Medical-Journals.com
UK
Europe
USA&Canada
Aust.&NZ
Asia
|
|
NEJM
Kanti R. Rai, M.B., B.S., and Nicholas Chiorazzi, M.D. The typical patient with chronic lymphocytic leukemia (CLL) presents to his physician at the time of a routine physical examination as an asymptomatic man who has reached the time in his life when he can enjoy the fruits of his working years. He learns from his physician that he has a somewhat enigmatic type of leukemia, the clinical course and outcome of which vary considerably from patient to patient, and therefore that his outlook is unpredictable. He may be told that some patients live with the disease for decades and die with (and not because of) the disease. He may also be told that other patients experience clinical difficulties soon after diagnosis, require frequent and often multiple forms of therapy, yet succumb to the illness within a few years. The inability to predict which clinical path this patient will follow presents a quandary for both the patient and his physician. Will he have a benign or an aggressive course? Will he require antileukemic therapy? If so, when, and what type of therapy should he receive? These uncertainties are made even more distressing because CLL remains incurable. For these reasons, the patient will probably be followed closely, without the immediate initiation of any antileukemic treatment. Clinical staging systems, introduced almost three decades ago by Rai et al.1 and Binet et al.,2 have been effective in classifying patients with CLL into broad prognostic groups. These classifications are based entirely on clinically determinable features that appear to correlate with the gross tumor burden and its effect on the normal functions of the bone marrow. These staging systems have provided a foundation on which clinicians build their management and therapeutic decisions. They have also provided an opportunity for reliable testing of new treatments in clinical trials through the exclusive enrollment of patients with similar prospects for survival. However, these systems fail to predict accurately the course of the disease in individual patients and do not take into account new discoveries about the molecular pathology of this disease. Inferences about the pathologic basis of CLL have been based on concepts proposed nearly four decades ago by Galton3 and Dameshek,4 who suggested that CLL is a homogeneous disease of long-lived, immunologically incompetent lymphocytes that turn over minimally and therefore accumulate over time. However, a succession of discoveries that began in the 1990s is calling into question several of these earlier concepts and is rapidly unraveling some of the biologic mysteries of the disease. In turn, these findings are having a clear effect on the accuracy with which an individual patient's clinical course can be predicted, as well as on the potential for improved approaches to treatment in the foreseeable future. It is now clear that CLL is heterogeneous, not only at the clinical level but also at the molecular and cellular levels. Cases can be divided into two subgroups on the basis of the presence or absence of somatic mutations in the specific immunoglobulin heavy-chain variable-region (IgVH) genes used by the leukemic cells. Mutations in these genes are somehow closely linked to the clinical course, since patients whose leukemic B cells express IgVH genes with somatic mutations fare much better than those without such mutations.5,6 The disparity in life span is very striking, with a median survival of more than 24 years and 6 to 8 years, respectively. Although a complete understanding of the link between IgVH mutations and clinical outcome still eludes us, one factor probably relates to the different chromosomal abnormalities associated with each of the two subgroups.7 The leukemic cells from patients with few or no IgVH mutations more frequently have cytogenetic changes that herald a poor clinical outcome (e.g., a 11q22�23 deletion, a 17p deletion, trisomy 12, or p53 dysfunction), whereas the cells from patients with biologically significant numbers of IgVH mutations more frequently have chromosomal changes associated with a benign clinical course (e.g., a 13q14 deletion). Several prospective clinical studies are currently under way to confirm the relation between IgVH mutational status and clinical outcome. In anticipation of the prospective confirmation of this relation, searches for rapid and technically simple ways to determine the IgVH mutational status of individual patients with CLL are proceeding. Expression of CD38, a membrane protein that marks cellular activation and maturation and that has signaling activity, often correlates with the presence of IgVH mutations.5 However, since this finding has varied among subsequent studies, CD38 is now viewed as an independent marker of the outcome,8 with clinical value of its own in foreshadowing the likely course of CLL. With the use of comparative gene-expression profiling,9 another signaling-associated molecule, the zeta-chain�associated protein 70 (ZAP-70), was recently found to be differentially expressed in the CLL subgroup without IgVH mutations and with poor outcomes. ZAP-70 is an enzyme that is normally expressed in T lymphocytes and that is critical for the activation of T cells by antigen. The surprising expression of this T-lineage gene in CLL cells, which derive from the B-cell lineage, has been confirmed by Chen et al.10 In this issue of the Journal, Crespo et al.11 provide a valuable contribution by demonstrating that the expression of ZAP-70 protein is limited to CLL cells with unmutated IgVH genes. They describe its detection by a relatively convenient and clinically available technology (multiparameter immunofluorescence flow cytometry). The sensitivity and specificity of these analyses are high, at 91 percent and 100 percent, respectively. These data agree in principle with those recently reported,10 although the use of immunofluorescence for detection permits a precise determination of ZAP-70 expression in individual cells. If, indeed, the mutational status of IgVH genes is confirmed as an accurate prognostic variable in prospective studies, then this method will enable clinical laboratories to quantify the ZAP-70+ cells in individual patients and to use this information in the clinical workup of patients with CLL. This immunofluorescence method will also allow the identification and isolation of these cells and the detailed study of their molecular characteristics. Especially relevant is the fact that Crespo et al. could readily detect ZAP-70 in leukemic cells, even when they were not separated from the other cellular components of whole blood. The potential utility of this approach is exemplified by the strong correlation of ZAP-70 expression, detected by immunofluorescence, with rapid clinical progression to more advanced stages of disease and with diminished survival times. Selective expression of ZAP-70 in patients with CLL who do not have IgVH mutations may also provide clues to the identification of factors that lead to a poor clinical outcome in these patients. In addition, it may provide important candidate targets for therapeutic advances. Protein tyrosine kinases, like ZAP-70, are enzymes that are essential for activating lymphocytes through their antigen receptors. Normally, these kinase activities are unleashed only after the receptor has engaged antigen. Once activated, they interact with other proteins in a signaling cascade that ultimately reaches the cell nucleus and induces cellular activation. ZAP-70 is essential for signaling through the T-cell antigen receptor. However, its "inappropriate" expression in CLL cells may in some way alter the action of another protein tyrosine kinase, Syk, found in B lymphocytes. Chen et al. have reported that with few exceptions, the activation of signaling proteins after the B-cell receptor binds antigen is more effective in cases of CLL with ZAP-70 expression and no IgVH mutations than it is in cases without ZAP-70 expression and with IgVH mutations.10 Others have shown signaling differences in cases of CLL categorized according to the expression of CD3812 or unmutated IgVH genes.13 These signaling differences could alter the standard activation pathway and lead to modifications in the biology of this subgroup of CLL cells. Therefore, enzymes such as ZAP-70, Syk, and other molecules important for the BCR-mediated activation pathway may become attractive therapeutic targets, as is the Bcr-Abl kinase that is now being targeted effectively in chronic myelogenous leukemia.14 We can expect that the newly recognized prognostic markers and their surrogates will soon allow physicians to offer individual patients with CLL a much more definitive projection of their clinical course than is currently possible. Moreover, it seems reasonable to hope that, in the not-too-distant future, molecular insights into the pathogenesis of CLL will begin to provide new and effective therapeutic options that are superior to today's all-too-frequent prescription of "watchful waiting."
From the North Shore�Long Island Jewish Research Institute, Manhasset, N.Y. References
|