Cell-mediated immunity in chronic lymphocytic leukemia

Although there has been a rapid increase in knowledge of the biology of the malignant B-lymphocyte in chronic lymphocytic leukemia (CLL), we still lack a clear understanding of the role of non-malignant lymphocytes in the pathogenesis of CLL. The first clinically significant immune defect in CLL is usually hypogammaglobulinemia, but qualitative and quantitative defects in T and NK cells in CLL patients have been recognized for decades [1], can occur early in the course of CLL, and have even been reported in patients with monoclonal B cell lymphocytosis [2]. However, the investigation of the clinical implications of these changes has only recently begun.

Patients with CLL have early and profound changes in blood T and NK cell counts. Absolute counts of both T and NK cell counts are increased. Among T cells, CD8+ cells are increased more than CD4+ cells resulting in a decrease in the CD4:CD8 ratio [3]. The absolute counts of regulatory T cells (T_reg) are also significantly increased [4]. While similar changes do occur with aging of the immune system, the magnitude of the changes in patients with CLL are clearly pathological.

The mechanisms of these changes in T and NK cells are not well understood. CLL cells interact directly with non-malignant lymphocytes, can act as antigen presenting cells, and secrete cytokines that could alter T and NK cell numbers, subtype ratios and function. In addition, the impaired humoral and cell-mediated immunity in patients with CLL increases the overall risk of infection and especially the risk of reactivation of latent cytomegalovirus (CMV) infection. CMV reactivation increases the levels of cytotoxic CD8+ and CD4+ T cells with a considerable number of these cells being CMV antigen specific [3,5]. However, there is no direct evidence of other infections altering T cell counts [5]. There is still no convincing data that the changes in T and NK cells counts seen in CLL comprise an effective immune response against tumor cells [3,5]. In contrast, the increased numbers of functionally normal T_reg could potentially inhibit any anti-tumor cell immune effect [4,6]. The available data thus show that non-malignant lymphocyte biology in patients with CLL is markedly disordered. The consequent immunodeficiency increases the risk of infection and possibly second malignancies, and is unlikely to provide effective anti-tumor immunity.

Because of the potential importance of abnormalities in non-malignant lymphocytes in patients with CLL, there could be a clinical value in quantifying these defects and correlating them with a clinical outcome. In this issue of Leukemia and Lymphoma, Gonzalez et al. report the T and NK cell counts in a cohort of patients with CLL and examine the relationship between lymphocyte subtype ratios and survival [7]. Their study shows that the ratios of CD4+ T cells, CD8+ T cells, and NK cells compared to CLL cells are significantly increased in earlier stage CLL and that higher CD4+ and CD8+:CLL ratios are independent predictors of survival on multivariate analysis. However, only the absolute CD8+ T cell count (but not the absolute CD4+ T cell count) was associated with better survival. The strengths of this study include the representative age at diagnosis, cohort size, duration of follow up, and the correlation the measured parameters with survival rather than time to first treatment. In addition, this study is a confirmation and extension of previously published work [8]. The major concern about this study is that the non-malignant lymphocyte:CLL cell ratio was more predictive of survival than the absolute lymphocyte counts. This ratio is obviously affected by the size of the circulating population of clonal CLL cells which usually increases with disease progression. The authors did not find that the absolute lymphocyte count was an independent predictive factor for survival on multivariate analysis. However, this does not exclude the possibility that the lower ratios of T to CLL cell counts that predicted a poorer survival were largely influenced by the effect of an increased CLL tumor burden on the size of the denominator of this ratio. An additional concern is the absence of data on the number of T_reg. The results of the multivariate analysis also have to be interpreted with caution because of the lack of inclusion of data on the most important biological prognostic factors (genetic defects detected by fluorescence in situ hybridization (FISH) and immunoglobulin variable heavy-chain (IgHV) mutation status) which could have altered the results. Despite these limitations, this study does provide additional data suggesting that studies of cell-mediated immunity in CLL could provide biologically and clinically important data.

Quantitative analysis of nonmalignant lymphocytes in patients with CLL could provide prognostic data with patients with a less damaged immune system having a better prognosis. However, we do not know whether this means that a less defective immune system can better control CLL and its complications or that less aggressive CLL results in less damage to the immune system. Distinguishing between these two possibilities could have important clinical consequences and could be especially important in evaluating the long-term effects of the currently used chemoimmunotherapy regimens that are highly toxic to all lymphocytes. Additional investigation of the role of cellular immunity in CLL will thus be important for improving many aspects of the treatment of patients with this disease.