Polyomavirus BK associated hemorrhagic cystitis (HC) is becoming an increasingly recognized complication of HSCT. The pathogenesis of BK-associated HC is not fully understood, and it is not possible to predict the clinical course of BK reactivation in patients following HSCT. Multiple co-factors could play a role in the development of HC after HSCT in patients (pts) with BK viruria (BKV) including the type of transplant autologous (auto) vs. allogeneic (allo), conditioning regimen, development of Graft-Versus-Host Disease (GVHD), conditioning intensity and source of donor stem cells (Matched unrelated [MUD] vs. HLA-identical sibling [SIB]).We retrospectively analyzed the incidence, risk factors and outcome of BK reactivation and HC in pts who had HSCT at the Univ. of Pittsburgh Cancer Institute during the last 2 years. BKV was determined using a quantitative real-time PCR assay considered positive for titers higher than 10 copies. In total, 247 pts were studied, 63% (155/247) received an auto and 37% (92/247) an allo HSCT. BKV was observed in 20% (49/247) of pts. 3% (5/155) of pts undergoing auto and 48% of pts undergoing allo HSCT (44/92) developed BKV. 18% of pts (9/49) developed HC, 1 in the auto and 8 in the allo group. Three pts died from complications related to HC. Of the allo HSCT pts, 50% (46/92) received Sib and MUD HSCT. Non-myeloablative conditioning regimens (Flu/TBI, Flu/Cy, Flu/Bu/ATG, TLI/ATG, Flu/Csa/MMF) were used in 39% pts (36/92) and whereas 61% (56/92) of pts underwent myeloablative conditioning (BuCy, TBI/Cy). In pts undergoing allo HSCT with myeloablative regimen, 52% (29/56) had BKV with 28% (8/29) having HC, while 42% (16/35) of pts after nonmyeloablative conditioning had BKV with no pt developing HC. Of the allo HSCT pts, GVHD data was available for 67 pts, 69% (46/67) had grade 1-4 acute GVHD whereas 31% (21/67) had no evidence of GVHD. In pts with no GVHD, BKV was seen in 38% (8/21) with no pt developing HC while 52% (24/46) pts with GVHD had BKV with 13% (3/24) developing HC. In patients undergoing MUD HSCT, BKV and HC were present in 43% (20/46) and 5% (1/20) respectively, whereas in the SIB group, BKV and HC developed in 52% (24/46) and 29% (7/24), respectively. As calculated by Fisher exact test BKV was significantly associated with allo HSCT (p< 0.001) and Cyclophosphamide-containing non-myeloablative conditioning in allo HSCT recipients (p=0.03). In contrast we did not observe a significant impact of ATG vs. non-ATG-containing regimen, non-ablative vs. myeloablative conditioning or presence vs. absence of GVHD in our cohort of allo-HSCT patients. However, development of BK-induced HC in BKV+ allogeneic HSCT patients was significantly associated with the presence of myeloablative conditioning (p=0.03). Based on these findings we conclude that 1. Development of BKV occurs predominantly following allogeneic HSCT. 2. Conditioning intensity is an important risk factor for the development of HC in patients with BKV. 3. Cyclophosphamide is an important co-factor for the development of BKV in patients undergoing non-myeloablative conditioning. We hypothesize that cyclophosphamide-induced bladder damage might not only contribute to BKV-associated HC, but also to BK reactivation. Therefore, prospective trials addressing the value of prophylactic or pre-emptive therapy for BKV should focus on patients undergoing myeloablative allogeneic HSCT.

Author notes

Disclosure: No relevant conflicts of interest to declare.