Thrombotic Microangiopathy Increases the Risk of Chronic Kidney Disease but Not Overall Mortality in Long-Term Survivors

Introduction: Thrombotic microangiopathy (TMA) is a known complication of allogeneic hematopoietic cell transplantation (HCT). Post-transplant TMA has been associated with acute kidney injury (AKI) and early mortality. However, the long-term kidney outcomes and survival in patients who recover from the disease have not been well characterized.

Methods: We performed a retrospective cohort study of adult allogeneic HCT recipients transplanted during 2006-2015 who survived to one-year (index date) and had follow-up at the Long-Term Follow-Up (LTFU) clinic at the Fred Hutchinson Cancer Research Center (FHCRC). Patients were classified as TMA or non-TMA based on whether a diagnosis was made within one-year post-transplant (BBMT 2019;25:570). Outpatient creatinine values obtained during LTFU visits were assessed over time and averaged at the distinct years post-transplant. Estimated glomerular filtration rate (eGFR) was calculated using the CKD-EPI formula. Chronic kidney disease (CKD) was defined as eGFR <60 mL/min/1.73m². Potential confounders included pre-HCT eGFR, prior autologous HCT, older age, female sex, black race, myeloablative conditioning (including high-dose total body irradiation), calcineurin/mTOR inhibitor exposure, development of AKI within 6 months, acute graft versus host disease (GVHD) within 6 months, and chronic GVHD within 12 months post-transplant. Pre-transplant hypertension and diabetes were not considered as confounders because they had no known association with TMA development. To assess the association between history of TMA and CKD over time among post-transplant survivors, generalized estimating equation (GEE) was used with exchangeable correlation, binomial family, and logit link, after adjustment for pre-index variables. GEE was chosen to model the longitudinal creatinine outcomes at discrete intervals and to help account for interval missingness. The adjusted odds ratio (OR), 95% confidence interval (CI), robust standard error (SE), and P-values were presented. Unadjusted Kaplan Meier (KM) analysis with landmark at 1 year was used to compare long-term overall survival.

Results: Among 2091 patients that underwent first allogeneic HCT, we identified 1151 patients who had survived at least one-year and had available long-term follow-up data (Figure 1). Fifty-seven patients were survivors who had a history of TMA within one-year post-transplant and 1094 did not. Outpatient creatinine data were available in decreasing number of patients each year for the first 5 years post-transplant. The median eGFR over time for the two groups was shown in Figure 2. At one-year post-transplant, 52% of TMA survivors had CKD versus 27% of non-TMA survivors. After adjusting for other potential confounders, a history of TMA was associated with an odds ratio of 2.62 (95% CI 1.25-5.52) for CKD at one-year post-transplant (Table 1). There was no appreciable change in CKD status over time (non-significant interaction for TMA x year). The adjusted covariates had the expected magnitude and significance of association with CKD development, whereas age, pre-transplant eGFR, acute GVHD, and early AKI had the strongest association. While TMA was significantly associated with short-term mortality, there was no association between history of TMA and long-term overall survival in KM analysis landmarked beyond year one, where the conditional 5-year survival was 71% in the TMA survivors and 74% in the non-TMA survivors (log rank P= 0.113).

Conclusions: In this study of 1151 post-transplant long-term survivors, we found that TMA survivors had higher risk of CKD post-transplant despite adjusting for key potential confounders. The overall eGFR had the largest decrease between pre-transplant and year-one post-transplant, with non-appreciable variation in subsequent years. While TMA patients were more likely to die early, in those who survived to one-year, their long-term mortality was similar to non-TMA patients. Limitations in the study include the lack of uniform follow-up for all transplant survivors and potentially unobserved confounders. Overall, our data suggest that TMA appears to be a time-limited systemic insult; although its damage to the kidney requires continued monitoring and management.