LAS VEGAS – The search is on for predictors of which cancer patients will experience treatment-induced cardiotoxicity, and an initial report from the PREDICT study has identified several.

One predictor is the patient’s type of cancer. In PREDICT, patients with lymphoma had a twofold greater risk of developing treatment-related cardiotoxicity, compared with those with breast cancer. Moreover, those with a cancer diagnosis other than lymphoma or breast cancer had a fivefold greater risk than breast cancer patients, Dr. Daniel J. Lenihan reported at the annual meeting of the Heart Failure Society of America.

“It’s important to know that many of the studies on cardiotoxicity risk have been done in breast cancer patients. There are a lot of other cancer patients out there,” observed Dr. Lenihan, professor of medicine and director of clinical research in the division of cardiovascular medicine at Vanderbilt University in Nashville.

PREDICT is a prospective, community-based study of 597 cancer patients undergoing anthracycline-based chemotherapy in 24 community oncology programs. It is primarily a study of the effectiveness of using cardiac biomarkers to predict cardiotoxicity, along with an analysis of the results of various forms of treatment of the cardiotoxicity.

During up to 12 months of follow-up 11% of PREDICT participants experienced a cardiac event, most commonly symptomatic heart failure or a greater than 10% drop in left ventricular ejection fraction, which took a patient from normal range to below normal. Another impressive finding was the substantial burden of conventional cardiovascular risk factors present at baseline in patients scheduled for anthracycline-based chemotherapy. In a multivariate logistic regression analysis, the higher a cancer patient’s cardiovascular risk factor level, the greater the likelihood of chemotherapy-related cardiotoxicity.

A baseline B-type natriuretic peptide (BNP) level in excess of 100 pg/mL was a powerful predictor of a chemotherapy-related cardiac event, with an associated 2.1-fold increased risk. As a predictor of cardiotoxicity during the study period, baseline BNP had a sensitivity of 35%, a specificity of 85%, a positive predictive value of 22%, and – most importantly – a negative predictive value of 92%, according to Dr. Lenihan.

Similarly, using as cutoffs either a baseline BNP greater than 100 pg/mL or a troponin greater than 0.05 ng/mL had a sensitivity of 60%, a specificity of 50%, a positive predictive value of 13%, and a negative predictive value of 91%, he continued.

Myocardial imaging as a tool for predicting which patients will develop cardiotoxicity during or after cancer therapy is another active area of investigation. Other investigators have shown that myocardial strain imaging holds considerable promise (J. Am. Coll. Cardiol. 2014;63:2751-68); however, it’s not terribly practical because many echocardiography laboratories balk at the idea of routinely performing serial strain imaging studies in all cancer patients, Dr. Lenihan said.

Practical predictors of increased risk for cancer therapy-related cardiotoxicity are sorely needed in order to identify candidates for prophylaxis with an agent such as dexrazoxane, which has been shown in a meta-analysis to reduce the risk of clinical or subclinical heart failure by 71% (Cochrane Database Syst. Review 2008; April 16:CD003917).

Another promising preventive approach was displayed in the Spanish OVERCOME trial, involving 90 patients undergoing intensive chemotherapy for malignant hemopathies. Those randomized to combined prophylaxis with enalapril plus carvedilol had a 6.7% rate of the composite endpoint of death, heart failure, or an LVEF below 45% at 6 months, compared with 24.4% in controls (J. Am. Coll. Cardiol. 2013;61:2355-62).

Also, predictors of increased risk are helpful in identifying cancer therapy–related cardiotoxicity early in its course, when aggressive treatment with standard heart failure medications such as beta blockers and ACE inhibitors is most likely to be beneficial.

“Everybody in cardiology is used to the concept that time is muscle: don’t let ischemia persist. Have a strategy to resolve it as soon as possible. That paradigm really can also apply to chemotherapy-related injury: the longer you leave it alone, the more permanent it becomes. Being able to detect it at its earliest stage is critically important,” according to Dr. Lenihan.

He cited as an example a study of 201 consecutive patients with anthracycline-induced cardiomyopathy conducted at the European Institute of Oncology in Milan. The conventional dogma is that anthracycline-induced cardiomyopathy is typically permanent, but this study showed that’s not true.

When treatment with enalapril and carvedilol was initiated within the first couple of months following the end of chemotherapy, 64% of patients experienced complete recovery of their LVEF. When the heart failure medications were commenced 3-4 months after completing chemotherapy, the LVEF recovery rate dropped to 28%. No complete recovery of LVEF occurred in patients who began enalapril plus carvedilol after 6 months (J. Am. Coll. Cardiol. 2010;55:213-20).

“There is a real opportunity here to improve the care of cancer patients and prevent heart failure,” Dr. Lenihan concluded.

He reported receiving research support from Singulex, Millenium, and Acorda as well as serving as a consultant to Onyx and Roche.

bjancin@frontlinemedcom.com

Who will get cancer treatment-induced cardiotoxicity?