Current Strengths and Weaknesses of ChatGPT as a Resource for Radiation Oncology Patients and Providers.

PURPOSE: Chat Generative Pre-Trained Transformer (ChatGPT), an artificial intelligence program that uses natural language processing to generate conversational-style responses to questions or inputs, is increasingly being used by both patients and health care professionals. This study aims to evaluate the accuracy and comprehensiveness of ChatGPT in radiation oncology-related domains, including answering common patient questions, summarizing landmark clinical research studies, and providing literature reviews with specific references supporting current standard-of-care clinical practice in radiation oncology. METHODS AND MATERIALS: We assessed the performance of ChatGPT version 3.5 (ChatGPT3.5) in 3 areas. We evaluated ChatGPT3.5's ability to answer 28 templated patient-centered questions applied across 9 cancer types. We then tested ChatGPT3.5's ability to summarize specific portions of 10 landmark studies in radiation oncology. Next, we used ChatGPT3.5 to identify scientific studies supporting current standard-of-care practice in clinical radiation oncology for 5 different cancer types. Each response was graded independently by 2 reviewers, with discordant grades resolved by a third reviewer. RESULTS: ChatGPT3.5 frequently generated inaccurate or incomplete responses. Only 39.7% of responses to patient-centered questions were considered correct and comprehensive. When summarizing landmark studies in radiation oncology, 35.0% of ChatGPT3.5's responses were accurate and comprehensive, improving to 43.3% when provided the full text of the study. ChatGPT3.5's ability to present a list of studies related to standard-of-care clinical practices was also unsatisfactory, with 50.6% of the provided studies fabricated. CONCLUSIONS: ChatGPT should not be considered a reliable radiation oncology resource for patients or providers at this time, as it frequently generates inaccurate or incomplete responses. However, natural language programming-based artificial intelligence programs are rapidly evolving, and future versions of ChatGPT or similar programs may demonstrate improved performance in this domain.

Enhancing radiotherapy response via intratumoral injection of the TLR9 agonist CpG to stimulate CD8 T cells in an autochthonous mouse model of sarcoma.

Radiation therapy is frequently used to treat cancers including soft tissue sarcomas. Prior studies established that the toll-like receptor 9 (TLR9) agonist cytosine-phosphate-guanine oligodeoxynucleotide (CpG) enhances the response to radiation therapy (RT) in transplanted tumors, but the mechanism(s) remain unclear. Here, we used CRISPR/Cas9 and the chemical carcinogen 3-methylcholanthrene (MCA) to generate autochthonous soft tissue sarcomas with high tumor mutation burden. Treatment with a single fraction of 20 Gy RT and two doses of CpG significantly enhanced tumor response, which was abrogated by genetic or immunodepletion of CD8+ T cells. To characterize the immune response to RT + CpG, we performed bulk RNA-seq, single-cell RNA-seq, and mass cytometry. Sarcomas treated with 20 Gy and CpG demonstrated increased CD8 T cells expressing markers associated with activation and proliferation, such as Granzyme B, Ki-67, and interferon-γ. CpG + RT also upregulated antigen presentation pathways on myeloid cells. Furthermore, in sarcomas treated with CpG + RT, TCR clonality analysis suggests an increase in clonal T-cell dominance. Collectively, these findings demonstrate that RT + CpG significantly delays tumor growth in a CD8 T cell-dependent manner. These results provide a strong rationale for clinical trials evaluating CpG or other TLR9 agonists with RT in patients with soft tissue sarcoma.

Enhancing radiotherapy response via intratumoral injection of a TLR9 agonist in autochthonous murine sarcomas.

Radiation therapy (RT) is frequently used to treat cancers, including soft-tissue sarcomas. Prior studies established that the toll-like receptor 9 (TLR9) agonist cytosine-phosphate-guanine oligodeoxynucleotide (CpG) enhances the response to RT in transplanted tumors, but the mechanisms of this enhancement remain unclear. Here, we used CRISPR/Cas9 and the chemical carcinogen 3-methylcholanthrene (MCA) to generate autochthonous soft-tissue sarcomas with high tumor mutation burden. Treatment with a single fraction of 20 Gy RT and 2 doses of CpG significantly enhanced tumor response, which was abrogated by genetic or immunodepletion of CD8+ T cells. To characterize the immune response to CpG+RT, we performed bulk RNA-Seq, single-cell RNA-Seq, and mass cytometry. Sarcomas treated with 20 Gy and CpG demonstrated increased CD8 T cells expressing markers associated with activation and proliferation, such as Granzyme B, Ki-67, and IFN-γ. CpG+RT also upregulated antigen presentation pathways on myeloid cells. Furthermore, in sarcomas treated with CpG+RT, TCR clonality analysis suggests an increase in clonal T cell dominance. Collectively, these findings demonstrate that CpG+RT significantly delays tumor growth in a CD8 T cell-dependent manner. These results provide a strong rationale for clinical trials evaluating CpG or other TLR9 agonists with RT in patients with soft-tissue sarcoma.

Loss of function of Atrx leads to activation of alternative lengthening of telomeres in a primary mouse model of sarcoma.

The development of a telomere maintenance mechanism is essential for immortalization in human cancer. While most cancers elongate their telomeres by expression of telomerase, 10-15% of human cancers use a pathway known as alternative lengthening of telomeres (ALT). In this work, we developed a genetically engineered primary mouse model of sarcoma in CAST/EiJ mice which displays multiple molecular features of ALT activation after CRISPR/Cas9 introduction of oncogenic KrasG12D and loss of function mutations of Trp53 and Atrx. In this model, we demonstrate that the loss of Atrx contributes to the development of ALT in an autochthonous tumor, and this process occurs independently of telomerase function by variation of mTR alleles. Furthermore, we find that telomere shortening from the loss of telomerase leads to higher chromosomal instability while loss of Atrx and activation of ALT lead to an increase in telomeric instability, telomere sister chromatid exchange, c-circle production, and formation of ALT-associated promyelocytic leukemia bodies (APBs). The development of this primary mouse model of ALT could enable future investigations into therapeutic vulnerabilities of ALT activation and its mechanism of action.

Atrx deletion impairs CGAS/STING signaling and increases sarcoma response to radiation and oncolytic herpesvirus.

ATRX is one of the most frequently altered genes in solid tumors, and mutation is especially frequent in soft tissue sarcomas. However, the role of ATRX in tumor development and response to cancer therapies remains poorly understood. Here, we developed a primary mouse model of soft tissue sarcoma and showed that Atrx-deleted tumors were more sensitive to radiation therapy and to oncolytic herpesvirus. In the absence of Atrx, irradiated sarcomas had increased persistent DNA damage, telomere dysfunction, and mitotic catastrophe. Our work also showed that Atrx deletion resulted in downregulation of the CGAS/STING signaling pathway at multiple points in the pathway and was not driven by mutations or transcriptional downregulation of the CGAS/STING pathway components. We found that both human and mouse models of Atrx-deleted sarcoma had a reduced adaptive immune response, markedly impaired CGAS/STING signaling, and increased sensitivity to TVEC, an oncolytic herpesvirus that is currently FDA approved for the treatment of aggressive melanomas. Translation of these results to patients with ATRX-mutant cancers could enable genomically guided cancer therapy approaches to improve patient outcomes.