Breast cancer impacts 2.3 million women annually and in 2022, an estimated 670,000 women died from breast cancer worldwide [1]. There is a significant unmet need for novel therapies to enhance the current standard of care treatments. Immune therapies have demonstrated success in many solid tumors and has potential to improve breast cancer therapeutics. CD3 bispecific molecules are a new class of targeted immune-therapeutics developed to redirect T cells to the tumor cells expressing specific cell surface antigens like B7H4 [2]. CD8+ cytotoxic T cells are essential effector cells of the immune system, and detecting their presence in the tumor microenvironment following CD3 bispecific treatment indicates a potential anti-tumoral response [3]. Here we present CD8–PET (positron emission tomography) as a biomarker to evaluate efficacy of B7H4-CD3 bispecific T cell engager (PF-07260437) in a humanized mouse model for breast cancer.
NSG mice carrying MDA-MB-468 breast cancer xenograft tumors expressing B7H4 were inoculated with human PBMCs 22 days post tumor implantation. This was demarcated as Day -6 relative to treatment initiation. B7H4-CD3 bispecific antibody or isotype control were administered SC at high (0.5 mg/kg), medium (0.05 mg/kg), and low (0.02 mg/kg) doses on Days 0 and 7.  To capture the kinetics of the pharmacodynamic (PD) effects of B7H4-CD3 treatment, T cell infiltration was assessed on Days 5 and 9 across two PET imaging study cohorts (designated A & B respectively). Cohort A mice received a single dose of bispecific antibody treatment, whereas cohort B mice were administered with 2 doses of bispecific antibody treatment. In both cohorts, the 89Zr-radiolabled CD8 minibody radiotracer was administrated 22 hours prior to image acquisition. PET imaging and ex vivo gamma counter data indicated a dose dependent response where average tumor uptake reached 6.89, 5.93, and 5.19 %ID/g in the high, medium, and low dose groups, respectively, in cohort A by gamma counting quantitation (figure 1). In cohort B, following the same escalating dosing regimen, tumor uptake reached 9.09, 7.41, and 6.15 % ID/g. In contrast, tumor uptake in the Isotype control treated tumors were at 5.13 and 5.85 % ID/g respectively for Cohort A and B. Furthermore, IHC staining from cohort B significantly correlated with PET imaging results, corroborating that higher CD8 T cell density (399 CD8α+ cells /mm2) was achieved in bispecific treated tumors compared to the isotype control (4.99 CD8α+ cells/mm2, p value =0.0001). 
Both PET imaging and IHC data demonstrate a clear dose-dependent response of B7H4-CD3 bi-specific antibody treatment. Further, comparing the two timepoints (Cohort A vs. Cohort B), intratumoral CD8 T cell infiltration was greater in cohort B, which is consistent with the expected course of immune response following two doses of the T cell engager treatment [4]. These results showcase 89Zr-radiolabled CD8 minibody as a PET biomarker to evaluate efficacy of PF-07260437 a B7H4-CD3 bispecific in a Breast cancer model. These data suggest that the evaluation of this biomarker in patients after PF-07260437 therapy could be valuable in assessing efficacy.  This study also highlights the merit of non-invasive CD8-PET as a biomarker and screening tool for discovery and development of novel bi-specific cell engagers and other novel IO therapies.
All procedures performed on animals were in accordance with regulations and established guidelines and were reviewed and approved by an Institutional Animal Care and Use Committee or through an ethical review process.

Image/Figure:
Click to View

Image/Figure Caption:

Figure 1.  89Zr-CD8-Mb PET imaging in breast cancer tumor xenografts as a biomarker following B7H4-CD3 bispecific antibody treatment. (a) MIP coronal PET/CT images scaled the same of MDA-MB-468 breast cancer tumors in NSG mice treated with PF-07260437 and isotype control. Cohort A (top) and B (bottom), with yellow circles indicating the tumor in the right shoulder region. (b) Bispecific treatment and PET/CT imaging time course of PBMC humanized NSG mice carrying  MDA-MB-468 tumors for both cohort A and B; where cohort A is subjected to PET/CT imaging following a single dose of PF-07260437 and cohort B is subjected to PET/CT imaging following 2 doses of PF-07260437. (c) Tumor ex-vivo gamma counting data of both cohort A and B post treatment on day 5 and 9.  %ID/cm3 or /g: % injected dose per cm3 or g of tumor tissue. Statistical analysis: Tukey’s multiple comparisons test; *, p value 0.0229; ***, p value <0.0001.
 

Author