CD19-CD28 (RO7443904) Combination with Glofitamab Enhances T-Cell Proliferation and Effector Function in Patients with Relapsed/Refractory Non-Hodgkin Lymphoma (R/R NHL)

Background: RO7443904 is a CD19-targeted bispecific that, in the presence of a T-cell receptor signal, co-stimulates T cells via CD28 agonism, thereby boosting their activation, proliferation, effector functions, and potentially reversing T-cell exhaustion. A Phase I dose-escalation study (NCT05219513) recently investigated the safety, efficacy, pharmacokinetics, and pharmacodynamics (PD) of RO7443904 combined with glofitamab in patients with R/R NHL. Preliminary clinical data shows promising clinical activity and good tolerability in R/R NHL (Dickinson et al, submitted to ASH 2024). Here, we report peripheral blood (PB) and tissue biomarker analyses to demonstrate mechanisms of action (MoA), dose relationship, and features associated with response.

Methods: Exploratory biomarker analyses included data from 28 patients with indolent (iNHL, n=9) or aggressive NHL (aNHL, n=19) treated with escalating dose levels of 0.15–14 mg RO7443904, starting after completion of glofitamab step-up dose, on Cycle 2 Day 8 (C2D8), and followed by combination with the fixed target dose of glofitamab (30mg), once every 3 weeks, from C3D1 for up to 12 cycles. Immune profiling of PB by flow cytometry and plasma cytokines analysis by ELLA were performed. Baseline tumor biopsies (n=21) were analyzed by immunohistochemistry/immunofluorescence, bulk mRNA- and targeted DNA-sequencing. To evaluate individual contributions, on-treatment PD changes during the first five cycles of treatment in patients with aNHL were compared to historical glofitamab monotherapy PD data generated in a phase I/II study (NCT03075696).

Results: Compared to glofitamab monotherapy, and in line with the constitutive expression of CD28 on CD4+ and CD8+ T cells, the combination significantly (p<0.01) boosted and sustained CD4+ (1.7-fold) and CD8+ (2.5-fold) T-cell proliferation (Ki67+) in PB, and preferentially increased the differentiation of naïve CD8+ to effector memory (T_EM) T cells by 1.5-fold. These effects are similar to what is observed after second-generation chimeric antigen receptor (CAR) T-cell therapies. RO7443904 in combination with glofitamab induced PD changes in PB at all tested doses. Specifically, in a bell-shaped dose-dependent manner, RO7443904 limited the PB expansion of potentially exhausted PD1+ CD8+ T_EM cells, which were previously found to expand in a dose-dependent manner following glofitamab monotherapy (Broeske et al, Blood Advances 2022). The optimal PD effects were observed within dose ranges that corresponded to efficacious RO7443904 exposures predicted by pre-clinical humanized mouse models (Sam et al, Blood 2024). Furthermore, within the doses tested, we detected no significant additional cytokine release following RO7443904 alone or in combination with glofitamab, aligning with the clinically observed safety profile.

Complete responses observed with the combination were independent of recurrent mutations identified in pre-treatment biopsies such as KMT2D, TP53, and CREBBP, or cell-of-origin. These responses were not associated with CD19+ B-cell abundance but correlated with a low tumor proliferation index, low infiltration of granulocytic myeloid-derived suppressor cells, and high total and CD28+ T-cell infiltration in pre-treatment biopsies. These trends were observed in aNHL but were more prominent in iNHL.

Conclusions: In this study, we have demonstrated for the first time the MoA of a CD28 costimulatory bispecific (RO7443904) in combination with a T-cell engaging bispecific (glofitamab) in humans. The addition of RO7443904 to glofitamab monotherapy significantly boosted and sustained the proliferation and effector function of T cells. This enhanced response appears to benefit patients with a less immunosuppressive tumor microenvironment and more active T-cell infiltration. Overall, our PD and biomarker findings support the rationale for this combination as an effective off-the-shelf therapy for patients with R/R NHL.

Acknowledgments: This study was sponsored by F. Hoffmann-La Roche Ltd. The authors would like to thank the patients who participated in this trial, their families, and their caregivers. Special thanks also to Iva Lelios, Abu Ali, Kat Reyskens, Isabelle Vogl, Katharina Menzel, Francesca Michielin, and Giuseppe Palladino for their contributions to this work.