Patients and Methods
Data from HER-2-overexpressing breast cancer patients aged ≥70 years treated since 2005 at the Breast Cancer Unit of Vall d'Hebron University Hospital were collected. Patients were considered for the study if they had received at least one dose of a trastuzumab-based regimen for the treatment of early or advanced disease and had carried out at least two LVEF assessments while on treatment. Patients who were withdrawn from trastuzumab therapy due to a cardiac event were also recorded. The LVEF was assessed either by a multiple-gated acquisition scanning or by echocardiography. Symptomatic cardiotoxicity was retrospectively assessed based on clinical records, which included medical history, physical examination and diagnostic tests (chest X-ray, electrocardiogram, 2D echo, radionuclide angiography, blood tests).
Cardiac events were categorized by the New York Heart Association (NYHA) functional classification system and then used to document symptomatic cardiotoxicity. Asymptomatic cardiotoxicity was defined as an absolute drop ≥10% resulting in a final LVEF <50% or any absolute drop >20%.
Trastuzumab was discontinued in patients who either developed cardiac symptoms or experienced a significant decrease in LVEF. In other cases, the decision to continue trastuzumab was considered on a case-by-case basis following strict cardiac monitoring. Re-treatment with trastuzumab in those patients who developed any cardiac event and recovered after discontinuation and appropriate cardiac therapy was evaluated carefully in each case.
The following cardiovascular risk factors (CRF) for the development of a cardiac event were retrospectively assessed: current or prior therapy for hypertension, diabetes and dyslipidemia; obesity (BMI >30 kg/m); prior anthracycline exposure (cumulative dose, type and time between the last anthracycline administration and the first trastuzumab dose); concomitant exposure to anthracyclines or taxanes; number of previous treatments; prior left-sided radiation therapy; baseline LVEF; smoking history and history of previous episodes of heart failure, cardiac arrhythmia, ischemia or valvular heart disease.
Statistical Analysis
Descriptive analysis of data was carried out according to standard methodology. Factors included in the univariate logistic regression model were history of cardiac toxicity, obesity, smoking history, hypertension, diabetes, dyslipidemia, cumulative dose of previous anthracyclines and age. Confounding and effect-modification variables were studied by multivariate analysis. SPSS v.15 for Windows (SPSS Inc., Chicago, IL) was used for the statistical analysis.