Immunity War: A Novel Therapy for Lymphoma Using T-cell Bispecific Antibodies Free

In this issue of Clinical Cancer Research, Bacac and colleagues investigated the preclinical efficacy of a bispecific antibody in the treatment of lymphoma (1). To date, immunotherapy for the treatment of non-Hodgkin lymphoma (NHL) has been primarily focused on monoclonal antibody (mAb) therapy, which has demonstrated both single-agent activity and substantial synergy with cytotoxic chemotherapy (2). Further development of immunotherapy has been mixed. Chimeric antigen T-cell therapy has demonstrated exciting activity with significant potential toxicities, and some preliminary successes have been seen with antibody–drug conjugates. However, checkpoint blockade and other novel approaches have been complicated by disappointing activity as well as off-target side effects. In this report on the novel bispecific antibody CD20-TCB, the authors demonstrate an asymmetric design and B-cell depletion pretreatment, which may decrease off-target effects, improve the safety profile, and potentially improve the response rate. This molecule represents a new approach for the treatment of relapsed and refractory NHL.

Therapeutic antibody design in B-cell malignancies has benefited greatly from our current understanding of rituximab's mechanism of action (2). This anti-CD20 mAb appears to have a multitude of downstream effects, including cell-independent apoptosis, complement-mediated cytotoxicity, and antibody-dependent cytotoxicity. Although similar anti-CD20 molecules such as obinutuzumab and ofatumumab have shown efficacy in the treatment of various B-cell malignancies (Fig. 1), the absence of a single identifiable signaling pathway complicates the development of further therapeutic small molecules and antibodies. Blinatumomab, a bispecific antibody consisting of a pair of anti-CD19 and anti-CD3 antibody fragments with a small-molecule linker, showed promise in preclinical models and in acute lymphocytic leukemia but has not demonstrated significant activity in NHL (2). In their article, Bacac and colleagues proposed an interesting modification to prior antibody design (1). Their primary protein contains a head-to-tail configuration of anti-CD20 to anti-CD3 antibodies, similar to that seen in blinatumomab. To this design, they add yet another anti-CD20 antibody across the hinge region, connected by a modified Fc chain, resulting in the novel antibody CD20-TCB. The authors then use the preclinical surrogates of B-cell depletion and cell lysis to compare CD20-TCB with a single-armed bispecific antibody and another control antibody with a single anti-CD3 and anti-CD20 antibody on each Fab arm. Using in vitro cytotoxicity assays, the authors determine that their CD20-TCB molecule is, on average, 40-fold more potent than the other 1:1 conformations across multiple malignant B-cell lines. The choice of these controls is important, as it demonstrates that both the head-to-tail conformation anti-CD20/anti-CD3 molecule and the duplicate anti-CD20 moiety on the opposing Fab arm are critical to the potency of this novel antibody. These data are intriguing, yet it is important to note that potency and toxicity are often correlative. Although cytokine levels are evaluated as a surrogate for toxicity, the clinical picture is far more complicated than this. The relative toxicity of CD20-TCB compared with existing NHL therapies requires further study.

The authors’ modifications to the Fc chain of their antibody present an interesting aside (1). As noted above, rituximab takes advantage of a number of nonspecific immunomodulatory effects to perform its antitumor effect (3). It is known, for example, that the IgG Fc domain activates cytotoxic activity thorough the Fc-γ receptor (FcγR) on a number of immune cells including natural killer (NK) cells, macrophages, and neutrophils. In addition, this region is known to activate complement through C1q and the complement cascade. Bacac and colleagues have abrogated both of these effects through modifications in the Fc domain, which further emphasizes that their molecule's mechanism of action is primarily through the recruitment of T cells. This is confirmed by their in vitro culture assay demonstrating that tumor lysis is dependent on T-cell recruitment and activation, with a preferential activation/expansion of CD8 cell populations. The dosing, efficacy, and toxicity of CD20-TCB need to be further evaluated in phase I clinical trials to better understand the consequences of the authors’ design on clinical management.

The cytokine release syndrome (CRS) associated with activation and release of inflammatory cytokines in the setting of supraphysiologic immune stimulation is a potential side effect of this and other bispecific antibodies, as well as chimeric antigen T cells (4). The CRS reaction typically occurs within minutes to hours after infusion initiation, with the incidence of the symptoms typically corresponding to the overall lymphocyte/malignancy burden. Symptoms may vary from fever and organ toxicity to cardiovascular dysfunction and organ failure, and are primarily driven through the proinflammatory actions of IL6 (5). Classically, this has been managed through dose escalation or step-up dosing (SUD), with treatment of severe symptoms incorporating corticosteroids or the anti-IL6R antibody tocilizumab. Here, the authors propose using obinutuzumab, a type II monoclonal CD-20 antibody, to deplete B cells and reduce cytokine release in response to CD20-TCB infusion (1). Even though the mechanism is sound, and there is a demonstrable decrease in cytokine peak following this method in comparison with SUD, verification in phase I studies will be critical. This is because, as noted, obinutuzumab is itself an anti-CD20 antibody with well-documented infusion reactions (5). The authors have somewhat anticipated this critique with an in vivo analysis of proinflammatory cytokine levels in their animal models. However, no data show a clear relationship between cytokine levels and symptom severity in individual patients, especially given that malignancy is itself a proinflammatory state. Thus, although these preclinical data are promising regarding treatment-related toxicity, further clinical validation is necessary before safety of this approach can be confirmed.

The data presented by Bacac and colleagues (1) represent an exciting step forward in the treatment of NHL. These data, demonstrating preclinically the superior potency of asymmetric bispecific antibody design, argue for the evaluation of asymmetric antibodies for current bispecific targets, especially those that use CD3-driven T-cell recruitment as their primary mechanism of action. Given the proven efficacy of blinatumomab in B-cell acute lymphoblastic leukemia (ALL), it would be interesting to recapitulate the authors’ design with a 2:1 anti-CD19/anti-CD3 bispecific antibody and evaluate its efficacy in B-cell malignancies. Another future direction may be to evaluate whether a mirrored antibody design with head-to-tail linkers on both Fab regions may lead to an even greater increase in in vitro efficacy and to evaluate the relative toxicity of that approach. These and other approaches could then be replicated across our known catalog of cell-surface targets, expanding our pool of potential therapeutics significantly. The CD20-TCB designed by Bacac and colleagues offers both exciting preclinical activity and new avenues for antibody development in NHL. For patients with relapsed and refractory NHL, these developments cannot come soon enough.

C.S. Diefenbach is a consultant/advisory board member for Genentech. No potential conflicts of interest were disclosed by the other author.

Conception and design: A. Prakash, C.S. Diefenbach

Writing, review, and/or revision of the manuscript: A. Prakash, C.S. Diefenbach