et al, Lancet Oncol, 2019 (6-yr follow-up); PMID: 31345626
de Boer SM et al, Lancet Oncol, 2018 (primary report); PMID: 29449193