Body Surface Area and Baseline Blood Pressure Predict Subclinical Anthracycline Cardiotoxicity in Women Treated for Early Breast Cancer

Abstract

Background and AimsAnthracyclines are highly effective chemotherapeutic agents which may cause long-term cardiac damage (chronic anthracycline cardiotoxicity) and heart failure. The pathogenesis of anthracycline cardiotoxicity remains incompletely understood and individual susceptibility difficult to predict. We sought clinical features which might contribute to improved risk assessment.MethodsSubjects were women with early breast cancer, free of pre-existing cardiac disease. Left ventricular ejection fraction was measured using cardiovascular magnetic resonance before and >12 months after anthracycline-based chemotherapy (>3 months post-Trastuzumab). Variables associated with subclinical cardiotoxicity (defined as a fall in left ventricular ejection fraction of ≥5%) were identified by logistic regression.ResultsOne hundred and sixty-five women (mean age 48.3 years at enrollment) completed the study 21.7 months [IQR 18.0–26.8] after starting chemotherapy. All received anthracyclines (98.8% epirubicin, cumulative dose 400 [300–450] mg/m2); 18% Trastuzumab. Baseline blood pressure was elevated (≥140/90mmHg, mean 147.3/86.1mmHg) in 18 subjects. Thirty-four subjects (20.7%) were identified with subclinical cardiotoxicity, independent predictors of which were the number of anthracycline cycles (odds ratio, OR 1.64 [1.17–2.30] per cycle), blood pressure ≥140/90mmHg (OR 5.36 [1.73–17.61]), body surface area (OR 2.08 [1.36–3.20] per standard deviation (0.16m2) increase), and Trastuzumab therapy (OR 3.35 [1.18–9.51]). The resultant predictive-model had an area under the receiver operating characteristics curve of 0.78 [0.70–0.86].ConclusionsWe found subclinical cardiotoxicity to be common even within this low risk cohort. Risk of cardiotoxicity was associated with modestly elevated baseline blood pressure–indicating that close attention should be paid to blood pressure in patients considered for anthracycline based chemotherapy. The association with higher body surface area suggests that indexing of anthracycline doses to surface area may not be appropriate for all, and points to the need for additional research in this area.

Anthracycline associated cardiotoxicity in a sub-Saharan African population - tertiary care experience

“…This was a study looking at factors associated with early anthracycline cardiotoxicity in a young, female, African population with breast cancer. The mean and median ages were comparable to similar studies in African Americans (12), local populations (41) and non-black cohorts (16). The large proportion of breast cancer patients reflects the predominant use of anthracyclines coupled with high prevalence.…”

“…With varying definitions, comparison between studies are difficult. However, our cardiotoxicity prevalence was higher than 22% in an Australian study (cardiotoxicity -relative reduction in GLS > 11%) (42), and 20.7% in a study of non-blacks (Cardiotoxicity ->5 percentage points drop in LVEF on cardiac magnetic resonance imaging (16). Our study was also higher than 21% in a study by Sawaya et al, who defined cardiotoxicity on transthoracic echocardiography as a drop in LVEF > 5 percentage points (to < 55% with symptoms) or a drop in LVEF > 5 percentage points (to < 55% regardless of symptoms).…”

“…Our average duration of 14.3 months, with approximately half having an evaluation at ≤ 1 year, suggests that cardiotoxicity may occur at any time, as mirrored by studies which looked at pathological changes after one dose (18), echocardiographic changes at seven days after completion (16) or echocardiographic parameters at > 1 year (12). After a sharp rise in the first two quarters in our cohort, progress was linear, with half of the cohort showing changes by approximately one year.…”

“…Heterogeneity exists in study definitions of anthracycline cardiotoxicity: absolute fall ≥ 5 percentage points in LVEF (16), absolute value of LVEF < 50% (41), absolute value of LVEF < 45% or heart failure (12) and > 11% relative GLS decline from baseline (42), with no assessment of GLS in an African population. With varying definitions, comparison between studies are difficult.…”

“…Earlier studies, however, give disparate results with no association with race (14) and there is generally poor representation of blacks in adult risk evaluation studies (15). Other cardiotoxicity predictors include female gender, concurrent and prior treatment with cardiotoxic agents, age (> 65 years or < 18 years ) (13), co-morbidities (renal failure, heart disease, systemic hypertension) (6) pre-treatment blood pressure and body surface area (BSA) (16). Epirubicin is less cardiotoxic compared to doxorubicin (17)(18)(19), and the final cardiotoxic profile when other nonanthracycline chemotherapy are co-administered may reflect a synergistic action (20).…”