Serum intestinal fatty acid–binding protein in the noninvasive diagnosis of celiac disease

Current diagnostic guidelines for celiac disease (CD) in pediatric patients require a duodenal biopsy if the IgA anti‐tissue transglutaminase (tTG) is below 10x the upper limit of normal (ULN). Additional markers may enable a noninvasive diagnosis in this group. Serum intestinal‐fatty acid‐binding protein (I‐FABP), a marker for intestinal epithelial damage, could be useful in this respect. A total of 95 children with a clinical suspicion of CD and tTG 1‐10x ULN were investigated. All had a duodenal biopsy and analysis of serum I‐FABP. A control group of 161 children with familial short stature and normal tTG was included. I‐FABP levels in the 71 patients with tTG 1‐10x ULN and biopsy‐proven CD (median 725 pg/mL) were not significantly different (p = 0.13) from the levels in the 24 patients with a tTG 1‐10x ULN but a normal biopsy (median 497 pg/mL). However, when combining tTG and I‐FABP levels, 11/24 patients could have been diagnosed noninvasively if tTG is ≥ 50 U/mL and I‐FABP ≥880 pg/mL or in 12/19 patients if tTG is ≥ 60 U/mL and I‐FABP ≥ 620 pg/mL. Therefore, addition of I‐FABP to the diagnostic procedure of CD may provide a noninvasive diagnosis in patients with a tTG ≥ 50 U/mL.     

Iron-deficiency anemia as the sole manifestation of celiac disease

We report on a 40-year-old woman with a 2-year history of iron-deficiency anemia of unknown origin. Repeated endoscopic investigations in the past had revealed no abnormality of the gastrointestinal system on macroscopic examination. Oral iron supplementation was shown to have no effect on serum iron levels and had no influence on the anemia. Upper gastrointestinal endoscopy performed at our hospital confirmed normal macroscopic findings. However, jejunal biopsies revealed subtotal villous atrophy of the mucosa of the small intestine. A strict gluten-free diet led to an increase in serum iron, resolution of the anemia, and restitution of normal mucosal architecture. Thus iron-deficiency anemia may be the lone manifestation of celiac disease.Abbreviations IgG Immuno globin G - IgM Immunoglobin M - IgA Immunoglobin A Correspondence to: U. Schmitz     

Direct cloning and tetramer staining to measure the frequency of intestinal gluten‐reactive T cells in celiac disease

Knowledge of the frequency of disease‐driving CD4⁺ T cells in lesions of chronic human inflammatory diseases is limited. In celiac disease (CD), intestinal gluten‐reactive CD4⁺ T cells, which recognize gluten peptides only in the context of the disease‐associated HLA‐DQ molecules, are key pathogenic players. By cloning CD4⁺ T cells directly from intestinal biopsies of CD patients, we found that 0.5–1.8% of CD4⁺ T cells were gluten reactive. About half of the gluten‐reactive T cells were specific for either the immuno‐dominant DQ2.5‐glia‐α1a or DQ2.5‐glia‐α2 epitopes, suggesting that direct visualization of gluten‐specific T cells could be possible. Assessed by flow cytometry, tetramer‐positive T cells were present in 10/10 untreated CD patients with a frequency of 0.1–1.2% of CD4⁺ T cells. Gluten‐specific T cells were also detectable in most treated CD patients (7/10). Moreover, the frequency of gluten‐specific T cells correlated with the degree of histological damage in the gut mucosa as scored by Marsh‐grading, and also with serum IgA anti‐transglutaminase 2 antibody levels. Tetramer staining of gluten‐reactive T cells in biopsy material is a useful tool for future studies of such cells in CD and could also potentially serve as a diagnostic supplement in selected cases.     

