FAS copy Number Variants Can Also Lead to ALPS, Besides Germline and Somatic Variants

Background: Autoimmune lymphoproliferative syndrome (ALPS) is characterized by childhood onset chronic nonmalignant lymphadenopathy, massive splenomegaly, multilineage cytopenias, and other autoimmune manifestations including uveitis, hepatitis, and nephritis. Typically, the biomarker profile of patients with ALPS includes elevated TCR αβ⁺ DNT cells, serum IgG, serum B12, serum IL-10, soluble FAS ligand (sFASL), and very low HDL cholesterol. Defects in the intrinsic FAS-signaling pathway leading to diminished apoptosis in physiological lymphocyte populations results in lymphocyte accumulation and lymphoproliferation. The molecular defect underlying most ALPS etiologies is attributed to heterozygous germline or somatic (limited to DNT cell subpopulation) pathogenic single nucleotide variants (SNV) in FAS. We describe copy number variants (CNVs) at the FAS locus underlying ALPS in 3 unrelated families.

Methods: Through the Centralized Sequencing Initiative at at the National Institute of Allergy and Infectious Diseases (NIAID), patients undergo genomic workup to identify molecular defects contributing to clinical phenotypes of immune system disorders. All patients receive exome sequencing and a subset of patients also receive array-CGH analysis.

Patients and Results: We performed exome sequencing on 96 patients with a clinical diagnosis of ALPS. For 35 patients with no evidence of molecular defect through exome, we performed CNV analysis. In this cohort, we identified three patients with a copy number variant involving the FAS locus. All patients presented with splenomegaly and lymphadenopathy in childhood with ages of onset ranging from 2 years old to 9 years old. All patients experienced autoimmune neutropenia and thrombocytopenia. They had biomarker evidence showing elevated serum B12 levels and elevated αβ⁺DNT cell populations. They were also found to have very low HDL cholesterol in early childhood ranging from 5-8mg/dL (normal range: 38-55 mg/dL). All patients had negative family histories for lymphoproliferative disorders and immunodeficiency. These patients had clinical presentations and biomarker profiles similar to ALPS patients with germline and somatic FAS variants.

Patient 1: We detected a ~1.03 Mb copy number loss encompassing all of FAS. Parental studies were not performed.

Patient 2: We detected a ~1.004 Mb copy number loss encompassing all of FAS. Parental studies showed this to be maternally inherited. In addition, prior karyotype testing of the bone marrow showed the same deletion.

Patient 3: We detected a ~0.044 Mb copy number loss encompassing exons 7-9 of FAS. Parental studies were not performed.

These results are consistent with the pathogenic nature of copy number variant losses involving FAS. The mechanism of disease in these patients is consistent with haploinsufficiency. In family 2, the mother harboring the FAS deletion is unaffected. This is consistent with prior observation of reduced penetrance within a family in ALPS.

Conclusion: These three cases harbored causative deletions in FAS in the presence of biomarkers indicative of ALPS-FAS and negative results for germline and somatic genetic variant testing. These patients demonstrate that copy number variant analysis should be pursued if there is robust clinical phenotype and biomarker evidence of ALPS as it can lead to an appropriate molecular diagnosis and definitive treatment when EXOME or NEXTgen panel-based FAS sequencing is inconclusive.