Background: Arsenic trioxide (ATO) has shown activity in relapsed/refractory multiple myeloma (MM). An uncommon but life-threatening side effect of ATO is arrhythmia such as torsade de pointes (TdP), especially when QT or QTc (heart-rate corrected) intervals >500 milliseconds (ms). We evaluated the incidence and predictors of QT and QTc prolongation in MM patients treated with ATO.

Methods: We reviewed prospectively collected demographic, laboratory and 766 weekly-ECG data on 48 patients treated in phase II trials with ATO. Patients received a median of 1.7 cycles (range 1–8) of 0.1–0.25 mg/Kg of IV ATO. Cycles consisted of 12 weeks of ATO (5-days/week for 1 week, then twice weekly for 11 weeks). Electrolytes were aggressively repleted to keep serum electrolytes as follows: K > 4 mEq/L, Mg > 2 mg/dL and Ca > 8.5 mg/dL. Endpoints included number of patients with QT and QTc prolongation (> 450 ms in males and > 470 ms in females), percentage of QT and QTc > 500 ms, and arrhythmia incidence. Predictors for QT and QTc prolongation were analyzed using a logistic regression model.

Results: Median patient age was 67.4 years. 85% were Caucasians, 64 % were males, and 37% achieved a response to ATO (CR or PR). Throughout the study, median values for K, Mg, Ca were 4.3 (4.0, 4.5), 2.1 (1.9, 2.2), and 9.1 (8.6, 9.4), respectively. Prolonged QT and QTc intervals at any time during the study occurred in 6/48 patients (12.5%) and in 25/48 patients (52%), respectively. ATO therapy was held based on ECG changes in 2 patients. One of whom had both QT and QTc > 500 ms and was found to be on a phenothiazine derivative; the other patient had only QTc > 500 ms and was found to have severe hypomagnesemia and hypocalcemia. After discontinuing the offensive drug and correcting electrolytes, therapy was restarted without any further QT or QTc prolongation. Overall, QT and QTc intervals returned to baseline with subsequent doses indicating the absence of a cumulative dose-effect. The factors influencing QT and QTc considered in the statistical analyses were age, race, gender, and response. In univariate and multivariate analyses, the only variable associated with QT and QTc prolongation was male gender (p<0.01). No TdP or any other ATO-related arrhythmia occurred.

Conclusion: ATO can result in prolongation of QT and to a higher degree QTc interval. QT and QTc are rarely prolonged > 500 ms. Males were more likely to develop QT and QTc prolongation. With repeated administration, a dose-cumulative effect does not seem to further prolong QT or QTc. ATO can be safely administered to patients with MM when electrolytes and concomitant medications are appropriately managed. Decisions based on ECG occurred in two patients during the first 12 weeks of therapy, utilizing 2/766 ECG’s (0.003%). Weekly ECG’s may therefore not be warranted in this clinical setting.

Author notes

Corresponding author