Mesenchymal tumors originate from mesenchymal cells and can be either benign or malignant, such as bone, soft tissue, and visceral sarcomas. Surgery is a cornerstone treatment in the management of mesenchymal tumors, often requiring complex procedures performed in high-volume referral centers. However, the COVID-19 pandemic has highlighted this need for alternative non-surgical approaches due to limited access to surgical resources. This review explores the role of non-surgical treatments in different clinical scenarios: for improving surgical outcomes, as a bridge to surgery, as better alternatives to surgery, and for non-curative treatment when surgery is not feasible. We discuss the effectiveness of active surveillance, cryoablation, high-intensity focused ultrasound, and other ablative techniques in managing these tumors. Additionally, we examine the use of tyrosine kinase inhibitors in gastrointestinal stromal tumors and hypofractionated radiotherapy in soft tissue sarcomas. The Sarculator tool is highlighted for its role in stratifying high-risk sarcoma patients and personalizing treatment plans. While surgery remains the mainstay of treatment, integrating advanced non-surgical strategies can enhance therapeutic possibilities and patient care, especially in specific clinical settings with limitations. A multidisciplinary approach in referral centers is vital to determine the optimal treatment course for each patient.

Background: Gastrointestinal stromal tumors (GISTs) present a complex clinical landscape, where precise preoperative risk assessment plays a pivotal role in guiding therapeutic decisions. Conventional methods for evaluating mitotic count, such as biopsy-based assessments, encounter challenges stemming from tumor heterogeneity and sampling biases, thereby underscoring the urgent need for innovative approaches to enhance prognostic accuracy. Objective: The primary objective of this study was to develop a robust and reliable computational tool, PROMETheus (Preoperative Mitosis Estimator Tool), aimed at refining patient stratification through the precise estimation of mitotic count in GISTs. Methods: Using advanced Bayesian network methodologies, we constructed a directed acyclic graph (DAG) integrating pertinent clinicopathological variables essential for accurate mitotic count prediction on the surgical specimen. Key parameters identified and incorporated into the model encompassed tumor size, location, mitotic count from biopsy specimens, surface area evaluated during biopsy, and tumor response to therapy, when applicable. Rigorous testing procedures, including prior predictive simulations, validation utilizing synthetic data sets were employed. Finally, the model was trained on a comprehensive cohort of real-world GIST cases (n=80), drawn from the repository of the Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Humanitas Research Hospital, with a total of 160 cases analyzed. Results: Our computational model exhibited excellent diagnostic performance on synthetic data. Different model architecture were selected based on lower deviance and robust out-of-sample predictive capabilities. Posterior predictive checks (retrodiction) further corroborated the model’s accuracy. Subsequently, PROMETheus was developed. This is an intuitive tool that dynamically computes predicted mitotic count and risk assessment on surgical specimens based on tumor-specific attributes, including size, location, surface area, and biopsy-derived mitotic count, using posterior probabilities derived from the model. Conclusions: The deployment of PROMETheus represents a potential advancement in preoperative risk stratification for GISTs, offering clinicians a precise and reliable means to anticipate mitotic counts on surgical specimens and a solid base to stratify patients for clinical studies. By facilitating tailored therapeutic strategies, this innovative tool is poised to revolutionize clinical decision-making paradigms, ultimately translating into improved patient outcomes and enhanced prognostic precision in the management of GISTs.

BACKGROUND Gastrointestinal stromal tumors (GISTs) present a complex clinical landscape, where precise preoperative risk assessment plays a pivotal role in guiding therapeutic decisions. Conventional methods for evaluating mitotic count, such as biopsy-based assessments, encounter challenges stemming from tumor heterogeneity and sampling biases, thereby underscoring the urgent need for innovative approaches to enhance prognostic accuracy. OBJECTIVE The primary objective of this study was to develop a robust and reliable computational tool, PROMETheus (Preoperative Mitosis Estimator Tool), aimed at refining patient stratification through the precise estimation of mitotic count in GISTs. METHODS Using advanced Bayesian network methodologies, we constructed a directed acyclic graph (DAG) integrating pertinent clinicopathological variables essential for accurate mitotic count prediction on the surgical specimen. Key parameters identified and incorporated into the model encompassed tumor size, location, mitotic count from biopsy specimens, surface area evaluated during biopsy, and tumor response to therapy, when applicable. Rigorous testing procedures, including prior predictive simulations, validation utilizing synthetic data sets were employed. Finally, the model was trained on a comprehensive cohort of real-world GIST cases (n=80), drawn from the repository of the Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Humanitas Research Hospital, with a total of 160 cases analyzed. RESULTS Our computational model exhibited excellent diagnostic performance on synthetic data. Different model architecture were selected based on lower deviance and robust out-of-sample predictive capabilities. Posterior predictive checks (retrodiction) further corroborated the model’s accuracy. Subsequently, PROMETheus was developed. This is an intuitive tool that dynamically computes predicted mitotic count and risk assessment on surgical specimens based on tumor-specific attributes, including size, location, surface area, and biopsy-derived mitotic count, using posterior probabilities derived from the model. CONCLUSIONS The deployment of PROMETheus represents a potential advancement in preoperative risk stratification for GISTs, offering clinicians a precise and reliable means to anticipate mitotic counts on surgical specimens and a solid base to stratify patients for clinical studies. By facilitating tailored therapeutic strategies, this innovative tool is poised to revolutionize clinical decision-making paradigms, ultimately translating into improved patient outcomes and enhanced prognostic precision in the management of GISTs.

