Immunotherapy, while showing promise, fails to cure the majority of cancer patients. Preclinical studies using transplanted tumor models have suggested a synergistic effect between radiotherapy (RT) and immunotherapy, specifically immune checkpoint inhibitors. In these models, RT acts as an in situ vaccine, stimulating antitumor immunity and eliminating tumors beyond the radiation field. These preclinical studies often report high cure rates with combined RT and checkpoint blockade. However, the use of transplant tumor models raises concerns, as these models may not accurately reflect the complexities of the tumor microenvironment and the immune response in patients. Transplanted tumors are not subjected to the same evolutionary pressures and immuno-surveillance experienced by primary tumors that develop in situ. This raises questions regarding the translational potential of findings from transplant models. The current study investigates the efficacy of combining RT and anti-programmed cell death-1 (PD-1) antibody treatment in a high-mutational-burden mouse model of sarcoma, comparing the response in both transplant and primary (autochthonous) tumor models. The goal is to understand the mechanisms underlying response and resistance to this combined therapy and to better understand the discrepancies between preclinical findings and clinical outcomes. This high-mutational load model was chosen due to the known correlation between high tumor mutational burden and response to immunotherapy in human cancers.

Previous research has extensively explored the potential of combining radiotherapy and immunotherapy. Studies using transplanted tumor models demonstrated that local RT can synergize with immune checkpoint inhibitors, generating systemic antitumor immune responses. This effect is often attributed to RT acting as an in situ vaccine, eliminating tumor cells and releasing tumor-associated antigens that stimulate T cell responses. However, these preclinical successes haven't fully translated to the clinic. A considerable body of literature highlights the limitations of transplanted tumor models, which often overestimate treatment efficacy compared to the clinical setting. The complex interplay between the tumor and the immune system, including factors like immune editing and the establishment of immune tolerance, is often simplified in transplant models. Understanding the intricacies of this interplay is critical for improving the effectiveness of immunotherapy and for identifying patients most likely to benefit from combined RT and immunotherapy.

The study utilized a novel high-mutational-burden mouse model of sarcoma, induced by deleting Tp53 (tumor protein p53) and injecting 3-methylcholanthrene (MCA). Primary tumors were established in immunocompetent mice, while a cell line derived from these primary tumors was used to generate transplant tumors in syngeneic mice. Mice bearing either primary or transplant tumors received different treatment combinations: anti-PD-1 antibody alone, RT alone, both anti-PD-1 and RT, or a combination of anti-PD-1 and anti-CTLA-4 with RT. Survival analysis was performed using the Kaplan-Meier method and log-rank test. To investigate the mechanisms of response and resistance, several high-throughput methods were employed. Whole-exome sequencing (WES) and bulk RNA sequencing were performed to identify genomic and transcriptomic differences between primary and transplant tumors. Immunogenic neoantigens were predicted and their expression levels were compared. To assess immune tolerance, a series of auto-transplantation experiments were performed, comparing tumor growth in donor mice (mice from which the tumor cell line was derived) and naïve syngeneic mice. To further characterize the immune microenvironment, CIBERSORTx was used to deconvolute bulk RNA-seq data and estimate the abundance of various immune cell types. Mass cytometry was used to quantify PD-L1+ macrophages. Finally, single-cell RNA sequencing (scRNA-seq) was performed on FACS-sorted CD45+ tumor-infiltrating immune cells to gain a more detailed understanding of the immune cell heterogeneity and their transcriptional responses to treatment.

The study revealed striking differences in the response to combined RT and PD-1 blockade between transplant and primary sarcomas. While transplant tumors were cured by this treatment, primary tumors remained resistant. Primary tumors exhibited a higher number of nonsynonymous mutations, indicating strong immune selection during their development. However, these tumors expressed a smaller fraction of neoantigens compared to transplant tumors, suggesting immune evasion mechanisms such as neoantigen downregulation. Auto-transplantation experiments demonstrated immune tolerance to the 'self' tumor in donor mice, highlighting the importance of tumor-immune co-evolution. Analysis of the immune microenvironment showed that transplant tumors resembled the most inflamed human sarcomas (SIC E), characterized by abundant CD8+ T cells and macrophages. In contrast, primary tumors displayed an immunosuppressive microenvironment, with fewer activated CD8+ T cells and a higher proportion of immunosuppressive myeloid cells. ScRNA-seq revealed that RT induced significant remodeling of myeloid cells in both primary and transplant tumors, but transplant tumors exhibited a stronger interferon response. In primary tumors, RT combined with PD-1 blockade resulted in a shift towards an antitumor myeloid phenotype, suggesting that the combination therapy could effectively reprogram immunosuppressive myeloid cells. The study also found that transplant tumors had a significantly higher number of activated CD8+ T cells compared to primary tumors, a difference that may contribute to the divergent response to therapy.

This study's findings highlight the limitations of using transplant tumor models in preclinical immunotherapy research. The observed differences in immune landscapes and treatment responses between transplant and primary sarcomas emphasize the critical role of tumor-immune co-evolution in shaping the tumor microenvironment and determining therapeutic efficacy. The development of immune tolerance in primary tumors represents a significant barrier to successful immunotherapy. The study suggests that patients with sarcomas that resemble the inflammatory immune phenotype of transplant tumors in this model (high CD8+ T cell and macrophage infiltration) may be more likely to benefit from combined RT and PD-1 blockade. However, the study also indicates that primary tumors, while resistant, exhibit some response to combined RT and PD-1 blockade through myeloid cell reprogramming. This suggests that strategies aiming to overcome immune tolerance, such as bone marrow transplantation or other methods to reset the immune system, may be explored in the context of primary tumors.

This study demonstrates that primary and transplant sarcomas have distinct immune microenvironments and responses to immunotherapy and RT. The findings highlight the limitations of transplant models and suggest that focusing on the immune landscape, specifically targeting immune tolerance, may be crucial for improving the effectiveness of immunotherapy in sarcoma patients. Future research should investigate the mechanisms of immune tolerance in primary sarcomas and explore strategies to overcome this resistance, possibly through immune system resetting approaches.

The study utilized a single mouse model of sarcoma. The findings may not be generalizable to all types of sarcomas or other cancer types. The study focused on a limited set of treatment combinations. Further research is needed to investigate the effects of other immunotherapy agents or treatment schedules. The sample size for some experiments, especially in the scRNA-seq analysis of primary tumors could be considered relatively small and future studies may benefit from larger cohorts to improve statistical power. Finally, the mouse model, while useful for mechanistic studies, may not completely reflect the complexity of human sarcoma immunology.