Human leukocyte antigen (HLA)-incompatible mean fluorescence intensity-selected platelet products have corrected count increments similar to HLA antigen-matched platelets.

Abstract

Donor specific antibody sum mean fluorescence intensity (MFI) values have been successfully used in transplant medicine to assess risk for organ rejection. However, little is known regarding whether MFI values could be similarly used to aid in platelet product selection. We have developed a novel protocol where MFI values are used to offer human leukocyte antigen (HLA)-incompatible platelet products when HLA antigen-matched products are not available. We aimed to evaluate the efficacy of this protocol. We performed a 4-year retrospective chart review for all patients who received at least one MFI-selected platelet product. A corrected count increment (CCI) was calculated for each transfusion event. A mixed effects model was used to investigate the association between CCIs for MFI-selected, HLA antigen matched, and random donor platelet transfusions. A random effects expectation-maximization regression tree was used to define the extent to which other patient variables, such as age, sex, and diagnosis impacted the CCI for each platelet transfusion. Twenty highly HLA alloimmunized patients received a total of 591 platelets. MFI-selected platelet (low MFI) transfusions had a significantly higher median CCI 0-6 hour post-transfusion (13,559, interquartile range [IQR]: 8275-18,736) compared to random donor platelets (2121, IQR: 0-10,368, p< 0.0001). There was no significant difference in median CCI between HLA antigen matched and MFI selected platelet transfusions (p= 0.2). Mixed effects and regression modeling revealed that MFI-selected platelet products had a significantly higher CCI than non-matched platelets, even when accounting for other significant patient variables. MFI-selected HLA-incompatible platelet products could provide a comparable alternative to traditional HLA antigen-matched platelet products.