Tricks of the trade: How to avoid histological pitfalls in celiac disease

Currently, the diagnosis of celiac disease (CD) is based upon the combination of raised serum anti-tissue transglutaminase or anti-endomysial antibodies and the presence of histological alterations of variable degree in the duodenal mucosa. Interpretation of duodenal biopsies is subjected to a number of variables, and the lack of standardization may cause diagnostic controversy or even misdiagnosis. The aim of this overview is to solicit a standardization of the procedures of biopsy taking, orientation, processing, staining and interpretation in order to avoid or minimize misinterpretation of duodenal biopsies. Based on a literature review and extensive personal experience, the appropriate methodology of duodenal biopsy taking, orientation, fixation, processing, staining and interpretation was thoroughly reviewed, and the most common and relevant errors and artifacts were identified. To maximize the diagnostic yield of duodenal biopsy in CD, multiple specimens are best taken from different sites of the duodenum during endoscopy, and careful visual inspection of the duodenal mucosa may help identify abnormalities related to villous atrophy. Biopsy handling and orientation are of utmost importance to avoid artifacts that may impair the pathologist's ability to detect pathology and normality. Immunostaining with anti-CD3 monoclonal antibody should be carried out, and a simplified histological classification may help distinguish atrophic from non-atrophic stages of CD enteropathy. Meticulous attention to biopsy orientation, handling and processing - together with the knowledge of the most common histological artifacts - is necessary to avoid a wrong histological interpretation which, in turn, may lead to misdiagnosis in CD.     

Hypercoagulability in celiac disease — An update

Celiac disease is a life-long autoimmune disease affecting multiple organs of genetically susceptible individuals. One of the extra intestinal manifestations of the disease is thromboembolic events like strokes, veins' thrombosis, and pregnancy losses. Hypercoagulable autoimmune diseases like lupus erythematosus and antiphospholipid syndrome, associated with celiac disease just add risk to the patients. Pathogenic predisposing avenues increasing the hypercoagulability in celiac disease are multiple: nutritional deficiencies (B12, folate, and vitamin K), genetic predisposition (MTHFR mutations), thrombophilic autoantibodies, hyperhomocysinemia, endothelial dysfunction and platelet abnormalities. Primary pharmacologic thromboprophylaxis or treating the predisposing factors should be considered on a personal basis.     

Immunopathogenesis and therapeutic approaches in pediatric celiac disease

Celiac Disease is an autoimmune enteropathy with increasing incidence worldwide in both adults and children. It occurs as an inflammatory condition with destruction of the normal architecture of villi on consumption of gluten and related protein products found in wheat, barley and rye. However, the exact pathogenesis is not yet fully understood. A gluten-free diet remains the main modality of therapy to date. While some patients continue to have symptoms even on a gluten-free diet, adherence to this diet is also difficult, especially for the children. Hence, there is continued interest in novel methods of therapy and the current research focus is on the promising novel non-dietary modalities of treatment. Here, we critically reviewed the existing literature regarding the pathogenesis of celiac disease in children including the role of in-utero exposure leading to neonatal and infant sensitization and its application for the development of new therapeutic approaches for these patients.     

Allergy prevalence in adult celiac disease

BACKGROUND: Celiac disease is considered to arise from an inappropriate T-cell-mediated immune response against ingested gluten in genetically predisposed people, whereas the T(H)2-type lymphocytes are mostly involved in IgE-mediated reactions. The matter of possible coexistence of T(H)1- and T(H)2-type diseases is still debated. OBJECTIVE: This study was aimed to evaluate the allergy prevalence in a large series of adults with untreated celiac disease and their families at the moment of diagnosis. We also evaluated whether 1 year of gluten-free diet had any effect on allergy prevalence in our cohort. METHODS: At the moment of celiac disease diagnosis a standardized questionnaire was administered for detailed information on presence and type of any allergy symptoms in 1044 adult patients with celiac disease, 2752 relatives, and 318 spouses. Those reporting any allergy underwent tests with dosage of serum levels of total IgE and search for serum specific IgE with a standard makeup of 20 antigens and PRICK tests in selected individuals. At follow-up visit patients with celiac disease were administered the same allergy questionnaire. RESULTS: One hundred seventy-three patients with celiac disease (16.6%), 523 relative (19%), and 43 spouses (13.5%) had at least 1 allergy (P=not significant). Atopic dermatitis was more frequent in patients with celiac disease (3.8%) and their relatives (2.3%) than in spouses (1.3%). The presence of allergy in general and atopic dermatitis was not affected by presence of overt malabsorption or duration of undiagnosed disease. Follow-up data showed no change in allergy prevalence in the cohort examined. CONCLUSION: Allergy prevalence in a large series of patients with celiac disease is not different from that of their relatives and spouses. However, atopic dermatitis was about 3 times more frequent in patients with celiac disease and 2 times more frequent in their relatives than in spouses. One year of gluten-free diet did not change allergy prevalence in the celiac group under investigation.     

