Cardiac effects 2 years after successful non-myeloablative human leukocyte antigen-matched related donor hematopoietic cell transplants in sickle cell disease.

Cardiopulmonary disease, particularly elevated tricuspid regurgitation velocity (TRV ≥2.5 m/s), remains a significant risk marker for early mortality in individuals with sickle cell disease (SCD). Indeed, diastolic dysfunction and elevated TRV are independent risk factors for mortality.1 Hematopoietic cell transplant (HCT) remains the only available curative therapy for SCD. The degree of organ impairment and co-morbidities in many adults with SCD preclude the use of myeloablative regimens due to the risk of chemotherapy-induced organ toxicity. However, our non-myeloablative (NMA) approach has demonstrated success in adults with SCD, including those with end-organ damage.2 We have previously shown that successful NMA allogeneic HCT improves cardiac size, markers of diastolic dysfunction, TRV, and N-terminal pro-B-type natriuretic peptide within the first year post-HCT.3 In this study, we focused on human leukocyte antigen (HLA)-matched related donors (MRD) and broadened our analyses to three centers with a predominantly adult cohort of patients with SCD, extending our follow-up period to 2 years. These three centers used the same NMA regimen with comparable outcomes.2 Patients with SCD underwent NMA MRD HCT at the National Institutes of Health (NIH, NCT00061568 or NCT02105766), University of Illinois, Chicago (UIC NCT01499888), and King Abdulaziz Medical City in Riyadh, Saudi Arabia (KAMCR, RC20/646/R). Routine laboratory parameters, as well as a transthoracic echocardiogram (TTE), were performed before HCT, and at 1 and 2 years post-HCT. Patients with a successful transplant who were free of dialysis or chronic transfusion therapy were included, and both a pre-HCT TTE and a TTE at least 1 year following HCT were required. All patients received alemtuzumab, 300 cGy total body irradiation, and sirolimus; some also received pentostatin and cyclophosphamide preconditioning. Transthoracic echocardiograms were performed using commercially available systems. Cardiac measurements were performed according to the American Society of Echocardiography guidelines1 A generalized estimating equation regression model approach was used to evaluate TTE and laboratory changes from baseline, adjusting for age, gender, and site. Given the number of outcomes tested, a more stringent p-value threshold of .005 was used for significance. Baseline assessments of differences between sites were conducted using Kruskal–Wallis tests for continuous outcomes and Fisher exact tests for categorical variables. The study population consisted of 123 patients (68 NIH, 19 UIC, 36 KAMCR) who had stable engraftment. The mean ± SD age of patients was 29 ± 10 years, and 49 (40%) were female. Eighty-seven percent of patients at the NIH and UIC had hemoglobin SS, while one-third of KAMCR patients had compound heterozygous hemoglobin S beta0 thalassemia. Patients at the NIH and UIC were also older (mean values of 31.0 years at NIH, 33.8 at UIC, 26.1 at KAMCR p = .02) and had a larger body surface area (BSA mean values of 1.8 m2 at NIH, 1.8 at UIC, 1.6 at KAMCR, p < .0001). The TRV was elevated at ≥2.5 m/s in 40% of patients at baseline and was higher in patients at the US sites (mean values of 3.5 m/s at NIH, 4.0 at UIC, 3.0 at KAMCR, p = .004), reflective of more severe disease. Hemoglobin improved significantly in the first year after HCT and continued to improve at 2 years (Table 1). Laboratory results showed improvements in lactate dehydrogenase, total bilirubin, and absolute reticulocyte count within the first year after HCT (Table 1). The left ventricular size, as measured by the left ventricular end-diastolic volume index/BSA (LVEDV/BSA), decreased at 1 year after successful HCT (baseline 76.6 mL/m2 ± 21.9, 1 year 65.9 mL/m2 ± 15.7, p < .0001) and remained stable at 2 years (60.7 mL/m2 ± 15.1, p = .03). Almost all patients have normal right ventricular function at baseline, 1 year, and 2 years post-HCT. Although there is a statistically significant reduction in the tricuspid annular plane systolic excursion (TAPSE) value post-HCT, it remains within the normal range. To assess the extent to which improvement in hemoglobin is responsible for this improved parameter after HCT, we performed correlative analyses. There was a moderate inverse correlation between the change in hemoglobin and the change in LVEDV/BSA at 1 year (r = −0.37; p = .0002) and 2 years (r = −0.3; p = .006) after HCT. There was no significant correlation between change in hemoglobin and change in TRV one and 2 years after HCT (r = −0.08, p = .55 at both timepoints). At baseline, there were 3 individuals with concentric remodeling, 12 individuals with eccentric left ventricular hypertrophy (LVH), and none with concentric hypertrophy out of a total of 122 individuals from all 3 sites (individuals with missing data were excluded). The number of individuals with eccentric LVH decreased to 8 by 1 year and 4 in the second year. LV mass index did not change in the first year (baseline 82.3 g/m2 ± 21.9, 1 year 82.7 g/m2 ± 20.4, p = .81); however, it