Introduction

The aim of a lymphoma classification is to provide a means of communication between those with a special interest in this group of diseases. The classification must be reproducible and clinically relevant, so that the results of treatment can be compared worldwide, and sufficiently flexible to allow the incorporation of new data. Finally, the classification should be histopatholog-ically based since it is the histopathologist who, almost always, makes the initial diagnosis. Traditionally, Hodgkin's disease (Hodgkin's lymphoma) and non-Hodgkin's lymphomas have been classified separately. This is a reflection of the specific identifying cell and limited morphological range of Hodgkin's disease as well as its distinctive clinical features. In comparison, the clinicopathological features of the non-Hodgkin's lymphomas are much more wide ranging and less distinct for any given entity. Not surprisingly, therefore, there have been only two classifications of Hodgkin's disease proposed since 1925 compared to more than 25 classifications of non-Hodgkin's lymphoma that have appeared in the same period.

To the early pathologists, the histological appearances of all non-Hodgkin's lymphomas were alike, consisting of replacement of the normal lymph node architecture by sheets of small or sometimes larger cells with dark-staining nuclei. It was clear, however, that not all cases behaved alike; the survival of patients with non-Hodgkin's lymphoma varied from a few months to many years. Pathologists were, therefore, under an increasing pressure from their clinical colleagues to predict the natural course of an individual case. Initial therapeutic success with Hodgkin's disease was followed by the emergence of more effective means of therapy for the non-Hodgkin's lymphomas. These therapies were not homogeneously effective in all the non-Hodgkin's lymphomas and the results clearly varied according to their histology. Consequently, clinicians began to demand much more precise and clinically relevant histological diagnoses. In 1966, in response to this, Rappaport formulated the first clinically relevant histological classification of non-Hodgkin's lymphomas. Broadly speaking, the Rappaport classification divided lymphomas into those composed of small cells and those composed of large cells. Each of these groups could be further subdivided into those with a follicular (or nodular) growth pattern and those that were diffuse. The follicular and small-celled tumors were clinically less aggressive; a better survival could, therefore, be predicted and, importantly, less potent and less toxic therapy was suitable for these cases. The converse applied to cases with a diffuse growth pattern, especially if composed of large cells. As histological techniques improved, allowing finer morphological discrimination between cells, more detailed classifications emerged. In parallel with these improvements it was becoming possible to establish the immunophenotype of lymphoma cells using immunohistochemical techniques. It soon became evident that the lymphoma cells were closely related to normal lymph node cells and that the cells of many non-Hodgkin's lymphomas recapitulated the cytology of normal lymphocytes, particularly the B cells of the follicle center. It was also clear, however, that there were an alarmingly wide variety of lymphoid neoplasms and a whole host of classifications based on these new concepts soon emerged. This caused so much confusion that a series of special international meetings were convened to decide on a single clinically relevant classification that could be used throughout the world. In the absence of any consensus, The United States National Cancer Institute convened a study to evaluate the competing classifications and the result was the compromise "Working Formulation for Clinical Use."1 It was stressed at the time that this "formulation," although based on histopathology, was a system for translation between the competing classifications and not a classification in its own right. However, it was rapidly accepted as such by patholo-gists, particularly in the United States where it became the classification of choice. The working formulation divided lymphomas into three grades based on their clinical behavior, according to their response to therapy prevalent in the late nineteen sixties and early seventies. Imprecise collective morphological terms such as "large cell" and "mixed small and large cell" were used to characterize individual entities. The result was that different clinicopathological entities were lumped together and as new entities were described, they merely became absorbed into this rather rigid system. The Working Formulation was incapable of incorporating the rapidly expanding amount of immunopheno-typic data on which pathologists were, nevertheless, increasingly relying for lymphoma diagnosis. The Working Formulation thus soon lost its main reason for existence, namely its clinical relevance.

The majority of European pathologists never accepted the Working Formulation and preferred to use the Kiel classification. This classification and its updated editions2 were based on immunophenotypic data dividing lymphomas into B- and T-cell types and, thereafter, into individual entities based principally on the similarity of their cells to normal lymphocyte variants. The lymphomas were designated low or high grade according to their cytological characteristics, rather than their predetermined clinical behavior and in keeping with established schemes for other tumors. Unlike the Working Formulation, the Kiel Classification had a sound biological basis and could easily be updated and maintain its clinical relevance. Criticisms that could be leveled at the Kiel Classification included overreliance on establishing the normal cell counterpart for each type of lymphoma, illogical oversplitting of some entities, and failure specifically to include extranodal lymphomas.

The use of different lymphoma classifications on either side of the Atlantic and the inherent defects in each contradicted the basic requirement of a classification, namely that it should provide a language for international communication, and threatened a return to the chaos of the 1970s. Moreover, new techniques and new concepts were emerging that urgently required incorporation into the principles underlying lymphoma classification. Developments in immuno-histochemistry meant that a cell lineage could confidently be assigned to most lymphomas and that many distinctive functional properties of the neo-plastic cells could be determined. Distinctive molecular genetic properties of the different disorders also began to emerge and some of these could be identified using simple immunohistochemical techniques. Another important development was the recognition that many lymphomas arose in extranodal sites and that the site of origin was often a significant clinical determinant.

0 0

Post a comment