Impact of the Current versus the Previous Diagnostic Threshold on the Outcome of Patients with Borderline Changes Suspicious for T-cell-mediated Rejection Diagnosed on Indication Biopsies

Background Since the borderline changes suspicious for acute T-cell-mediated rejection (BL) category was broadened, there has been a debate regarding the right threshold for tubulitis (t) and interstitial inflammation (i) scores. Methods We studied a first cohort of 111 patients with BL found on an indication biopsy between 2006 and 2016 and compared those with scores of t1i0 (BLt1i0) to those with higher scores (BL≥t1i1). A second cohort of 56 patients with BL was used for external validation. We used a composite endpoint of death-censored graft failure or doubling of the serum creatinine level postbiopsy. Results In the first cohort, sixty-eight percent (75/111) of the BL cases fell in the BLt1i0 group. The occurrence of the composite endpoint was 5% and 14% for BLt1i0 and BL≥t1i1, respectively. In contrast, the endpoint occurred in 5% of nonrejectors and 21% of patients with T-cell-mediated rejection (TCMR). In the validation cohort, 8% versus 36% of BLt1i0 and BL≥t1i1 reached the endpoint, respectively. Multivariable Cox modeling revealed that BLt1i0 patients had a prognosis similar to that of nonrejectors (adjusted hazard ratio, HR 0.6, 95% confidence interval 0.1 to 2.2, P = 0.40) but better than that of patients with BL≥t1i1 (HR 3.8, 95% confidence interval 1.3 to 11.5, P = 0.02). Sensitivity analyses restricted to death-censored graft loss or using time post transplant as the time of reference provided similar results. Conclusions In summary, patients with BLt1i0 have a different prognosis to that of BL≥t1i1 patients, which brings into question the current diagnostic thresholds.