Predictors, Risk Factors and Outcomes of Patients with SAA with Anti-HLA Class I Antibodies Treated with Immunosuppressive Therapy

Introduction: Platelet refractoriness, defined as persistently inadequate post-transfusion count increments, can be immune or non-immune. Human leukocyte antigen (HLA) Class I antibodies are the usual immune cause. The presence of HLA platelet antibodies (Abs) in hematologic disease has been associated with poorer clinical outcomes, graft failure in hematopoietic stem cell transplant, increased bleeding, and decreased survival.

In immune severe aplastic anemia (SAA), literature addressing prophylactic transfusion of platelets is lacking, but most recommend transfusion with bleeding at a count <10 or <20 K/ µL if febrile or during ATG (anti-thymocyte globulin) administration. Chronic platelet transfusion is a risk for development of HLA Abs; nevertheless, some chronically transfused patients do not become alloimmunized while others do so rapidly, even after a single platelet transfusion. We analyzed our SAA patients to document the rate of HLA alloimmunization and its possible association with disease characteristics and outcomes after immunosuppressive therapy (IST).

Methods: Clinical records from 444 SAA patients treated at the NIH Clinical Center from 2000 to 2022 were retrospectively reviewed. Demographics included: age, sex, HLA Class 1 Abs status, platelet refractoriness, peripheral blood counts, PNH clone, treatment and relapse, evolution to myeloid malignancy or chromosome abnormality, and overall survival (OS). Patients were tested for HLA Abs if they met 1 or more of 3 clinical criteria: 1) history of HLA Abs; 2) history of or current platelet transfusion refractoriness; 3) work up for bone marrow transplantation. Clinical refractoriness was defined as lack of platelet increment of >5 K/µL or a corrected count increment (CCI) of >5 K/µL. Patient lacking evidence of refractoriness were presumed anti-HLA negative and not tested, and those that tested negative for HLA Ab were transfused platelets unmatched for any HLA antigens.

Logistic regression was used to analyze associations between age, sex, blood counts, PNH clone, and disease severity, and Kaplan Meier curves and Cox proportional hazards model for survival. Data were analyzed using R, and p<0.05 was considered statistically significant.

Results: Demographic data were collected for 444 patients (Table.1). HLA Ab testing was performed in 155 patients while 289 didn't meet criteria for testing. Of 444 patients, 99 (22%) tested positive for HLA Class 1 Abs. Seventy-three of 99 patients responded to HLA matched platelets, whereas, 17 remained refractory and for 9 data were unavailable. Univariate analysis predicted male sex to be protective against HLA Abs formation (OR=0.59,p= 0.02), while more severe disease (VSAA, OR= 2.51, p<0.001) and lower baseline blood counts predicted alloimmunization. Low platelet counts (p<0.001) were most significant followed by absolute reticulocyte (ARC) and absolute neutrophil counts (ANC) with p=0.005. PNH was not associated with refractoriness, although data were incomplete. Only HLA Class 1 Ab status and age, and sex were strong predictors of survival in both univariate and multivariate analysis. ANC, ARC, and disease severity were significant predictors of survival in the univariate but not multivariate model.

Median time to relapse was 2.4 years with a relapse rate of 36% at median follow-up. Rate of relapse was 46% in alloimmunized patients vs.34% in the Ab negative group, but this difference was not statistically significant (p=0.2). OS for the entire cohort was 84% at median follow up 4 years; HLA Ab-negative patients had a significantly higher OS compared with positive (86% and 75% respectively; p = 0.02). Rates of clonal evolution were also higher in the HLA Ab positive group but not statistically significant (25% vs. 13%, p=0.08).

Conclusions: We report a large cohort of IST-treated SAA patients, comparing outcomes with and without HLA alloimmunization. Although SAA patients are usually heavily transfused platelets, we observed a rate of HLA alloimmunization of 22%. OS was decreased in HLA alloimmunized patients. Presence of HLA ab and age were strong predictors of survival. ANC, ARC and disease severity were predictors of survival only in the univariate model. Relapse and clonal evolution may be increased when HLA Abs are present. Overall, the presence of HLA Abs is associated with poorer outcomes in SAA patients.

No relevant conflicts of interest to declare.