Cardiac events among patients with sarcoma treated with doxorubicin by method of infusion: A real‐world database study

BACKGROUND: Administration of doxorubicin by continuous intravenous (CIV) infusion, versus bolus (BOL) administration, has been proposed to mitigate the risk of cardiac events. This study used real‐world data to explore the association between mode of doxorubicin administration and duration of treatment, time‐to‐treatment failure (TTF), and cardiac events. METHODS: Occurrence of cardiac events after initiation of BOL versus CIV doxorubicin for sarcoma in the International Business Machines MarketScan claims database were compared. Duration of doxorubicin treatment, TTF, and time‐to‐first‐cardiac event (TCE) were evaluated using Kaplan–Meier method and unadjusted and adjusted Cox regression models. RESULTS: A total of 196 patients were included in the BOL group and 399 in the CIV group. In unadjusted analyses, there were significant differences between BOL versus CIV for duration of doxorubicin treatment (median 1.4 vs. 2.1 months, p = .002), TTF (median 8.8 vs. 5.6 months, p = .002), and TCE (medians not reached, p = .03). Adjusting for baseline covariates, only TTF remained significant (hazard ratio: 0.71, 95% confidence interval 0.59–0.86, p = .0004), favoring BOL. CONCLUSIONS: While the risk of cardiac complications was higher with BOL in unadjusted analysis, the risk was no longer present in the adjusted analysis. While we cannot draw causal inferences due to the retrospective, nonrandomized study design, these data suggest that replacing BOL with prolonged CIV administration has not been effective as a strategy to mitigate cardiac events, given community standards of oncologic practice.