Non-invasive monitoring of lymphocyte response to glucocorticoid-induced TNF receptor-related protein agonism with an anti-CD3ε PET probe

Objectives Noninvasive methods that enable real-time detection and quantification of endogenous immune cells, without the need for blood draws or invasive biopsies, have great utility in immuno-oncology.1,2 The ability to monitor immune response during the course of treatment could potentially enable early determination of treatment efficacy, inform dosing regimen, and support mechanism of action studies. The immunotherapeutic DTA-1, an agonistic antibody against the murine glucocorticoid-induced TNF receptor-related protein (GITR), is proposed to elicit anti-tumor efficacy by two mechanisms (1) receptor agonism on T effector cells resulting in the activation and infiltration of lymphocytes and (2) FcγR-mediated Treg depletion.3,4 The aim of this study was to non-invasively image tumor infiltration of CD3+ lymphocyte response to treatment with murinized DTA-1 (mDTA-1).

Methods The 2C11 anti-mouse CD3ε antibody underwent pepsin digestion to generate a F(ab’)2 fragment that was modified with NOTA-NCS via non-specific lysine conjugation and radiolabeled with 64Cu. We’ve previously shown the resulting 64Cu-NOTA-αmuCD3ε probe retains antigen specificity in vitro and in vivo.5 Balb/c mice were inoculated with 3 x 105 CT26 cells mid flank and on day 14 subjects were randomized by tumor volume and received a treatment of either 300 µg of mDTA-1, or PBS. Two and nine days post treatment subjects were injected i.v. with 100 - 200 µCi of 64Cu-NOTA-αmuCD3ε and PET/CT images acquired 24 h post injection. A subset from each group was sacrificed post imaging and tumors excised for immunohistochemistry (IHC) and flow cytometry analysis. Subjects were kept on study for a total of 42 days and tumor measurements taken twice a week to monitor treatment efficacy.

Results High intra-tumoral uptake of 64Cu-NOTA-αmuCD3ε probe was observed in both the PBS control (SUVmax 1.69 ± 0.46) and mDTA-1 treated subjects (SUVmax 1.76 ± 0.65) at 17 days post innoculation; highlighting the probe’s ability to visualize CD3+ tumor infiltrating lymphocytes (TILs). At day 24, a significant increase in intra-tumoral probe uptake was observed in both groups; with SUVmax values of 3.21 ± 0.69 and 3.67 ± 1.18 for PBS and DTA-1 treated groups respectively. IHC and flow cytometry confirmed the presence of CD3+ lymphocytes in excised tumors; however no difference in CD3 expression was observed between day 17 and 24. Additionally, PET imaging did not detect an increase in TILs in response to DTA-1 treatment, an observation that was corroborated by flow cytometry and IHC.

Conclusions In the current study, tumor infiltrating lymphocytes were successfully imaged using a probe against the T cell marker CD3. Treatment with the GITR agonist antibody mDTA-1 inhibited tumor growth; however, PET imaging did not show an increase in tumoral CD3+, and this observation was corroborated by parallel IHC and flow cytometry endpoints. The 50 % increase in 64Cu-NOTA-αmuCD3ε tumor accumulation observed between days 17 and 24 in both the DTA-1 treated and control groups was likely dominated by nonspecific uptake resulting from ruptured vasculature and the enhanced permeability and retention effects. Overall, mDTA-1 treatment does not appear to increase CD3+ lymphocyte infiltration. Given the robust anti-tumor efficacy of mDTA-1 therapy in mice, CD3 imaging may not a viable biomarker for predicting response in patients treated with GITR agonist antibodies.