Salvage hysterectomy for persistent residual cervical cancer: assessment of prognostic factors

Abstract

In this multicenter retrospective cohort study of 99 patients who underwent salvage hysterectomy for residual disease in the uterine cervix following the completion of definitive radiotherapy for cervical cancer across 25 Japan Clinical Oncology Group-affiliated centers from 2005-2014, (i) time duration from the completion of definitive radiotherapy to the diagnosis of residual disease in the uterine cervix, (ii) salvage hysterectomy surgical margin status, and (iii) extent of residual disease, were independently associated with progression-free survival (PFS). Specifically, (i) time duration to identify residual disease of >62 days was associated with decreased PFS compared to ≤62 days (4-year rates 21.8% vs. 55.0%, adjusted-hazard ratio [aHR]=2.69, 95% confidence interval [CI]=1.55-4.67); (ii) presence of tumor in the surgical margin of hysterectomy specimen was associated with 4 times increased risk of disease progression compared to tumor-free surgical margin (4-year PFS rates 0% vs. 45.3%, aHR=4.27, 95% CI=2.20-8.29); and (iii) hazards of disease progression was 4.5-fold increased when the residual disease extended beyond the uterine cervix compared to residual disease within the uterine cervix only (4-year PFS rates 11.1% vs. 50.6%, aHR=4.54, 95% CI=2.60-7.95). In the absence of these 3 prognostic factors, 4-year PFS rate reached nearly 80% (78.6%, SAL-HYS criteria). In sum, these data suggested that early detection of persistent, residual disease following definitive radiotherapy for cervical cancer may be the key to improve survival if salvage hysterectomy is considered as a tailored treatment option. Ideal surgical candidate would be uterine cervix-contained disease and assurance of adequate tumor-free surgical margin.

Keywords: Cervical Cancer; Disease Persistence; Hysterectomy, Salvage; Neoplasm, Residual; Radiotherapy; Survival.

Fig. 1. PFS associated with salvage hysterectomy. Kaplan-Meier survival curves for PFS are shown based on (A) time to residual diagnosis, (B) hysterectomy margin status, (C) presence of extra-cervical disease, and (D) “SAL-HYS” criteria. Log-rank test for p-values. SAL-HYS criteria was assessed among 94 cases with complete information for the independent prognostic factors shown in Table S2, grouped based on the extent of prognostic factors: (i) time to diagnosis^* (score=1 for >62 days, and score=0 for ≤62 days), (ii) hysterectomy surgical margin status (score=1 for tumor involvement, and score=0 for no tumor involvement), and (iii) residual tumor site (score=1 for cervix/extra-cervix, and score=0 for cervix only). Survival estimates are shown in Table S3.

C, cervical disease alone; EC, extra-cervical disease; Neg, negative; PFS, progression-free survival; Pos, positive. ^*Time duration from the completion of definitive radiotherapy to the date of residual cervical disease diagnosis.