Anti-CD27 Enhances Lymphoma Immunotherapy through Profound Myeloid Cell Recruitment

Direct-targeting monoclonal antibodies (mAb) such as anti-CD20 mAb are thought to elicit their anti-tumor function through antibody-dependent cellular phagocytosis (ADCP) mediated by myeloid cells (monocytes and macrophages), with little involvement of T cells. In contrast, immunomodulatory mAbs to TNFR superfamily members, CD27, OX40 and CD137, function by augmenting T-cell responses.

We examined the therapeutic potential of combining anti-CD20 mAb with a panel of immunomodulatory mAbs (OX40, CD137, CD27, TIGIT, GITR, CTLA4, PD-1). In the syngeneic BCL₁ B-cell lymphoma mouse model only an agonistic mAb to CD27, provided a synergistic effect when combined with anti-CD20. Anti-CD20 and anti-CD27 mAb individually provided modest therapeutic benefit (median survival 33 days and 62 days, respectively), but mice treated with the combination survived beyond 100 days. Similar synergistic survival benefit was observed in another B-cell lymphoma model, A31, and in BCL₁-bearing human CD27 transgenic mice, when anti-CD20 was combined with varlilumab, an anti-human CD27 mAb currently under clinical investigation.

We observed that in mice treated with anti-CD27, there was an early and substantial increase in intra-tumoral monocyte, neutrophil and macrophage infiltration. CD27 is expressed constitutively on T and NK cells but not myeloid cells or the tumor itself. To investigate whether CD27 agonism promotes intra-tumoral myeloid cell infiltration through T cells, we depleted T cells in the BCL₁model. Surprisingly, CD4 or CD8 T-cell depletion had no effect on the survival of anti-CD20 and anti-CD27-treated mice, suggesting that the remaining CD27+ immune effector cells, NK cells, are required. To further probe the relative importance of these two sub-sets, we performed experiments in γ chain knockout mice, where activatory FcγR are not expressed. Here, anti-CD27 mediated T-cell activation can still occur via crosslinking from the inhibitory FcγRII, but effector function through NK cells, mediated through activatory FcγR, is abrogated. In this model, the therapeutic benefit of anti-CD27 was completely abolished, thereby supporting the requirement for NK cells.

We hypothesize that anti-CD27 stimulates CD27+ NK cells to release chemokines that draw myeloid cells into the tumor, where they subsequently perform augmented anti-CD20 mediated ADCP.

These data demonstrate the clear therapeutic potential of combining direct targeting and immunomodulatory mAb but that the therapeutic mechanism of action may differ to that expected; here involving a previously unheralded effect of anti-CD27 on myeloid infiltration. Based upon these data, we have implemented a phase II clinical trial examining rituximab and varlilumab in B-cell lymphoma, which will commence recruitment shortly.