INTRODUCTION

Hematopoietic cell transplantation (HCT) represents a potential cure for various malignant and non-malignant hematological diseases¹. When there is no HLA-identical donor, a haploidentical one is a valid option, usually a first-degree relative². According to the recommendations of the Blood and Marrow Transplant Clinical Trials Network (BMT CTN), donation from parents, siblings, or other relatives who share a single HLA-A, -B, -C, and -DRB1 haplotype typed by intermediate or high-resolution with the recipient is defined as haploidentical related³. This type of donation carries the risk of primary graft failure^4-6, rejection, and acute graft-versus-host disease (GVHD) in patients allosensitized against HLA antigens. Furthermore, guidelines from the NMDP/CIBMTR state that DSA leads to detrimental effects after HCT⁷. Therefore, it is necessary to test for specific antibodies in the recipient against donor HLA antigens (donor-specific antibodies, DSA) before haplo-HCT^3,8.

In non-myeloablative conditioning regimens, the receptor’s T and NK cells, responsible for the cellular immune response, can cause bidirectional alloreactions, as in the case of GVHD^2,5. With the selective depletion of T cells, the risk of GVHD and graft failure in haploidentical transplants decreases² at the cost of increased relapse; however, the trend is to use allografts without manipulation⁵. In developing countries, there is also the prohibitive cost of T cell depletion.

Haploidentical transplantation has been related to a higher frequency of graft rejection and GVHD⁹. Anti-HLA donor-specific antibodies (anti-HLA DSA) can generate graft failure in 2-15% of cases. DSA specificity is routinely investigated in solid-phase assays, such as LIFECODES™ single antigen assay (LSA). The results are expressed in mean fluorescence intensity (MFI) values. These assays provide a practical, specific, and sensitive approach since they contain a higher density of the specific antigen per bead for evaluating the risk of graft rejection mediated by antibodies through the virtual crossmatch^10,11. The clinical decisions derived from the DSA testing results are based on certain MFI thresholds. Despite their importance, no international consensus exists on DSA cutoff levels⁶. Moreover, fluorescence is not calibrated against a standard, making interlaboratory standardization difficult. A DSA threshold ≥ 1000 MFI is recommended in the European Bone Marrow Transplant (EBMT) guidelines¹².

MATERIAL AND METHODS

The files and electronic records of patients who attended a university hospital hematology reference center, self-identified as Hispanic mestizos with malignant or non-malignant hematological pathologies, undergoing their first haploidentical transplant of hematopoietic progenitors obtained from the peripheral blood after mobilization with subcutaneous granulocyte colony-stimulating factor, 10 µg/Kg/day for 4 days, collected on day 5 by a single large volume leukapheresis, from June 2017 to December 2020 were analyzed. All individuals had a single antigen assay to detect anti-HLA DSA carried out within 1 month before transplantation. The Ethics and Research Committee at the institution approved the study, and informed consent was waived due to the study’s retrospective nature.

The conditioning regimen was administered in the outpatient clinic as described¹³. It consisted of fludarabine 25 mg/m² and cyclophosphamide (Cy) 350 mg/m² from days −5 to −3, and melphalan, 70-100 mg/m² on days −2 and −1, with or without 2 Gy of total body irradiation. In aplastic anemia cases, melphalan was not used; instead, Cy 50 mg/kg/day on days −2 and −1 was administered.

Posttransplant GVHD prophylaxis included Cy, 50 mg/kg/day on day +3 and +4, mycophenolic acid 1 g/day from day +5 to day +35, and oral CsA through day +18; CsA levels were measured weekly, adjusting for a target level of 150-250 ng/mL, and later tapered over 30-60 days.

Collected data included high-resolution HLA typing of recipients and donors by sequencing-based typing and anti-HLA antibodies (LSA, LIFECODES™ single antigen, Immucor, Waukesha, WI, USA) present in recipients’ serum before transplant, hematological diagnosis, disease status at the time of transplant, conditioning scheme received, the dose of CD34+ cells/kg infused, and ABO compatibility. Days to the recovery of neutrophils and platelets, the development of cytokine release syndrome (CRS), GVHD, and death in the first 100-day post-transplant, among other relevant clinical variables, were also entered in the database.

RESULTS

Information from 70 HLA haploidentical patient/donor pairs who received mobilized HCT in an outpatient setting from June 2017 to December 2020 was analyzed. Patients’ characteristics are described in table 1. Thirty-eight patients (54.3%) were male, and the median age was 32 years (15-38).

Regarding donors, 58.6% were males, and the median age was 32 years (24-43); the kinship of the donor with the recipient was brother/sister in 30 cases (42.9%), son/daughter in 13 (18.6%) and father/mother in 27 (38.6%). The donor and recipient match in 16 cases was female /female (22.9%), female/male in 13 (18.6%), in 16 cases, male /female (22.9%), and in 25 cases, the recipient and donor were males (35.7%).

Two MFI thresholds, DSA ≥ 500 and DSA ≥ 1000, were evaluated for HLA loci A, B, C, and DR. In the study group, 13 (18.6%) DSA-positive patients were found for the threshold DSA ≥ 500 MFI, while for the DSA ≥ 1000 MFI cut-off, only 4 (5.7%) cases were encountered. Of the 13 cases with ≥ 500 MFI, 6 (46.2%) were men, and 7 (53.8%) were women (Table 1).

