Body composition measures as predictors of toxicity in patients with metastatic breast cancer receiving PARP inhibitors.

Abstract

e13010 Background: Poor body composition metrics (BCM), such as high adiposity or low muscle mass, are associated with inferior cancer outcomes. Poly ADP-ribose polymerase inhibitors (PARPi) revolutionized the treatment of metastatic breast cancer (MBC) with germline BRCA mutation but can cause substantial toxicity. The recommended dose of PARPi is fixed and neither weight nor BCM are taken into consideration during initial prescription or during dose reductions. We aimed to investigate the association between BCM and toxicity and tolerability to PARPi in MBC patients. Methods: We identified all MBC patients treated with PARPi between 2015-2022 at Tel Aviv Sourasky Medical Center. Using computerized tomography (CT) images taken for the evaluation of disease burden, BCM such as: skeletal muscle area (SMA) and subcutaneous adipose tissue area (SAT), were measured at the level of the third lumbar (L3) vertebrae. Low SMA was defined as < 107.2 cm2 and low SAT was defined as < 190 cm2. Receiver operating characteristic (ROC) curves were used to define these thresholds. Sarcopenia was defined as skeletal muscle index (SMI = SMA/height2) < 41 based on literature. Fisher exact tests and Mann-Whitney U tests were used for statistical analyses. Results: We identified 30 patients with MBC who were treated with PARPi and met the study requirements at a median age of 54 (range 29-87). Median duration of PARPi treatment was 214 days (range 35-921). Among them, 19 (63%) were sarcopenic, 22 (73%) had low SMA, 10 (33%) had low SAT and 17 (57%) suffered from Grade 2-4 adverse events (AE). Patients with sarcopenia had non-significantly more Grade 3-4 AE (42% vs. 18%) and more dose reduction or dose delay events (42% vs. 27%) compared to non-sarcopenic patients. Patients with low SMA had borderline significantly more Grade 3-4 AE, dose reduction or dose delay events (55% vs. 13%, p = 0.065), compared to patients with high SMA. Patients with low SAT had significantly more Grade 3-4 AE (60% vs 20%, P = 0.036), compared to patients with high SAT. Moreover, patients with Grade 2-4 AE had significantly lower SAT compared with patient with < Grade 2 AE (median 183 vs. 298 cm2, p = 0.008). Conclusions: Our study implies an association between BCM and PARPi toxicity and tolerability. Skeletal muscle mass and adipose tissue may be taken into consideration in PARPi treatment and dosing decisions. A larger study is needed to further examine this hypothesis.