Impact of Concomitant Chemotherapy on the Effects of Treatment Prolongation in Cervical Cancer

Abstract

Purpose/Objective(s)Prior reports have demonstrated adverse effects of treatment prolongation on pelvic control and survival outcomes in cervical cancer patients treated with radiation therapy alone (RT). In this study, we seek to assess whether treatment prolongation has similarly adverse effects in the setting of concomitant chemoradiation therapy (CRT).Methods/MaterialsFrom 1989 to 2009, 480 consecutive cervical cancer patients treated at a single institution with RT or CRT for curative intent were retrospectively analyzed. Relapse was defined as in-field or out-of-field, with patients censored at the time of first relapse. The effects of treatment duration (TD) on in-field and out-of-field relapse rates, disease-free survival (DFS) and overall survival (OS) were assessed continuously and categorically (TD <55 vs. TD ≥55 days) within the separate RT and CRT cohorts. Covariates included age, histology, FIGO stage, baseline Karnofsky performance status, brachytherapy dose-rate, and cumulative dose to point M (assuming α/β = 10). Cumulative incidence of relapse rates, DFS, and OS were estimated with Kaplan-Meier analysis; comparisons used log-rank statistic. Multivariate analysis used the Cox proportional hazards model.ResultsThree hundred seventy-two patients (RT n= 206; CRT n= 166) were assessable with median follow-up of relapse-free patients of 4.2 years (RT 4.4 years, CRT 4.2 yrs; p = 0.807). TD was longer in the RT cohort (median 55 days, range 35-99) vs. the CRT cohort (median 51, range 35-92) (p=0.001). In the CRT cohort, TD≥55 days was not predictive of in-field relapse (p=0.247), out-of-field relapse (p=0.669), DFS (p=0.406) or OS (p=0.202). Similar findings were observed when TD was assessed continuously in the CRT cohort. In the RT cohort, TD≥55 days predicted for higher in-field relapse rates (p=0.006), worse DFS (p=0.014), and worse OS (p=0.016) but not for out-of-field relapse rates. When assessed as a continuous variable in the RT cohort, TD was associated with DFS (p=0.002) and OS (p=0.009) and trended to significance for predicting in-field relapse rates (p=0.056). In the RT cohort, FIGO stage also predicted for in-field relapse, DFS and OS (p<0.001 for all) but was closely correlated with TD (p<0.001). After adjustment for FIGO stage, TD≥55 days lost statistical significance in predicting in-field relapse (p=0.741), DFS (p=0.887) and OS (p=0.892) in the RT cohort. Similar findings were observed when TD was assessed continuously.ConclusionsWith CRT, treatment prolongation does not adversely impact in-field relapse, out-of-field relapse, DFS, or OS outcomes. With RT alone, although on univariate analysis treatment prolongation ≥55 days is associated with higher in-field relapse and worse DFS and OS, these findings are attributable to the confounding association between TD and FIGO stage. Purpose/Objective(s)Prior reports have demonstrated adverse effects of treatment prolongation on pelvic control and survival outcomes in cervical cancer patients treated with radiation therapy alone (RT). In this study, we seek to assess whether treatment prolongation has similarly adverse effects in the setting of concomitant chemoradiation therapy (CRT). Prior reports have demonstrated adverse effects of treatment prolongation on pelvic control and survival outcomes in cervical cancer patients treated with radiation therapy alone (RT). In this study, we seek to assess whether treatment prolongation has similarly adverse effects in the setting of concomitant chemoradiation therapy (CRT). Methods/MaterialsFrom 1989 to 2009, 480 consecutive cervical cancer patients treated at a single institution with RT or CRT for curative intent were retrospectively analyzed. Relapse was defined as in-field or out-of-field, with patients censored at the time of first relapse. The effects of treatment duration (TD) on in-field and out-of-field relapse rates, disease-free survival (DFS) and overall survival (OS) were assessed continuously and categorically (TD <55 vs. TD ≥55 days) within the separate RT and CRT cohorts. Covariates included age, histology, FIGO stage, baseline Karnofsky performance status, brachytherapy dose-rate, and cumulative