The Association between Transplant-Associated Thrombotic Microangiopathy and Calcineurin Inhibitor and Sirolimus Levels

Introduction: Thrombotic microangiopathy (TMA) is a known complication of allogeneic hematopoietic cell transplantation (HCT). An ongoing debate is whether TMA is caused by acute graft-versus-host disease (GVHD) or calcineurin inhibitor (CNI)/sirolimus. Previous studies have not examined the impact of CNI/sirolimus average trough levels while accounting for the confounding effect of GVHD. In the current study, we examined the association between time-varying levels of CNI/sirolimus and the onset of TMA among 2105 adult allogeneic HCT recipients.

Methods: We performed a retrospective cohort analysis of allogeneic HCT recipients during 2006-2015 at Fred Hutchinson Cancer Research Center who received CNI/sirolimus for GVHD prophylaxis. TMA (outcome) was ascertained and validated as previously described (Blood 130;Suppl:666). Acute GVHD (confounder) was defined and graded according to established criteria. Consecutive CNI/sirolimus trough levels (exposure) were obtained and interval values between checks were imputed via linear interpolation. We followed patients from the time of hematopoietic cell infusion until the onset of TMA and censored patients at day 100 or >1 week of missing drug levels. To examine the association between the "type" of immunosuppressant and TMA, CNI/sirolimus exposure over time was characterized as a discrete categorical variable (tacrolimus alone, cyclosporine alone, or sirolimus + CNI). To examine the impact of the "level" of immunosuppressant on TMA, CNI/sirolimus exposure was both defined as an average of the previous 7-day trough levels (continuous variable) as well as a discrete time-above-peak variable (binary variable for tacrolimus >15 ng/mL, cyclosporine >450 ng/mL, or sirolimus >10 ng/mL). Extended Cox regression models were built to examine the time-varying association between CNI/sirolimus exposures and TMA after adjusting for GVHD.

Results: In the current study, 2105 adult allogeneic HCT recipients contributed 173,171 person-days and 150 cases of TMA. The median follow-up time was 94 days and trough blood levels were checked on average 3x per week. Initial GVHD prophylaxis regimen for patients included 54% on tacrolimus (n=1135), 40% on cyclosporine (n=837), and 6% on a combination of sirolimus + CNI (n=133). Nearly no one received sirolimus alone without CNI. Eight percent of patients (n=162) had at least one switch in the immunosuppression regimen within 100 days. The median (IQR) trough levels for tacrolimus, cyclosporine, and sirolimus were 9 ng/mL (7-12), 283 ng/mL (206-372), and 5 ng/mL (4-7), respectively. Acute GVHD developed in 64% of patients including 53% grade 2 (n=1115), 9% grade 3 (n=185), 2% grade 4 (n=52).

In the analysis adjusted for both discrete time-varying CNI/sirolimus exposure and GVHD, higher grades of GVHD had strong associations with TMA (Table 1: grade 2 HR 2.24, P=0.001; grade 3 HR 12.38, P<0.001; grade 4 HR 26.04, P<0.001). While cyclosporine and tacrolimus had a similar risk for TMA (HR 1.23, P=0.230), sirolimus + CNI regimen was associated with a slightly higher risk compared to CNI alone (HR 1.76, P=0.051).

In the analyses of individual drug levels adjusted by GVHD, higher trough levels of tacrolimus and cyclosporine (either as a linear average level or binary time-above-peak) were not associated with an increased risk of TMA (Table 2). However, higher sirolimus trough levels were associated with an appreciable risk of TMA (HR 1.44, P=0.001 for every 1 ng/mL increase in sirolimus average trough level; HR 3.23, P=0.164 for time above 10 ng/mL).

Conclusion: In allogeneic HCT patients, acute GVHD was strongly associated with the onset of TMA independent of ongoing exposures of CNI/sirolimus. We did not find an association between CNI recipients (tacrolimus versus cyclosporine) or CNI levels and the risk of TMA. However, when combined with CNI, higher sirolimus trough levels were linearly associated with a higher risk for TMA. Limitations in this observational study include the inability to adjust for concomitant medications and conditions that might have altered drug levels and inability to adjust for target drug levels since recognized risk factors might have led clinicians to target higher drug levels due to higher risks of GVHD. These findings suggest that prevention and treatment of GVHD and avoidance of high sirolimus levels will decrease the risk of TMA.

View largeDownload slide

View largeDownload slide

Close modal