Persistent BK Virus Viremia (pBKV) and Kidney Injury At Day 100 After Pediatric Allogeneic Hematopoietic Stem Cell Transplant (SCT)

Overview: pBKV (≥10,000 copies/ml) after kidney transplant is associated with kidney injury. We hypothesized that pBKV contributes to lower day 100 estimated kidney function (eGFR) post-SCT.

Methods: Retrospective cohort analysis of 100 SCT recipients prospectively enrolled to study thrombotic microangiopathy (TMA). Plasma BK PCR testing was performed for hematuria or elevated creatinine (creat) or on stored samples in those with no clinical indication. We included allogeneic SCTs with ≥1 BK plasma PCR from each of day: 0-14, 15-85, and 100±14. The outcome was day 100 eGFR using serum cystatin C (cysC, muscle mass-independent eGFR) and creat. pBKV was defined as ≥4 consecutive weekly PCRs ≥10,000 copies/ml. We tested associations between pBKV and clinical data, laboratory results, and medication exposures (coded yes/no in first 100 days). Multivariate regression assessed eGFR, pBKV, and covariates with a univariate p<0.2.

Results: Of 72 included subjects, 7 (10%) had pBKV and 65 (90%) did not: 7/65 had transient BKV ≥10,000 copies/ml on a median of 2 measures. Selected univariate risk factors for pBKV are shown in the Table. There was no difference in age, sex, diagnosis, pre-SCT nuclear GFR, total body irradiation, acute GVHD, steroid prophylaxis, cyclosporine level, VOD, grade ≥2 cystitis, CMV, adenoviremia, or dialysis. By univariate analyses, the pBKV group had significantly lower day 100 cysC, but not creat-based eGFR (Table) and more likely received cidofovir (p<0.0001). On multivariable analyses, T-cell modulating antibody therapy trended (p=0.13) towards an association with pBKV. pBKV was associated with lower cysC eGFR, but this was no longer significant after adjusting for cidofovir.

Conclusions: T-cell modulating antibody therapy may be a risk factor for BKV. pBKV possibly contributes to lower day 100 cysC eGFR, but we are unable to conclude an independent association as most with pBVK also received nephrotoxic cidofovir. Creat-based eGFR was no different, perhaps because of low muscle mass after SCT. Future studies will follow-up past day 100, temporally assess covariates (cidofovir), and uniformly measure BKV exposure in all subjects.

	Persistent BK viremia n=7	No persistent viremia n=65	p-value*
Age, years	6 (3-19)	8 (2-14)	0.57
Reduced intensity vs  myeloablative	71%	38%	0.12
Un-related donor vs related	100%	71%	0.18
Fludarabine prep	86%	49%	0.11
Alemtuzumab prep	71%	37%	0.11
T-cell modulating antibody GVHD therapy	43%	12%	0.07
TMA	71%	37%	0.11
Pre-SCT nuclear GFR	111 (105-139)	122 (104-144)	0.64
Day 100 eGFR equations
CysC
Zappitelli	55 (42-64)	83 (62-103)	0.02
Rule	46 (35-55)	73 (53-93)	0.02
CKiD cysC	54 (44-62)	76 (60-90)	0.02
Creat
Bedside CKiD	104 (68-116)	100 (83-122)	0.86
CysC and creat
Full CKiD	68 (64-78)	89 (72-97)	0.06

Data as median (25^th-75^th percentile) or %

GFR in ml/min/1.73m², normal >90

CKiD, Chronic Kidney Disease in Children study

*Mann-Whitney U or Fischer exact