Standard Single Antigen HLA Luminex Panels Predict Most, but Not All, Antibody Reactivity Against Alleles in Extended...

Current Luminex platforms cannot represent the totality of polymorphism in human HLA. Although supplemental, extended panels are available, they are not universally applied and their diagnostic value is uncertain. It is unclear whether standard bead kits can accurately predict the reactivity seen in extended panels. This study investigated, when using the standard panel alone, whether a closest related bead strategy or an eplet pattern strategy might predict reactivity seen in an extended panel kit (Explex). In the closest related bead strategy, each bead in the extended panel was compared to a standard kit bead in the same serotype and/or had the fewest amino acid differences. Using the closest bead method the standard bead panel predicts reactivity for the majority of beads in the Explex panel (30/54 for Class I and 16/24 for Class II). In the eplet pattern strategy, the reactivity of beads in the Explex panel was predicted from proposed eplet assignment of the standard panel. This strategy showed modest accuracy for predicting positive Explex beads, with greater accuracy for Class II (25/26) compared to Class I (49/62). This strategy had similar accuracy for highly sensitised cohorts (19/22 for Class II, and 27/32 for Class I) when patterns of eplet reactivity could be identified. We found that closely-related beads and eplet patterns (when identifiable) can predict positive reactivity in unrepresented alleles with moderate success. However, we detail several exceptions where additional panel testing was found to be helpful. Our findings support selective use, rather than universal application of extended panels in our local population.