BK Virus by PCR From Peripheral Blood At Day 21 After Allogeneic Transplant Predicts Risk for Hemorrhagic Cystitis

Introduction: Hemorrhagic Cystitis (HC) is a well-recognized complication of hematopoietic stem cell transplantation (HSCT) and is frequently associated with BK virus reactivation. While self-limited, it is associated with significant morbidity. We hypothesized that serum PCR testing for BK virus (BKV) before and after HSCT might predict development of HC. To test our hypothesis, we conducted a retrospective analysis of patients who underwent allogeneic HSCT from 2005 to 2012 at our institution.

Methods: We identified 162 patients who underwent allogeneic HSCT, 124 received myeloablative conditioning (MA) and 38 received reduced intensity conditioning (RIC). The MA regimen consisted of fludarabine 50 mg/m²/day IV for 5 days and busulfan 3.2 mg/kg/day IV for 4 days with or without total body irradiation (TBI) of 200 cGy/day for 2 days (FB4 or FB4/TBI). The RIC regimen (FB2) consisted of fludarabine 30 mg/m²/day IV for 5 days and busulfan 3.2 mg/kg/day IV for 2 days. Ninety-six patients received matched unrelated donor HSCT (MUD), 60 received matched related donor (MRD) and 6 received unrelated cord blood (UCB). Serum BKV screening was performed in 70 patients before HSCT and in 63 patients at day +21. Patients with urinary symptoms were tested for BKV in urine by PCR.

Results: Overall 36/162 patients (22%) developed HC. Twenty-six of the 36 (72%) were positive for BKV. The median time to development of HC from transplant was 37 days. The odds of developing HC was 4.2 fold higher in patients receiving MA conditioning as compared to RIC (p=0.01; 95% CI=1.22-14.69). The severity of HC was not increased with the addition of TBI (p=0.8). Graft source had no influence on the incidence of HC (p=0.6).  HC developed in 12/23 patients (52%) who were positive for BKV at day +21 and in only 4/40 patients (10%) who were negative. Patients who tested positive for BKV at day +21 were 9.8 times more likely to develop HC than those who tested negative (p=0.0005; 95% CI =2.63-36.67). Testing at day +21 had a negative predictive value (NPV) of 90% (p=<0.0001; 95% CI=76-97%).

Conclusion:  HSCT using MA conditioning significantly increases the likelihood of developing HC compared to RIC. While pre-HSCT serum testing of BKV had no significant predictive value, testing at day +21 helped identify a cohort of patients at higher risk for development of HC. A clinical trial of antiviral pre-emptive therapy might be warranted for patients who test positive on day +21.