Several faces of refractory coeliac disease type 2

Enteropathyassociated tcell lymphoma (EATL) is a rightly feared complication of coeliac disease (CeD) often presenting as the first manifestation of CeD or a complication of preexisting refractory coeliac disease type 2 (RCD2). Cording et al in Gut have provided in exquisite detail the molecular landscape of RCD2 and the resultant EATL. It provides not so much of a road map to guide us, but rather a sketch of what has gone wrong in this severe complication of CeD. The sequence of CeD leading to RCD2 and finally to EATL provides a rare opportunity for so much confusion. Is RCD2 a chronic inflammatory disease? Most certainly yes. Is it a lymphoma? Perhaps not yet? Thanks to this paper and another recent paper we have a clearer picture of the genotoxic somatic aberrations that occur in and cause expansion and spread of these rare lineage cells. Short of the oftenlethal lymphoma, RCD2 itself produces substantial morbidity and even mortality. Intraepithelial lymphocytes (IELS) can be quite cytotoxic causing significant damage to the epithelium that has persisted despite a glutenfree diet. These aberrant IELS drive the destructive nature of the disease and its cytotoxicity. The face of RCD2 for the gastroenterologist: gastroenterologists looking after these patients are often faced with an ill patient who has very advanced malnutrition as well as a propensity to infection. This face of CeD requires the clinician to confirm preexisting CeD, then detect and treat concurrent conditions such as small intestinal bacterial overgrowth, exocrine pancreatic insufficiency, microscopic colitis, motility changes, lactose or other maldigestive problems. Vigorous management of nutritional deficiencies including protection of bone health is crucial. Often, gastroenterologists will institute an antiinflammatory approach to the disease on the basis that it has produced chronic inflammation. In the author’s experience, the early and aggressive use of topical budesonide formulated to be delivered to proximal small bowel often improves symptoms, corrects malnutrition, and in many cases suppresses inflammation. In addition to this clinical management, it is also critical to verify that the patients are glutenfree by expert dietary review and perhaps allied with testing stool or urine for gluten immunogenic peptides. The face of RCD2 for the laboratory: histopathologically, RCD2 looks no different than unhealed CeD except the IELS may permeate further down the crypts than usual. By applying straightforward immunohistochemical techniques to FFPE samples, the pathologist can identify aberrant IELS by showing that at least 50% of the IELS have lost their usual surface markers or 25% have expression of the NKp48 receptor. However, this approach is not sensitive. The next step is to look for clonal tcell receptor gene rearrangement, however, this test is quite prone to false positives. Laboratories must undertake and interpret tcell gene rearrangement studies with great care to avoid overdiagnosis. Polyclonal and oligoclonal responses are common. Monoclonal tcell gene rearrangements may also occur in patients who do not have RCD29. If fresh tissue is available, flow cytometry is more sensitive for identifying an increased percent of aberrant IELS. However, demonstrating significant numbers of clonally expanded aberrant cells is insufficient evidence for EATL or perhaps even RCD2. These cells are cytologically bland, permeating the epithelial surface and the lamina propria but lack a tumefactive appearance. Indeed, they are present in a tissue that has quite heterogeneous inflammation with many other immunocytes present. Flow cytometry, which has been used to characterise these aberrant cells, is dependent on being able to obtain enough numbers of IELS from the small endoscopic biopsies typically provided. The percentage of IELS that are aberrant in RCD2 may be as low as 20% of IELS. The isolation protocols and surface markers may differ from what a clinical immunology laboratory uses for lymphoma characterisation from tissue such as lymph nodes. This has limited the technique to those few laboratories with a special interest in RCD2. The interpretation of this flow cytometry also requires an intimate knowledge of the phenotypes of the aberrant cells and the intraepithelial lymphocytes they are derived from. Flow cytometry should be done and interpreted by labs that can handle the limitations of tiny samples presented by gut mucosal biopsies and interpreted in the light of what is known about the biology of the intraepithelial lymphocyte from which these cells arise. The Face of RCD2 to the Haematologist\Oncologist: The haematologist is faced with an often sick, symptomatic patient, often a poor performance status whose molecular markers are suspicious for lymphoma but without evidence of a mass effect or bulky disease. It is important to recognise that even if the clonally expanded aberrant cells have spread beyond the gut, the patient may not yet have aggressive EATL. The standard approach to treating patients with tcell lymphomas which often requires a very aggressive multidrug regimen is often not possible due to the poor performance status of the patient. Nor are the aberrant cells particularly responsive to standard chemotherapy. Even when a patient who has developed EATL has a generally good performance status, a targeted approach may be more fruitful especially when standard options have failed. This needs to follow an aggressive programme to reduce inflammation and its consequences. These aberrant cells are hard to kill and a near myeloablative therapy with stem cell support has been necessary to suppress the disease and rescue the patients. However, even this very aggressive regimen while initially suppressive of the disease by killing rapidly proliferating cells but not the aberrant cells that provide a reservoir for relapse, may not produce a durable response with relapse occurring in many patients. Cordinget al identify over 80% of RCD2 patients harbour somatic gain of function(GOF) mutations in the JAK/ STAT3 pathway with fully 1⁄2 having an identical p.G1097 mutation in the JAK1 kinase. It is the unhappy marriage of these GOF mutations with those in the nuclear factor kappa B (NFKB) pathways that seems to drive the emergence of frank lymphoma. These mutations and the in vitro effects of suppression by JAK/STAT pathway and proteosome inhibitors provide a rationale for using these agents to control or retard the progress of the disease. Due to the rarity of the disease and the heterogeneity of the somatic mutations, we need to tailor treatment to these molecular aberrations. Even humble steroids especially topical budesonide, which suppress symptoms and reverse malabsorption also inhibit cell Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota, USA