In this issue of Blood, Arora et al report substantial progress in developing an approach to exploit and condition the tumor microenvironment (TME) to enhance rituximab (RTX)-mediated killing of cancer cells.1 Extensive literature has been reported showing that RTX requires immune effector functions to kill targeted CD20-positive B cells,2 and these killing mechanisms have been demonstrated to include complement-mediated cytotoxicity, antibody-dependent cellular phagocytosis mediated by macrophages, and natural killer (NK) cell–mediated antibody-dependent cellular cytotoxicity (ADCC). Because of high tumor burdens found in B-cell lymphomas, all of these tumor-killing mechanisms can be saturated or exhausted,3-7 thus compromising RTX efficacy (see table). A key question, therefore, must focus on how to enhance or restore killing of RTX-targeted B cells by the different pathways. The present report of Arora et al extends the work of George Weiner’s group1 in this area by examining the action of CD4+ T cells with respect to their potential to increase the ability of NK cells to execute ADCC of RTX-targeted cells.
Almost 20 years ago, Bowles and Weiner reported that NK cell–mediated ADCC of RTX-opsonized B cells induces substantial downregulation of NK-cell CD16, the Fcγ receptor that engages B-cell–bound RTX and is clearly required to initiate ADCC.4 More recently, Weiner's group examined an in vitro model of RTX-mediated killing of Raji cells and found that contact between CD4+ T cells and NK cells was needed to enhance and/or restore the ADCC activity of NK cells.8 The key mediator in this reaction is interleukin 2 (IL-2), presumably secreted by T cells in contact with the NK cells. This observation has now been extended into both a novel humanized mouse model and a clinical correlative study.
In the humanized mouse model, the TME is constructed on the basis of injection of target Raji B cells accompanied by a mixture of human peripheral blood mononuclear cells. The TME is fashioned to contain human NK cells with or without an admixture of human CD4+ T cells. Fine-needle aspirates (FNAs) of tumors growing in the mice were taken before and after RTX treatment and then analyzed by flow cytometry using an array of probes. The methods and results are clearly presented, and the experiments are well controlled; there was no tumor cell elimination in the absence of RTX. NK cells were found to be essential to promote RTX-mediated elimination of targeted Raji cells, and NK cell–mediated ADCC was considerably better in the presence of the CD4+ T cells. The positive effect of the CD4+ T cells was also evident based on considerably slower tumor growth and enhanced survival curves of RTX-treated mice that had CD4+ T cells in the TME. As expected, after NK cell–mediated ADCC due to RTX, the NK-cell levels (CD56+ cells) in the TME decreased and the remaining NK cells manifested downregulation of both CD16 and CD25, indicative of an exhausted state. However, when the TME included CD4+ T cells, NK-cell levels increased, and the cells retained and expressed considerably higher levels of CD16 and CD25, indicating they were far from exhausted and could be ready to engage in additional ADCC. Thus, addition of CD4+ T cells into the TME appears to be an innovative strategy to enhance and restore ADCC of RTX opsonized target B cells by NK cells.
The provocative findings in the mouse model are reinforced by parallel clinical correlative studies that included analyses of FNA of tumors from patients with B-cell lymphoma treated with RTX and bendamustine. FNAs of patient tumors were taken before and after RTX treatment and analyzed by flow cytometry using the same cocktails of interrogating monoclonal antibodies (mAbs) as used in the mouse model studies. Projection of the mouse model studies to the clinical study suggests that higher levels of CD4+ T cells in the TME of the patients should lead to better outcomes. Indeed, in those patients with the highest levels of CD4+ T cells in the TME, there were substantial increases in expression of CD25 and CD16 in the NK cells, and elimination of B cells (based on the CD19 marker) tended to be more effective in tumors that had higher levels of CD4+ cells. The investigators infer that these positive findings are likely a consequence of the action of IL-2 secreted by the CD4+ T cells.
The implications of these results in the 2 models are considerable because they presage development of enhanced therapies that are based on RTX infusion. Simple use of infused IL-2 along with RTX may not be at all straightforward because of several problems,9 but as noted by the investigators, RTX-based strategies, including bispecific mAbs that can engage CD3 on T cells and CD19 on B cells, would appear to be among the promising possible new approaches. Selection of the subsets of patients for RTX therapy based on high levels of CD4+ cells in the TME would be reasonable.
This interesting and important report appears to have its origins based on a basic science article published almost 20 years ago.4 Successful extension and testing of this work in the clinic is a likely next step.
Conflict-of-interest disclosure: R.P.T. declares no competing financial interests.
