(P 420) COMPREHENSIVE MOLECULAR PROFILING AND TUMOR MICROENVIRONMENT CHARACTERISTICS OF RADIATION-INDUCED SARCOMA.

Objective: A late-onset complication of radiotherapy (RT), radiation-induced sarcoma (RIS) is a rare tumor associated with chemotherapy resistance and poor prognosis. Although immune checkpoint inhibitors (ICI) are recommended for several soft tissue tumors, their efficacy has not been demonstrated in RIS. Therefore, identifying effective therapies for RIS remains an unmet need. Comprehensive molecular testing including whole exome sequencing (WES) and RNA-seq was performed to characterize the tumor microenvironment (TME) composition and uncover potential genomic drivers of RIS to guide therapeutic decision-making.

Methods: Sarcoma samples and clinical data from an internal database were screened to match RIS criteria, including a history of prior RT, a period of latency after RT (>2 years), and histologic confirmation of sarcoma. WES and RNA-seq were performed as a part of the BostonGene Tumor Portrait® test for 9 RIS samples. Retrospective analysis of the test results revealed transcriptomic and genomic alterations associated with RIS, and samples were assigned RNA-based immune-enriched (IE), immune-enriched, nonfibrotic (IE/M), fibrotic (F), or desert (D) TME subtypes. which we previously described in Bagaev et al.

Results: Histological subtypes included angiosarcoma (n = 3), sarcoma not otherwise specified (NOS, n = 3), osteosarcoma (n = 1), and malignant peripheral nerve sheath tumor (n = 2, Table 1). The median duration between the initial RT and current sarcoma was 10 years (3-33 years). The cohort included 5 women and 4 men with a median age of 53 years (34-80 years) at the time of RT. The most frequent primary tumor was breast cancer (n = 5).
During transcriptomic analysis, five samples (56%) contained IE TME subtypes characterized by low levels of stromal and fibrotic elements and high levels of immune infiltrate, including cytotoxic effector cells. The remaining four patients (44%) had fibrotic (F) TMEs with minimal immune infiltration and high content of stromal components.
Six patients were treated with systemic therapy prior to molecular testing. All six patients with treatment data received chemotherapy, and of these patients, additional therapies included single-agent pazopanib (n = 2), combination chemotherapy plus ICI (n = 2). Response data were available for two patients with IE TMEs who received ICIs with clinically positive responses. Of these two patients, one exhibited stable disease during the 7 months of advanced disease treatment, and the other patient received ICI and taxane neoadjuvant therapy for 4 months prior to surgery.
WES was performed to assess for somatic alterations that could serve as potential drivers or targets, such as genes associated with oncogenic pathways or DNA damage response (Figure 1). Transcriptomic analysis identified high MET expression in 6 patients and high CD274 (PD-L1) expression in 4 patients. Low tumor mutational burden and microsatellite stability were observed in all samples. Potential clinical trials were matched for all 9 RIS cases based on these molecular findings.

Conclusion: BostonGene's Tumor Portrait® test identified clinically relevant findings for 9 RIS patients, who received integrated WES and RNA-seq testing. Over 50% of RIS patients had an IE TME subtype, which indicated the probability of a favorable response to ICI. Integrated WES and RNA-seq analysis can be used to select targeted molecular-based therapies and identify clinical trials for RIS patients.