Abstract
Abstract only
4084
Background: The SHARP trial showed that S improves survival in advanced HCC. Full dose (FD) S at 400mg bid can be difficult to tolerate, so some clinicians begin with a reduced dose (RD) & escalate as tolerated to maximum dose. The purpose of this study was to determine whether starting dose or dose intensity of S affects survival. Methods: All patients treated with S for HCC from 01/2008 to 06/2016 in British Columbia, Alberta, Ontario (Princess Margaret Cancer Centre & Sunnybrook Odette Cancer Centre), were included. Patient demographics, clinical, tumor characteristics, S starting dose & mean dose intensity were collected & analyzed. Patients were dichotomized into starting FD or RD of S. Mean dose intensity was categorized into > 75%, 50-75% & < 50%. Survival outcomes were assessed with Kaplan-Meier curves & compared with the log-rank test. A Cox-proportional hazard model was constructed with starting dose, dose intensity & relevant clinical & pathologic factors to assess their impact on survival. Results: We included 681 patients. Median age 64 years, 80% men, 37% East Asian, & most frequent causes of liver disease were hepatitis B (33%) & C (29%). ECOG performance status prior to starting S was 0 in 30% & 1 in 60%. Most patients were Childs-Pugh A (86%) at start of S. Overall median survival was 9.1 months (m). S was started at FD in 42% of patients & 31% had a dose intensity > 75%. The median survival for starting FD & RD was 9.4 m & 8.9 m, respectively (p = 0.15). The median survival for a dose intensity > 75% was 9.5 m, 50-75% was 12.9 m & < 50% was 7.1 m (p = 0.005). In multivariate models that adjusted for demographic, stage, performance status, AFP, prior treatment, toxicity & liver function, starting dose (HR 1.1, 95%CI 0.86-1.3, p = 0.51) & dose intensity (50-75% HR 0.93, 95% CI 0.73-1.2; < 50% HR 0.89, 95% CI 0.69-1.1, p = 0.65) were not predictors of survival. Conclusions: Based on our multi-center database, starting HCC patients on a RD of S may be a reasonable since it does not appear to compromise survival. Patients receiving a dose intensity of S at 50-75% appear to have a superior median survival, though this is not significant after controlling for baseline characteristics.