Management of High-Grade Lymphomas

Introduction

In the United States “high-grade non-Hodgkin’s lymphomas” generally refer to three histological subtypes of lymphoma categorized in the Working Formulation for Clinical Usage, published over 16 years ago.[1] The classification into low-, intermediate- or high-grade lymphomas was mainly based on differences in survival, and the three histological subtypes with the worst prognosis were defined as high-grade (immunoblastic lymphoma, lymphoblastic lymphoma, and small-noncleaved-cell lymphoma, which were associated with 5-year survival rates of 32%, 26%, and 23%, respectively). Small-noncleaved-cell lymphoma was further classified into Burkitt’s and non-Burkitt’s lymphomas.

High-Grade Non-Hodgkin’s Lymphoma

The decision to label immunoblastic lymphoma, lymphoblastic lymphoma, and small-noncleaved-cell lymphoma as high-grade has been criticized on a number of grounds.

First, the term “grade” is usually used in the context of specific histopathological criteria rather than clinical outcome.

Second, in the study that led to the Working Formulation for Clinical Usage, there was little or no attempt to take into consideration either the treatment used (although most patients received an anthracycline-containing regimen), or the prognostic factors that we know to be critically important to the outcome of therapy. Even in 1982, better results than those recorded in the study that led to the Working Formulation for Clinical Usage had been reported, at least in children, for small-noncleaved-cell and lymphoblastic lymphoma [2-6], so that the present, paradoxical situation in which the survival rates of these high-grade lymphomas are much better than the survival rates of either low-grade or intermediate-grade lymphomas were, even then, beginning to become apparent.

Third, although differences in survival among the three grades defined in the study that led to the Working Formulation for Clinical Usage were statistically significant, there was clear overlap in the survival rates of lymphomas classified as high and intermediate grades, overlap that was not always reflected accurately by the survival rates at 5 years. Clearly, the decisions as to where to draw the line between high-grade and intermediate-grade was somewhat arbitrary, as indeed, is the histological distinction between small-noncleaved-cell lymphoma of non-Burkitt’s type and large-cell lymphoma.[7] Chromosomal translocations 8;14(q24;q32) (which might be considered defining characteristics of Burkitt’s lymphoma), are also observed in subsets of Burkitt-like lymphoma and large-B-cell lymphoma, particularly in children.[8]

Identifying High-Grade Lymphomas

The problem of the definition of a high-grade lymphoma has been compounded by the identification, with the help of immunohistochemical and molecular analyses, of new entities, some of which (or some histological variants of which) would be classified as high-grade in the Working Formulation for Clinical Usage. And, finally, immunoblastic lymphoma cannot be reproducibly separated from large-B-cell lymphoma, and there is disagreement among pathologists as to whether it constitutes a separate entity.[9]

The identification of prognostic groups has been greatly refined since the early 1980s, particularly by the detailed analysis leading to the “International Prognostic Index,” and such prognostic factors work for all histologies. However, there is an increasing recognition that each entity that can be precisely defined behaves differently, and that the tendency to lump tumors together for treatment purposes may not be wise. Certainly there is no a priori reason to suppose that different biological entities will respond similarly to the same therapy–the reverse is often the case, particularly in children. Unfortunately, while histology remains the dominant means of making diagnoses, distinctions between biological entities within a histological category will remain difficult at best, and the hypothesis that genetically different tumors require different therapies for optimal results will remain untested. Thus, the use of broad risk groups that encompass multiple entities as a continuing basis for the determination of therapy is of increasingly questionable validity.

There can be little doubt that the biological nature of the tumor cell, in addition to the molecular genetic lesions that have modified normal cellular behavior during the process of lymphomagenesis, are the primary determinants of treatment outcome, although research is urgently needed to determine whether molecular genetic markers (particularly those that influence proliferation and apoptosis) provide a more effective basis for treatment decisions than histology (other factors, such as tumor burden and effective drug concentrations will, doubtless, remain of significance, too). Moreover, as novel therapies are introduced, (eg, monoclonal antibody therapy such as anti-CD20 (rituximab [Rituxan]), or therapies targeted towards specific molecular lesions or viral proteins present in the tumor cells), the identification of the presence of specific therapeutic targets is likely to become at least as important as histology as a basis for therapeutic decisions.

With these considerations in mind, this article will deal with the management of the three tumors, 1) Burkitt’s lymphoma, 2) the B-cell tumors from the Working Formulation for Clinical Usage’s category of non-Burkitt’s lymphoma (renamed in the more recent Revised European American Lymphoma (REAL) classification [10] “Burkitt-like” lymphoma), and 3) lymphoblastic lymphoma. These tumors were defined on exclusively histological grounds in the Working Formulation for Clinical Usage, but are specified as B-cell neoplasms (in the case of Burkitt’s lymphoma and Burkitt-like lymphoma), or neoplasms of precursor T or B cells in the case of lymphoblastic lymphoma, in the REAL classification.

It is important to recognize that Burkitt-like lymphoma overlaps considerably with both Burkitt’s lymphoma and large-B-cell lymphoma and indeed, the dividing lines between these three “entities” cannot be reproducibly made. It seems probable that Burkitt-like lymphoma is not truly a separate diesease, but is composed of a mixture of Burkitt’s lymphoma and large-B-cell lymphoma, the relative proportion of large-B-cell lymphoma increasing with age. However, Burkitt-like lymphoma appears, in general, to have a clinical behavior pattern and response to therapy similar to Burkitt’s lymphoma (and will be included with Burkitt's lymphoma for the purposes of this article) such that its large-B-cell lymphoma element is likely to consist predominantly of a subtype of large-B-cell lymphoma. For therapeutic purposes it may remain appropriate to separate this category from large-B-cell lymphoma in some way. However, in children, large-B-cell lymphoma has a similar prognosis to Burkitt’s lymphoma and Burkitt-like lymphoma when treated with the same chemotherapy regimens.[2-4] Whether this would apply to the large-B-cell lymphomas of adults is not known. All three of the neoplasms dealt with in this article have a peak incidence in the first 2 decades of life and are the predominant non-Hodgkin’s lymphomas of childhood and adolescence, but account for only a few percent of cases of adult non-Hodgkin’s lymphoma.