Fat accumulation in the kidney, or renal steatosis, in obesity and diabetes mellitus type 2 (DM2) is speculated as a contributing factor for pathogenesis of obesity-related and diabetic nephropathy¹. While renal biopsy is currently the only accepted method for evaluating renal steatosis, its utility as a clinical or research tool is limited due to procedural risks. Recently, a quantitative magnetic resonance imaging (qMRI) technique has become available for noninvasive imaging-based tissue fat quantification. Initially developed for hepatic steatosis, the technique has subsequently been applied in the pancreas, bone marrow, and kidneys². Its utility in the assessment of renal steatosis in obesity and DM2 is however unknown to date. In this pilot study, we assessed the feasibility of detecting the differences of renal fat quantity between non-diabetic and diabetic subjects using qMRI. This study was a retrospective data analysis of prospectively acquired data in three IRB-approved, HIPAA-compliant studies, in which human subjects underwent multiecho gradient echo imaging (mDixon-Quant) for fat fraction (FF) quantification on 3T whole body systems (Philips Achieva or Ingenia, Philips Healthcare, Best, the Netherlands) covering the liver and the kidneys. Study 1 included patients with known DM2 (n=19), study 2 new diagnosis of pancreatic cancer (n=10), and study 3 new diagnosis of renal cancer (n=40), with pooled patient population of N=69. Patient’s age, sex, body mass index (BMI) and DM2 status were recorded. Multiple axial mDixon-Quant images were acquired at slice thickness of 4-6 mm depending on the study protocol, 6 TEs (TE1=1.05 ms, DTE=0.8ms), TR 6.8 ms, and flip-angle of 2-3°, using 16-/32-element abdominal phased array coil (Achieva/Ingenia) with SENSE factor 2 when possible. In patients who could not fit into the magnet bore with a phased-array coil, built-in body coil was used with reduced number of slices. PDFF were measured within manually segmented regions-of-interest (ROIs) of the upper or inter-polar regions of the kidneys, the right hepatic lobe, and the spleen, and mean PDFF were recorded for each ROI. In 40 patients with renal cancer, PDFF was measured only in the non-tumor-bearing kidney. In the remaining 29 patients, PDFF was measured in both kidneys.The study population was comprised of non-DM2 and DM2 cohorts of 40 and 29 patients, respectively. The age±SD, sex (male:female), and BMI were 60.3±11.5 yrs, 29:11, and 29.7±6.2 kg/m2 for non-DM2 cohort, and 55.3±8.3 yrs, 16:13, and 36.8±9.6 kg/m2 for DM2 cohort. In 29 subjects with FF measurements in both kidneys, the agreement between right and left was analyzed by Bland-Altman Analysis; the mean difference was 0.021% with standard deviation 1.655% and the 95% limits of agreement [-3.264, +3.223]%. Figure 1 shows that Pearson’s correlation (r) analysis between renal FF and BMI (left), liver FF and BMI (middle), and renal FF and liver FF (right). Both renal and liver FF are weakly but significantly correlated with BMI, but renal and liver FFs are uncorrelated. Figure 2 shows the Wilcoxon rank sum test showing statistically significant renal FF elevation in the DM2 cohort compared to the non-DM2 cohort but no statistically significant difference in splenic FF between the two cohorts. Multiple linear regression analysis (Table 1) shows that DM2 is independently associated with elevated renal FF even after correction for the effects of age, sex, BMI, and liver FF. These findings are similar to histological and biochemical findings in the nephrectomy specimen in human kidneys¹.Renal steatosis quantification by qMRI is technically feasible. Renal FF is elevated in DM2, and may represent a new biomarker for diabetes and chronic kidney disease. However, further prospective validation is needed.