It is altogether appropriate that the gut should possess efficient immunological defences given the interface between its antigen-rich contents and the permeable mucosa. More than that, however, the gut is an important immunological organ in its own right. In this respect, the gut mucosa contains more lymphocytes than all other organs of the immune system combined. It is estimated that in humans a metre of intestine contains 10¹⁰ lymphocytes, which is almost equivalent to the number contained in the spleen.1 In the light of this quantitative difference and the fact that intestinal lymphoid tissues are subject to constant immunostimulation, it might be expected that a larger proportion of lymphomas would arise from the intestine than the peripheral lymphoid system but, curiously, the contrary is true. Although the gastrointestinal tract is the commonest site of extra-nodal lymphoma in Western countries, only 10% of lymphomas overall arise there.2 There are, however, significant differences in the types of lymphoma that arise in the gut compared with the peripheral lymphoid tissues. Thus, both Hodgkin’s disease and follicular lymphoma, among the commonest tumours of peripheral lymph nodes, only rarely arise in the gut. Most gastrointestinal lymphomas are gastric B-cell lymphomas of mucosa associated lymphoid tissue (MALT) type that arise from non-native lymphoid tissue acquired by the gastric mucosa usually following Helicobacter pyloriinfection.3 A smaller proportion of B-cell lymphomas arises in the intestine, presumably from native MALT. T-cell lymphomas arising in the intestine, although frequently featured in single case reports, are uncommon4 but include at least one specific subtype, namely enteropathy associated T-cell lymphoma, which occurs as a complication of coeliac disease.5 In this issue (see page 662), Carbonnel and colleagues describe four patients with low grade T-cell lymphoma characterised by diffuse infiltration of the intestinal mucosa and suggest that these cases comprise a new distinctive clinicopathological entity.

In formulating the revised European American lymphoma (REAL) classification,6 which is in the process of evolving into the new World Heath Organization classification, the International Lymphoma Study Group took the view that the classification should consist as far as possible of a list of real disease entities. The criteria by which an “entity” was defined were morphology, phenotype and genotype, normal cell counterpart and clinical features including site of origin. Although all these criteria were taken into account in defining entities, the emphasis was not necessarily the same for each. Some lymphoma entities are consistently homogeneous by all the criteria whereas others although morphologically homogeneous, might not display a reproducible phenotype and not conform to any known normal cell counterpart but be characterised by distinctive clinical features. It was hoped that defining distinctive diseases in this way would lead to greater insight into lymphoma biology and to the eventual development of properly tailored therapies. Although this system works admirably for B-cell lymphomas, it has proved especially difficult to collect the T-cell lymphomas into distinctive entities. With a few exceptions, these relatively rare tumours tend to be morphologically and phenotypically heterogeneous and their genetic features are poorly characterised. Establishing a normal cell counterpart often proves impossible and, equally, their clinical features tend to vary from case to case. Not surprisingly, most T-cell lymphomas, whether nodal or extranodal, including those arising in the intestine, end up being classified under the category “T-cell lymphoma unspecified” in the REAL classification.

Enteropathy associated T-cell lymphoma is one of the exceptions and is the only intestinal T-cell lymphoma that is recognised as a distinct entity in the REAL and WHO classifications. Although morphologically heterogeneous, the lymphoma demonstrates a relation to normal intraepithelial T-lymphocytes (IEL) by virtue of its phenotype, which is similar but not identical to that of the major IEL population, and the tendency of the neoplastic cells to accumulate widely within the intestinal epithelium.7 No distinctive genotypic features have been described. The most important defining feature of enteropathy associated T-cell lymphoma is its association with coeliac disease. Have Carbonnel et alnow described a second distinctive intestinal T-cell lymphoma? Morphologically, their cases seem to be remarkably homogeneous, their low-grade cytology being unusual for T-cell lymphomas which are more often comprised of pleomorphic medium-sized and large cells. The wide distribution of the tumour cells throughout the lamina propria of the intestinal mucosa is another unusual characteristic. In common with enteropathy associated T-cell lymphoma, this suggests a specific association with native gut lymphoid tissue, in this case the lamina propria rather than the intraepithelial T cells. The phenotype of the lymphomas, although similar in all cases, is not distinctive but would be in keeping with origin from normal CD4+ positive lamina propria T cells. Although genotypic abnormalities are described in only two cases, in which they were notably different, studies of more cases may yet reveal homogeneous changes. Finally, the distinctive slow relentless course and prolonged survival manifested by all four patients is unusual for T-cell lymphomas, which are usually clinically aggressive, and suggests a degree of clinical homogeneity.

According to the principles of the REAL and WHO classifications Carbonnel et al would seem to have described a subtype of intestinal T-cell lymphoma that constitutes a reproducible “entity”. Their findings will need to be confirmed by further reports of similar cases which, on past experience, will very likely be forthcoming. Any advance that helps to rationalise the heterogeneous assortment of T-cell neoplasms is to be welcomed.