BK virus nephropathy (BKN) after kidney transplantation is often associated with TCMR-like histology, but pathologists are uncertain about whether this is TCMR. We studied the histologic and molecular phenotype of BKN in kidney allografts in 703 biopsies for clinical indication, including 25 with BKN, 67 with TCMR but no BKN. Time of biopsy post Tx was similar in both groups (BKN 11±9, TCMR 12±23 months). GFR was lower in patients with BKN (34±14 vs 45±19 ml/min).

All biopsies with BKN showed interstitial inflammation but with varying intensity. 9/26 biopsies with BKN (35%) met histologic criteria for TCMR (i-score ≥2, t-score ≥2). BKN biopsies with additional histologic signs of TCMR had higher molecular TCMR scores than BKN alone (0.63±0.48 vs 0.07± 0.27, p=0.0348), similar to those in pure TCMR (0.54±0.47, p=0.847) (Fig.1). BKN biopsies with high TCMR score had higher expression of inflammatory and injury gene sets and previously defined molecular rejection scores compared to BKN biopsies with low TCMR scores.

In BKN and in TCMR, GFR at biopsy was negatively correlated with tissue injury gene sets (r=-0.465 and r=-0.417). In BKN, GFR also correlated negatively with expression of TCMR score (r=0.442) and inflammatory gene sets (representing T cells (r=-0.464) and macrophages (r=-0.467)). Comparison of gene expression between BKN and other diseases yielded very few differentially expressed genes. A molecular classifier that was built to distinguish biopsies with TCMR from BKN failed, probably because many BKN have TCMR.

In summary, typical TCMR is common in BKN but is often not diagnosed. Thus BKN must be classified as two populations: BKN alone vs BKN+TCMR, which are not readily distinguished by histology. Although there is no molecular definition of BKN per se, the molecular definition of TCMR reliably distinguishes these populations. BKN with TCMR features manifests worse function than BK alone because it has two diseases impairing function. Whether treatment of TCMR in the context of BKN is helpful or harmful is not known, and must determined in trials. Molecular identification of TCMR will help case selection for such studies.

« Back to 2013 American Transplant Congress