Abstract
BK virus associated hemorrhagic cystitis (HC) is a common complication post allogeneic hematopoietic cell transplantation (HSCT). A number of retrospective studies have suggested a variety of risk factors for HC following HSCT including busulfan-containing myeloablative conditioning regimens, unrelated donors, and the occurrence of graft versus host disease (GVHD). However, these risk factors were not observed consistently and have led us to review the incidence and risk factors for BK-HC confirmed by microbiologic test at our institution.
We retrospectively reviewed 339 patients who received HSCT between 2009 and 2012 at the Princess Margaret Cancer Centre, Toronto Canada. The diagnosis of BK-HC was based on clinical features, urinalysis and detection of BK virus by urine PCR. The cumulative incidence of BK-HC was calculated considering competing risks and Gray’s method was used for univariable analysis. To confirm the results of the univariable analysis, we conducted a multivariable analysis using the Fine-Gray method and EZR software.
Out of 339 patients, 79 patients (23.3%) developed BK-HC. The median onset was 47 days (range 16-749) after HSCT. The median time to resolution of symptoms was 21 days (range 4-260). The clinical grades were as follows: grade 1 (n=14, 17.9%), 2 (n=34, 43.0%), 3 (n=22, 27.8%) and 4 (n=8, 10.1%). Sixty-two patients (83.7%) out of 74 had urine BK virus PCR testing which showed a high titer of BK virus PCR (mean 2.56x1010 (SD 7.41x1010) and median 6.19x109). The cumulative incidence of BK-HC at 2 years was 24.0% (95% CI, 19.4-28.9). There was no difference in transplant characteristics between the HC and non-HC groups, in terms of age, gender, gender match, diagnosis, HLA match and donor type, stem cell source, T-cell depletion, GVHD prophylaxis and CMV serostatus. However there was a higher incidence of BK-HC in the group of patients receiving myeloablative conditioning (62/208, 29.8%) vs (17/132, 12.9%, p<0.001) in the reduced intensity conditioning group. In the univariable model three risk factors for BK-HC were identified: myeloablative conditioning (31.2% in myeloablative vs 12.4% in reduced intensity; p<0.001), HLA mismatch (34.9% in mismatched vs 22.0% in matched, p=0.027), CMV viremia (28.6% in CMV+ vs 19.1% in CMV-; p=0.023) and acute GVHD (30.2% in aGVHD vs 12.0% in without aGVHD; p<0.001). Unrelated donor or T-cell depletion did not increase the risk of HC in univariable analysis. Multivariable analysis confirmed that these 3 factors were independent risk factors for BK-HC: myeloablative conditioning (p<0.001; HR 2.38, 95% CI, 1.35-4.17), CMV viremia (p=0.012; HR 1.77, 95% CI, 1.10-2.83) and aGVHD (p<0.001; HR 2.26, 95% CI, 1.23-4.13). HLA mismatch was not an independent risk factor. The 2-year overall survival (OS) of the study cohort was 51.6%. The development of BK-HC did not affect transplant outcomes with respect to OS (p=0.446). We observed a higher incidence of 1-year non-relapse mortality (NRM) in the BK-HC group 39.7% vs 24.2% in the non HC group, but this was not confirmed in the multivariable analysis.
Symptomatic BK-HC is a frequent complications following HSCT with a 2-year cumulative incidence of 24.0%. Three risk factors for BK-HC were identified including myeloablative conditioning, CMV viremia and aGVHD. While this infection did not adversely impact survival, it is associated with significant morbidity and constitutes an unmet need given the lack of effective therapy.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