Fatal CNS vasculopathy in a patient with refractory celiac disease and lymph node cavitation

Celiac disease is an enteropathy occurring in genetically predisposed individuals due to a dietary intolerance to gluten. Patients with celiac disease may develop a neurological disorder of unknown cause, although autoimmune mechanisms are suspected. We report on a 56-year-old man with celiac disease, who became refractory to a gluten-free diet and died of a rapidly progressive encephalopathy. Magnetic resonance imaging indicated focal lesions of the cerebellum and brainstem, and electrodiagnostic studies suggested an axonal neuropathy. Autopsy revealed a flattened small-bowel mucosa with intraepithelial lymphocytosis, a spectrum of degenerative changes of the intra-abdominal and mediastinal lymph nodes, including cavitary degeneration, and splenomegaly. Histologically, the lymph nodes showed pseudocyst formation and lymphocytic vasculitis with fibrinoid necrosis, and sections of the brain exhibited fibrinoid degeneration of small blood vessels, sparse perivascular lymphocytic infiltrates, and perivascular ischemic lesions. Identical T-cell clones were identified in the duodenum, stomach, lymph nodes, and spleen. This patient had an unusual neurological disorder related to a vasculopathy, probably mediated by a circulating neoplastic clone of activated T cells.     

Anti-cardiolipin and anti-beta 2-glycoprotein I antibodies in celiac disease

Aims

To determine the frequency of anti-cardiolipin (aCL) and anti-β2-glycoprotein I antibodies (aβ2GPI) in celiac disease (CD) patients.

Patients and methods

Sixty-three untreated CD patients and 40 healthy blood donors (HBD) were studied. IgG, IgA and IgM aCL and aβ2GPI were detected by Elisa.

Results

The frequency of antiphospholipid antibodies (aPL) (aCL and/or aβ2GPI) was significantly higher in CD patients (12 out of 63) than in HBD (two out of 40) (19% vs 5%, P = 0.04). Six CD patients out of 63 (9.5%) and one HBD out of 40 (2.5%) had aCL. Ten CD patients (15.9%) and two HBD (5%) had aβ2GPI. Only aβ2GPI-IgA was significantly more frequent in CD patients than in HBD (14.3% vs 2.5%, P = 0.048). In CD patients, aβ2GPI-IgA (nine out of 63) was significantly more frequent (14.3%) than aβ2GPI-IgG (1.6%) and IgM (1.6%) (P = 0.008). In CD patients, the frequency of aCL-IgA and IgM was 6.3% (four out of 63) and aCL-IgG were not detected. Simultaneous presence of positive antibodies was found in four CD patients: one patient had four aPL, one had three aPL and two had two aPL. The four patients who had aCL-IgA had also aβ2GPI-IgA and three of them had a titer higher than 50 units. Among nine patients with aβ2GPI-IgA, four had a titer higher than 100 units. The highest titers were found in adults.

Conclusions

aPL and particularly aβ2GPI-IgA are frequent in CD. The significance of these antibodies has to be determined.     

High frequency of low-risk human leukocyte antigen class II genotypes in latent celiac disease

Human leukocyte antigen (HLA) class II genotypes in latent celiac disease, a clinical variant of celiac disease (CD) have been scarcely studied. The aim of this work was to investigate whether latent CD and CD share similar frequencies of HLA class II genotypes. HLA class II genotypes of CD patients compared with controls were subdivided in the following at-risk groups: DQB1*02/DQB1*02 (43.0%, odds ratio [OR] 8.02, p < 0.0001), DQB1*0302/DQB1*02 (12.2%, OR 2.77, p = 0.0002), DQB1*02/DQB1*X (39.2%, OR 1.23, p = 0.1903), DQB1*0302/DQB1*X (3.4%, OR 0.35, p = 0.0064) and DQB1*X/DQB1*X (0.8%, OR 0.01, p = 0.0001) where X is neither DQB1*0302 nor DQB1*02. Next, HLA class II genotypes of 21 latent CD patients were compared with the above at-risk groups. Only one latent CD patient (4.8%) was found in the high risk DQB1*02/DQB1*02 group, three (14.3%) were DQB1*0302/DQB1*02, one (4.8%) was DQB1*0302/DQB1*X and the remaining 16 (76.2%) showed the DQB1*02/DQB1*X genotype. Noteworthy, the only 1 patient with the DQB1*02/DQB1*02 high risk genotype did not carry the DR3-DQB1*02/DR3-DQB1*02 or the DR3-DQB1*02/DR7-DQB1*02 but the uncommon DR3-DQB1*02/DR4-DQB1*02 genotype. These data suggest that latent CD is prevalently associated with low-risk HLA class II genotypes.     