improved significantly by 2 years (75.5 g/m2 ± 20.2, p < .0001). The left atrial volume index decreased in the first year (baseline 38.1 mL/m2 ± 13.8, 1 year 28.3 mL/m2 ± 8.5, p < .0001) with no further significant change in the second year (26.5 mL/m2 ± 8, p = .03). Diastolic filling parameters were seen to parallel the volume changes with baseline E/A ratio decreasing in the first year (baseline 1.7 ± 0.6, 1 year 1.5 ± 0.5, p < .0001) then remaining unchanged at 2 years (1.4 ± 0.5, p = .03). In the first year after HCT, TRV decreased significantly (baseline 2.5 ± 0.4; 1 year 2.3 ± 0.4, p < .0001) and remained stable in the second year (2.3 ± 0.3, p = .01). At the 2-year time point, 28% of patients had an elevated TRV ≥2.5 m/s compared to 40% at baseline. Although the mean TRV in this elevated TRV subgroup was 2.8 m/s at baseline, it improved to 2.5 m/s at 1 year (p < .0001), and 2.4 m/s by 2 years, p < .0001. As expected, the right ventricular systolic pressure (RVSP) showed a similar trend of a significant reduction in the first year (baseline 30.6 ± 10.7, 1 year 26.4 ± 7, p < .0001) and remained improved at the 2-year timepoint (26.1 ± 7.8, p < .0001). Our findings from three independent centers demonstrate significant improvements in cardiac size, mass, diastolic function, and TRV up to 2 years after NMA HCT. Volume overload secondary to the anemia in SCD is in part responsible for cardiac remodeling.4 As the anemia resolves and high flow states decrease following HCT, many cardiac changes also reverse. LV volume and LV mass index decreased significantly in the first year and these improvements continued in the second year. Following HCT, there was a statistically significant increase in diastolic blood pressure (BP) at 1 and 2 years post-HCT. The systolic BP also slightly increased, which approached but did not meet our stringent definition of significance (p < .005). The increase in BP was likely related to the weight gain post-HCT, as reflected by the rise in BSA. Others have shown a positive correlation between BP and body weight, and this association may be even more pronounced in diastolic BP.5 Correlative analyses to evaluate the role of improved hemoglobin in changes in cardiac parameters demonstrate a moderate negative correlation between change in hemoglobin and change in LVEDV at 1-year post-HCT. The chronic impact of anemia reversal on oxygen delivery and cardiac parameters remains unclear. The improvements reported here are likely not due to improved hemoglobin alone. Decreased sickling events and ischemia–reperfusion injury, improved blood flow, and improved myocardial fibrosis may contribute to the improved cardiac parameters seen after HCT. Further study to elucidate the mechanism of change is required. Cardiac structural and functional improvements following HCT have significant implications. Abnormal diastolic function and elevated TRV are linked with early mortality in SCD, and both of these parameters improved following HCT. The mean TRV decreased in the entire group and also the subgroup with elevated TRV at baseline to the normal range by 2 years post-HCT. Left ventricular size and hypertrophy are closely linked to diastolic function and pulmonary pressures and improved. This three-center study confirms our prior report on reversing the rheologic cardiomyopathy in SCD patients undergoing HLA-matched and haploidentical HCT.3 We now see that LV dimension and LV mass index continue to improve up to 2 years following HCT from HLA-MRD. Despite one recent report on hydroxyurea therapy leading to reductions in LV dilation and hypertrophy in children,6 HCT is the only therapy in adults that has been shown to reverse cardiac abnormalities. Limitations of our study include some baseline differences in the patient population across the three centers. We were limited to small sample sizes and only 2 years of follow-up after HCT. Other conditioning regimens were not evaluated since we limited the analysis to NMA HLA-MRD HCT. Moreover, we did not have access to additional laboratory (e.g., brain natriuretic peptide), TTE (e.g., strain), and functional capacity (e.g., 6-min walk distance), variables that may have been informative. In conclusion, we have shown for the first time that cardiac morphology and TRV improve 2 years after HLA-matched sibling HCT for SCD. As diastolic dysfunction and TRV have been associated with early mortality in adults with SCD, our results suggest that successful HCT may impact survival. A larger multicenter study with longer follow-up is indicated. This research was supported by the Intramural Research Program of the National Heart, Lung, and Blood Institute (NHLBI), National Institutes of Health, and the Cooperative Study of Late Effects for SCD Curative Therapies (COALESCE, 1U01HL156620-01, NHLBI). The content of this manuscript is solely the responsibility of the authors and does not necessarily reflect the official views of the National Heart, Lung, and Blood Institute, National Institutes of Health, or the United States Department of Health and Human Services. The authors declare no conflicts of interest. Requests for data should be sent to the corresponding author at [email protected].