Impact of CT-Derived Body Composition Analysis on the Performance of GFR Estimating Equations in Patients with Cancer

Sarcopenia and obesity are common in patients with cancer and may reduce the accuracy of estimated glomerular filtration rate (eGFR) equations. We evaluated the performance of recommended eGFR equations based on creatinine or cystatin C, and novel GFR markers β2-microglobulin (B2M) and β-trace protein (BTP) according to body composition derived from computed tomography (CT). Prospective cohort study of adult patients with solid tumors recruited between May 2015 and October 2017 who had a CT scan within 90 days of measured GFR (mGFR) using plasma clearance of 51Cr-EDTA. eGFR was calculated with the CKD-EPI equations using either creatinine (eGFRCR); cystatin C (eGFRCYS); creatinine and cystatin (eGFRCR-CYS); creatinine and B2M (eGFRCR-B2M), cystatin, B2M, and BTP (eGFRCYS-B2M-BTP); or creatinine, cystatin, B2M, and BTP (eGFRCR-CYS-B2M-BTP). Bias was assessed as the median of the differences between mGFR and eGFR. Accuracy was assessed as the percentage of estimates that differed by more than 30% from the mGFR (1-P30). 1-P30 < 10%, 10-20%, and > 20% are considered optimal, acceptable, and poor accuracy, respectively. Skeletal muscle index (SMI) was quantified on CT and calculated by dividing the skeletal muscle cross-sectional area by the patient's height squared. Of 465 patients included, 157 (34%) met criteria for sarcopenia. Bias varied by magnitude of SMI. In patients with sarcopenia, the accuracy of eGFRCR and eGFRCYS was poor (1-P30 42.0% [95% CI 34.4, 49.6] and 20.4% [95% CI 14.0, 26.8], respectively). eGFRCR-CYS had acceptable accuracy (1-P30: 14.0 [8.3, 19.1] %) whereas eGFRCYS-B2M-BTP and eGFRCR-CYS-B2M-BTP had optimal accuracy (1-P30: 7.0 [3.2, 10.8] % and 8.3 [3.8, 12.3] %, respectively). Obesity did not significantly affect bias or accuracy. GFR estimates based on eGFRCR and eGFRCYS are not sufficiently accurate in patients with cancer and sarcopenia. Body composition analysis can identify patients in need of more accurate GFR assessment.