Visualization and quantification of in vivo homing kinetics of myeloid-derived suppressor cells in primary and metastatic cancer.

Myeloid-derived suppressor cells (MDSCs) are immunosuppressive cells of the myeloid compartment and major players in the tumor microenvironment (TME). With increasing numbers of studies describing MDSC involvement in cancer immune escape, cancer metastasis and the dampening of immunotherapy responses, MDSCs are of high interest in current cancer therapy research. Although heavily investigated in the last decades, the in vivo migration dynamics of MDSC subpopulations in tumor- or metastases-bearing mice have not yet been studied extensively. Therefore, we have modified our previously reported intracellular cell labeling method and applied it to in vitro generated MDSCs for the quantitative in vivo monitoring of MDSC migration in primary and metastatic cancer. MDSC migration to primary cancers was further correlated to the frequency of endogenous MDSCs. Methods: Utilizing a 64Cu-labeled 1,4,7-triazacyclononane-triacetic acid (NOTA)-modified CD11b-specific monoclonal antibody (mAb) (clone M1/70), we were able to label in vitro generated polymorphonuclear (PMN-) and monocytic (M-) MDSCs for positron emission tomography (PET) imaging. Radiolabeled PMN- and M-MDSCs ([64Cu]PMN-MDSCs and [64Cu]M-MDSCs, respectively) were then adoptively transferred into primary and metastatic MMTV-PyMT-derived (PyMT-) breast cancer- and B16F10 melanoma-bearing experimental animals, and static PET and anatomical magnetic resonance (MR) images were acquired 3, 24 and 48 h post cell injection. Results: The internalization of the [64Cu]NOTA-mAb-CD11b-complex was completed within 3 h, providing moderately stable radiolabeling with little detrimental effect on cell viability and function as determined by Annexin-V staining and T cell suppression in flow cytometric assays. Further, we could non-invasively and quantitatively monitor the migration and tumor homing of both [64Cu]NOTA-αCD11b-mAb-labeled PMN- and M-MDSCs in mouse models of primary and metastatic breast cancer and melanoma by PET. We were able to visualize and quantify an increased migration of adoptively transferred [64Cu]M-MDSCs than [64Cu]PMN-MDSCs to primary breast cancer lesions. The frequency of endogenous MDSCs in the PyMT breast cancer and B16F10 melanoma model correlated to the uptake values of adoptively transferred MDSCs with higher frequencies of PMN- and M-MDSCs in the more aggressive B16F10 melanoma tumors. Moreover, aggressively growing melanomas and melanoma-metastatic lesions recruited higher percentages of both [64Cu]PMN- and [64Cu]M-MDSCs than primary and metastatic breast cancer lesions as early as 24 h post adoptive MDSC transfer, indicating an overall stronger recruitment of cancer-promoting immunosuppressive MDSCs. Conclusion: Targeting of the cell surface integrin CD11b with a radioactive mAb is feasible for labeling of murine MDSCs for PET imaging. Fast internalization of the [64Cu]NOTA-αCD11b-mAb provides presumably enhanced stability while cell viability and functionality was not significantly affected. Moreover, utilization of the CD11b-specific mAb allows for straightforward adaptation of the labeling approach for in vivo molecular imaging of other myeloid cells of interest in cancer therapy, including monocytes, macrophages or neutrophils.