Orchestration of the circadian clock and its association with Alzheimer's disease: Role of endocannabinoid signaling

Circadian rhythms are 24-hour natural rhythms regulated by the suprachiasmatic nucleus, also known as the "master clock". The retino-hypothalamic tract entrains suprachiasmatic nucleus with photic information to synchronise endogenous circadian rhythms with the Earth’s light-dark cycle. However, despite the robustness of circadian rhythms, an unhealthy lifestyle and chronic photic disturbances cause circadian rhythm disruption in the suprachiasmatic nucleus’s TTFL loops via affecting glutamate and γ-aminobutyric acid-mediated neurotransmission in the suprachiasmatic nucleus. Recently, considerable evidence has been shown correlating CRd with the incidence of Alzheimer's disease. The present review aims to identify the existence and signalling of endocannabinoids in CRd induced Alzheimer's disease through retino-hypothalamic tract- suprachiasmatic nucleus-cortex. Immunohistochemistry has confirmed the expression of cannabinoid receptor 1 in the suprachiasmatic nucleus to modulate the circadian phases of the master clock. Literature also suggests that cannabinoids may alter activity of suprachiasmatic nucleus by influencing the activity of their major neurotransmitter γ-aminobutyric acid or by interacting indirectly with the suprachiasmatic nucleus’s two other major inputs i.e., the geniculo-hypothalamic tract-mediated release of neuropeptide Y and serotonergic inputs from the dorsal raphe nuclei. Besides, the expression of cannabinoid receptor 2 ameliorates cognitive deficits via reduction of tauopathy and microglial activation. In conclusion, endocannabinoids may be identified as a putative target for correcting CRd and decelerating Alzheimer’s disease.     

HLA-G and susceptibility to develop celiac disease

The Human Leukocyte Antigen-G has immunomodulatory function and its expression has been associated with several diseases. In our study we analyzed HLA-G polymorphisms in order to evaluate their possible association with susceptibility to celiac disease development. A total of 420 celiac patients and 509 controls were genotyped for HLA-G polymorphisms. We sequenced 800 bp upstream the ATG codon (5′ upstream regulatory region) and the whole 3′ untranslated region of the HLA-G gene, whereas the ΔC deletion at exon 3 was detected by RFLP-PCR.Five polymorphisms (namely −477 C>G, −369 C>A, 14 bp del/ins, 3187 A>G, 3196 C>G) and one haplotype (TCGGTACGAAITCCCGAG) were significantly more frequent in celiac patients than controls and associated with increased disease susceptibility. The 14 bp I/I, 3187 G/G, 3196 G/G genotypes and TCGGTACGAAITCCCGAG haplotype, were still significantly associated with increased disease susceptibility (and in addition also the 3003 C/C genotype) when the analysis was restricted to patients and controls presenting the DQ2.5 or DQ8 HLA-DQ celiac disease risk haplotypes.Our findings indicate an association between HLA-G gene polymorphisms and susceptibility to celiac disease development, suggesting that HLA-G molecule is possibly involved in the pathogenesis of the disease.     

Bone-specific antibodies in sera from patients with celiac disease: characterization and implications in osteoporosis

Osteopenia and osteoporosis are well-known complications detected in celiac disease patients with still obscure pathogenesis. In the present study we investigated the presence of circulating anti-bone autoantibodies in patients with celiac disease and explored their role in the associated bone disease. We evaluated serum samples from 33 patients at the time of diagnosis and from 20 of them after treatment. Sera from patients with inflammatory bowel disease (n = 9), nonceliac osteoporotic (n = 18), and healthy individuals (n = 10) were used as controls. The presence of IgA specific anti-bone antibodies was first investigated using indirect immunofluorescence on cryosections of fetal rat tibia (20-day pregnancy). Furthermore, samples were homogenized and total tissue extracts were subjected to Western blot analysis to confirm immunoreactivity. At diagnosis, sera from 51.5% (17/33) of celiac patients had antibodies that recognized antigenic structures in chondrocytes and the extracellular matrix along mature cartilage, bone interface, and perichondrium of fetal rat bone. Among controls, only two osteoporotic patients showed very low titles of anti-bone autoantibodies. The immunostaining was localized in areas where an active mineralization process occurred and was similar to the distribution of the native bone tissue transglutaminase. The frequency of patients with positive baseline titers of anti-bone antibodies diminished significantly after treatment (P = 0.048). Western blot assays confirmed the presence of autoantibodies in sera from patients with a positive immunofluorescence staining. Autoantibodies recognized a major protein band on tissue extracts with a molecular weight of 77–80 kDa, which could be displaced when sera were preadsorbed with human recombinant tissue transglutaminase. We provide original evidence that patients with celiac disease have IgA-type circulating autoantibodies against intra- and extracellular structures of fetal rat tibia. Our findings suggest that these antibodies recognize bone tissue transglutaminase as the autoantigen, and based on the localization of the immunoreactivity we speculate that they might have an active role in the pathophysiology of celiac disease-associated bone complications.     

