- © 2000 by American Society of Clinical Oncology
Grading in Marginal-Zone Lymphomas
To the Editor:
With great interest we read the recent article by Harris et al1 reporting on the results of a meeting of a clinical advisory committee convened to review the proposed World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissue and to advise pathologists on its clinical utility.
We cannot agree in full to the recommendation on marginal-zone lymphoma. Since 1992 we have studied clinical features and treatment results in primary gastric lymphoma in two consecutive prospective trials. Considering the results of our first study with a median observation time of 55 months, we would like to stress the importance of a grading in marginal-zone lymphoma by the proportion of large cells. Two retrospective studies published in the early 1990s were contradictory in the outcome in high-grade lymphoma with simultaneous low-grade components (SLGC).2,3 In our first interim analysis, we could demonstrate no significant difference between low-, high-, and high-grade lymphoma with SLGC, neither in event-free (EFS) nor in overall survival,4 but in a later evaluation, high-grade non-Hodgkin’s lymphoma (NHL) with SLGC was a significant prognostic factor in EFS (P = .0168), and we therefore stressed the importance of grading because of its role in developing treatment strategies.5 In Fig 1, the current data for EFS are shown from our first study, the results of which will be submitted for publication soon.
Fig. 1. EFS stratified by histologic subtype, stages I and II, every treatment.
Even if there are not any SLGCs in biopsies taken from the stomach, we strongly advise not to subsume these NHL as diffuse large B-cell lymphoma, because SLGCs might have been overlooked through sampling errors. This is confirmed by the analysis of relapses in high-grade NHL showing low-grade lymphoma of mucosa-associated lymphoid tissue type (data to be published). Because of these findings and the data from our study, a special terminology should be used for large-cell lymphoma of the stomach. In our opinion, these NHL need a different treatment strategy compared with nodal diffuse large B-cell lymphoma, combining radio- and chemotherapy to cure both possible components of these lymphomas.
Response
- N.L. Harris,
- E.S. Jaffe,
- H.K. Mueller-Hermelink,
- Jacques Diebold,
- A.K. Lister,
- C.A. Bloomfield and
- Peter Isaacson
- Massachusetts General Hospital BostonMA
- National Institutes of Health BethesdaMD
- University of Wurzburg WurzburgGermany
- Hopital Hotel-Dieu ParisFrance
- St. Bartholomew’s Hospital LondonUnited Kingdom
- Ohio State University Comprehensive Cancer Center ColumbusOH
- University College and Middlesex School of Medicine LondonUnited Kingdom
In Reply:
We believe that Koch et al have not correctly interpreted our recommendations concerning grading of extranodal marginal-zone/mucosa-associated lymphoid tissue (MALT) lymphoma.1 When both low-grade and high-grade lymphoma are present in the same specimen, the recommendation of the World Health Organization classification is that both diagnoses be made separately (eg, (1) diffuse large B-cell lymphoma; (2) extranodal marginal-zone B-cell lymphoma of MALT type). The pathologist is advised to comment further on the relative proportions of each component of the tumor present in the specimen. We believe that this recommendation is identical to the suggestion of Koch et al, who suggest that a simultaneous low-grade component should always be noted in the diagnosis of high-grade lymphoma of the stomach. However, the Clinical Advisory Committee report concluded that grading of extranodal marginal-zone lymphoma (low-grade lymphoma of MALT type) according to the relative proportion of large cells scattered within the low-grade component is not warranted at present, until further information is available regarding its prognostic significance. The data presented by Koch et al do not address this issue.
Additionally, we do not believe that a special terminology for de novo diffuse gastric large B-cell lymphomas (without a low-grade component) is warranted, as the site itself is sufficient to identify these lymphomas for subset analysis and further clinical and biologic studies.
