BASELINE CLINICAL, ECHOCARDIOGRAPHIC, AND BIOMARKER PREDICTORS OF EARLY NT-PROBNP AND GALECTIN-3 DETERIORATION DURING ANTHRACYCLINE CHEMOTHERAPY: A PLACEBO-ARM SECONDARY ANALYSIS

Abstract

Background: Anthracycline cardiotoxicity may begin with subclinical biomarker deterioration before overt left ventricular dysfunction develops. Whether routine baseline variables can identify patients who subsequently develop early biomarker worsening remains uncertain. This exploratory secondary analysis evaluated whether baseline clinical, echocardiographic, and biomarker characteristics were associated with 4-month changes in NT-proBNP and galectin-3 during anthracycline-based chemotherapy. Methods: This analysis included placebo-treated participants from a prospective, randomized, double-blind, placebo controlled trial evaluating cardioprotection during anthracycline therapy. Only the placebo arm was analyzed to assess the natural course of biomarker changes without active cardioprotective treatment. The final cohort included 45 patients with preserved baseline cardiac function and paired baseline and 4-month NT-proBNP and galectin-3 measurements. Candidate baseline predictors included age, sex, body surface area, left ventricular ejection fraction, E/A ratio, NT-proBNP, and galectin-3. Associations with absolute biomarker changes were assessed using Spearman correlation and exploratory simple linear regression. Results: NT-proBNP increased from 47.96 ± 45.98 pg/mL to 73.62 ± 55.24 pg/mL, with a mean change of +25.67 ± 38.43 pg/mL. Galectin-3 increased from 379.1 ± 158.8 pg/mL to 438.7 ± 186.2 pg/mL, with a mean change of +59.56 ± 117.1 pg/mL. Any increase occurred in 77.8% of patients for NT-proBNP and 68.9% for galectin-3. No predefined baseline clinical, echocardiographic, or biomarker variable was significantly associated with ΔNT-proBNP or ΔGalectin-3. Conclusion: Early NT-proBNP and galectin-3 deterioration was common during anthracycline chemotherapy despite preserved baseline cardiac function. Routine baseline variables showed limited ability to identify patients who later developed biomarker worsening, supporting serial biomarker surveillance and dynamic risk assessment.