Extranodal lymphoma: a reappraisal.

Approximately one-third of non-Hodgkin lymphomas (NHLs) arise primarily from sites other than lymph nodes, spleen or the bone marrow and even from sites which normally contain no native lymphoid tissue [1, 2]. Indeed, extranodal lymphomas can arise in almost every organ [1, 2]. However most of the presentations appears to be clustered in a few sites: skin, stomach, brain, small intestine and—when included in the reports—the Waldeyer’s ring [3, 4]. The extranodal lymphomas represent a challenge in routine lymphoma diagnosis, due to the variety of histological types, molecular abnormalities and clinical pictures that can be present. Correct diagnosis and appropriate treatment of extranodal lymphoma are also complicated by the relative rarity of many of these tumours. Moreover, in comparison with nodal presentation, Band T-cell lymphomas diagnosed at extranodal sites may have quite different outcomes and may frequently require different therapeutic approaches due to specific organ-related problems. The definition of primary extranodal lymphoma is controversial, particularly in the presence of both nodal and extranodal disease. Strict criteria were first proposed in 1961 by Dawson, who defined primary gastric lymphoma as presentation with main disease manifestation in stomach, with or without involvement of regional lymph nodes. Later these criteria were extended to allow for contiguous involvement of other organs (e.g. liver, spleen) and for distant nodal disease providing that the extranodal lesion was the presenting site and, after routine staging procedures, constituted the predominant disease bulk, to which primary treatment must be directed [3]. The inclusion among primary extranodal lymphomas of cases presenting with stage III and IV is also questionable and several authors consider only stage I and II presentation as primary extranodal disease. Since many extranodal lymphomas have the potential to disseminate, this approach may, however, lead to an imperfect representation. On the other hand, extranodal involvement in a disseminated disease may represent a secondary spread. Any chosen definition inevitably introduces a selection bias; in a recent Dutch study the frequency of extranodal NHL fluctuated from 20% to 34% depending on the definition criteria adopted [5]. The incidence of extranodal NHL in Western countries has increased substantially in the last 40 years [4, 6]. This may in part be due to improved diagnostic procedures (particularly in brain and gastrointestinal lymphomas) and changes in classification systems, but much of the change is real and the AIDS epidemic in the 1980s does not completely explain this rise [4]. The aetiology of extranodal lymphomas appears to be multifactorial and includes immune suppression, infections, both viral and bacterial, and exposure to pesticides and other environmental agents. Nevertheless, despite the considerable progress made in the understanding of MALT lymphoma and its relationship to bacterial infections [7], the precise cause of most lymphoid neoplasms remains unknown [1–3]. The proportion of NHL presenting at extranodal sites comprises 25–50% of new lymphoma cases with important geographic variations that may be explained by the variability of the reporting criteria (variable definitions of primary extranodal disease) as well as by different types of data source (referral cancer centres versus tumour registry). True geographic differences are, however, present, for example, the incidence of Epstein–Barr virus and human T-cell lymphotropic virus 1-associated T-cell lymphomas is higher in Asia than in Europe and North America. The histological spectrum of extranodal lymphomas somehow differs from that of nodal lymphomas. Nearly half of the extranodal cases are of diffuse large cell histology. Aggressive subtypes, mainly diffuse large B-cell lymphomas (DLBCLs) are predominant in NHL of central nervous system (CNS), testis, bone and liver while in the gastrointestinal tract a large spectrum of histological disease entities can be seen, comprising DLBCL, MALT lymphoma (including the immunoproliferative small intestinal disease), Burkitt’s lymphoma, enteropathy-associated T-cell lymphoma, mantle cell lymphoma and follicular lymphoma. Although extranodal lymphomas are not rare, the frequency of involvement of any particular site is not high enough for a single institution to answer the major question about their natural history and proper therapy. Attempting to overcome these difficulties, in the late 1990s, the International Extranodal Lymphoma Study Group (IELSG) was created to provide an adequate network to study the extranodal lymphomas. This international collaboration has originated a number of retrospective and prospective trials aimed to clarify the management issues distinct to extranodal presentations (http://www.ielsg.org). Whether or not extranodal lymphomas as a whole have a worse overall survival in comparison with that of nodal cases The author reports no relationships with companies whose products or services are mentioned in this manuscript.