Risk Factors and Outcomes According to Age at Transplant with an HLA Identical Sibling for Sickle Cell Disease

Background: Hematopoietic stem cell transplant (HSCT) from an HLA identical (HLAid) sibling donor is a well-established curative therapy for sickle cell disease (SCD). However, the ideal age to perform HSCT in SCD patients remains controversial. We report the outcomes after HLAid sibling HSCT for SCD according to patient's age at the time of HSCT as well as their pre-transplant characteristics.

Methods: We retrospectively analyzed 727 HLAid sibling HSCTs performed between 1986 and 2015 in 98 EBMT centres. Patients were divided in three groups according to their age at the time of HSCT: <5 year-old (n=176): group 1, 6-15 year-old (n=431): group 2 and >15 years old (n=120): group 3. Primary endpoint was probability of 3-year overall survival (OS) according to age group.

Results: Median follow-up was 46 (range: 0.5-346) months. Median age at HSCT was 4 (range: 0.9-5) years in patients <5 years, 10 (range: 6-14) years in patients 6-15 years and 17 (range: 15-39) years in patients >15 years. Gender distribution did not significantly differ among the three groups. The most common SCD genotype was homozygous (SS) in the three groups (95% vs 92% vs 84%, p=0.004). Positive cytomegalovirus (CMV) serology was less frequent in group 1 (63% vs 79% vs 83%, p<0.001. Also, patients in group 1 were less treated with hydroxyurea (37% vs 60% vs 77%, p<0.001), received less than 20 red blood cell units transfusions (32% vs 43% vs 58%, p=0.002) and presented less than 3 SCD complications before HSCT (11% vs 19% vs 39%, p<0.001). For all age groups, vaso occlusive crisis (VOC), acute chest syndrome (ACS) and cerebral vasculopathy were the most reported SCD complications, and the main indications for HSCT were VOC, cerebral vasculopathy, ACS and nephropathy.

The proportion of patients with Karnofsky performance status (KPS) <80 did not differ between groups. Graft source distribution was bone marrow (BM) in 74%, cord blood (CB) in 23% and peripheral blood (PBSC) in 3% of cases in group 1; BM in 85%, CB in 11% and PBSC in 4% of cases in the group 2; BM in 79%, CB in 1% and PBSC in 20% of cases in group 3 (p<0.001). Conditioning regimen was myeloablative in 97% of cases in group 1, 96% in group 2 and 83% in group 3 (p<0.001), and included in vivo T cell depletion, mostly antithymocyte globulin (ATG), in 74%, 86% and 90% of the cases, respectively (p<0.001). In all groups, graft versus host disease (GvHD) prophylaxis was mainly cyclosporine- A with methotrexate±mycophenolate mofetil. ABO and gender match between donor and recipient did not significantly vary among groups.

Neutrophil engraftment at day +60 was 96%+1 in the first group, 98%+1 in the second group and 95%+2 in the third group (p=0.009). Four patients in group 1, 10 patients in group 2 and 6 patients in group 3 had primary graft failure. The first group experienced less acute GvHD (aGvHD) within 100 days after HSCT (9%+2 vs 18%+2 vs 17%+4, p=0.022) and less chronic GvHD (cGvHD) within 3 years after HSCT (9%+2 vs 12%+2 vs 20%+4, p=0.006). Three-year OS was significantly better in group 1 (99%+1 vs 95%+1 vs 88%+3, p<0.001) as well as 3-year event free survival (EFS) (96%+2 vs 92%+1 vs 84%+4, p=0.001). In a subset of patients with available chimerism data (n=405), in group 1 and group 2 65% were full donor, 32% mixed chimeras and 3% autologous reconstitution whereas in group 3 46% were full donor, 49% mixed chimeras and 5% autologous reconstitution. A univariate risk factor survival analysis was performed for group 3 (only group with sufficient events for analysis), but no patient, donor or HSCT risk factors influencing OS were identified.

Conclusion: Patients transplanted at a young age have a better 3-year OS and 3-year EFS, with lower incidence of aGvHD and cGvHD. These findings outline the importance of early referral to HSCT for SCD patients. Strategies to further evaluate pre HSCT, HSCT and donor factors that may influence survival and adverse outcomes could help improve HSCT results.