Targeted immunotherapy in Hodgkin lymphoma

In this issue of Blood, Rothe et al introduce a new principle of targeted Hodgkin lymphoma (HL) immunotherapy in their report from a phase 1 study of the bispecific anti-CD30/CD16A antibody construct AFM13.¹ 

It has been 15 to 20 years since Küppers et al²  demonstrated that the Hodgkin Reed-Sternberg (RS) cell is in fact the clonal cell of B-cell origin from which HL derives. However, the RS cell is a very atypical B cell that has lost most of the usual B-cell identity and gained some important other functions. The clonal cell makes up <1% of the tumor mass, but through effective organization of the much more numerous surrounding immune cells, it is able to generate a highly aggressive and potentially lethal malignancy. Also, through cytokine and chemokine activity, as well as through more direct interaction with the microenvironment, the RS cell exerts a number of mechanisms that enable the tumor to escape the normal immune system of the host.³  One example of this is the absence of specific cytotoxic T cells and natural killer (NK) cells in the cellular landscape of HL.⁴ 

Assuming that the ability of HL to escape the normal immune response is central to tumor survival and growth, modulating the immune system and its response to the disease is a logical novel therapeutic strategy. Different immunological approaches are currently the focus of clinical development in HL: (1) so-called checkpoint inhibition, eg, the promising activity of monoclonal antibodies targeting programmed cell death protein 1 (nivolumab⁵  and pembrolizumab⁶ ); (2) modulation of the immune status and tumor environment, eg, lenalidomide⁷ ; or (3) the direct engagement of cytotoxic immune effector cells, such as cytotoxic T or NK cells, to mediate tumor cell lysis. The most prominent example of the last is the engineering of T cells with chimeric antigen receptors (CAR-T cells). Another example is the bispecific antibody AFM13, which targets the CD30 receptor present on the surface of RS cells and via the CD16a antibody selectively attracts and activates NK cells.

Using the CD30 receptor as a target for HL therapy is not at new concept. Stein et al⁸  demonstrated this antigen on the surface of RS cells >30 years ago. Because it is almost consistently expressed on RS cells of classical HL and at the same time relatively specific to those cells, it is an attractive therapeutic target. However, naked anti-CD30 antibodies have shown little or no antitumor activity.⁹  Brentuximab vedotin is approved for the treatment of relapsed and refractory HL. This drug is an antibody-drug conjugate that combines a cytotoxic agent (monomethyl auristatin E) with an anti-CD30 antibody through a cleavable linker molecule.¹⁰  Thus, it uses the CD30 antigen not to modulate the function of the intracellular domain of CD30 but to achieve high-concentration delivery of chemotherapy within and around the RS cells. In the same way, AFM13 uses the CD30 receptor as a geographical target without exerting any clear influence on the transmembrane protein itself.

The study by Rothe et al is a phase 1 study and as such is not primarily designed to assess efficacy.¹  The drug seems generally well tolerated, and the maximal tolerable dose was not reached. Although AFM13 treatment resulted in a significant NK-cell activation and decrease of soluble CD30 in peripheral blood, the demonstrated activity is not very impressive at first, with an overall response rate of 11.5% (all partial response [PR]) and an overall disease control rate (PR + stable disease [SD]) of 61.5% in a very heavily pretreated cohort of patients. However, because this was a first-in-human experience, a proportion of the patients were treated at extremely low doses (starting at 700 times less than the final dose level of 7 mg/kg). The activity was notably higher in the higher dose levels of >1.5 mg/kg (23% PR and 77% PR + SD) even though toxicity remained stable during dose escalation. The majority of patients only received 1 treatment cycle (4 weekly infusions), and no patient received >2 cycles. Furthermore, the weekly dosing schedule is hardly optimal for this construct, with a half-life of 19 hours. A subsequent phase 2 study will take these considerations into account and shed more light on the actual efficacy of this promising compound and targeted immunotherapy in general.