Pelvic Exenteration for Advanced Nonrectal Pelvic Malignancy

Abstract

Objective: To determine factors associated with outcomes following pelvic exenteration for advanced nonrectal pelvic malignancy.

Background: The PelvEx Collaborative provides large volume data from specialist centers to ascertain factors associated with improved outcomes.

Methods: Consecutive patients who underwent pelvic exenteration for nonrectal pelvic malignancy between 2006 and 2017 were identified from 22 tertiary centers. Patient demographics, neoadjuvant therapy, histopathological assessment, length of stay, 30-day major complication/mortality rate were recorded.

The primary endpoints were factors associated with survival. The secondary endpoints included the difference in margin rates across the cohorts, impact of neoadjuvant treatment on survival, associated morbidity, and mortality.

Results: One thousand two hundred ninety-three patients were identified. 40.4% (n = 523) had gynecological malignancies (endometrial, ovarian, cervical, and vaginal), 35.7% (n = 462) urological (bladder), 18.1% (n = 234) anal, and 5.7% had sarcoma (n = 74).

The median age across the cohort was 63 years (range, 23–85). The median 30-day mortality rate was 1.7%, with the highest rates occurring following exenteration for recurrent sarcoma or locally advanced cervical cancer (3.3% each). The median length of hospital stay was 17.5 days. 34.5% of patients experienced a major complication, with highest rate occurring in those having salvage surgery for anal cancer.

Multivariable analysis showed R0 resection was the main factor associated with long-term survival. The 3-year overall-survival rate for R0 resection was 48% for endometrial malignancy, 40.6% for ovarian, 49.4% for cervical, 43.8% for vaginal, 59% for bladder, 48.3% for anal, and 48.1% for sarcoma.

Conclusion: Pelvic exenteration remains an important treatment in selected patients with advanced or recurrent nonrectal pelvic malignancy. The range in 3-year overall survival following R0 resection (40%–59%) reflects the diversity of tumor types.