Anthracycline-induced arterial toxicity: structural and inflammatory assessments in non-human primates and lymphoma patients using 18F-FDG positron emission tomography

Abstract Background Anthracyclines used in cancer treatment are associated with arterial injury. Histopathological insights have been limited to small animal models. The role of inflammation in the arterial toxic effects of anthracycline is unclear in humans. Purpose 1) To evaluate anthracycline-associated aortic fibrosis in a non-human primate model; 2) To assess the effect of anthracycline upon aortic inflammation in patients treated for lymphoma. Methods 1) Five female African Green monkeys (AGM) received doxorubicin (30–60 mg/m²/biweekly intravenously, cumulative dose: 240 mg/m²). Blinded histopathologic analyses of collagen deposition in the ascending aorta were performed after necropsy 15 weeks after the last doxorubicin dose (~12 months equivalent in humans) and compared to 5 age-matched healthy, untreated AGMs. 2) Analysis of the thoracic aorta of 101 patients with diffuse large B-cell lymphoma (DLBCL), before and after doxorubicin exposure, was performed using 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT). The primary outcome was the change in maximal tissue-to-background ratio (TBRmax) of the thoracic aorta. Univariate analyses of aortic uptake including baseline demographics, aortic calcification and cardiovascular (CV) disease were performed. Results In AGMs (age 13±1y), doxorubicin exposure was associated with greater aortic fibrosis than untreated controls (collagen deposition in doxorubicin treated was 6.23±0.88% vs. 4.67±0.54% in controls; p=0.01; Figure 1). PET/CT from 101 patients with lymphoma treated with anthracyclines (mean age 64±12y; 47% female, median cumulative doxorubicin dose 300mg/m²) were analysed. The mean time from starting anthracycline to follow up PET/CT was 4.8 months and from finishing anthracycline to follow up PET/CT was 1.2 months (IQR 0.9-1.6). There was no change in aortic TBRmax after anthracycline exposure (pre-doxorubicin TBRmax 1.46±0.16 vs post-doxorubicin TBRmax 1.44±0.14, p=0.14; Figure 2). The absence of change in TBRmax was consistent in univariate analyses including age, sex, lymphoma stage, baseline CV risk and in those with and without pre-existing cardiovascular disease or aortic calcification. Further assessment including TBRmean, ‘active segments’ and ‘most diseased segment’ analyses yielded consistent results. Conclusions In a large animal model, with a dosing regimen similar to that used for the treatment of patients with lymphoma, anthracycline exposure was associated with aortic fibrosis. In patients with lymphoma, anthracycline exposure was not associated with aortic inflammation assessed by 18F-FDG-PET/CT. Further research is required to elucidate the time course and mechanisms of anthracycline-associated arterial fibrosis in order to inform strategies to prevent this vascular toxicity and its clinical consequences.