Abstract: FR-PO1190
Urinary CXCL10 Detects Active Tubulointerstitial Inflammation in Kidney Transplant Rejection Missed by Donor-Derived Cell-Free DNA
Session Information
- Transplantation: Clinical - Rejection, Biomarkers, and Pharmacology
October 23, 2026 | Location: Exhibit Hall A, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Transplantation
- 2002 Transplantation: Clinical
Authors
- Paulus, Amber B., Virginia Commonwealth University, Richmond, Virginia, United States
- Seelam, Stephen Rithvik, Virginia Commonwealth University, Richmond, Virginia, United States
- Sinhmar, Pawan, Virginia Commonwealth University, Richmond, Virginia, United States
- Christensen, Johanna L., Virginia Commonwealth University, Richmond, Virginia, United States
- Brown, Andrew, Virginia Commonwealth University, Richmond, Virginia, United States
- Gupta, Gaurav, Virginia Commonwealth University, Richmond, Virginia, United States
Background
Donor-derived cell-free DNA (dd-cfDNA) is less sensitive for TCMR than ABMR: ABMR is a microvascular process that releases dd-cfDNA into the circulation, whereas TCMR is primarily a tubulo-interstitial process. Urinary CXCL10, an IFN-γ–inducible chemokine produced by tubular epithelium that recruits CXCR3+ effector T cells, signals tubulo-interstitial inflammation and may detect TCMR — and ABMR with tubulo-interstitial features — that dd-cfDNA misses. A biomarker can only appear as accurate as the reference standard used to judge it; dd-cfDNA correlates better with the Molecular Microscope Diagnostic System (MMDx) than histology. We evaluated urinary CXCL10 at a published 3 ng/mmol creatinine threshold in kidney transplant recipients (KTRs) with dd-cfDNA <1% and MMDx-confirmed rejection.
Methods
Single-center convenience cohort of 19 adult KTRs with indication biopsy May 2023–Aug 2025, dd-cfDNA <1%, urine from biopsy day, and MMDx-confirmed rejection (10 ABMR, 9 TCMR/Mixed). Urinary CXCL10 was measured at One Lambda Laboratories by a CLIA-validated LDT, blinded to histology, dd-cfDNA, and MMDx. Cases were stratified by CXCL10 >3 vs ≤3 ng/mmol (n=11 high; n=8 low). MMDx tubulo-interstitial (TCMR Score, t>1, i>1) versus microvascular (ABMR Score, g>0, ptc>0, MVI) features and histology were compared.
Results
Urinary CXCL10 was above threshold in 11/19 (58%) dd-cfDNA-occult MMDx rejections (median 4.88 [IQR 3.95–21.55] vs 0.65 [0.54–0.84] ng/mmol, high vs low); dd-cfDNA did not differ between strata (0.40% vs 0.53%; P=0.12). CXCL10-high cases had higher tubulo-interstitial inflammation across both TCMR and ABMR: TCMR Score (0.43 vs 0.19; P=0.047), tubulitis>1 (0.62 vs 0.28; P=0.039), and interstitial inflammation>1 (0.41 vs 0.21; P=0.057) were elevated; microvascular features were comparable (ABMR Score, g>0, ptc>0, MVI; all P>0.6). Histology under-diagnosed rejection in 7/19 (37%) overall and 5/11 (45%) of CXCL10-high cases.
Conclusion
In KTRs with dd-cfDNA <1% and MMDx-confirmed rejection, urinary CXCL10 >3 ng/mmol identified active tubulo-interstitial inflammation, regardless of Banff diagnosis. This reflects complementary biology: dd-cfDNA tracks endothelial injury (ABMR axis); CXCL10 tracks T-cell trafficking into the tubulo-interstitium. Optimal CXCL10 thresholds remain to be defined. Prospective validation is underway (NCT07415876).
Funding
- Commercial Support – One Lambda (Thermo Fisher)