Elution of antitransglutaminase antibodies from duodenal biopsies: a novel approach in the diagnosis of celiac disease

Celiac disease (CeD) is a disease more prevalent and multisymptomatic than was earlier recognized. Whereas prompt initiation of gluten-free diet (GFD) is beneficial in relieving the symptoms, an accurate CeD diagnosis is necessary also to avoid years of restricted diet on uncertain grounds. We propose a new diagnostic method, based on elution of deposited antibodies against transglutaminase (anti-tTG) from duodenal biopsies in patients with symptoms and screening serology analyses suggestive of CeD. The eluates were analyzed in a Phadia 250 fluoroimmunoassay, demonstrating elevated concentrations of anti-tTG in CeD patients, corresponding to serology and histopathology findings. In one case histology was inconclusive, displaying only unspecific inflammation, but eluted anti-tTG was positive. This patient has clinically improved following GFD. We conclude that our novel method represents a new tool in the diagnostic work up in CeD. The detection of deposited anti-tTG at the site of inflammation appears to provide a high sensitivity and specificity using a technique that is quick, simple and reliable. Further studies are needed for optimization and elucidation of suitable applications for this elution method.     

Antibodies to muscle and ganglionic acetylcholine receptors (AchR) in celiac disease

Background: About 2.5% of patients with idiopathic peripheral neuropathy or idiopathic dysautonomia have underlying celiac disease (CD). Antibodies to ganglioside have been reported in CD patients with neuropathy. No data are so far available on the presence in CD of acetylcholine receptor (AChR) antibodies. Muscle AChR antibodies are found in patients with myasthenia gravis, and ganglionic AChR antibodies in patients with autoimmune autonomic neuropathy.

Objective: To determine the frequency of AChR antibodies in CD patients and assess possible correlations with neurological manifestations.

Methods: Seventy CD patients (16 M, 54 F, mean age 36 years) underwent neurological and electrophysiological evaluation. AChR antibodies were detected with radioimmunoprecipitation assay. Sera from 15 age-matched patients with systemic lupus erythematosus (SLE) and 10 with Sjogren syndrome were studied as controls.

Results: None of our CD patients complained of autonomic symptoms or fatigable weakness. Borderline titres (0.03–0.05 nmol/l) of ganglionic AChR antibodies were present in 4 patients, one affected with type I diabetes and one with subclinical neuropathy. Three of the 4 patients underwent cardiovascular autonomic function tests, which showed no abnormalities. Low levels of ganglionic AChR antibodies (0.05–0.10 nmol/l) were found in 2 SLE control patients, one of whom had a severe sicca complex. Muscle AChR antibodies (>1.0 nmol/l) were found in two CD patient and one control patient with SLE. Neither had symptoms or signs of myasthenia gravis.

Discussion and conclusions: CD is occasionally associated with neurologic disease, and with antibody reactivity to neuronal antigens. None of our CD patients had autonomic failure or significant levels of ganglionic AChR antibodies. Two CD patient and one control with SLE had muscle AChR antibodies without clinical evidence of myasthenia. The presence of antibodies in CD and in SLE patients may reflect a non-specific autoimmune response in these patients or may indicate subclinical autoimmune autonomic and neuromuscular involvement.     

Mouse models of intestinal inflammation as tools to understand the pathogenesis of inflammatory bowel disease

Mouse models of intestinal inflammation resemble aspects of inflammatory bowel disease in humans. These models have provided important insights into mechanisms that control intestinal homeostasis and regulation of intestinal inflammation. This viewpoint discusses themes that have emerged from mouse models of intestinal inflammation including bacterial recognition, autophagy, the IL‐23/Th‐17 axis of inflammation as well as the role of negative regulators. Many of the pathways highlighted by model systems have been identified in recent genome‐wide association studies in human validating the relevance of mouse models to human inflammatory bowel disease. Understanding of the complex biological mechanisms that lead to intestinal inflammation in mouse models may help to define targets for treatment of human diseases.     

