Predicting Acute Cardiac Events in Patients With Breast Cancer Receiving Radiotherapy

In a Dutch study reported in the Journal of Clinical Oncology, van den Bogaard et al found that a model including the volume of the left ventricle receiving 5 Gy (LV-V5) might provide improved prediction of acute cardiac events in patients with breast cancer receiving radiotherapy. Prior findings indicated that the mean heart dose is predictive of risk. Anne P.G. Crijns, MD, PhD, of the University Medical Center Groningen, is the corresponding author of the Journal of Clinical Oncology article.

The current study involved 910 consecutive women treated with radiotherapy after breast-conserving surgery for stage I to III invasive adenocarcinoma or carcinoma in situ between January 2005 and December 2008 at the University Medical Center Groningen. The primary endpoint was the cumulative incidence of acute cardiac events within 9 years of follow-up. The mean heart dose and dose-distribution parameters of cardiac substructures were obtained from three-dimensional computed tomography planning data.

Predictive Models

A total of 30 patients had an acute cardiac event during the median follow-up of 7.6 years. The median mean heart dose was 2.37 Gy (range = 0.51–15.25 Gy), with the cumulative incidence of acute cardiac events increasing by 16.5% per Gy (P = .042). Analysis of dose-volume parameters showed that LV-V5 was the most important prognostic parameter on univariate analysis (hazard ratio [HR] = 1.016, P = .016).

Substituting LV-V5 for mean heart dose in a normal tissue complication probability (NTCP) model consisting of mean heart dose, age, and risk factors (yes or no) improved the c-statistic from 0.79 to 0.80. Among the multivariable NTCP models tested, the most optimal included LV-V5 (HR = 1.017, P = .041), age (HR = 1.065, P = .010), and weighted acute cardiac event risk score (0.8 for diabetes, 1.4 for hypertension, and 1.8 for history of ischemic cardiac events) per patient (HR = 2.036, P = .001); the model yielded a c-statistic of 0.83 (95% confidence interval = 0.75–0.91), which proved significantly better than the mean heart dose model (P = .042).

The investigators concluded: “A significant dose-effect relationship was found for [acute cardiac events] within 9 years after [radiotherapy]. Using [mean heart dose], the relative increase per Gy was similar to that reported in [a] previous study. In addition, LV-V5 seemed to be a better predictor for [acute cardiac events] than [mean heart dose]. This study confirms the importance of reducing exposure of the heart to radiation to avoid excess risk of [acute cardiac events] after radiotherapy for [breast cancer].” ■

van den Bogaard V, et al: J Clin Oncol. January 17, 2017 (early release online).




click me