Kidney Week

Abstract: FR-OR002

Association of CT Body Composition Features with eGFR and Its Longitudinal Slope Among Patients Treated with Molecularly Targeted Cancer Therapies

Session Information

• Advancing Onconephrology Precision: Biomarkers, Therapeutic Targets, and Patient Outcomes
  October 23, 2026 | Location: Room 401, Convention Center
  Abstract Time: 04:40 PM - 04:50 PM

Category: Onconephrology

• 1600 Onconephrology

Authors

• Ziolkowski, Susan, Stanford University School of Medicine, Stanford, California, United States

Susan Ziolkowski, MD, MS, FASN

• Parvathinathan, Gomathy, Stanford University School of Medicine, Stanford, California, United States

Gomathy Parvathinathan, MS

• Long, Jin, Stanford University School of Medicine, Stanford, California, United States

Jin Long, PhD

• Mottez, Clemence, Stanford University, Stanford, California, United States

Clemence Mottez

• Stedman, Margaret R., Stanford University School of Medicine, Stanford, California, United States

Margaret R. Stedman, PhD, MPH

• Anand, Shuchi, University of California Davis, Davis, California, United States

Shuchi Anand, MD, MS

Background

Computed tomography (CT) scans are routinely obtained in patients with cancer and may provide opportunistic biomarkers of kidney health. Low muscle mass can falsely elevate creatinine-based eGFR despite underlying kidney dysfunction, potentially masking patients at risk for subsequent decline in kidney function. Central obesity is associated with lower eGFR and albuminuria.

Methods

We identified adults initiating molecularly targeted cancer therapies with CT imaging within 6 months before to 30 days after treatment initiation. Automated segmentation quantified skeletal muscle, subcutaneous (SAT), visceral (VAT), and intermuscular adipose tissue (IMAT), and psoas muscle density (PMD) at the L3 level; skeletal muscle index (SMI) was calculated as muscle area/height2. Linear mixed-effects models evaluated associations between CT measures with baseline eGFR, and longitudinal eGFR slope.

Results

The cohort included 1,335 patients (57% female, 21% diabetes) with a mean (SD) age of 60.5 (14) years and a baseline mean (SD) eGFR of 87.8 (23.9). 55% of patients had sarcopenia and 52% of patients had low PMD. At baseline, SMI, SAT, VAT, and IMAT were negatively associated with eGFR. PMD was not associated with baseline eGFR. In longitudinal models, low PMD was associated with greater 1-year eGFR decline compared with normal PMD (difference in slope: −4.22 mL/min/1.73 m2/year; 95% CI, −7.8 to −0.60; p=0.02).

Conclusion

CT-derived measurements of skeletal muscle and fat mass are associated with lower eGFR. Low PMD, a marker of poor muscle quality and myosteatosis, is associated with eGFR decline despite no association with baseline eGFR. CT-derived features may help identify patients with cancer at risk for future kidney dysfunction.

Funding

• NIDDK Support