ALPS or not?

In this issue of Blood, Molnár et al propose a diagnostic classifier to facilitate the rapid and accurate diagnosis of childhood autoimmune lymphoproliferative syndrome (ALPS).¹ 

Clinicians faced with a challenging case of generalized lymphadenopathy or splenomegaly with or without autoimmune cytopenias/other autoimmune manifestations have a long list of potential diagnoses from which to choose. Included are self-limited viral infections, Evan’s syndrome, ALPS, and Rosai-Dorfman disease, as well as more serious conditions, such as underlying immunological disorders like common variable immunodeficiency, Wiskott-Aldrich syndrome, X-linked lymphoproliferative disease, and malignancies.²  The conundrum is to differentiate true ALPS cases from other diagnoses with overlapping laboratory findings, especially elevated double-negative T cells (DNTs). Molnár et al have provided some valuable observations to help with these challenges.

ALPS is a clinically and genetically heterogeneous group of disorders characterized by underlying immune dysregulation secondary to a defect in FAS-mediated lymphocyte apoptosis. It typically manifests as chronic (>6 months) lymphadenopathy, splenomegaly, or hepatomegaly, polyclonal hypergammaglobulinemia, and refractory multilineage cytopenias (most commonly Coombs-positive autoimmune hemolytic anemia and immune thrombocytopenia). Patients may have other autoimmune disorders and a significantly increased risk of lymphoma. The median age of presentation is 2.7 to 3 years.^3,4  More than 70% of those with ALPS have defined germline and somatic mutations in FAS (TNFRSF6, CD95, and APO1), FASLG, and CASP10, although the penetrance and expressivity are variable among family members with the same mutations. Some 20% to 30% of patients remain genetically undefined. Genetic testing of other family members is therefore important. Treatment options range from observation to short- or long-term immunosuppression with ≥1 drug, including corticosteroids, immunosuppressive agents, mycophenolate mofetil, cyclosporine, sirolimus, and rituximab.² 

Typical laboratory findings include elevated circulating CD3⁺TCRαβ⁺CD4⁻CD8⁻ T lymphocytes (αβ-DNTs), an abnormal in vitro lymphocyte apoptosis assay, and elevated serum interleukin-10 (IL-10), IL-18, soluble FAS ligand (sFASL), and/or vitamin B12 levels.^5-7  αβ-DNTs between 1.0% and 1.5% of total lymphocytes can be seen in healthy individuals or as a reactive phenomenon in other autoimmune disorders. The normal range for these cells varies by age and between laboratories. Lymphopenic patients or those who are ultimately diagnosed with ALPS but have been exposed to high-dose steroids or sirolimus before or at the time of testing can have a high rate of false-positive or false-negative results. Therefore, the αβ-DNT numbers need to be interpreted with caution.^2-8 

A definitive diagnosis of ALPS requires the presence of 2 required criteria (>6-month history of nonmalignant, noninfectious lymphadenopathy or splenomegaly and elevated DNTs) plus 1 primary accessory criterion (defective lymphocyte apoptosis or mutations in FAS, FASLG, or CASP10), whereas a probable diagnosis is based on the presence of both required criteria plus 1 secondary accessory criterion (elevated sFASL, IL-10, vitamin B-12, or IL-18; consistent immunohistological findings; autoimmune cytopenias and polyclonal hypergammaglobulinemia; or family history of nonmalignant/noninfectious lymphoproliferation with or without autoimmunity).⁸  A patient with combination of elevated αβ-DNTs with elevation of 1 of the secondary accessory criteria (vitamin B-12, IL-10, IL-18, or sFASL level) has an 85% to 97% probability of having a FAS mutation, depending on the biomarker used and level of αβ-DNTs. What has been lacking are validated diagnostic classifiers to be able to accurately differentiate ALPS from the pretenders, especially in cases who do not carry the FAS mutation or have an atypical clinical presentation.^5-7 

Molnár et al retrospectively collected clinical, immunological, and genetic data from 215 patients with a clinical suspicion of ALPS. They divided patients into 3 clinical groups (definite, suspected, or unlikely ALPS) based on an updated ALPS 2010 diagnostic protocol (updated according to the 2019 working definitions of the European Society for Immunodeficiencies). Consistent with previous reports, the authors observed significantly higher αβ-DNTs in all patients with definite ALPS (median, 3.95%; range, 1.8% to 23.0%), although the positive predictive value of this test remained low by itself. An abnormal in vitro apoptosis assay was highly specific for patients with definite ALPS and was also the only biomarker that showed a significant difference between the definite and suspected ALPS groups. Although this test is not widely available and may be unsuitable for routine clinical use, it continues to be a valuable aid in diagnosis and management of these patients.

Given these challenges, the authors sought to define an optimal biomarker combination of αβ-DNTs, in vitro apoptosis assay, and sFASL level. Although all combinations showed a significant difference between the definitive and unlikely ALPS groups, the authors found that normal αβ-DNTs and a normal in vitro apoptosis assay could essentially rule out ALPS (see table). The caveat is that they were unable to evaluate the sensitivity or specificity of additional biomarkers combinations, such as immunoglobulin G, vitamin B-12, IL-10, and IL-18, because of a very small number of available samples. Additionally, they were unable to define a biomarker combination to differentiate between the definite and suspected ALPS groups.

The limitations of the study include its retrospective nature, lack of availability of outcome data, inability to correlate specific abnormal biomarkers with clinical presentation, and missing information on treatment received and disease course. The authors were also not able to evaluate the sensitivity or specificity of biomarkers such as vitamin B12 level and others. However, the findings of this study do support the value of the biomarkers defined in revised 2010 diagnostic criteria for ALPS and provide data on additional reliable biomarker combinations that could expedite the diagnosis and treatment of patients with ALPS, without waiting for a genetic diagnosis. This report provides valuable criteria with which to define patients who truly have ALPS and those who do not.