E23523 Background: How sarcoma metastasize is unknown. VETC has been described as an epithelial-to-mesenchymal-transition independent process of metastasis: endothelium covers neoplastic clusters allowing tumor dissemination. It has also been shown to be a predictive of tyrosine kinase inhibitors (TKI) response. The objectives of the present study are: 1) to assess the presence of VETC in sarcoma; 2) to model its prognostic role. Methods: The study was retrospective. We selected 54 cases of sarcomas (6 DDLPS, 10 GIST, 6 retroperitoneal LMS, 9 MLPS, 8 MPNST, 10 SFT, 5 UPS); of them 31 were metastatic (M1 group), 23 were not (M0 group, defined as least 5 years of negative follow-up). High risk GIST and SFT, and grade 3 FNCLCC were considered high grade. VETC was assessed with CD31 immunohistochemistry and defined as a continuous endothelial lining around tumor clusters. We used Bayesian probabilistic modeling to detect small effects and multilevel hierarchical modeling to reduce overfitting. Models were fit using Stan and R. CI were computed with the HPDI method for 89%. The whole posterior probability (PP) was used for calculations. Results: The two groups (M0 & M1) were substantially homogeneous: the CI of the contrast PP (CPP) included 0 in each histology for sex, age, size, and grade; VETC was more expressed in M1 SFT cases with almost all the CPP mass above the 0. Also, UPS and GIST showed VETC to be more present in metastatic diseases with 79% and 78% respectively of the CPP mass above 0 (see table). Metastasis free survival (MFS) analysis showed that VETC in SFT and UPS with a coefficient of 2.42 (CI 0.73 – 4.65) and 1.94 (CI 0.16 – 3.67); for the latter the median was reached with a PP density of median MFS of 65 months (mo) (standard deviation, SD: 74 mo) for VETC- Vs 11 mo (SD: 14 mo) for VETC+. Similarly, disease specific survival analysis showed – in SFT and UPS – that VETC was negatively associated with prognosis with coefficients of 1.24 and 2.34; however the CI covered slightly the 0 (-0.54 – 3.96 and -0.09 – 5.73, respectively). Conclusions: We found VETC in sarcoma and highlighted its prognostic role in SFT and UPS.

Background: The antigen processing machinery (APM) plays a critical role in generating tumor-specific antigens that can be recognized and targeted by the immune system. The status of the APM in sarcomas is not well characterized. Methods: We investigated 126 patients with 8 types of bone and soft tissue sarcoma operated between 2001-2021. Tissue microarrays mapped 11 specific areas in each case. The presence/absence of APM protein was determined through immunohistochemistry. Bayesian networks were used. Results: All investigated sarcomas had some defects in APM. The least damaged component was HLA Class I subunit β2-microglobulin and HLA Class II. The proteasome LMP10 subunit was defective in leiomyosarcoma (LMS), myxoid liposarcoma (MLPS), and dedifferentiated liposarcoma (DDLPS), while MHC I transporting unit TAP2 was altered in undifferentiated pleomorphic sarcoma (UPS), gastrointestinal stromal tumor (GIST), and chordoma (CH). Among different neoplastic areas, high-grade areas showed different patterns of expression compared to high lymphocytic infiltrate areas. Heterogeneity at the patient level was also observed. Loss of any APM component was prognostic of distant metastasis (DM) for LMS and DDLPS and of overall survival (OS) for LMS. Conclusion: Sarcomas exhibit a high degree of defects in APM components, with differences among histotypes and tumoral areas. The most commonly altered APM components were HLA Class I subunit β2-microglobulin, HLA Class I subunit α (HC10), and MHC I transporting unit TAP2. The loss of APM components was prognostic of DM and OS and clinically relevant for LMS and DDLPS. This study explores sarcoma molecular mechanisms, enriching personalized therapeutic approaches.