Proteomic biomarkers differentiate reversible from irreversible renal dysfunction in liver transplant recipients

Josh Levitsky¹, Daniel Salomon², Peter Langfelder³, Stephen Horvath³, Chunfa Jie¹, Ed Wang¹, Talia Baker¹, Lorenzo Gallon¹, Sunil Kurian², Tony Mondala², Karri Ballard⁴, Ralph McDade⁴, John Friedewald¹, Michael Abecassis¹.

¹Comprehensive Transplant Center, Northwestern University, Chicago, IL; ²The Scripps Research Institute, Department of Molecular and Experimental Medicine, La Jolla, CA;

³UCLA Biostatistics, Los Angeles, CA; ⁴Rules-Based Medicine, Austin, TX, USA.

Renal dysfunction (RD) has a major impact on post-liver transplant (LT) survival. Therefore, biomarkers that can predict the natural history of RD in this patient population, beyond clinical data, risk factors and serum creatinine, would be useful. We have utilized a novel Luminex-based, multi-analyte proteomic plasma/urine panel in two LT settings. First, we compared plasma and urine protein arrays in LT recipients with reversible RD (n=7; 56±9 years, eGFR at LT 25±10 mL/min) to age/etiology-matched controls with no RD (n=9; 56±6 years, eGFR at LT 77±20 mL/min) at the time of, and one month post LT. Only 2/14 plasma and 0/16 urine proteins showed differences between RD and no RD and the 2 proteins (TFF3, OPN) normalized with recovery of RD. No proteins typically associated with acute renal failure (KIM-1, NGAL, ß2M) were elevated in the group with transient RD. Second, we compared plasma protein arrays in LT recipients (>3 years post) with chronic kidney disease - CKD (n=33; 62±3 years, eGFR 44±5mL/min) to matched controls with no CKD (n=31; 54±5 years, eGFR 78±8 mL/min). Ten plasma proteins (4/13 from kidney injury panel; 6/189 from broader discovery panel) were significantly elevated in the CKD group in separate test and validation cohorts (TFF3, a1M, FABP, ß2M, Factor VII, ApoLH, CgA, ApoLCIII, CD40, and CystC). These data show that pre LT reversible RD is associated with elevation of no urine, and only two plasma proteins, both of which resolve with renal recovery after LT. In contrast, in LT recipients who developed CKD, 10 plasma proteins were significantly elevated and the changes predicted eGFR. These results suggest a role for novel proteomic panels in both predicting recovery of acute RD and in the development of CKD following LT, thus potentially informing therapeutic decisions in this population. Further validation is now underway.