Molecular profile-based adjuvant treatment for women with high-intermediate risk endometrial cancer (PORTEC-4a): results of a randomised, open-label, phase 3, multicentre, non-inferiority trial

Abstract

Background: PORTEC-4a investigated molecular risk profile-based individualised adjuvant treatment for women with high-intermediate risk endometrial cancer, aiming to reduce both overtreatment and undertreatment while optimising locoregional control.

Methods: PORTEC-4a was a randomised, open-label, phase 3, multicentre, non-inferiority trial, conducted across eight European countries. Women (aged ≥18 years and with a WHO performance score of 0-2) with early stage high-intermediate risk endometrial cancer were eligible. Patients were randomly assigned post-surgery in a 2:1 ratio to either adjuvant treatment according to their molecular integrated risk profile or to standard vaginal brachytherapy. Allocation used a biased-coin minimisation with stratification for participating centre, grade, and lymphadenectomy. Adjuvant treatment in the molecular-profile group in case of favourable profile (POLE-mutated or no specific molecular profile [NSMP]-CTNNB1 wildtype) was observation, for intermediate profile (mismatch repair deficient or NSMP-CTNNB1 mutated) was brachytherapy (21 Gy in three fractions of 7 Gy given at 5 to 7 day intervals), and for unfavourable profile (p53 abnormal or substantial lymphovascular space invasion or L1 cell adhesion molecule overexpression) was pelvic radiotherapy (45·0-48·6 Gy in 1·8-2·0 Gy fractions, 5 days per week). The primary endpoint was overall 5-year cumulative incidence of vaginal recurrence as first event. Kaplan-Meier, Cox model, and cumulative incidence with competing risks were used for final analysis in the intention-to-treat population. Patient advocates were involved during grant application and trial conduct. The trial is registered with the Netherlands Trial Registry (NTR5841), the ISRCTN registry (ISRCTN11659025), and ClinicalTrials.gov (NCT03469674), and follow-up is ongoing.

Findings: Between June 1, 2016, and Dec 24, 2021, 569 patients were enrolled in PORTEC-4a. After the addition of 23 favourable patients out of PORTEC-4, the final combined PORTEC-4a cohort consisted of 564 eligible and evaluable patients (367 in the molecular profile group and 197 in the standard group). All patients were female, the median age was 69·0 years (IQR 63·0-73·5), and data on race and ethnicity were not collected. Median follow-up was 58·1 months (IQR 40·7-63·6). In the molecular profile group 168 (46%) patients had a favourable profile, 148 (40%) had an intermediate profile, and 51 (14%) had an unfavourable profile. The 5-year cumulative incidence of vaginal recurrence was 4·5% (95% CI 2·23-6·76) in the molecular profile group and 1·6% (0·00-3·32) in the standard group (HR 2·71 [95% CI 0·79-9·34]). The upper-bound of the one-sided confidence interval of the difference (5·3%) was below the predefined-equivalence margin of 7·0% (p_{non-inferiority}=0·005). The second primary analysis in patients with a favourable molecular profile showed 5-year vaginal recurrence of 4·1% (95% CI 0·81-7·37) in the molecular profile group versus 0·9% (0·00-2·78) in the standard group (HR 3·97 [95% CI 0·48-32·95]). Adverse events were mainly grades 1-2, with grade 3 or above related genitourinary toxicities in four (1%) of 367 versus four (2%) of 197, without substantial differences between groups. Five serious adverse events occurred, of which one was possibly related to treatment (vaginal scar dehiscence). No treatment-related deaths occurred.

Interpretation: Individualised adjuvant treatment by molecular integrated risk profile is safe and effective for patients with high-intermediate risk endometrial cancer; it spared 46% of patients with a favourable profile from adjuvant treatment, and reduces both overtreatment and undertreatment.

Funding: KWF Dutch Cancer Society.