Upregulation of KIR3DL1 gene expression in intestinal mucosa in active celiac disease

Killer cell immunoglobulin-like receptors (KIRs) modulate natural killer (NK) and T-cell function by human leukocyte antigen class I interaction and have been implicated in celiac disease (CD). Qualitative expression of 16 KIR genes was determined in biopsies from 22 CD patients at diagnosis and after >2 years on a gluten-free diet (GFD). Quantitative expression analysis of KIR2DL4, KIR3DL1, KIR3DL3, and KLRC2 (a marker of an NK-reprogrammed T-cell subpopulation augmented in CD) was performed in 35 additional CD biopsy pairs and 14 non-CD control biopsies. No specific KIR expression profile was observed in CD. KIR3DL1 was more frequently expressed in active CD compared with GFD (p = 0.0312) and controls (p = 0.0008), with slightly increased levels in active disease. KLRC2 was overexpressed in active (p = 0.0037) and GFD (p = 0.0469) patients compared with non-CD controls and coexpressed with KIR3DL1. Results suggest the participation of KIR3DL1 overexpression in the overall immune activation seen in CD mucosa, which could be partly explained by the NK-like T-cell subpopulation increase.     

Automated Marsh-like classification of celiac disease in children using local texture operators

Automated classification of duodenal texture patches with histological ground truth in case of pediatric celiac disease is proposed. The classical focus of classification in this context is a two-class problem: mucosa affected by celiac disease and unaffected duodenal tissue. We extend this focus and apply classification according to a modified Marsh scheme into four classes. In addition to other techniques used previously for classification of endoscopic imagery, we apply local binary pattern (LBP) operators and propose two new operator types, one of which adapts to the different properties of wavelet transform subbands. The achieved results are promising in that operators based on LBP turn out to achieve better results compared to many other texture classification techniques as used in earlier work. Specifically, the proposed wavelet-based LBP scheme achieved the best overall accuracy of all feature extraction techniques considered in the two-class case and was among the best in the four-class scheme. Results also show that a classification into four classes is feasible in principle however when compared to the two-class case we note that there is still room for improvement due to various reasons discussed.     

Celiac disease risk stratification based on HLA-DQ heterodimer (HLA-DQA1 ~ DQB1) typing in a large cohort of adults with suspected celiac disease

BackgroundsPatients with celiac disease (CeD) carry the major histocompatibility complex class II, HLA-DQ2 or DQ8 haplotype; the absence of these haplotypes excludes a diagnosis of CeD. While the most common and highest risk HLA haplotypes in CeD have been established, the risk profiles of the less common and equivocal HLA haplotypes need further refinement. The aim of this study was to use a large national patient cohort to further stratify the risk gradient of HLA-DQ haplotypes.MethodsThe study cohort included 24,339 adult patients with suspected CeD and immunoglobulin (Ig)A sufficiency (total IgA ≥ 70 mg/dL) whose samples were assessed at Mayo Clinic Laboratories for HLA-DQ genotyping, total IgA, and tissue transglutaminase (tTG)-IgA. Data from a subset of the patients who had duodenal biopsies were analyzed to determine the risk gradient of CeD. Logistic regression models were used to evaluate the risk gradient and to calculate odds ratios (ORs) for being positive to CeD serology according to different HLA-DQ2 and DQ8 heterodimers.ResultsOf the 24,339 patients, 55% (n = 13,456) expressed HLA-DQ2 or DQ8 heterodimers. Compared with patients who had non-permissive HLA-DQ heterodimers, patients who had HLA-DQ2 homozygosity (HLA-DQ2.5/DQ2.5, HLA-DQ2.5/DQ2.2, or HLA-DQ2.2/DQ2.2) showed increased odds for tTG-IgA positivity (OR = 96.9; 95% CI, 58.3–147.9). Interestingly, the odds for patients who were compound heterozygous for HLA-DQ2.5 and HLA-DQ8 were similar to those for HLA-DQ2.5 heterozygotes. However, a single HLA-DQ2.2 haplotype (without HLA-DQ8, DQ2.2 heterozygous) was not associated with tTG-IgA positivity. These findings were confirmed in a subset of patients (n = 738) who had duodenal biopsies performed in addition to CeD serologic testing.DiscussionThis large national reference laboratory cohort study demonstrated that HLA-DQ2.2 heterozygous is not associated with positive tTG-IgA serology, suggesting the reclassification of this haplotype as non-permissive for CeD.