Phenotypic characterization of patients with rheumatologic manifestations of common variable immunodeficiency

Abstract

Patients with common variable immunodeficiency (CVID) have a higher incidence of rheumatologic disorders. To delineate this clinical association, we investigated the phenotypic features of patients with CVID affected by these conditions.

Methods: We conducted a retrospective analysis of 870 pediatric and adult patients with CVID included in the United States Immunodeficiency Network (USIDNET) registry. Outcomes included clinical characteristics (age, gender, ethnicity, rheumatologic diagnosis, and comorbidities), infectious history and basic immunophenotype (serum immunoglobulin levels, CD19+ B cells, and CD4/CD8 ratio) in patients with CVID and rheumatologic disorders compared to those with non-inflammatory CVID. Demographic and clinical data were compared using chi-square, Fisher's exact or Wilcoxon-Mann-Whitney tests. Non-parametric tests, single and multiple logistic regression models were used to evaluate the relationship between CVID-associated rheumatologic disorders and basic immunophenotypic parameters (IgA, IgM, CD19+ B-cell counts, and CD4/CD8 ratios).

Results: Physician-reported rheumatic diseases were present in 5.9% of patients with CVID (n = 51) included in the registry. Although CVID affects both sexes equally, and patients are of predominantly White-Caucasian ethnicity, there were more females (3.3:1 female to male ratio) and increased proportion of non-white patients in the rheumatologic disease group (p < 0.05). Specific disorders included: inflammatory arthritis (n = 18), Sjogren's syndrome (n = 11), SLE (n = 8), Raynaud's syndrome (n = 8), vasculitis (n = 9), MCTD (n = 3), and other (n = 5). In about one-third of patients, a rheumatologic condition was associated with an additional inflammatory complication or malignancy. In regards to the immunophenotype parameters compared (CD19+ B-cell counts, CD4/CD8 ratio, IgA, and IgM), no significant differences were demonstrated between the two groups.

Conclusion: Our findings highlight the coexistence of primary antibody immunodeficiencies and systemic rheumatologic disorders, describe the spectrum of rheumatologic manifestations, and contrast differences in relevant demographic, clinical and immunophenotype parameters in the largest registry of CVID patients in the U.S. In spite of its limitations, our study details the intersection of systemic autoimmunity and CVID and provides valuable insights into these two groups of disorders. Further delineating the link between systemic autoimmunity and humoral immunodeficiencies can provide novel insights into the immune abnormalities underlying these related conditions.

Keywords: Autoimmunity; CVID; Primary Antibody Deficiency; Rheumatologic Disease.

Figure 1.

Flow Diagram shows selection of patients included in the analysis. Nine patients (n=9) had an additional primary immunodeficiency (PID) diagnosis including Transient Hypogammaglobulinemia of Infancy (n=7), Autoimmune Lymphoproliferative syndrome (n=1) and X-linked Lymphoproliferative Syndrome Type 2 (n=2). There were 274 patients with inflammatory complications of CVID listed. Specific inflammatory complications included systemic autoimmunity (n=51), organ-specific autoimmunity (n=85) including those with autoimmune cytopenias, malignancy (n=87) and other inflammatory complications (n=120) which included granulomatous disease, inflammatory lung disease and non-malignant lymphoproliferation.

Figure 2.. Immunophenotypical differences between non-inflammatory CVID and CVID-associated rheumatologic disease groups.

A. CD19+ cell counts by age and disease group. B. Comparison of CD19+ B-cell counts between patients with non-inflammatory CVID and those with CVID-associated rheumatologic disease. C. CD4/CD8 ratio by age and disease group. D. CD4/CD8 ratio values in non-inflammatory CVID and CVID-associated rheumatic disease groups. E and G: Serum immunoglobulin levels (IgA, IgM) by age and disease group. F and H: Serum immunoglobulin levels (IgA, IgM) in the non-inflammatory CVID and CVID-associated rheumatologic disease groups (outliers are omitted). Logistic regression analyses, adjusted by age at the time of the test, failed to reveal any significant associations between CD19+ cells, CD4/CD8 ratio and immunoglobulin levels (IgA, IgM) and CVID-associated rheumatologic disease.

Figure 3.. Memory B-cell phenotype.

Total and IgM+ memory B-cells are below median values for patients with non-inflammatory CVID in patients with rheumatologic manifestations. Statistical comparison was not performed as there were insufficient observations in the CVID-rheum group.