Association of anthropometric variables with therapy-induced cardiotoxicity in women with breast cancer: a pilot study for a randomized clinical trial.

Background: Doxorubicin (DOX) has been widely used in the treatment of breast cancer, but it is directly associated with late-onset cardiovascular disease (CVD). Whether anthropometric, food intake or other risk factors together with DOX-based chemotherapy can increase the risk of developing cardiotoxicity remains uncertain. We examined the association between anthropometric variables with doxorubicin-induced cardiotoxicity in women with breast cancer.

Methods: Twenty-six women (53.7 ± 9.6 y) undergoing DOX-based chemotherapy (408.3 ± 66.7 mg/m) participated in the study. We collected data on body composition (bioimpedance), dietary intake (24 h) and cardiac function (echocardiographic assessment of left ventricular ejection fraction, LVEF). All measurements were taken at baseline, one month of treatment completion and one-year follow-up after start of treatment. DOX-induced cardiotoxicity was defined as ≥ 10% absolute decrease in LVEF. Thus, the participants were then grouped as DOX-induced (DIC) or non-DOX-induced (non-DIC) cardiotoxicity. Data are shown as mean ± SD (standard deviation). We performed comparisons between the two groups using Student's t-test for independent samples or Generalized Estimating Equations (groups + 3 evaluation time points) with Bonferroni post-hoc test. Lastly, the correlations were analyzed using Pearson correlation; p < 0.05 for all tests.

Results: At baseline the participants' body mass index (BMI) was 29.9 ± 7.9 kg/m and LVEF was 67.4 ± 6.2%. Seven of them (26.9%) developed therapy-induced cardiotoxicity (ΔLVEF - 3.2 ± 2.6%; p < 0.001). Postmenopausal status and family history of CVD were more prevalent in the DIC group than non-DIC group. We found no consistent BMI changes in the groups over time. Interestingly, the non-DIC group showed a small increase in visceral fat at treatment completion and increased waist circumference at one-year follow-up compared to baseline. These same changes were not seen in the DIC group. We also observed a pattern of correlation of some anthropometric variables with LVEF: the more unfavorable the body composition the more pronounced the LVEF decrease at one-year follow-up, though not associated with cardiotoxicity.

Conclusions: Our study did not provide sufficient evidence to support that anthropometric variables, food intake or other risk factors increase the risk of developing cardiotoxicity. However, there are apparent trends that need to be further investigated in larger samples.