Blood Transfusion Support for Sickle Cell Patients During Haematopoietic Stem Cell Transplantation: a Single‐institution Experience

Sickle cell disease (SCD) is the most prevalent monogenic disease in the world.1 In Brazil, SCD affects approximately 30 000 people, with 3 500 new patients born each year.2 SCD is characterized by haemolytic anaemia and organ injury. Treatment for SCD includes red blood cell (RBC) transfusion. Approximately 10% of SCD patients in high-income countries receive chronic RBC transfusion.1 Alloimmunization, iron overload, and transmission of infectious agents are a concern.3 Alloimmunization affects a variable percentage of patients depending on the extent of RBC matching, 0–7% for extended matching to 5–14% for Rh and Kell matching and 27–75% for ABO-D matching alone.4 The presence of anti-erythrocyte antibodies (AEAs) was found to be a marker of disease severity, as alloimmunized patients had a higher prevalence of chronic pain, osteonecrosis, end-organ damage and shorter life expectancy (54 vs. 65 years).5 To date, the only curative treatment for SCD is haematopoietic stem cell transplantation (HSCT).6 There are many barriers to HSCT, such as finding a suitable donor, patient age and the presence of severe sequelae. Despite these concerns, the number of HSCTs for SCD is increasing in many countries, including Brazil. ABO mismatch between patient and donor, which occurs in approximately 30% of HSCT, can cause haemolysis and red cell aplasia.7 Major mismatch increases RBC requirements, especially with high IgG antibody titres against donor erythrocytes.8 Here, we evaluated the impact of AEA and ABO mismatch on overall survival (OS) and on transfusion requirements until days 30 and 180 from transplantation. We reviewed medical charts of 48 (24 ≥ 18 years) consecutive patients with SCD (SS 37; Sβ0 9; and SC 2) who underwent related sibling human leucocyte antigen (HLA)-identical HSCT (46 myeloablative; two reduced intensity) in a single institution, between 2003 and 2019. A total of 46 patients (95·8%) received bone marrow (BM) and two received peripheral blood as graft source. Patients with major/bidirectional and minor/bidirectional ABO mismatch received BM processed to remove RBC and/or plasma. Primary outcome was OS. Other outcomes were RBC transfusion requirement according to the presence of AEA and to ABO mismatch. RBC was transfused to maintain haemoglobin around 90–100 g/l. Extended RBC phenotype matching was performed. Number of RBC units transfused was recorded until days 30 and 180 from the HSC infusion. Transfusion numbers before HSCT were not recorded as most patients were followed at other institutions, which provided no detailed information on this issue. Nineteen (39·6%) patients had preparative RBC exchange before HSCT (1 manual; 18 automated) to achieve HbS ≤ 30%. Anti-A/B iso-haemagglutinin removal (plasmapheresis) was performed when patient presented major/bidirectional ABO mismatch and antibody titre ≥ 32 (until titre < 32). Overall survival estimates were generated using Kaplan–Meier curves (log-rank test to compare groups). A Mann–Whitney test was employed to compare data from two groups. Statistical significance was defined when P < 0·05. Statistical analyses were performed employing GraphPAD Prism8 software (GraphPad Software, San Diego, CA, USA). This study was approved by the research ethics board. Forty-eight SCD patients, with a median (range) age of 15·5 (7–38) years, underwent HSCT. One patient dropped out fiour months after transplantation. OS was 89·58% (43/48), with a median (range) follow-up of 4·2 years (15 days-16·8 years). Five deaths occurred, 1·2 years (15 days-2·9 years) after transplantation, four of which in the AEA group (P = 0·0007; Fig 1). The ABO-mismatched group presented a tendency to worse OS than the matched group (P = 0·0977). Patients with age ≥ 16 years presented similar OS to the group of patients < 16 years of age (P = 0·6022). Ten (20·83%) patients had AEA before transplantation, fourt of them also had autoantibodies (none developed new antibodies after transplantation). In only one patient the antibodies disappeared two years after transplantation. The other patients continued to present antibodies three months to two years after transplantation. Two patients with three AEA each and one patient with one auto and one AEA underwent three sessions of plasmapheresis followed by rituximab (375 mg/m2) plus immunoglobulin (2 g/kg). One of them died on day 15. None had overt haemolysis. No clinically significant transfusion reaction after HSCT was reported. Nineteen (39·58%) patients had ABO mismatch with their donors, of which six (12·5%) had major, four (8·33%) bidirectional, and nine (18·75%) minor. Five of 46 (10·87%) patients with available data (one died on day 15 and one dropped out) had secondary engraftment failure, 1/10 and 4/36 with and without AEA (P > 0·9999; Table I). All patients (n = 48) Match (n = 29) Mismatch (n = 19) Major/bidi (n = 10) Minor (n = 9) 0·0143** 0·0124* 0·024** 0·029* 0·77& 0·34# In this study OS was 90%, lower than previously reported, but our patients were older (15·5 vs. 9·4 years).6 In another study, it was shown that the group of patients, with a median age of 17·6 years, had a four-year OS of 88%, similar to that reported here.9 SCD is a heterogenous disease, which has prompted the search for biomarkers of disease severity, such as the presence of AEA. Non-transplanted SCD patients with AEA have lower OS.5 Here, we show that transplanted patients with AEA presented lower OS. Some authors found that ABO mismatch was associated with lower survival,10 whereas others did not.11 For SCD patients, the influence of ABO mismatch on survival is not defined. Here, we observed a tendency to lower OS in ABO mismatch. Chronic inflammatory status of SCD patients might potentiate the additional aggression by ABO mismatch. Our patients received similar numbers of RBC units to those previously reported,12 in which AEA patients had higher requirements, a finding not observed here, perhaps because our patients received fully compatible transfusion. Patients with major/bidirectional ABO mismatch required more RBC transfusions, similar to findings shown for other diseases.13 The explanation is the persistence of anti-donor iso-haemagglutinins, thwarting engraftment. Transplanted SCD patients with AEA present lower survival rates. AEA should be considered a major prognostic marker for HSCT. Major/bidirectional ABO mismatch is associated with an increased requirement of RBC transfusions. This study was supported by a grant from the São Paulo Research Foundation (Center for Cell-Based Research, CTC-CEPID-FAPESP, Process 2013/08·135-2) and by INCTC (CNPq Process number 465·539/2014-9). The authors declare to have no potential conflicts of interest regarding the present work.