Patterns of adjuvant radiotherapy use in early-stage endometrial cancer: A retrospective study

INTRODUCTION

Endometrial cancer is the most common gynaecologic malignancy in developed countries and the second most common in developing nations. The majority of patients present with early-stage disease, for whom surgery is the cornerstone of treatment, followed by risk-adapted adjuvant therapy. Several landmark randomised trials have shaped the current approach to adjuvant radiotherapy (RT) in early-stage endometrial cancer. The Gynecologic Oncology Group 99 (GOG-99) trial demonstrated that pelvic radiotherapy significantly reduced locoregional recurrence in intermediate-risk disease, though without a survival benefit.^[1] Similarly, the Postoperative radiotherapy in endometrial cancer trial 1 (PORTEC 1) trial confirmed that postoperative pelvic RT decreases local recurrence rates compared with surgery alone, but again did not show an overall survival advantage.^[2] The subsequent PORTEC-2 trial established that vaginal brachytherapy (VBT) alone achieves equivalent local control to external beam radiotherapy (EBRT) in high-intermediate-risk patients, with fewer gastrointestinal and genitourinary toxicities.^[3] For patients with high-risk features, the PORTEC-3 trial provided evidence supporting adjuvant chemoradiation over radiotherapy alone.^[4] In addition, the GOG-249 trial compared EBRT with VBT plus chemotherapy, further informing the management of high-intermediate and high-risk patients.^[5]

On the basis of this evidence, current international guidelines recommend tailoring adjuvant RT according to clinicopathologic risk factors, including lymphovascular space invasion (LVSI), depth of myometrial invasion, histologic grade, and subtype.^[6,7] LVSI, in particular, has been validated in pooled analyses of the PORTEC trials as an independent predictor of recurrence risk.^[7]

Although molecular classification is now increasingly incorporated into endometrial cancer management, particularly with the recognition of prognostic subgroups such as POLE ultramutated, mismatch repair–deficient (MMRd), p53-abnormal, and no specific molecular profile (NSMP) categories.^[8,9]

Widespread adoption remains limited in low- and middle-income countries. Until such genomic stratification becomes routine, clinicopathologic features remain central to adjuvant treatment decisions.

In this context, we conducted a retrospective study to evaluate patterns of adjuvant radiotherapy use in early-stage endometrial cancer at our tertiary cancer centre, focusing on the distribution of RT modalities and the key clinicopathologic factors driving treatment selection in the pre-molecular classification era.

MATERIAL AND METHODS

This retrospective study was conducted at a tertiary cancer centre and included patients with histologically confirmed FIGO stage I–II endometrial carcinoma who underwent primary surgical management between 2018 and 2023. A total of 96 patients were identified. Clinical and pathological parameters collected included patient age, histologic subtype, tumour grade, lymphovascular space invasion (LVSI), depth of myometrial invasion (MMI), and FIGO stage. Each patient was further stratified into risk categories (low, intermediate, high-intermediate, and high) based on established international guideline criteria.^[6]

All patients underwent total abdominal hysterectomy with bilateral salpingo-oophorectomy (TAH+BSO). Variations in surgical staging were recorded, including the extent of lymphadenectomy and additional procedures. Of the 96 patients, four underwent TAH+BSO alone, 49 underwent TAH+BSO with pelvic lymphadenectomy (PLND), and 15 underwent TAH+BSO with both pelvic and para-aortic lymphadenectomy (PLND+PALND). The remaining 28 patients underwent TAH+BSO with additional adjuvant interventions, including EBRT, VBT, and/or chemotherapy integrated into the treatment plan.

Adjuvant radiotherapy (RT) modalities were categorised into four groups: VBT only, EBRT only, combined EBRT plus VBT, or none/unspecified. Associations between clinicopathologic features and RT modality selection were assessed descriptively. The distribution of RT modality was summarised according to risk category, stage, LVSI status, and MMI depth using tabulated data [Tables 1–4]. A bar chart was also generated to illustrate modality distribution across risk groups [Figure 1].

Close

Figure 1: RT modality distribution by risk group. RT: Radiotherapy, VBT: Vaginal brachytherapy, EBRT: External beam radiotherapy.

Export to PPT

RESULTS

All 96 patients in the cohort underwent total abdominal hysterectomy with bilateral salpingo-oophorectomy (TAH+BSO), with variations in the extent of lymphadenectomy and additional adjuvant procedures. Of these, four patients underwent TAH+BSO alone, 49 underwent TAH+BSO with pelvic lymphadenectomy (PLND), and 15 underwent TAH+BSO with both pelvic and para-aortic lymphadenectomy (PLND+PALND). The remaining 28 patients underwent TAH+BSO with additional adjuvant interventions such as EBRT, VBT, and/or chemotherapy integrated into the treatment plan.

The majority of patients in the low- and intermediate-risk categories received adjuvant VBT following surgery, while those in the high-intermediate and high-risk groups more frequently underwent combined EBRT+VBT. Table 1 summarises the distribution of RT modality by risk category, showing that VBT predominated in low-risk patients (73%) and remained the most common modality in intermediate-risk patients (65%), whereas EBRT+VBT was used in 81% of high-intermediate-risk patients and in 67% of high-risk patients.

When analysed according to stage as illustrated in Table 2, VBT was most commonly used in stage IA patients (68.6%), while the use of EBRT+VBT increased with stage progression, being administered to 42% of stage IB and all stage II patients. Analysis by LVSI status as illustrated in Table 3 revealed that patients with positive LVSI were more frequently treated with EBRT-containing regimens compared with those without LVSI. Similarly, deeper myometrial invasion (≥50%) was associated with greater use of EBRT+VBT, whereas VBT predominated in cases with <50% invasion [Table 4].

