Surveillance of in situ tumor arrays reveals early environmental control of cancer immunity

The immune phenotype of a tumor is a key predictor of its response to immunotherapy1-4. Patients who respond to immune checkpoint blockade generally present with tumors that are infiltrated by activated T cells, a tumor-immune phenotype referred to as immune inflamed5-7. However, not all immune inflamed tumors respond to therapy, and in addition the majority of patients presents with tumors that lack T cells ( immune desert) or that exclude T cells in the periphery of the tumor islet ( immune excluded)8. Despite the importance of these tumor-immune phenotypes in patients, little is known about their development, heterogeneity or dynamics due to an inability to model these features pre-clinically. Here, we describe an approach designated STAMP (skin tumor array by micro-poration), which combines in vivo noninvasive, high-throughput time-lapse imaging with excisional biopsies and next generation sequencing to characterize the establishment of the immunological niche and follow its evolution during immunotherapy. STAMP involves the seeding of dozens to hundreds clonal tumors in the superficial dermis of a single mouse ear that can be visualized in situ over weeks to months. Using this approach, we found that genetically identical tumors could display surprisingly different immune phenotypes. Although individual tumors of the same array were populated by the same T cell clonotypes, regression or progression of individual tumors were associated with distinct patterns of spatial organization of the T cells. In situ imaging of 14K tumors revealed that immune phenotypes were not static over-time but could rather evolve with tumor growth and response to treatment. Therapy-induced or spontaneous early conversion to the immune inflamed phenotype correlated with tumor regression and enhanced cytotoxic T cell activity. Therefore, STAMP provides a flexible approach to study the relationship between tumor evolution, immune cell dynamics, and tumor microenvironment with therapeutic response.