Role of NT-proBNP in Late-Onset Anthracycline-Induced Cardiotoxicity Screening in Adult Survivors of Pediatric Cancer

PURPOSE: Anthracyclines are chemotherapeutic agents with well-characterized cardiotoxic effects that can occur years after treatment. Current screening techniques in the cancer survivorship population are centered upon noninvasive cardiac imaging; however, this is effective in identifying cardiomyopathy only after significant remodeling has occurred. This study is a continuation of efforts to identify early biomarkers, such as NT-proBNP, and characteristics of adult survivors of pediatric cancer at increased risk of progression to non-ischemic cardiomyopathy (NICM). METHODS: In this retrospective chart review, data was collected for 190 patients at the UT Southwestern After Cancer Experience clinic over the duration of their follow-up care (mean, 4.1 patient yrs, mean of 15 yrs after chemotherapy completion). Patient groups were determined based on absence of NICM (group 1, n=118), those who developed NICM during observation (group 2, n=16), and patients with pre-existing NICM (group 3, n=56). NICM was defined as ejection fraction (EF) <55%. These groups were compared based on demographic data, cardiovascular risk factors, maximum recorded NT-proBNP, and ΔEF ((Max EFMin EF )/ Max EF). Multivariate regression analysis for ΔEF was performed. RESULTS: Patients with established NICM were found to have a significantly younger age at diagnosis, greater time from chemotherapy completion, and larger cumulative anthracycline dosage (CAD) as compared to those without NICM (p=0.022, p=0.006, p=0.006). Mean greatest NT-proBNP was 69.4 ng/mL, 206.2 ng/mL, and 302.2 ng/mL for groups 1,2, and 3 respectively. Though group 3 was significantly elevated from group 1 (p=0.043), those who developed NICM (group 2) did not significantly vary from either subgroup. Group 2 did have a significantly elevated ΔEF as compared to group 1 and 3 (p <0.002). Regression analysis of ΔEF yielded 4 variables with limited predictive value (R2=0.35) with NT-proBNP most heavily weighted (partial R2= 0.23). CONCLUSIONS: NT-proBNP was not clearly implicated in early identification of patients at increased risk of progression to NICM. Traditional cardiovascular risk factors, such as HTN, elevated cholesterol, or low HDL were not supported as predictive measures as well. Further prospective data with a larger cohort would be beneficial in clarify the distinguishing characteristics of cancer survivors at elevated risk of developing late-onset NICM.