CT scan versus bioelectrical impedance spectrometry sarcopenia assessment to predict chemotherapy toxicity in early breast cancer.

Abstract

12078 Background: There is growing literature on the impact of the progressive loss of skeletal muscle known as sarcopenia in women with early breast cancer (EBC). Imaging techniques such as Computed Tomography (CT) scans are the most studied methods to detect sarcopenia in patients with cancer; however, most patients with EBC do not require routine CT scans. Bioelectrical impedance spectrometry (BIS) is a method that does not expose patients to radiation and provides measurements to assess sarcopenia instantaneously. The extent to which BIS-assessed sarcopenia correlates with CT-assessed sarcopenia is uncertain. It is also unclear whether sarcopenia detected with either method can be associated with chemotherapy tolerance To address these questions, we evaluated the correlation between CT and BIS sarcopenia and associations of sarcopenia by each method with chemotherapy related outcomes in patients with EBC. Methods: This retrospective study identified patients with EBC who received chemotherapy (ACT, TC, TH, THP), underwent BIS analyses (ImpediMed) at any point between diagnosis and treatment completion and also received a CT abdomen including L3 level. Axial CT L3 segments were analyzed using Slice-O-Matic software. CT L3, and BIS generated the Skeletal Muscle (SM). Skeletal Muscle Index (SMI) was then calculated to assess for sarcopenia: BIS SMI= (SM in kg)/ (patient height, m2) and CT L3 SMI (SM cm2/ height m2). Patients were divided into normal SMI, and sarcopenia (if BIS was used SMI <6.75 kg/m2 and if CT SMI <40 cm2/m2). Patient medical records were reviewed for patient characteristics, grade 3 and 4 toxicity, including neuropathy symptoms, chemotherapy dose reductions, early treatment discontinuation or need for hospitalization during chemotherapy treatment. Spearman's rank coefficient was used to assess the correlation between CT SMI and BIS SMI. Multivariable logistic regression was used to associate sarcopenia with outcomes, controlling for age and BMI. Results: 361 patients received chemotherapy, underwent BIS; of those 171 had L3 CT scans. The correlation between L3 CT SMI and BIS SMI was r= 0.64 p<0.0001. In the multivariable model, sarcopenia assessed by L3 CT scan was associated with chemotherapy dose reduction (CI 0.04-0.60 p=0.01) and neuropathy (CI 0.10-0.95 P=0.04) but not with grade 3 or 4 chemotherapy toxicity (CI 0.13-1.09 p=0.07), early treatment discontinuation (p=0.3) or hospitalization (p=0.60). Sarcopenia assessed by BIS was associated with grade 3 or 4 chemotherapy toxicity (CI 0.2-0.83 p=0.01), neuropathy (CI 0.1-0.51 p=0.0003), early treatment discontinuation (OR 0.26 CI 0.11-0.80 p=0.001), and hospitalization (CI 0.12-0.68 p=0.004). Conclusions: There is a strong correlation between L3 CT scan and BIS screened sarcopenia in patients with EBC. Sarcopenia detected with either CT L3 or BIS is associated with worse chemotherapy tolerance.