Non-exenterative surgical management of recurrent endometrial carcinoma: patient characteristics and outcomes

Abstract

<h3>Objectives:</h3> To determine what clinicopathologic variables are associated with survival outcomes for patients with recurrent endometrial cancer selected for non-exenterative surgery. <h3>Methods:</h3> All patients with recurrent endometrial cancer who underwent primary surgery between 1/1/2009 and 12/31/2017 at our center were retrospectively identified. We included patients with disease of endometrioid, serous, clear cell, and mixed histologies. Appropriate statistical tests were used. <h3>Results:</h3> We identified 350 patients with recurrent endometrial cancer; 57 (16%) were selected for secondary cytoreductive surgery (SUR) and 240 (69%) were medically managed (MM) with chemotherapy and/or radiation therapy (RT). The remaining patients were treated with hormonal therapy or did not undergo further treatment at recurrence. Median age was 62 years (range, 39-83) for the SUR and 66 years (range, 28-90) for the MM group (p<0.001). Thirty-four patients (60%) in the SUR group and 110 (46%) in the MM group had disease of endometroid histology (p=0.04). Thirty-three (58%) and 101 (42%), respectively, had stage I disease at primary diagnosis (p=0.01). Thirty-eight (67%) and 122 (51%), respectively, had undergone RT as adjuvant treatment at primary diagnosis (p=0.04). Forty-two (74%) and 116 (48%), respectively, had single-site recurrent disease (p<0.001). The most common sites of recurrence for those selected for surgery involved extra-pelvic disease in 42 patients (74%), of which an upper abdominal mass was the most frequent site of disease in 14 patients (33%), followed by lung disease in 8 patients (19%). Among SUR patients, median OR time was 145 min (11-405), estimated blood loss was 100 mL (5-1800), length of stay was 1.5 days (0-13), and 2 patients (4%) experienced a grade 3 complication. Forty-two (74%) achieved a complete gross resection and 43 (75%) underwent postoperative chemotherapy and/or RT. Patients in the SUR group had a median OS of 58 months (95% CI, 33.1-not reached) compared to 24 months (95% CI, 23-30) for those in the MM group (p<0.001). On multivariate analysis,surgery was an independent predictor of improved survival. The MM group had an HR of death of 2.0 (95% CI, 1.2-3.3; p<0.001). Among the 57 patients selected for surgery, tumor size, multisite disease, site of recurrence, and residual disease were not associated with improved post-relapse survival (p=0.259, 0.688, 0.999, 0.929, respectively). <h3>Conclusions:</h3> Medical management with chemotherapy and/or RT is the most common treatment approach for first recurrence of endometrial cancer; however, surgery was associated with acceptable perioperative outcomes and may contribute to improved long-term survival in highly selected patients. Location and multi-site recurrence should not preclude non-exenterative surgical resection if deemed safe and feasible.