At a median follow-up of 36 months (range, 24–96 months), no recurrences or deaths were observed in the cohort. These findings highlight the strong association between clinicopathologic features—particularly stage, LVSI, and depth of myometrial invasion—and the selection of adjuvant RT modality.

DISCUSSION

This study demonstrates that adjuvant radiotherapy (RT) decisions in early-stage endometrial cancer at our centre were strongly risk-adapted and aligned with international guidelines. In low- and intermediate-risk patients, the predominance of vaginal brachytherapy (VBT) reflects evidence from PORTEC-2, which showed equivalent local control and lower toxicity compared with external beam radiotherapy (EBRT).^[3,10] In contrast, patients with high-intermediate and high-risk disease more frequently received EBRT+VBT, a pattern consistent with recommendations from PORTEC-3^[4] and the GOG-249 trial,^[5] both of which support broader pelvic coverage for patients with higher recurrence risk.

The observed association between lymphovascular space invasion (LVSI) and the use of EBRT-containing regimens further reinforces findings from pooled PORTEC analyses identifying substantial LVSI as an independent predictor of recurrence.^[7] Similarly, the consistent use of EBRT+VBT in all stage II patients mirrors consensus guidelines such as the ESMO-ESGO-ESTRO recommendations,^[6] which emphasise pelvic nodal and parametrial coverage in this subset.

Our surgical patterns, dominated by total abdominal hysterectomy with bilateral salpingo-oophorectomy (TAH+BSO) and frequent lymphadenectomy in higher-risk patients, also align with international recommendations for comprehensive staging.^[6,11,12] The integration of surgical extent with histopathologic features into RT decision-making reflects a coordinated multidisciplinary approach and underscores the value of individualised care pathways in early-stage disease.

An important strength of this study is its demonstration that, even in the absence of molecular profiling, conventional clinicopathologic risk factors—particularly stage, depth of myometrial invasion, and LVSI—remain powerful determinants of adjuvant treatment. This is highly relevant for low- and middle-income countries (LMICs), where molecular classification is not yet routinely available. Nonetheless, increasing evidence highlights the prognostic and predictive value of molecular subtypes such as POLE ultramutated, mismatch repair–deficient (MMRd), p53-abnormal, and NSMP categories, which are now incorporated into the updated WHO classification and international guidelines.^[8,9] Ongoing studies such as PORTEC-4a are evaluating the integration of molecular profiles into adjuvant decision-making, potentially enabling further personalisation of therapy.^[13]

Another important consideration is treatment-related toxicity and survivorship. Randomised trials have consistently shown that limiting EBRT to patients with higher risk reduces gastrointestinal and hematologic toxicity and improves quality of life.^[3,10,14] The preference for VBT in lower-risk groups in our study is therefore in keeping with the evidence-based effort to minimise morbidity without compromising oncologic outcomes. Patient-reported outcome measures (PROMs) from PORTEC-2 and other studies further highlight the importance of quality-of-life considerations in selecting RT modalities.^[14,15]

Our findings are also consistent with real-world data from multicentric registries, which confirm that guideline-directed, risk-adapted adjuvant RT results in excellent outcomes with low recurrence rates.^[11,16] The absence of recurrences in our cohort at a median follow-up of 36 months is reassuring and comparable to international series, although longer follow-up remains essential to evaluate the durability of control, late radiation sequelae, and survival endpoints.^[17]

Taken together, these results emphasise the continuing relevance of clinicopathologic risk stratification while providing a valuable benchmark for future integration of molecular and genomic classifiers. As access to molecular testing expands in India and other LMICs, hybrid models combining clinical, pathologic, and molecular factors may offer the optimal balance between oncologic safety and toxicity reduction.^[9,13,18,19]

TAKE HOME MESSAGE

At our centre, adjuvant RT selection in early-stage endometrial cancer demonstrated strong adherence to risk-adapted, guideline-concordant practices. Treatment decisions were primarily influenced by stage, LVSI status, and depth of myometrial invasion, in line with prognostic factors validated in multiple landmark trials. The predominance of VBT in low and intermediate-risk groups, and the consistent use of EBRT-containing regimens in high-intermediate and high-risk patients, reflects evidence-based tailoring aimed at optimising locoregional control while minimising toxicity. Analysis of surgical patterns revealed that the majority of patients underwent total abdominal hysterectomy with bilateral salpingo-oophorectomy, with lymphadenectomy more frequently performed in higher-risk cases. Comprehensive surgical staging was associated with greater use of EBRT-containing regimens, highlighting the importance of nodal assessment in guiding adjuvant therapy.

Importantly, the absence of recurrences in this cohort, despite the lack of molecular profiling, highlights the continued utility of conventional clinicopathologic stratification when rigorously applied. These results reinforce the relevance of guideline-driven therapy in settings without routine access to genomic classifiers, providing a real-world benchmark against which molecular-era practices can be evaluated.

Future studies should focus on integrating molecular subtyping (e.g., POLE mutation, p53 status, MMR deficiency) into risk assessment, potentially refining indications for pelvic RT versus brachytherapy. Longer follow-up is needed to assess the durability of disease control, late toxicity, and quality-of-life outcomes, ensuring that treatment personalisation maintains both oncologic safety and functional well-being.

Even in the absence of molecular testing, risk-adapted adjuvant radiotherapy guided by conventional clinicopathologic factors ensures excellent outcomes in early-stage endometrial cancer. Stage, LVSI, and myometrial invasion remain key drivers of treatment choice in real-world practice.