br Details of Cardiac Toxicities
Details of Cardiac Toxicities
A total of 255 (7.9%) patients developed some form of a cardiac toxicity, among which 153 (4.7%) patients developed some form of arrhythmia and 62 (1.9%) patients developed an ischemic event. The remaining 40 patients developed non-specified forms of cardiac adverse events.
Comparing bevacizumab-based regimens with non-bevacizumab based regimens, bevacizumab-based regimens were more likely to be associated with any cardiac toxicity (9.8% vs. 6.1%; P < .001) and ischemic events (2.9% vs. 1%; P < .001); but not arrhythmias (5.4% vs. 4.1%; P ¼ .088). More specifically, the absolute number of ischemic events among patients receiving bevacizumab-containing regimens is 46 patients, and the absolute number of ischemic events among nonebevacizumab-containing regimens is 16 patients. Likewise, comparing panitumumab-based regimens with nonepanitumumab-based regimens, panitumumab-based regimens were more likely to be associated with any cardiac toxicity (11.5% vs. 6.6%; P < .001) and arrhythmias (7.5% vs. 3.7%; P < .001), but not ischemic events (1.8% vs. 2%; P ¼ .752).
Arrhythmias were graded as grade 1 in 74 patients, grade 2 in 43 patients, grade 3 in 24 patients, grade 4 in 3 patients, and grade 5 (fatal) in 9 patients. The median time to develop arrhythmia was 15 days. Among patients who Aztreonam developed arrhythmias, 33 patients developed atrial fibrillation, 3 patients developed atrial flutter, 7 patients developed supraventricular tachycardia, 5 patients devel-oped ventricular arrhythmia, and 48 patients developed unspecified tachycardia. Two patients developed atrioventricular/bundle branch block, 4 patients developed sinus bradycardia, and 10 patients developed unspecified bradycardia. The remaining 41 patients developed non-specified forms of arrhythmias.
Ischemic events were graded as grade 1 in 18 patients, grade 2 in
5 (fatal) in 4 patients. The median time to develop an ischemic
Table 3 Multivariate Logistic Regression Analysis for Factors Predicting Any Cardiac Toxicity
Parameters Odds Ratio (95% CI)
Concurrent panitumumab treatment
Concurrent bevacizumab treatment
Abbreviations: CI ¼ confidence interval; HR ¼ hazard ratio.
event was 74 days. Among patients who developed ischemic events, 37 patients had an attack of angina, 11 patients had a myocardial infarction, and the remaining 14 patients had non-specified forms of ischemic heart disease.
Predictors of Cardiac Toxicity
The following factors were evaluated in univariate logistic regression analysis as potential predictors of any cardiac toxicity: age at diagnosis, gender, race, body mass index, ECOG score, pre-existing diabetes mellitus, pre-existing hypertension, bevacizumab-containing regimens, and panitumumab-containing regimens. Among these factors, only age, pre-existing hypertension, bevacizumab-containing regimens, and panitumumab-containing regimens were predictive of cardiac toxicity (P < .05). When these 4 factors were enrolled in multivariate logistic regression model, only bevacizumab-containing regimens (P ¼ .002) and panitumumab-containing regimens (P < .001) were predictive for the occurrence of cardiac toxicity (Table 3).
A similar set of factors were evaluated in univariate logistic regression analysis as predictors of arrhythmias, and among these factors only body mass index, pre-existing hypertension, and panitumumab-containing regimens were predictive of arrhythmias (P < .05). When these factors were included into a multivariate lo-gistic regression model, only panitumumab-containing regimens were predictive of the occurrence of arrhythmias (P < .001) (Table 4).
Likewise, the same set of factors was evaluated in univariate logistic regression analysis as predictors of ischemic events. Among these factors, only age, pre-existing hypertension, and bevacizumab-containing regimens were predictive of ischemic events (P < .05). When these factors were included in multivariate logistic regression analysis, only bevacizumab-containing regimens were predictive of ischemic events (P ¼ .004) (Table 5).