Gastric Cancer Screening in Common Variable Immunodeficiency

Individuals with common variable immunodeficiency (CVID) have an increased risk of gastric cancer, and gastrointestinal lymphoma, yet screening for premalignant gastric lesions is rarely offered routinely to these patients. Proposed screening protocols are not widely accepted and are based on gastric cancer risk factors that are not applicable to all CVID patients. Fifty-two CVID patients were recruited for screening gastroscopy irrespective of symptoms or blood results and were compared to 40 controls presenting for gastroscopy for other clinical indications. Overall, 34% of CVID patients had intestinal metaplasia (IM), atrophic gastritis or moderate to severe non-atrophic gastritis, which can increase the risk of gastric cancer, compared to 7.5% of controls (p < 0.01). Focal nodular lymphoid hyperplasia, a precursor lesion for gastrointestinal lymphoma, was seen in eight CVID patients (16%), one of whom was diagnosed with gastrointestinal lymphoma on the same endoscopy. High-risk gastric pathology was associated with increased time since diagnosis of CVID, smoking, Helicobacter pylori, a low-serum pepsinogen I concentration, and diarrhea, but not pepsinogen I/II ratio, iron studies, vitamin B12 levels or upper gastrointestinal symptoms. There was a lower rate of detection of IM when fewer biopsies were taken, and IM and gastric atrophy were rarely predicted by the endoscopist macroscopically, highlighting the need for standardized biopsy protocols. The prevalence of premalignant gastric lesions in patients with CVID highlights the need for routine gastric screening. We propose a novel gastric screening protocol to detect early premalignant lesions and reduce the risk of gastric cancer and gastric lymphoma in these patients.

A dual-fixed neutrophil substrate improves interpretation of antineutrophil cytoplasmic antibodies by indirect immunofluorescence

OBJECTIVES

To determine whether the addition of a formalin-fixed neutrophil substrate could improve interpretation and prediction of autoantigenic specificity in antineutrophil cytoplasmic antibody (ANCA) testing.

METHODS

Routine diagnostic samples sent for ANCA testing were analyzed prospectively on a dual substrate of both ethanol- and formalin-fixed neutrophils. Positive samples on ethanol-fixed neutrophils were deemed "typical" if formalin-fixed neutrophils also stained, and "atypical" if not. Indirect immunofluorescence (IIF) results were correlated with antimyeloperoxidase (MPO) and anti-proteinase 3 (PR3) results with an enzyme-linked immunosorbent assay (ELISA).

RESULTS

Of 1,426 samples, 201 from unique patients were ANCA-positive (200 on IIF, 1 on ELISA alone). Thirty-two (45%) of 71 typical ANCA staining patterns were positive for either an anti-MPO or anti-PR3 antibodies, whereas only one (0.8%) of 129 atypical patterns was ELISA-positive, in a patient without systemic vasculitis. Only one (3%) of 34 ELISA-positive samples had a negative IIF-ANCA (1/1,426 patients, 0.07%), and this patient did not have vasculitis.

CONCLUSIONS

Concomitant staining on formalin fixation of IIF-positive ethanol-fixed ANCA samples improves the interpretation of ANCA testing and is predictive of vasculitis autoantigens MPO and PR3.

Macaque liver substrate for evaluating dense fine speckled-like patterns on HEp2010 cells

Antinuclear antibody (ANA) detection by indirect immunofluorescence (IIF) is instrumental in the evaluation of systemic autoimmune diseases (SADs). The dense fine speckled (DFS) ANA staining predominantly associates with anti-DFS70, an autoantibody that is thought to exclude the presence of SAD. However, the DFS pattern may mask the presence of other ANA patterns that may be clinically relevant. Our laboratory uses the HEp2010 substrate which contains both HEp2 and liver substates. The aim of this study was to determine whether negative liver nucleus immunofluorescence could exclude the presence of antibodies to extractable nuclear antigens (ENA) in sera with DFS-like patterns. One hundred consecutive sera samples suspicious for DFS pattern, along with 15 sera of control patterns (positive metaphase bars) were included in the study. Each sample was examined separately on HEp2010 (Euroimmun) and liver by two independent readers. Anti-DFS70 was assessed by line and chemiluminescent immunoassays. DFS-like sera were more likely to be liver nucleus-negative compared with control sera. Of the liver-negative sera, 61/64 (95.3%) were deemed anti-ENA negative. Using the liver substrate in the evaluation of anti-ENA had a sensitivity of 90.0% and a negative predictive value of 95.4%. In our cohort, concurrent evaluation of sera with the liver substrate helped rule out the presence of other anti-ENA. This technique may be a safeguard for DFS-like ANA patterns that may mask underlying anti-ENA on IIF.