SARC-F as a screening tool to detect computed tomography-based sarcopenia and myosteatosis among older adults with cancer

Abstract

Background: The European Working Group on Sarcopenia in Older People (EWGSOP) recommends SARC-F as a tool for identifying sarcopenia among older adults. However, the role of SARC-F among older adults with cancer remains unexplored. We aimed to evaluate the diagnostic utility of SARC-F to identify those with sarcopenia, or low muscle mass (using skeletal muscle index [SMI]), and myosteatosis (using skeletal muscle density [SMD]) from computed tomography (CT) imaging and the association of SARC-F with all-cause mortality.

Methods: Older adults (≥60 years) presenting for initial consultation at UAB medical oncology clinic who underwent geriatric assessment were enrolled in a prospective cohort study. We identified study participants who completed SARC-F screening and had available CT imaging within 60 days of study enrollment. Using single-slice CT images at the L3 vertebral level, we computed SMI and SMD using published methods. Sarcopenia and myosteatosis were defined using published cutpoints. We calculated the sensitivity and specificity of SARC-F for detecting low muscle mass and low muscle density using published thresholds. Finally, we computed the impact of SARC-F and CT measures on overall survival using Kaplan-Meier curves and Cox regression models, after adjusting for age, sex, cancer type, and cancer stage.

Results: We identified 212 older adults with a median age of 68.8 years; with 60.8% males, 76.6% whites, and pancreatic cancer (21.2%) being the most common malignancy. In the overall cohort, 30.7% had abnormal SARC-F using published cutpoints. SARC-F ≥ 4 had a sensitivity of 35% and a specificity of 76% to identify low muscle mass. SARC-F ≥ 4 had a sensitivity of 38% and a specificity of 74% to identify low muscle density. Those with SARC-F ≥ 4 and low SMI/SMD had worse survival compared to those with low SMI/SMD alone. Incorporating SARC-F improved survival prognostication beyond SMI and SMD (HR = 3.1; p < 0.001; Harrel's C from 0.73 to 0.76).

Conclusions: SARC-F as a screening tool has limited diagnostic utility for identifying older adults with low muscle mass and/or density. However, SARC-F retains prognostic value independent of CT-based muscle measures in predicting mortality among older adults with cancer.

Keywords: SARC-F; cancer; computed tomography; myosteatosis; sarcopenia.