Background: Soft tissue sarcoma is a rare form of cancer primarily affecting connective tissue in adults. Standard chemotherapy includes the anthracycline doxorubicin which has previously been shown in a variety of cancer models to induce immunogenic cell death (ICD). ICD is driven, in part, by the upregulation and/or translocation of molecules such as calreticulin and annexin I, known as damage associated molecular patterns (DAMPs). We aimed to investigate the ICD capacity of anthracyclines doxorubicin and mitoxantrone and explore activity in combination with anti-PD1 and CTLA-4 checkpoint inhibitors.Methods: We treated a variety of human osteosarcoma cell lines and MCA-205 mouse fibrosarcoma cells with doxorubicin and mitoxantrone and measured expression of DAMPs and immunosuppressive markers via multiplex flow cytometry. We also measured anti-tumor activity of doxorubicin with and without PD1/CTLA4 blockade in the murine sarcoma model MCA-205. After tumors were harvested, we analyzed tumor-infiltrating lymphocytes to determine the effects of treatment on phenotype, quantity, activation status, function, clonality, and expression of immune checkpoint proteins.Results: We found expression of Annexin I and TRAIL-R2 on the cell surface of osteosarcoma cell lines treated with anthracyclines and significant inhibition of tumor growth with the combination of doxorubicin plus PD1/CTLA4 blockade compared to either doxorubicin or checkpoint blockade alone. Profiling of TILs revealed a marked increase in CD8:CD4 ratio and increased expression of checkpoint proteins including PD1, 41BB and TIM3 in the combination of doxorubicin plus checkpoint blockade. Finally, increased preponderance of the TCRVb 5.1-5.2 clonotype was observed with the combination in TIL, which was not observed in spleen or peripheral blood samples.Conclusion: We have observed early evidence of improved anti-tumor effects with combination doxorubicin plus PD1/CTLA4 along with reshaping of the TIL repertoire, suggesting that doxorubicin may improve immunogenic cell death and supporting clinical investigation in human sarcoma patients. Citation Format: Nicholas R Therrien, Kyle Powers, Lindsey Kemp, Cristiam Moreno Tellez, Breelyn A Wilky. Evaluation of doxorubicin and PD-1/CTLA-4 immune blockade combination therapy in the MCA-205 murine model [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 B64.

LBA11504 Background: Cabozantinib (C) has been combined successfully with either PD-1 or CTLA-4 inhibition in cancer clinical trials. The combination of nivolumab (N) and ipilimumab (I) previously was evaluated in soft tissue sarcoma (STS) and demonstrated activity as well as an acceptable safety profile. Given these data, we hypothesize that C in combination with both I and N is a therapeutic strategy that will be more effective than C alone in metastatic STS that lack translocations. Additionally, we hypothesized that C priming would enhance the efficacy of combination I and N at the time of crossover. Methods: This is a 2:1 randomized phase 2 clinical trial evaluating the overall response rate (RR) of C 40mg orally daily in combination with I (1mg/kg IV) / N (3mg/kg) for 4 doses Q3W and then maintenance N 480mg Q4W compared to C 60mg oral alone, with crossover. Secondary endpoints include progression free survival (PFS), disease control rate (DCR), RR in crossover, quality of life by FACT-G7, RR by iRECIST and safety. Correlative and biomarker analysis are preplanned. Key patient selection includes ECOG 0-1, 1-2 lines of prior therapy, and sarcomas that lack of translocations. The trial was balanced for leiomyosarcoma (LMS), liposarcoma and UPS. Results: 69 patients were randomized to C+I/N and 36 patients were randomized to C alone. 19 patients crossed over to C+I/N at progression. 54/105 patients had LMS. RR of C+I/N was 11% (5 PR and 2 CR), while the RR of C was 6% (2 PR and 0 CR) (p = NS). C+I/N responding histologies included LMS, angiosarcoma, epithelioid sarcoma, and myxofibrosarcoma. C responding histologies included 2 PRs in LMS. There were also 2 LMS PRs in crossover to C+I/N. The median PFS for C+I/N was 5.4 months and for C was 3.8 months (p = 0.016). The DCR for C+I/N was 80% (41 SD, 5 PR, 2 CR), and 42% for C (11 SD, 2 PR) (p = 0.0004). The most common grade 3-4 adverse events affecting > 10% of patients included hypothyroidism, diarrhea, mucositis, oral dysesthesia, nausea, vomiting, elevated AST and ALT, anorexia, dysgeusia, headache, pruritis, maculopapular rash, and hypertension for C+I/N and hypothyroidism, diarrhea, oral dysesthesia, fatigue, palmar-plantar erythrodysesthesia, and hypertension for C. Conclusions: The combination of C+I/N was superior to C for the treatment of non-translocation STS for DCR and PFS. Most frequent responding histology was LMS. Correlative work is ongoing. Clinical trial information: NCT04551430 .