NON-CELIAC GLUTEN SENSITIVITY IN CHILDREN WITH HEADACHE: A FOCUS ON CLINICAL AND IMMUNOLOGICAL ASPECTS

BACKGROUND Headaches frequently occur in childhood, often with a relevant burden on quality of life. The identification of any underlying treatable causes or precipitating factors should be a primary endpoint in headache management. Headache is one of the most frequently reported complaints in patients with gluten-related disorders, particularly Non-celiac Gluten Sensitivity (NCGS), although prevalence of NCGS in headache patients in childhood is unknown. Moreover, etio-pathogenetic bases for headache in NCGS are poorly understood. AIMS OF THE STUDY The study aimed: 1) to explore the prevalence and the clinical picture of NCGS in a cohort of pediatric headache patients with serological markers of gluten sensitivity (native anti-gliadin antibodies - AGA); 2) to analyze gene expression profiles in peripheral blood mononuclear cells (PBMCs) from pediatric headache patients with NCGS; 3) to validate some of gene array results by the measurement of soluble mediators in sera of patients, and to determine serum levels of Tranglutaminase 6 (TG6) antibodies, markers of neurological gluten-related disorders. PATIENTS AND METHODS Patients aged <18 years and evaluated for headache at the Child Neuropsychiatry Unit, University Hospital of Verona, Italy, in the period 1/03/13 – 31/07/15 were considered for recruitment in the study. Inclusion criteria were: 1) to be diagnosed with a type of primary headaches according to the ICHD III-beta criteria; 2) to accept determination of total IgA, AGA IgG and IgA, anti-tranglutaminase 2 IgA antibodies (TGA), and specific IgE to wheat, gluten, gliadin on serum samples. After informed consent, patients were investigated with a complete blood count, biochemical analyses and evaluation of liver, renal and thyroid function and dosage of total IgA and TGA. Patients were excluded if one of the following conditions occurred: fulfilled criteria for secondary headaches or cranial nerve neuralgias; ongoing treatments for headache introduced or modified less than 2 months before; total or partial IgA deficiency, according to reference values for age; to have tested positive for TGA (suspected diagnosis of celiac disease) or IgE to wheat, gluten, gliadin (suspected diagnosis for wheat allergy); presence of enteropathy at duodenal biopsy (performed only in AGA-positive patients). Determination of AGA was performed. According to the diagnostic algorithm for NCGS, AGA-positive patients underwent a 3 months – period on gluten-free diet (GFD) followed by reintroduction of dietary gluten for 3 months (gluten challenge). AGA-negative patients continued on their normal, gluten-containing diet. The headache clinical course was evaluated by means of standardized scales and questionnaires. PBMCs were collected from NCGS patients in active phase of the disease for gene expression profiling. Based on results of gene expression profiling, soluble CTLA-4 (sCTLA-4) and soluble gp130 (sgp130) were chosen to be determined in sera of headache patients. Soluble CD25 (sCD25), chosen as a biomarker for T lymphocyte activation, and TG6 IgA and IgG were also measured in sera. RESULTS Seventeen patients were recruited with migraine or tension-type headache: 11 subjects (4 males, 7 females; mean age, 10.6 years; age range, 7.9-12.5 years) tested positive for AGA, 6 children (2 males, 4 females; mean age, 11.6 years; age range, 9.1-13.8 years) were AGA-negative. All AGA-positive patients had duodenal biopsy on gluten-containing diet: None of the subjects displayed histological signs of enteropathy. AGA-positive patients underwent GFD followed by gluten challenge. Six out 11 AGA-positive patients (55%) received the diagnosis of NCGS, due to clinical improvement on GFD and relapse on gluten challenge. As for clinical picture, no headache features or migraine-associated syndromes resulted to be typical of NCGS patients. Gastrointestinal symptoms occurred more frequently in NCGS patients (50%) than in other patients. Extra-intestinal symptoms (fatigue, depression, anemia, long-lasting low-grade fever, dermatitis, failure to thrive) were more frequent in AGA-positive patients and very similar to those reported in NCGS cohorts, although they could not distinguish NCGS patients among other AGA-positive patients. Three AGA-positive patients out 11 (27%), i.e. 1/6 (17%) AGA-positive NCGS patients and 2/5 (40%) AGA-positive no-NCGS patients tested positive for TG6 IgA. None of the subjects tested positive for TG6 IgA in the AGA-negative group. No one tested positive for TG6 IgG either. Patients with high levels of TG6 antibodies did not show clinical amelioration on a 3 months - period on gluten-free diet. The gene expression profiling of PBMCs from NCGS patients documented up-regulation, in the active phase of the disease, of a high number of genes related to T- and B-lymphocyte activation and up-regulation of genes related to Th17 cell subset and type I Interferon response. Levels of sCTLA-4 were significantly higher in headache AGA-positive patients than in AGA-negative subjects (mean levels 14.46±3.74 ng/ml vs 0.53±0.67 ng/ml; p=0.001), and sCTLA-4 levels markedly reduced on gluten-free diet in all subjects, despite NCGS diagnosis was not confirmed in some of them. The difference in mean levels at recruitment and on GFD was statistically significant in NCGS patients (mean levels 14.46±3.74 ng/ml vs 4.28±3.46 ng/ml; p=0.036). Serum levels of sgp130 were not significantly different at recruitment in NCGS patient, as compared to other AGA-positive subjects or to AGA-negative headache patients, neither levels of this soluble mediator significantly changed on gluten-free diet. Serum levels of sCD25 were similar in NCGS patients, no-NCGS patients and AGA-negative subjects. CONCLUSIONS A subgroup of children or adolescents fully satisfying criteria for primary headaches has Non-celiac Gluten Sensitivity. Experimental data from NCGS patients suggest the existence of an activated immune response with a type-1 Interferon signature. High levels of serum sCTLA-4 in AGA-positive patients may be a further clue for gluten sensitization and involvement of adaptive immunity. The presence of TG6 antibodies in sera may predict a poor clinical response to gluten-free diet upon a 3 months – period, thus justifying a dietary treatment for 6-12 months at least, as recommended for patients with gluten ataxia who tested positive for TG6 antibodies.