Blood-based test for diagnosis and functional subtyping of familial Mediterranean fever

Abstract

Background and objective: Familial Mediterranean fever (FMF) is the most common monogenic autoinflammatory disease (AID) worldwide. The disease is caused by mutations in the MEFV gene encoding the inflammasome sensor Pyrin. Clinical diagnosis of FMF is complicated by overlap in symptoms with other diseases, and interpretation of genetic testing is confounded by the lack of a clear genotype-phenotype association for most of the 340 reported MEFV variants. In this study, the authors designed a functional assay and evaluated its potential in supporting FMF diagnosis.

Methods: Peripheral blood mononuclear cells (PBMCs) were obtained from patients with Pyrin-associated autoinflammation with an FMF phenotype (n=43) or with autoinflammatory features not compatible with FMF (n=8), 10 asymptomatic carriers and 48 healthy donors. Sera were obtained from patients with distinct AIDs (n=10), and whole blood from a subset of patients and controls. The clinical, demographic, molecular genetic factors and other characteristics of the patient population were assessed for their impact on the diagnostic test read-out. Interleukin (IL)-1β and IL-18 levels were measured by Luminex assay.

Results: The ex vivo colchicine assay may be performed on whole blood or PBMC. The functional assay robustly segregated patients with FMF from healthy controls and patients with related clinical disorders. The diagnostic test distinguished patients with classical FMF mutations (M694V, M694I, M680I, R761H) from patients with other MEFV mutations and variants (K695R, P369S, R202Q, E148Q) that are considered benign or of uncertain clinical significance.

Conclusion: The ex vivo colchicine assay may support diagnosis of FMF and functional subtyping of Pyrin-associated autoinflammation.

Keywords: familial mediterranean fever; fever syndromes; inflammation.

Figure 1

Diagnosis of familial Mediterranean fever (FMF) using a functional assay. (A) Peripheral blood mononuclear cells from healthy donors (n=48) and patients with FMF (n=43) were treated for 5 hours with either Clostridium difficile toxin A (TcdA) alone or with TcdA in combination with colchicine before culture supernatants were analysed for interleukin (IL)-1ß and IL-18, and the TcdA+colchicine over TcdA ratio for each cytokine was calculated. Data are combined from multiple experiments. (B) For both parameters, the receiver operating characteristic (ROC) curve was calculated, as well as the area under curve (AUC). (C) For both parameters, the Youden index was calculated to determine the most appropriate cut-off point, given the sum of sensitivity and specificity being maximum.

Figure 2

Functional stratification of patients with Pyrin-associated autoinflammatory disease correlates with MEFV genetic variants. (A) Combined representation of interleukin (IL)-1β and IL-18 ratios of the ex vivo colchicine assay with peripheral blood mononuclear cells from healthy donors (n=48) and the patient group composed of patients with MEFV gene variants that presented with either a familial Mediterranean fever (FMF) phenotype (n=43) or with autoinflammatory features not compatible with FMF (n=8). Cut-off points as determined by the Youden index are indicated. (B) Analysis of variance for the patient group represented by the p value of the F test. Regression analysis for potential correlation between the assay and clinical response to colchicine tested at 5% significance level. (C) Representation of the functional assay with patient data being separated based on MEFV variants.

Figure 3

Functional stratification of familial Mediterranean fever (FMF) patients from healthy donors, pyogenic arthritis, pyoderma gangrenosum, and acne (PAPA), and mevalonate kinase deficiency (MKD) patients. Peripheral blood mononuclear cells from controls, patients with FMF, and patients with PAPA (A) or patients with MKD (B) were treated for 5 hours with either Clostridium difficile toxin A (TcdA) alone or TcdA in combination with colchicine before culture supernatants were analysed for interleukin (IL)-1ß and IL-18, and the TcdA+colchicine over TcdA ratio for each cytokine was calculated.

Figure 4

Functional familial Mediterranean fever (FMF) screening in human whole blood. Fresh, undiluted whole blood from controls (n=8) and patients with FMF(n=7) was treated for 24 hours either with Clostridiumdifficiletoxin A (TcdA) alone or TcdA in combination with colchicine before culture supernatants were analysed for interleukin (IL)-1ß and IL-18, and the TcdA+colchicine over TcdA ratio for each cytokine was calculated. Results are combined from four independent experiments.