Author information
1 Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy. Electronic address: francesco.tovoli2@unibo.it.
2 Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy.
3 Department of Medical Oncology, Istituto Scientifico Romagnolo per Lo Studio e Cura Dei Tumori (IRST) IRCCS, Meldola, Italy; Department of Oncology and Haematology, University Hospital of Modena, Modena, Italy.
4 Department of Internal Medicine, Degli Infermi Hospital, Faenza, Italy.
5 Department of Oncology and Haematology, University Hospital of Modena, Modena, Italy.
6 Unit of Radiology , Department of Diagnostic Medicine and Prevention, Sant'Orsola Hospital, University of Bologna, Bologna, Italy.
Abstract
BACKGROUND &AIMS:
Sorafenib is associated with multiple adverse events (AEs), potentially causing its permanent interruption. The impact of the physicians experience on the management of these AEs and the relative implications on clinical outcomes are unknown. We verified if the AEs management changed over time and if these modifications impacted on treatment duration and overall survival (OS).
METHODS:
We analysed the prospectively collected data of 338 consecutive patients who started sorafenib between January 2008 and December 2017 in three tertiary care centres in Italy. Patients were divided according to the starting date: Group A (2008-2012; n=154), and Group B (2013-2017, n=184). Baseline and follow up data were compared. In the OS analysis, patients who received second-line treatments were censored when starting the new therapy.
RESULTS:
Baseline characteristics, AEs, and radiological response were consistent across groups. Patients in Group B received a lower median daily dose (425 vs 568 mg/day, p<0.001) due to more frequent dose modifications. However, treatment duration was longer (5.8 vs 4.1 months, p=0.021) with a trend toward a higher cumulative dose in Group B. Notably, the OS was also higher (12.0 vs 11.0 months, p=0.003) with a sharp increase in the 2-year survival rate (28.1 vs 18.4%, p=0.003) in Group B. The multivariate time-dependent Cox regression confirmed later period of treatment as an independent predictor of survival (HR 0.728, 95%CI 0.581-0.937, p=0.013). Unconsidered confounders were unlikely to affect these results at the sensitivity analysis.
CONCLUSIONS:
experience in the management of sorafenib-related AEs prolongs treatment duration and survival. This factor should be considered in the design of future randomised clinical trials including a sorafenib treatment arm, as an underestimate of sample size may derive.