Impact of Thrombotic Microangiopathy on Renal Outcomes and Survival after Hematopoietic Stem Cell Transplantation

Abstract

Transplantation-associated thrombotic microangiopathy (TA-TMA) is a serious complication of hematopoietic stem cell transplantation (HSCT). We characterized the incidence, risk factors, and long-term outcomes associated with TA-TMA by performing a comprehensive review of all adult patients (n = 1990) undergoing allogeneic HSCT at the Dana Farber Cancer Institute/Brigham and Women's Hospital between 2005 and 2013. Using the City of Hope criteria, we identified 258 patients (13%) with "definite" TMA and 508 patients (26%) with "probable" TMA. Mismatched donor transplantation (subdistribution hazard ratio [sHR], 1.79; 95% confidence interval [CI], 1.17 to 2.75; P = .007), sirolimus-containing graft-versus-host disease prophylaxis (sHR, 1.73; 95% CI, 1.29 to 2.34; P < .001), myeloablative conditioning (sHR, 1.93, 95% CI, 1.38 to 2.68; P < .001), and high baseline lactate dehydrogenase (LDH) level (sHR, 1.64; 95% CI, 1.26 to 2.13; P < .001) were associated with definite TMA. Moreover, positive cytomegalovirus serostatus (sHR, 1.41; 95% CI, 1.16 to 1.71; P < .001), high and very high disease risk index (sHR, 1.48; 95% CI, 1.12 to 1.96, P = .007), and high baseline LDH level (sHR, 1.25; 95% CI, 1.05 to 1.49; P = .011) were associated with probable TMA. In multivariable analyses, definite and probable TMA were each independently associated with higher mortality (HR, 5.24; 95% CI, 4.43 to 6.20 and HR, 2.12; 95% CI, 1.84 to 2.44, respectively), and long-term kidney dysfunction (HR, 5.43; 95% CI, 4.61 to 6.40 and HR, 2.20; 95% CI, 1.92 to 2.51, respectively). Definite and probable TMA were also independently associated with an increased risk of nonrelapse mortality and shorter progression-free survival. Our findings indicate that TA-TMA is common following HSCT and is independently associated with increased risk of death and kidney dysfunction.

Keywords: Allogeneic; Hematopoietic stem cell transplantation; Renal outcome; Survival; Thrombotic microangiopathies.

Figure 1. TMA onset and incidence

(A) Time to TMA onset within first 6 months after HSCT. Most of the patients developed TMA within the first month after transplantation. (B) Figure displays incidence of TMA over the study period. The incidence of definite TMA ranged from 9.2–19% and incidence of probable TMA ranged from 18.8–30%.

Figure 2. Cause of death according to TMA

(A) Among patients with definite TMA, 46% of them died because of the primary disease. (B) Among patients with probable TMA, 59% of them died because of the primary disease. (C) Among patients without TMA, 65% of them died because of the primary disease and non-relapse related mortality was lower compared to patients with definite and probable TMA. (D) Among all patients, 61% of them died because of primary disease.

Figure 3. Renal outcome according to TMA

(A) The cumulative incidence of RRT at 6 months was the highest among patients with definite TMA and lowest among patients without TMA (p<0.001). (B) The cumulative incidence of kidney dysfunction at 6 months was the highest among patients with definite TMA and lowest among patients without TMA (p<0.001).

Figure 4. Hazard ratios for OS according to TMA and Grade II–IV aGVHD

Patients with TMA had higher mortality compared with patients with Grade II–IV aGVHD. Among patients with TMA, patients with definite TMA had higher mortality compared with patients with probable TMA. TMA and Grade II–IV aGVHD showed an additive effect on risk for mortality. Patients with both definite TMA and Grade II–IV aGVHD showed the highest mortality compared with patients without TMA or Grade II–IV aGVHD.

Figure 5. Landmark analysis at 6 months post-HSCT

Numbers under the graphs represent patient number at risk. (A) Landmark analysis for OS at 6 months post-HSCT for patients with no TMA, probable TMA and definite TMA. The No TMA group had the best results for OS and the definite TMA group had the worst results for OS (p<0.001). (B) Patients who developed TMA after relapse were excluded from the analysis. Patients with definite TMA had the worst PFS outcome compared with patients with probable or no TMA (p=0.047). (C) Patients with definite TMA had the highest probability for non-relapse mortality (NRM) (p<0.001). (D) After excluding patients who developed TMA after relapse, the group with definite TMA showed significantly lower risk of developing relapse compared with patients with probable TMA and without TMA (p=0.025).

Figure 6. Overall survival from onset of TMA: patients with TMA only

Numbers under the graphs represent patient number at risk. (A) Patients with definite TMA had significantly lower OS after the onset of TMA (p<0.001). (B) Sirolimus-containing aGVHD prophylaxis was associated with better OS in the definite TMA group compared with patients with definite TMA without sirolimus exposure. Sirolimus exposure was not associated with a significant difference in OS for patients with probable TMA. (C) Patients who experienced TMA after relapse showed poorer OS regardless of TMA severity (p<0.001).