dose to point M (assuming α/β = 10). Cumulative incidence of relapse rates, DFS, and OS were estimated with Kaplan-Meier analysis; comparisons used log-rank statistic. Multivariate analysis used the Cox proportional hazards model. From 1989 to 2009, 480 consecutive cervical cancer patients treated at a single institution with RT or CRT for curative intent were retrospectively analyzed. Relapse was defined as in-field or out-of-field, with patients censored at the time of first relapse. The effects of treatment duration (TD) on in-field and out-of-field relapse rates, disease-free survival (DFS) and overall survival (OS) were assessed continuously and categorically (TD <55 vs. TD ≥55 days) within the separate RT and CRT cohorts. Covariates included age, histology, FIGO stage, baseline Karnofsky performance status, brachytherapy dose-rate, and cumulative dose to point M (assuming α/β = 10). Cumulative incidence of relapse rates, DFS, and OS were estimated with Kaplan-Meier analysis; comparisons used log-rank statistic. Multivariate analysis used the Cox proportional hazards model. ResultsThree hundred seventy-two patients (RT n= 206; CRT n= 166) were assessable with median follow-up of relapse-free patients of 4.2 years (RT 4.4 years, CRT 4.2 yrs; p = 0.807). TD was longer in the RT cohort (median 55 days, range 35-99) vs. the CRT cohort (median 51, range 35-92) (p=0.001). In the CRT cohort, TD≥55 days was not predictive of in-field relapse (p=0.247), out-of-field relapse (p=0.669), DFS (p=0.406) or OS (p=0.202). Similar findings were observed when TD was assessed continuously in the CRT cohort. In the RT cohort, TD≥55 days predicted for higher in-field relapse rates (p=0.006), worse DFS (p=0.014), and worse OS (p=0.016) but not for out-of-field relapse rates. When assessed as a continuous variable in the RT cohort, TD was associated with DFS (p=0.002) and OS (p=0.009) and trended to significance for predicting in-field relapse rates (p=0.056). In the RT cohort, FIGO stage also predicted for in-field relapse, DFS and OS (p<0.001 for all) but was closely correlated with TD (p<0.001). After adjustment for FIGO stage, TD≥55 days lost statistical significance in predicting in-field relapse (p=0.741), DFS (p=0.887) and OS (p=0.892) in the RT cohort. Similar findings were observed when TD was assessed continuously. Three hundred seventy-two patients (RT n= 206; CRT n= 166) were assessable with median follow-up of relapse-free patients of 4.2 years (RT 4.4 years, CRT 4.2 yrs; p = 0.807). TD was longer in the RT cohort (median 55 days, range 35-99) vs. the CRT cohort (median 51, range 35-92) (p=0.001). In the CRT cohort, TD≥55 days was not predictive of in-field relapse (p=0.247), out-of-field relapse (p=0.669), DFS (p=0.406) or OS (p=0.202). Similar findings were observed when TD was assessed continuously in the CRT cohort. In the RT cohort, TD≥55 days predicted for higher in-field relapse rates (p=0.006), worse DFS (p=0.014), and worse OS (p=0.016) but not for out-of-field relapse rates. When assessed as a continuous variable in the RT cohort, TD was associated with DFS (p=0.002) and OS (p=0.009) and trended to significance for predicting in-field relapse rates (p=0.056). In the RT cohort, FIGO stage also predicted for in-field relapse, DFS and OS (p<0.001 for all) but was closely correlated with TD (p<0.001). After adjustment for FIGO stage, TD≥55 days lost statistical significance in predicting in-field relapse (p=0.741), DFS (p=0.887) and OS (p=0.892) in the RT cohort. Similar findings were observed when TD was assessed continuously. ConclusionsWith CRT, treatment prolongation does not adversely impact in-field relapse, out-of-field relapse, DFS, or OS outcomes. With RT alone, although on univariate analysis treatment prolongation ≥55 days is associated with higher in-field relapse and worse DFS and OS, these findings are attributable to the confounding association between TD and FIGO stage. With CRT, treatment prolongation does not adversely impact in-field relapse, out-of-field relapse, DFS, or OS outcomes. With RT alone, although on univariate analysis treatment prolongation ≥55 days is associated with higher in-field relapse and worse DFS and OS, these findings are attributable to the confounding association between TD and FIGO stage.