Longterm survival and late effects following hematopoietic cell transplantation for sickle cell disease into a modern era: A center for international blood and marrow transplant research (CIBMTR) analysis

Abstract Survival following curative hematopoietic cell transplantation (HCT) for sickle cell disease (SCD) is excellent using matched related donors (MRD), particularly in young patients. Alternative donor outcomes continue to improve, but there is paucity of data on long-term survival and late effects. All patients reported to CIBMTR post 1st HCT for SCD from 1/1996-3/2022 without graft failure (n=10) were included. Descriptive analysis of patient and HCT factors was conducted. Cumulative incidence (CI) of late effects (LEs) was estimated 7 years (yrs) post-HCT and number and type of SCD complications was computed. Overall survival (OS), event-free survival (EFS; alive without rejection), and graft-versus-host disease (GVHD), rejection-free survival (GRFS) probabilities were calculated using Kaplan-Meier method. Multivariable Cox regression analysis was used to evaluate risk factors related to LEs, OS, and EFS. Statistical significance was set at p0.05. A total of 1013 SCD patients underwent 1st HCT at 112 centers at median age of 12 yrs (range, 0-59), of whom 47% (N=475) were female and 87% (N=883) had HbSS. Most common indications were recurrent vaso-occlusive crisis (VOC; 39%, N=398), stroke (15%; N=153), and acute chest syndrome (ACS; 14%, N=141). Most common stem cell source was bone marrow (BM; 69%; N=700) then peripheral blood (PB; 19%; N=194). Donor source was MRD (53%; N=535), mismatched related donor (MMRD; 21%; N=208), matched unrelated donor (MUD; 11%; N=116), and MMUD (3%; N=29). Conditioning was myeloablative (MAC) in 66% (N=672), busulfan-based in 50% (N=508), TBI-based in 24% (N=246), and melphalan-based in 23% (N=234). GVHD prophylaxis included tacrolimus in 34% (N=341), cyclosporine in 32% (N=322), and post-transplant cytoxan (PTCy) in 15% (N=152). With median follow up of 60 mo (range, 3-248), 86% (N=872) remained cured (HbS≤50% and no SCD symptoms) and most remained free of acute (70%, N=710) or chronic (74%, N=752) GVHD. Most had no SCD complications post-HCT (74%; N=186); those reported were osteonecrosis (4%, N=40), VOC (1%, N=10), nephropathy (1%, N=10), stroke (1%, N=9), ACS (1%, N=7), and priapism (1%, N=3). At 7 yrs post-HCT, most common LEs were liver toxicity (CI: 10%; 95% CI: 8-12%), pulmonary toxicity (8%; 6-10%), diabetes (6%; 5-8%), gonadal dysfunction (6%; 4-8%), AVN (4%, 3-6%), and renal failure (RF) requiring dialysis (2%; 1-3%). Sixteen patients (2%) developed malignancy post-HCT. Specific LEs were significantly associated with older age at HCT (HR for AVN=6.3, diabetes=5.0, gonadal dysfunction=9.1, pulmonary toxicity=2.1; statistically defined cutpoints of ≥13, ≥10, ≥8, and ≥11 yrs, respectively), hydroxyurea pre-HCT (HR for AVN=4.5), recipient female sex (HR for gonadal dysfunction=11.7, pulmonary toxicity=1.6), recipient CMV+ status (HR for gonadal dysfunction=2.4), year of HCT (diabetes; HR=2.4 for 2011-2015, ref: 1996-2010), alternative donor (diabetes: HR for MMRD=2.4, MUD=3.4, MMUD=3.4, CB=3.2, ref: MRD; RF: HR=9.3 for MUD, ref: MRD), PB or CB source (RF; HR=2.9 and 4.4, ref: BM), horse ATG (gonadal dysfunction, HR=2.9, ref: no ATG), grade III-IV acute GVHD (HR for AVN=2.8, diabetes=6.4, RF=7.2, pulmonary toxicity=2.4), and chronic GVHD (AVN, HR=2.7); risk for gonadal dysfunction was significantly reduced in RIC/NMA (HR=0.13). Eight-eight patients (9%) died, with primary causes organ failure (36%; N=32), infection (27%, N=24), GVHD (15%; N=13), and SCD (7%; N=6). At 1 and 7 yrs post-HCT, OS, EFS, and GRFS were 95% (95% CI: 94-97%) and 90% (88-92%), 89% (87-91%) and 83% (80-85%), and 68% (65-71%) and 63% (59-66%), respectively. Older age at HCT (OS: ≥10 yrs, HR=4.7, EFS: ≥13 yrs, HR=2.5), alternative donor (OS: MUD, HR=2.5; EFS: HR for MMRD=1.8, MUD=2.4, MMUD=2.1; ref: MRD), and grade III-IV acute GVHD (HR for OS=6.1, EFS=2.6) were significantly associated with inferior survival; additionally, graft source (HR for PB=1.7 and CB=3.2, ref: BM) was significantly associated with inferior EFS and more recent era (HR for 2011-2015=0.51, 2016-2020=0.25, 2021+=0.17; ref: 1996-2010) with improved EFS. In a modern cohort with increased alternative donors, RIC/NMA, and PTCy, most patients were alive, cured of SCD, and free of LEs post-HCT. Outcomes including specific LEs are superior in patients who undergo HCT at a younger age, using MRD BM, and in those who remain GVHD free. Future strategies to improve GVHD prevention are likely to improve access and long-term outcomes post-HCT for SCD.