Normal LRBA expression in a patient with LRBA deficiency: Understanding the clinical test

Abstract

An 11-year-old girl presented as a new immunology patient with a history of celiac disease, inflammatory bowel disease, asthma, chronic bilateral otitis, autoimmune hemolytic anemia, and chronic splenomegaly. In addition, she had generalized lymphadenopathies and chronic lung disease (bronchiectasis, ground glass infiltrates, and nodular opacities). Relevant laboratory findings included anemia, neutropenia, elevated ESR (62), positive ANAs (1:320), poor antibody responses and hypogammaglobulinemia, slightly elevated αβ+double negative T-cells (1.6%) and serum soluble CD25 levels (9560 pg/mL). Parents are consanguineous and genetic testing reveals a homozygous pathogenic variant in LRBA (c.5980C>T). This is a novel defect predicted to cause a premature translational stop signal (p. Arg1994*), resulting in an absent or disrupted protein. The patient responded favorably to abatacept and subcutaneous IgG replacement with resolution of lymphadenopathies, organomegaly and cytopenias, improvement of infections and gastrointestinal symptoms, and decrease in inflammatory markers and serum soluble CD25 levels. Unexpectedly, during her evaluation, flow cytometry to assess intracellular LRBA using a polyclonal rabbit antibody (Sigma-Aldrich) showed protein expression in 100% of T-cells and 99% of B-cells. Since the patient’s clinical and genetic findings and her response to abatacept indicated LRBA deficiency, we reassessed LRBA expression using a different assay with an anti-LRBA monoclonal antibody recognizing the protein N-terminal (Clone 2D4.1, Milipore). This assay showed essentially absent expression of LRBA in T-cells (1.7% in CD4+ and 11.5% in CD8+), B-cells (3.6%), and NK-cells (0.8%), fitting with both our genetic findings and clinical phenotype. Hence, assays detecting LRBA expression may have discrepant results and understanding the technicalities of how they are performed is critical for their interpretation and avoiding potential pitfalls. For instance, it is possible that epitopes targeted by a polyclonal antibody could still be identified on a mutated LRBA protein, yielding normal protein phenotyping levels in LRBA deficient patients. Presence of protein also does not equate to functional protein but a validated functional clinical test for LRBA is yet unavailable. In conclusion, this case underscores the importance of using genetic and functional data in the context of the clinical phenotype for IEI diagnosis.