Acute kidney injury (AKI) is a frequent complication after hematopoietic stem cell transplantation (HSCT). Prior studies have shown a negative impact on survival when AKI develops early after HSCT, but have primarily included patients receiving regimens associated with a high degree of toxicity, in particular total body irradiation based-regimens. Fludarabine/ i.v. busulfan (FluBu4) is a myeloablative regimen with reduced extra-hematologic toxicity. We sought to examine the impact of AKI on survival in patients receiving FluBu4.

We retrospectively analyzed 91 consecutive adult patients who underwent allogeneic stem cell transplant and conditioning with FluBu4 between January 2003 and December 2016. The FluBu4 regimen consisted of daily fludarabine 40 mg/m2 and i.v. busulfan (AUC of 4800 mmol-min per dose) over 4 days. All patients received GVHD prophylaxis with tacrolimus and methotrexate. We observed a high incidence of AKI (defined as creatinine clearance below 60 ml/min) within 90 days of HSCT (n=62, 68%). The mean of the lowest measured creatinine clearance was 39.4 ml/min (± 14.2) in the AKI group and 81.7 ml/min (± 16.26) in the non-AKI group (p<0.0001). Baseline characteristics including disease risk index (DRI) and hematopoietic cell transplant-comorbidity index (HCT-CI) were similar between the two groups. However, patients in the AKI group were older (48.5 versus 29 years, p<0.0001). The mean tacrolimus concentration in the first 90 days after HSCT was similar between the two groups (9.2 versus 9.1 ng/mL, p=0.87).

In our cohort, we found that the median time to neutrophil (16 versus 15 days, p=0.5) and platelet (16 versus 16 days, p=0.16) engraftment was similar between the AKI and non-AKI group. Patients that experienced AKI within the first 90 days of HSCT had significantly worse overall survival (OS) (26.3 months versus median not reached, p=0.04) and progression-free survival (PFS) (19.6 months versus median not reached, p=0.02). When patients who died within the first 90 days were excluded and survival was measured from Day 90, AKI still predicted for poor OS (46.8 months versus not reached, p=0.03) and PFS (43.2 months versus not reached, p=0.02). The incidence of grade III-IV acute graft-versus-host disease was similar between the two groups. However, patients who developed AKI had a significantly higher cumulative incidence of relapse (21 (34%) versus 4 (14%), p=0.03) making this the major driver of reduced OS and PFS. In an age-adjusted multivariate analysis, AKI was associated with significant decreases in both OS (p=0.047) and PFS (p=0.033).

We next analyzed the impact of early AKI on the development of chronic kidney disease (CKD) at one year. In all patients surviving more than one year, we found that 13 (34%) of 38 patients in the AKI group and 1 (4.5%) of 22 patients in the non-AKI group developed CKD (p=0.02). OS measured from 1 year post transplant was significantly shorter in patients with CKD (39.3 months versus median not reached, p=0.002).

This is the first study evaluating the impact of AKI on outcomes in patients undergoing HSCT with FluBu4. We observed a high incidence of AKI during the first 90 days that predicted for worse OS and PFS. We hypothesize that the significantly higher relapse rate observed in the AKI group is due to increased tacrolimus exposure and diminished graft-versus-leukemia effect. Although mean tacrolimus levels were comparable between groups, these measurements are likely not an accurate estimation of cumulative tacrolimus exposure. Finally, early AKI also leads to a significant increase in CKD at one year, which in our cohort is associated with late mortality after HSCT.


Patel: Celgene: Consultancy, Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.