Banks Changes in Diagnosis of Non-Hodgkin ' s Lymphomas over Time Updated
P. Banks
Abstract:The question is raised as to whether changes in criteria for the diag nosis of non-Hodgkin's lymphomas, both clinical and pathological, have changed over the past four decades in sufficient scale to create a spuri ous increase in the apparent incidence of this grouping of disease. In the decade of the 1970s refinement in histomorphological criteria for the diagnosis of Hodgkin's disease resulted in as many as 10-15% of cases which previously would have been diagnosed as Hodgkin's disease being diagnosed instead as non-Hodgkin's lymphoma. Other considerations, including distinction of non-Hodgkin's lymphomas from leukemias and plasma cell myelomas, and recent recognition of angioimmunoblastic lymphadenopathy and extranodal "pseudolymphomas" as variant forms of non-Hodgkin's lymphoma, appear to have added only marginally to the total of reported cases. It is concluded that the increase in reported incidence of non-Hodgkin's lymphoma cannot be explained on the basis of changes in diagnostic criteria. Introduction One important consideration in judging the validity of re ported increase in incidence of non-Hodgkin's lymphomas per tains to the criteria used for diagnosing this grouping of disease. If criteria, clinical and pathological, have changed during the time period, then the apparent change in incidence could be spurious. Because our understanding and classification of the non-Hodgkin's lymphomas have been revolutionized by appli cations of new concepts and methods of immunology over the past two decades, a detailed review of this issue is warranted. Closely Related Malignancies In the case of some lymphoid malignancies, their designation as malignant lymphoma depends primarily on the topographic distribution of disease. Modern immunophenotyping studies have confirmed the unity of processes which can be considered either leukemia or lymphoma, according to whether the process is widely dispersed (leukemic) or localized as tumor-forming masses (lymphoma) (1). Common examples include low-grade small lymphocytic lymphoma and chronic lymphocytic leuke mia (2), and lymphoblastic lymphoma and acute lymphoblastic leukemia (3). Similarly, immunosecretory neoplasms of either low or high grade can be considered plasmacytic lymphoma or plasmacytoma (myeloma in bony tissues). The distinction is generally straightforward on the basis of clinical findings, in particular the anatomy of the disease process; however, in a small minority of cases there are overlapping features (4). Confusion has occurred when different terminologies are ap plied by various pathologists to entities in these paired groups. For example, in Europe the term "B-chronic lymphocytic leu kemia" of the Kiel classification is used for all low-grade neo plasms of small lymphocytic composition lacking immun osecretory differentiation, even when neither bone marrow nor peripheral blood is involved (5). For the purposes of this Workshop, the question to be asked is whether over the past four decades there has been any trend among clinicians, as well as pathologists, toward considering these malignancies lymphomas rather than leukemias (or plasma cell neoplasms). The answer is definitely no. Indeed, to the contrary, with advancement in our understanding of the pathobiology of these processes, pathologists, oncologists, and hematologists together have learned the importance of staging evaluation early in the evaluation of patients suffering such diseases. In current practice patients will often undergo initial peripheral blood and bone marrow analysis prior to resection or biopsy of lymph nodal or mediastinal tumor, on the basis of clinical suspicion that the process may have a leukemic (or myelomatous) component which would avoid the need for sur gery. Therefore, trends in clinical practice should be working to reduce rather than increase the reporting of this grouping as lymphoma (6). Distinction from Hodgkin's Disease Since Hodgkin's disease is the grouping of lymphomas in contradistinction from all others, i.e., the "non-Hodgkin's" lymphomas, and because it is a relatively common lymphomatous malignancy in North America and Europe, this represents an obvious source for recruitment of cases. Have the criteria for distinguishing Hodgkin's from non-Hodgkin's lymphomas changed? The answer is yes, without doubt. In the simpler, traditional classification of lymphomas which prevailed prior to 1970, some cases considered to be Hodgkin's disease would instead be classified as non-Hodgkin's lymphoma based on more modern concepts (see Fig. 1). While the presence of large Reed-Sternberg tumor giant cells was long emphasized as the essential criterion for establishing the diagnosis of Hodgkin's disease histologically, the Rappaport classification system in 1966 (7), as well as a separate study directed toward this issue itself (8), emphasized the benign-appearing tissue background cellularity surrounding such tumor cells as specific for Hodgkin's disease. This allowed inclusion of cases with large tumor giant cells in the setting of a histologically neoplastic background into the non-Hodgkin's lymphomas. More re cent studies utilizing immunophenotyping methods have iden tified variant forms of T-cell lymphomas (9) and so-called anaplastic large-cell lymphomas (10), which would earlier have been classified as Hodgkin's disease (see Fig. 2). In aggregate, these changes in pathological criteria have resulted in recruit ment of about 10-15% of the cases which would have been diagnosed as Hodgkin's disease 30 to 40 years ago into the grouping of non-Hodgkin's lymphomas today. While some small percentage of switchover from Hodgkin's disease is the result of special immunological studies available only in the last decade, most of this change was effected in the early 1970s on the basis of more precise histomorphological criteria and clin ical correlation. Therefore, whatever increase in the reporting of non-Hodgkin's lymphomas derives from earlier assignment to Hodgkin's disease is confined, for the most part, to the era from 1950 to 1970 and not in the recent decades. Newly Recognized Variant Forms of Lymphoma 1Presented at the National Cancer Institute Workshop. "The Emerging Epi demie of Non-Hodgkin's Lymphoma: Current Knowledge Regarding Etiological Factors." October 22-23. 1991. Bethesda. MD. Although the Working Formulation for classifying nonHodgkin's lymphomas has proved to be a useful and practical 5453s American Association for Cancer Research Copyright © 1992 on June 22, 2011 cancerres.aacrjournals.org Downloaded from CHANGES IN LYMPHOMA DIAGNOSIS