Transglutaminase 6 antibodies are not yet mainstream in neuro-coeliac disease

Worldwide 1% of the population suffers from a gluten related disorder (GRD) that severely impacts ones life, and no therapeutic treatment, except a gluten-free diet, is available. GRD’s are a group of immune mediated diseases triggered by the ingestion of gluten. Well known is coeliac disease (CD) that presents with enteropathy. Some patients present exclusively with extra-intestinal manifestations such as skin involvement in dermatitis herpetiformis (DH). Over the last fifty years, patients with gluten sensitivity presenting with neurological symptoms have been described more and more. These symptoms include e.g. cerebellar ataxia and peripheral neuropathies and were thus far considered to be a rare manifestation of GRD. Thus, the diagnosis of neuro-coeliac disease is not yet based on objective and verifiable clinical and/or pathological criteria set by international working groups, such as those of the international Society for the Study of Celiac Disease, the European Society of Pediatric Gastroenterology and Nutrition or Neurological Societies. It would be recommended that groups working on neuro-coeliac disease bundle their forces to obtain a large international cohort of patients to standardize the clinical and pathological picture of this disease.

The Sheffield group, led by professor Hadjivassiliou, is one of the leading groups working on neuro-coeliac disease. In the past they have shown that cerebellar ataxia in gluten sensitive patients coincides with white matter abnormalities and neuropathology. Moreover, they found that Transglutaminase (TG) 6 antibodies are overrepresented in the serum of these patients []. Serum markers in CD include the presence of TG2 antibodies, and in DH elevated levels of TG3 antibodies can be found [, ].

In a previous issue of this journal, Panagiotis Zis et al. present a cross-sectional study on 28 patients with peripheral neuropathies, i.e. sensory ganglionopathy, symmetrical axonal neuropathy or mononeuritis multiplex. The notion that TG6 antibodies were enhanced in cerebellar ataxia patients, and that TG6 is considered to be a Transglutaminase more specific to the central nervous system, they now questioned whether TG6 antibodies were elevated in gluten sensitive patients with peripheral neuropathy. The prevalence of TG6 antibodies was 50% compared to 4% in the healthy population, suggesting that TG6 involvement is not confined to the central nervous system alone, but also can affect the peripheral nervous system. In addition, in CD more than 95% of patients share the HLA DQ2 or DQ8 haplotype, whereas of the patients in this study only 62% shared those HLA haplotypes. Moreover, TG2 antibodies were also elevated in this population [].

As mentioned the so-called neuro-coeliac disease manifestations are not well studied by neurologists and neuroscientists and therefore not yet part of mainstream surveillance. In dermatology for DH and gastroenterology for CD, biopsies are easily obtainable and thereby performed in daily practice. Brain biopsies are non-existent for this diagnosis and post-mortem autopsy material is scarce. Awareness of neuro-coeliac disease is the only clue for diagnosis at this moment. The majority of gluten ataxia patients in our coeliac centre present themselves already as wheel-chair victims. The availability of a serum marker identifying neuro-coeliac patients at an earlier stage of disease would be very helpful. Transglutaminase 6 is suspected to play a key role in the neurological manifestations of GRD in literature well described by the pioneer research group from Sheffield for gluten ataxia, now expanding to neuropathy. After one year of gluten-free diet, TG6 antibody titres were almost undetectable []. However, the causative link between neuro-coeliac disease and circulating TG6 antibodies is questionable []. Aside from gluten ataxia, elevated TG6 antibodies have been observed in patients with other neurological diseases such a Multiple Sclerosis and Amyotrophic Lateral Sclerosis [, ]. Interestingly, the prevalence of DQ2 and DQ8 haplotypes was also increased compared to TG6 negative patients in these studies. Prospective multicentre studies on ALS, MS and neuropathy cases with testing of the whole spectrum of GRD related antibodies are necessary. A gluten-free diet is not a panacea for neurological diseases and should be restricted to well defined risk-groups.

Evidently, the question arises what is exactly measured by TG6 antibodies? In which neurological diseases are titres elevated? Are elevated titres, and at which levels, specific for so called neuro-coeliac disease?

In GRD’s, like CD and DH, 95% of patients are DQ2 and/or DQ8 positive. What is the role of non-DQ2/non-DQ8 genes in TG6 antibody positive patients? Clear cut off values should be calculated based on multi-centre studies with Big Data in order to use TG6 antibodies as a possible marker in daily practice of mainstream neurology.

Multiple commercial TG6 antibody assays (Zedira^©, MyBioSource^©) are available at the moment in addition to the house made assay used in the present study. The use of positive and negative standards is needed to standardize and compare the different batches to one another. Another question is whether the TG6 antibody titre relates to the severity of GA, and if elimination of these antibodies could prevent further damage in neuro-coeliac disease. Interestingly, in the study of Panagiotis Zis the TG2 and TG6 antibody titres were quite similar, resulting in the question whether TG2 antibodies could serve as a marker in neurology, in particular neuropathy patients? For such studies we need to evaluate TG2 and TG3 antibody titres versus TG6 antibodies in CD and DH patients versus so-called Neuro-coeliac Risk Groups. Why is TG6 different from TG2 in GRD? We need better answers for our daily practice. In neuro-coeliac disease, especially in gluten ataxia, humeral immune mechanisms have been strengthened as explanation for disease. In addition, perivascular IgA deposition in the cerebellum was recognised. However, lymphocyte infiltrates in the brain of these patients suggest a direct cytotoxic effect in the neuropathogenesis in neuro-coeliac disease. Insufficient results after plasmapheresis suggest that a direct antibody-antigen reaction might not be the only pathogenic event []. Immunosuppressive treatment options might be considered in neuro-coeliac disorders to stop further damage.

In conclusion, the diagnosis and follow-up of coeliac disease is still based on biopsies. A standardized well validated approach is necessary. In neuro-coeliac disease, awareness for coeliac disease in the differential diagnosis in a neurology outpatient clinic should be improved. The introduction of TG6 or TG2 antibody testing and scientific evaluation should be encouraged. Proper scientific interest is mandatory to prevent a long delay in daily practice and to consider better treatment options than only a gluten free diet.

The lack of a reliable gold standard hinders both the clinician in daily practice, as medical research overall. Legitimate cases get overlooked, or cases get mistakenly included which leads to dispersion of data and a worst case scenario where valuable data will be overseen entirely. Leading to the diagnosis of neuro-coeliac disease being missed based on low TG6 antibody titres, and patients without neuro-coeliac disease but with high TG 6 titres to be needlessly put on a gluten-free diet. We should be careful not to throw the baby out with the bathwater. Big data are mandatory to solve this problem in the next years.