Urinary cell mRNA profiling of kidney allograft recipients: Development of a portable protocol for noninvasive diagnosis of T cell mediated rejection and BK virus nephropathy

BackgroundWe developed urinary cell mRNA profiling for the noninvasive diagnosis of acute T cell mediated rejection (TCMR) and BK virus nephropathy (BKVN), two significant post-transplant complications. Our profiling protocol for the multicenter Clinical Trial of Transplantation-04 (CTOT-04) study consisted of centrifugation of urine to prepare cell pellets, washes, addition of an RNA preservative, storage at -800 C and shipment in cold containers to our Gene Expression Monitoring Core for total RNA isolation and quantification of mRNA copies in RT-qPCR assays. To simplify profiling, we developed a filter-based protocol (ZFBP) that eliminated the need for centrifugation, RNA preservative, storage at -800 C, and shipment in cold containers for mRNA profiling; furthermore, kidney allograft recipients could be trained to perform the filtration of urine at home using the filter and post the urinary cell lysate containing the total RNA at ambient temperature to our Core for profiling. We have now refined ZFBP and investigated the diagnostic performance characteristics of a protocol designated as Weill Cornell Hybrid Protocol (WCHP).MethodsTotal RNA was isolated from kidney allograft biopsy matched urines from kidney allograft recipients using a filter-based protocol complemented by a silica-membrane based cartridge for mRNA enrichment (WCHP). Absolute copy numbers of CD3ε mRNA, CXCL10 mRNA and 18S ribosomal RNA, components of the CTOT-04 three gene TCMR diagnostic signature, and urinary cell BKV VP 1 mRNA copy number were measured using RT-qPCR assays. Mann-Whitney test, Fischer exact test and receiver operating characteristic (ROC) curve analysis were used for data analyses.ResultsUrinary cell three gene TCMR diagnostic signature in urines processed using the WCHP discriminated kidney allograft recipients with TCMR (n=12 biopsies from 11 patients) from those without TCMR (n=29 biopsies from 29 patients). The median (25th and 75th) score of the CTOT-04 three gene TCMR diagnostic signature was -0.448 (-1.664, 0.204) in the TCMR group and -2.542 (-3.267, -1.365) in the No Rejection biopsy group (P=0.0005, Mann-Whitney test). ROC curve analysis discriminated TCMR group from the No Rejection group and the area under the ROC curve (AUROC) was 0.84 (95% Confidence Intervals [CI], 0.69 to 0.98) (P<0.001), and TCMR was diagnosed with a sensitivity of 67% (95% CI, 35 to 89) at a specificity of 86% (95% CI, 67 to 95) using the CTOT-04 validated cutpoint of -1.213 (P=0.0016, Fischer exact test). BKV VP1 mRNA copy number in urines processed using the WCHP discriminated patients with BKVN (n=7) from those without BKVN (n=29) and the AUROC was 1.0 (95% CI, 1.00 to 1.00) (P<0.0001) and BKVN was diagnosed with a sensitivity of 86% (95% CI, 42 to 99) at a specificity of 100% (95% CI, 85 to 100) with the previously validated cutpoint of 6.5 × 108 BKV-VP1 mRNA copies per microgram of total RNA (P<0.0001, Fischer exact test).ConclusionUrine from kidney allograft recipients processed using the WCHP predicted TCMR and BKVN. WCHP represents not only a significant advance towards portability of urinary cell mRNA profiling but also improved patient management by minimizing their visits for urine collection.