Cutting out radiotherapy appears to be safe while improving quality of life in patients with POLE-mutated endometrial cancer.
Molecular classification may be predictive of outcomes with radiotherapy and help guide adjuvant treatment decision-making for patients with stage I endometrioid endometrial cancer, according to findings from a study published in Journal of Clinical Oncology.
Across a cohort of patients with molecularly classified endometrial cancer who were included in the intent-to-treat population of the PORTEC-1 trial (n = 484), there was no disease recurrence in patients with POLE-mutated endometrial cancer. Additionally, the rate of vaginal recurrence was 6.4%, and pelvic recurrence was noted in 4.9% of those with mismatch repair deficient (dMMR) disease. The corresponding rates were 24.4% and 6.2% for patients with endometrial cancer harboring a p53 abnormality and 10.0% and 2.2% in those with no specific molecular profile, respectively.
In the PORTEC-1 trial population, 5-year locoregional recurrence-free survival was 97.4% (95% CI, 95.3%-99.5%) for those receiving external beam radiotherapy (EBRT) and 88.3% (95% CI, 84.3%-92.5%) for those who did not receive adjuvant therapy (P = 6.0x10–4). The corresponding 5-year locoregional recurrence-free survival rates were 95.7% (95% CI, 91.0%-100.0%) and 90.3% (95% CI, 83.2%-98.0%) for those with dMMR endometrial cancer (P = .52), 96.2% (95% CI, 89.0%-100.0%) and 72.2% (95% CI, 54.2%-96.2%) in those with endometrial cancer harboring a p53 abnormality (P = .15), and 98.3% (95% CI, 96.1%-100.0%) and 87.7% (95% CI, 82.4%-93.4%) in those with no specific molecular profile (P = 4.6x10–4).
Among 396 patients with molecularly defined endometrial cancer included in the phase 3 PORTEC-2 trial (NCT00376844), the 5-year vaginal recurrence-free survival rate was 97.8% among those receiving vaginal brachytherapy (VBT) and 97.9% with EBRT (P = .98). Additionally, the 5-year pelvic recurrence rate in each respective arm was 94.6% and 98.9%, respectively (P = .010).
In the PORTEC-2 trial population, no patients with POLE-mutated endometrial cancer experienced pelvic recurrence. The 5-year pelvic recurrence-free survival rates among those receiving EBRT and VBT, respectively, were 100.0% vs 96.2% (95% CI, 91.1%-100.0%) in those with dMMR disease (P = .20), 100.0% and 68.6% (95% CI, 48.8%-96.3%) in those with disease harboring a p53 abnormality (P = .055), and 98.2% (95% CI, 95.8%-100.0%) and 97.2% (95% CI, 94.2%-100.0%) in those with no specific molecular profile (P = .38).
“Results of this study support decisions on adjuvant therapy,” the study authors wrote. “Omitting radiotherapy is safe in [patients] with POLE-mutated [endometrial cancer] and will reduce toxicity, improve quality of life, and reduce health care utilization and costs.”
They continued, “[Patients] with stage I [endometrial cancer with no specific molecular profile] have significant benefit from adjuvant radiotherapy. Here, VBT is the treatment of choice as it is as effective as EBRT but has a much milder toxicity profile. This implies that assessment of the molecular classification is needed to provide [patients] with stage I [endometrioid endometrial cancer] with the most suitable adjuvant treatment strategy.”
Investigators of this study assessed data from the PORTEC-1 trial evaluating pelvic EBRT vs no adjuvant therapy among patients with early-stage intermediate-risk endometrial cancer. Analysis also included findings from the PORTEC-2 trial comparing VBT with EBRT in early-stage, high-/intermediate-risk endometrial cancer.
Outcomes of the primary analysis of this study included locoregional recurrence-free survival by treatment across molecular classes in the PORTEC-1 and PORTEC-2 studies. Investigators also conducted a secondary analysis in which they pooled data from the PORTEC-1 and PORTEC-2studies to evaluate locoregional recurrence-free survival rates.
In the PORTEC-1 and PORTEC-2 populations, the respective median patient ages at time of random assignment were 67 years and 69 years. Most patients in each respective population had grade 1 to 2 endometrioid disease (82.9% and 86.9%), stage IB disease (58.3% vs 81.8%), and no lymphovascular space invasion (82.0% and 77.5). Across the 4 molecular classes, most patients in each population had no specific molecular profile (54.8% and 58.6%) followed by dMMR disease (28.3% and 27.8%), disease with p53 abnormality (8.3% and 7.6%), and POLE-mutated disease (8.7% and 6.1%).
In the combined cohort of 880 patients, 5-year locoregional recurrence-free survival rates were 97.1% (95% CI, 95.5%-98.7%) with EBRT, 93.0% (95% CI, 89.4%-96.7%) with VBT, and 88.3% (95% CI, 84.3%-92.5%) with no adjuvant treatment (P = .0001). The corresponding rates with each respective treatment were 94.2% (95% CI, 90.1%-98.5%), 94.2% (95% CI, 87.9%-100.0%), and 90.3% (95% CI, 83.2%-98.0%) for the dMMR group (P = .74). Additionally, the corresponding values were 96.9% (95% CI, 91.0%-100.0%), 64.3% (95% CI, 44.8%-92.3%), and 72.2% (95% CI, 54.2%-96.2%) in the p53 abnormality group (P = .048), as well as 98.3% (95% CI, 96.6%-100.0%), 96.2% (95% CI, 92.7%-99.9%), and 87.7% (95% CI, 82.4%-93.4%) in the no specific molecular profile group (P = 5.0x10–5).
Based on a sensitivity analysis, cumulative rates of pelvic and locoregional recurrences were comparable to the estimates of locoregional and pelvic recurrence-free survival as determined with the Kaplan-Meier method.
Horeweg N, Nout RA, Jürgenliemk-Schulz IM, et al. Molecular classification predicts response to radiotherapy in the randomized PORTEC-1 and PORTEC-2 trials for early-stage endometrioid endometrial cancer. J Clin Oncol. Published online July 24, 2023. doi:10.1200/JCO.23.00062