Reply to V. Pitini et al

Abstract

We appreciate the comments of Pitini et al regarding our recent report of 7-year follow-up of cardiac function in National Surgical Adjuvant Breast and Bowel Project protocol B-31, which is evaluating the benefits and risks of adding trastuzumab to standard chemotherapy for the adjuvant treatment of women with early stage human epidermal growth factor receptor 2–positive breast cancer. We agree that a dynamic evaluation of cardiac function, for example by echocardiogram, provides more detailed assessment of myocardial function than multigated acquisition (MUGA) scans, and clearly all imaging modalities carry a margin of error. It is worth pointing out that for the first 1,000 patients enrolled onto B-31, the primary end point of the protocol was not the clinical efficacy of trastuzumab but cardiac safety. As such, patients with any significant cardiac condition were excluded from participation, including those with any arrhythmia requiring medication. The cardiac risk score applies only to those patients who would have met the eligibility requirements of the protocol from which it was developed. As we mentioned in the Methods section, MUGA scans were chosen over echocardiograms in order to write rules for holding or stopping trastuzumab and to minimize observer variability across institutions. We also agree that the accuracy of the baseline left ventricular ejection fraction (LVEF) measurement is critical, not only for the cardiac risk score, but also because it serves as the comparison point for subsequent LVEF measurements in determining whether to hold or continue trastuzumab. For instance, if a patient is dehydrated when her baseline LVEF study is performed, the value obtained may be higher than a true reflection of her level of systolic function, and the difference between baseline and subsequent measurements then may be exaggerated, in some cases potentially affecting decisions about whether to hold or continue trastuzumab. Finally, we welcome the important work that is being done using biomarkers such as ultrasensitive troponin, and we cite two such studies in our discussion. In the end, however, any indicators of potential cardiac risk, whether by risk score, biomarkers, or genetics, must be integrated into clinical decision algorithms regarding optimal use of trastuzumab in the treatment of patients with breast cancer.