Abstract A29: Uncovering transcriptional signatures of drug resistant tumor cells: Mechanisms of therapeutic resistance and opportunities for combination therapies

Abstract Soft tissue sarcomas (STS) are tumors of mesenchymal origin and include multiple sub-types such as undifferentiated pleomorphic sarcoma (UPS) which is the most frequent classification of STS. UPS is one of the most aggressive and recurrent sarcomas. The current standard of care for UPS patients remains surgical resection, radiotherapy, and chemotherapy, but many patients still develop recurrent tumors after these interventions. In addition, anti-PD-1 therapy has only been successful in a small fraction of patients with high levels of tumor infiltrating B and T cells, while immune excluded tumors showed poor responses. To understand why these tumors are frequently recurrent, it is essential to understand the characteristics of resistant tumor cells following standard treatment regimens in both immune excluded and immune infiltrated tumor microenvironments. We generated multiple mouse models that reflect the genetic alterations frequently found in patients via overexpression of oncogenes Ccne1 or Vgll3 in p53KO mesenchymal stem cells. Classification of the immune microenvironment of these models revealed that these driver genetics promote an immune excluded or immune infiltrated tumor microenvironment, respectively. We leveraged these two tumor models to understand the mechanisms of therapeutic resistance to common agents doxorubicin or anti-PD-1. We found that despite higher levels of baseline tumor infiltrating lymphocytes in Vgll3 tumors, anti-PD-1 failed to reduce tumor growth. Doxorubicin treatment resulted in a modest reduction of tumor volume in Ccne1 tumors, and more significantly in immune infiltrated Vgll3 tumors. To understand the transcriptional profiles of resistant tumor cells we harnessed single-cell RNA-sequencing (scRNA-seq) to characterize the least responsive model to determine if treatments had an impact on tumor cell expression profiles and which tumor clusters persist after treatment. scRNA-seq analysis revealed that different clusters of tumor cells were differently affected by anti-PD-1 and doxorubicin. While doxorubicin mainly affected tumor cells expressing collagens and matrix associated adhesion proteins, treatment with anti-PD-1 selectively reduced tumor cells enriched in interferon signaling pathways. Interestingly, tumor cells expressing high levels of extracellular matrix (ECM) remodeling genes remained unaffected and, in some cases, were enriched under both treatments. These results may suggest that tumor cells capable of ECM remodeling may shield tumor cells from chemotherapy agents and immune cells from immune checkpoint inhibitors and promote tumor recurrence. Therefore, these results highlight the need for investigation of combination therapies targeting extracellular matrix proteins in addition to immune checkpoint blockade. Citation Format: Marina T Broz, Marco DeSimone, Emily Ko, Roberta Piras, Jlenia Guarnerio. Uncovering transcriptional signatures of drug resistant tumor cells: Mechanisms of therapeutic resistance and opportunities for combination therapies [abstract]. In: Proceedings of the AACR Special Conference: Tumor Immunology and Immunotherapy; 2022 Oct 21-24; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2022;10(12 Suppl):Abstract nr A29.