Quantitative CD3 PET Imaging Predicts Tumor Growth Response to Anti-CTLA-4 Therapy

Abstract

Immune checkpoint inhibitors have made rapid advances, resulting in multiple Food and Drug Administration (FDA)-approved therapeutics that have markedly improved survival. However these benefits are limited to a minority sub-population that achieves a response. Predicting which patients are most likely to benefit would be valuable for individual therapy optimization. T cell markers such as CD3 represent a more direct approach than pre-treatment biopsy or genetic screening to monitoring tumor immune response, by directly examining active recruitment of T cells responsible for cancer cell-death. This approach could be especially effective as numerous different therapeutic strategies emerge, decreasing the need for drug-specific biomarkers and instead focusing on T cell infiltration, which has been previously correlated with treatment response. A CD3 positron emission tomography (PET) imaging agent targeting T cells was synthesized to test the role of such imaging as a predictive marker. The ⁸⁹Zr- p-isothiocyanatobenzyl-deferoxamine (DFO)-CD3 PET probe was assessed in a murine tumor xenograft model of colon cancer anti-cytotoxic T lymphocyte antigen-4 (CTLA-4) immunotherapy. Imaging on day 14 revealed two distinct groups of mice, stratified by PET signal intensity. Although there was no statistical difference in tumor volume on the day of imaging, in the high PET uptake group subsequent measurements revealed significantly smaller tumors than both the low-CD3 PET uptake group and untreated control. In contrast, low-CD3 PET uptake and untreated control mice demonstrated no statistical difference in size. These findings indicate that high-CD3 PET uptake in the anti-CTLA-4 treated mice is correlated with subsequent reduced tumor volume, and is a predictive biomarker of response.