Abstract PS1-06-17: A retrospective analysis of capecitabine-induced imaging hepatic steatosis in metastatic breast cancer: prevalence and resolution

Abstract Background: Capecitabine is a valid monotherapy option in multiple metastatic breast cancer (mBC) subtypes. While case reports of capecitabine-related hepatic steatosis have been published, its prevalence remains unknown. Traditional monitoring relies primarily on biochemical tests, which may not fully capture hepatic changes. Our aim was to evaluate the prevalence of capecitabine-related hepatic steatosis and determine whether routinely performed 18FFDG PET/CT could quantify hepatic changes during capecitabine treatment, including in patients without visually evident steatosis. Methods: We conducted a retrospective analysis of patients with mBC treated with capecitabine between 2014 and 2021 at Gustave Roussy. Clinico-biological variables (age, BMI, metastatic sites, therapy line, dihydropyrimidine dehydrogenase (DPD) testing, biochemical liver tests) at baseline were collected. Baseline PET/CT, first evaluation (180 days), and last PET/CT scans under capecitabine treatment were analyzed to assess hepatic steatosis prevalence. Hepatic steatosis was defined by visual assessment as a decreased liver attenuation relative to spleen and/or increased hepatic vessel visualization on unenhanced CT. Finally, hepatic changes were quantified using standardized uptake values (SUV) on PET images and Hounsfield units (HU) on CT images. Results: The study cohort included 183 patients with mBC. The median age at capecitabine initiation was 66.3 years (range 51-96), and median BMI was 24.2 (range 14.7-39.1). The most frequent metastatic sites were bone (67%), liver (40%) and lymph nodes (33%), with a median of 2 metastatic sites per patient. 86% of patients received capecitabine as monotherapy. Capecitabine was administered across multiple treatment lines: 1st line (14%), 2nd line (32%), 3rd line (31%) and = 4th line (23%). When available, DPD testing revealed partial deficit in 16/138 patients (12%). Median treatment duration was 298 days (range 80-2162). Sequential imaging analysis was available for 122 patients with both baseline and first evaluation PET/CT scans, and 77 patients at capecitabine discontinuation. No patients showed hepatic steatosis at baseline. At first evaluation, hepatic steatosis was observed in 3/122 patients (2.5%). All 3 patients with steatosis underwent dose reduction due to other capecitabine toxicities (hand-foot syndrome in 2 patients, general condition deterioration requiring weight-based dose adjustment in 1 patient), resulting in resolution of CT steatosis in 2 patients. At final PET/CT, steatosis was present in 4/77 patients (5.2%) and resolved in 3 patients on follow-up imaging after changing treatment line. All patients who developed visible steatosis received capecitabine monotherapy. Quantitative analysis revealed small magnitude hepatic changes during capecitabine treatment: mean HU liver decreased by -2.3 units (95%CI -4.0 to -0.6, p=0.007) and SUVmean liver increased by +0.2 units (95% CI +0.1 to +0.2, p0.001). These quantitative changes occurred independently of baseline liver metastases presence (p=0.470 for HU, p=0.881 for SUVmean). Among clinico-biological variables, only baseline bilirubin showed significant correlation with HU changes (p=0.044), while other liver function tests, BMI, and DPD status showed no significant association. Conclusions: This study provides a first estimate of capecitabine-induced hepatic steatosis prevalence, and initial evidence supporting the feasibility of using follow-up 18FFDG PET/CT to assess its occurrence and reversibility after dose reduction or discontinuation. Further studies are needed to evaluate the potential impact of systematic steatosis evaluation on imaging follow-up, including shorter follow-up intervals and preemptive dose adjustments. Citation Format: A. Daverio, T. Ben Ahmed, J. M Ribeiro, C. Bousrih, E. Rassy, M. Mokdad-Adi, S. Morbelli, B. Pistilli, D. Deandreis, A. Viansone, T. Henry. A retrospective analysis of capecitabine-induced imaging hepatic steatosis in metastatic breast cancer: prevalence and resolution abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS1-06-17.