The median MFI observed for DSA anti-HLA-A was 148 (102.3-231.7), HLA-B 186 (114.7-284.2), HLA-C, 229 (159.7-332.5), and HLA-DR, 113 (62.5-215.7). The characteristics of DSA-positive cases are detailed in table 2.

DISCUSSION

This study analyzed the characteristics of haploidentical HCT and its association with engraftment and adverse clinical events in the context of anti-HLA MFI thresholds ≥ 500 and ≥ 1000. A graft malfunction can be related to the presence of anti-HLA DSA with greater relevance in haploidentical transplants, and anti-HLA DSA at high MFI values can have a role in antibody-mediated rejection, engraftment delay, and poor graft survival^6,8,15.

The incidence of anti-HLA DSA in our cohort was 18.6%. This rate compares to a study that found 18% (22 of 122)⁴. The previous reports have found rates from 13.9% (11 of 79 patients) (11), 14.2% (18 of 134 patients)⁸, to 21% (5 of 28 patients)¹¹. A history of pregnancy in DSA-positive women at the two thresholds explored was comparable, 23.1% (DSA ≥ 500 MFI) and 25% (DSA ≥ 1000 MFI). Other groups have documented that DSA is significantly higher in patients with a history of pregnancy^16,17; a higher percentage of women (86%) than in our cohort (45.7%) has been reported⁴.

A test is usually considered positive with DSA values ≥ 1000 MFI, but this cutoff point varies between transplant centers^2,8,18). The poor functioning of the graft and the rejection mediated by antibodies are accentuated when the MFI values are ≥ 5000^5,8. Due to this variability, it is recommended that each HLA typing laboratory establish its cutoff values¹⁸. The European Society for Blood and Marrow Transplantation (EBMT) guide for the detection and treatment of DSA² strongly recommends local laboratory assay validation and standardization for the detection of DSA, including MFI cutoffs. The same guideline recommends that the anti-HLA DSA test be performed within 1 month before the transplant, allowing time to select the best-related donor in the presence of high DSA levels.

The association between the hematological diagnosis and the presence of anti-HLA DSA has been observed in other studies, being more frequent with AML and MDS. Ciurea et al. found 76% versus 55%⁴, and Bramanti et al. 58% versus 33% in these diagnoses⁸. This relationship is probably related to sensitization by transfusion of blood products, as intensive support treatment for these diseases is required^19,20. In our patients, the administration of platelet concentrates was significantly higher in the DSA-positive groups and more notable in those with DSA ≥ 1000 MFI. After AML, severe aplastic anemia presented a higher incidence of sensitization against HLA antigens, and ATG was added to the conditioning regimen to minimize graft rejection²¹.

In Mexico, 78% of HCTs are carried out in public institutions²²; however, the test is not performed routinely due to budget and infrastructure limitations in the health system²³. Our findings are important as they can help determine in which cases the search for DSA is clinically relevant and to establish a meaningful MFI cutoff.

Graft failure has been reported in patients with DSA at 2000 MFI and higher²⁴; however, this study is the first reporting clinically relevant adverse outcomes for thresholds of 1000 MFI and lower, with the incidence of infection being significantly higher in the DSA-positive group with a threshold ≥ 500 MFI; it is possible that post-transplant infection was associated to the use of Cy; however, the finding that it was statistically associated to MFI values ≥ 500 suggests that there was a relationship between anti-HLA DSA and infection.

Infection in the 1^st year after haploidentical HCT is associated with high mortality, increasing the probability of graft failure or delayed engraftment^25,26. Infection prevalence was recently reported in our patients receiving an outpatient haplo-HCT; it was the cause of death in 49.4%, with severe infections in the pre-engraftment period developing in 22.4% of the recipients; 14.9% were viral and 12.1% fungal. The 100-day and 2-year cumulative incidence of infection-related mortality (IRM) was 15%; fungal agents contributed to infection-related mortality in 33.3%²⁶.

The major adverse clinical event in both thresholds in our cohort was death during the 1^st year after HCT, seen in 40%, comparable to other reports. This finding was not observed during the 1^st month and was part of late-stage events. Chimerism and the success of myeloid and platelet grafting were analyzed at 1 month and after 1 year. Graft failure and mortality in the first 100 days were not related to a DSA ≥ 500 or ≥ 1000 MFI; 1st-year post-HCT mortality, however, was associated to both cutoffs.

As expected, patients with graft failure did not present acute or chronic GVHD. Their mortality in the 1^st 100 days and 1 year after HCT was 62.5% and 75%, respectively, underscoring the severity of this event.

In conclusion, Class I and Class II anti-HLA DSA present in a cohort of Hispanic haploidentical transplant recipients at two MFI cutoff levels were not associated with delayed engraftment or graft failure; however, the presence of DSA at these levels was associated with infection and 1^st-year mortality, supporting the need to establish clinically meaningful thresholds for local reporting of assay results at histocompatibility laboratories.