Abstract 4367220: Time-updated spatial ventricular gradient orientation is associated with anthracycline-associated heart failure

Background: Anthracycline use is limited by cardiotoxicity resulting in heart failure (HF). The spatial ventricular gradient (SVG) is a vectorcardiographic (VCG) measurement of myocardial electrical heterogeneity that is associated with adverse outcomes including sudden death and incident HF. Objective: Determine if serial assessment of SVG direction is associated with anthracycline-associated HF. Methods: Retrospective analysis of adult cancer patients treated with anthracyclines in 1992-2019 at a single institution. Patients with prior HF were excluded. Standard 12-lead ECGs obtained within 6 months before and up to 5 years after anthracycline initiation were analyzed. VCGs were constructed, and SVG vector coordinates were obtained by integrating under the X, Y, and Z QRST complexes. Multivariable competing-risks regression was performed with SVG azimuth (the angle of the SVG in the transverse plane) included as a time-varying covariate with sine/cosine transformation to account for the circular nature of SVG azimuth, incident HF as the main outcome, and death as a competing outcome. Results: Among 889 patients (51% male, mean age at drug initiation 58±15 yrs), 70% had hematologic malignancies, 12% breast cancer, and 18% other malignancies. Incident HF developed in 97 patients (11%), and 305 patients (34%) died within 5 yrs of treatment. In unadjusted analyses, time-updated SVG azimuth was associated with incident HF (joint p=0.003). After adjustment for known risk factors for anthracycline-associated HF, including age, sex, hypertension, CAD, diabetes, cumulative doxorubicin equivalent dose, heart rate and QRS duration, SVG azimuth remained significantly associated with incident HF (joint p=0.045) with adjusted sub-distribution hazard ratios (SHRs) ranging from 0.47 to 2.13 across the circular spectrum (4.5-fold SHR variation). The Figure shows how SHRs for incident HF varied significantly by SVG azimuth. Notably, the anterior/leftward orientation, (see Figure green area) was associated with a reduced risk of HF with SHRs ~0.5-0.6, while the posterior/rightward orientation, (see Figure red area) was associated with a ~2-fold increased risk for incident HF. Conclusion: SVG azimuth is independently associated with increased risk of anthracycline-associated HF. Measuring how SVG azimuth changes during anthracycline treatment may enhance early identification of patients at risk for HF and inform preventive cardioprotective strategies during cancer therapy.