Use of Arsenic Trioxide in Acute Promyelocytic Leukemia Leads to Cardiotoxicity in African-American Patients.

Arsenic trioxide (ATO) is an effective therapy for acute promyelocytic leukemia (APL), but cardiac side effects have been described. We reviewed the charts of all 77 patients given ATO at M.D. Anderson from 1998 to present. Our objectives were to: (1) identify patients at relatively high risk of developing arrhythmias after receiving ATO and (2) evaluate the sensitivity and specificity of the electrocardiographic QTc interval in predicting subsequent arrhythmias. The median age of the patients was 54 years. 5% were African-American. 70% received ATO for treatment of APL, 18% for myelodysplastic syndrome, and 12% for other hematological malignancies. The dose of ATO was 0.15 mg/kg daily for 60 days. If, after this time, response was observed ATO was given for an additional 6 months, on a 2 weeks on, 2 weeks off schedule. The dose was lowered from 0.15 mg/kg in the event of acute toxicity, however the African-American patients discussed below were on the standard 0.15 mg/kg dose when they experienced arrhythmias.

Arrhythmias were observed in 4 of the 77 patients (5%, exact 95% CI 1– 13%). Of these 4 patients, 2 developed ventricular tachycardia and 2 atrial tachycardia. In 3 of the 4 patients, these arrhythmias led to discontinuation of ATO and its replacement by mylotarg. Three of the 4 African-Americans given ATO developed arrhythmias vs. only 1/73 non African-Americans (Fisher exact p value, 0.002). The patients who developed arrhythmias were younger (median age = 32) than those who did not (median age = 57) and had no history of cardiac disease, electrolyte abnormalities, or use of medications that would alter their response to ATO relative to the other patients. African-Americans tended to be younger than non African-Americans: 3/4 were age < 40 vs. 19/73 non African-Americans. However, since the incidence of arrhythmias in patients age < 40 was 3/3 for African-Americans vs. 0/19 for others (p= 0.0006), it seems that race, not age, is the principal predictor of arrhythmias. Similarly, although all 4 African-Americans had APL vs. 50/73 others, the respective incidence of arrhythmias in African-American APL patients was 3/4 vs. 1/50 in non African-American APL patients (p = 0.0006). Furthermore, arrhythmias were seen on average 8 days after ATO induction in the 3 African-Americans, while the Caucasian patient did not develop arrhythmias until 45 days after ATO induction. Although all patients were on weekly to bimonthly ECG monitoring, QTc interval prolongation (> 430 ms) was not observed in the patients who developed serious cardiac arrhythmias. None of the 3 patients with QTc prolongation had a documented arrhythmia. Thus, QTc monitoring had zero sensitivity and zero specificity for eventual arrhythmia development. Despite uncertainty as to mechanism, physicians should be aware that ATO may cause arrhythmias in young African Americans. Our data also calls into question the value of QTc monitoring, at least in patients given ATO.