Retroperitoneal leiomyosarcoma (RLS) is a rare and aggressive subtype of soft tissue sarcoma with limited population-level evidence guiding surgical decision-making. This study aimed to assess the prognostic value of surgery in patients with RLS using a large real-world cohort and advanced analytical methods. Patients diagnosed with RLS between 2000 and 2019 were identified from the Surveillance, Epidemiology, and End Results (SEER) database. Propensity score matching (PSM) was used to balance baseline variables. Overall survival (OS) and cancer-specific survival (CSS) were analyzed using Kaplan–Meier curves and Cox proportional hazards models. Random survival forests (RSF) were applied to evaluate variable importance and model robustness. A total of 1041 patients were included, of whom 817 (78.5%) underwent surgery. Before matching, significant imbalances were observed in age, grade, and SEER stage. After 1:1 PSM (159 matched pairs), covariate balance was substantially improved. Surgery was associated with significantly improved survival (OS: HR = 0.34, 95% CI: 0.26–0.45; CSS: HR = 0.34, 95% CI: 0.25–0.46; both P < 0.001). High-grade tumors and advanced SEER stage remained independent adverse prognostic factors. RSF consistently ranked surgery, stage, and grade as the most important predictors of survival. Surgical resection status was strongly associated with survival in SEER-based analyses, but this association is subject to substantial unmeasured confounding by resectability, anatomic extent, and patient fitness; therefore, results should be interpreted as prognostic rather than causal and highlight the need for multidisciplinary assessment in high-volume sarcoma centers.
To evaluate the association of surgery with survival outcomes in patients with small cell carcinoma of the digestive system (SCCDS) using data from the Surveillance, Epidemiology, and End Results (SEER) database and machine-learning-based analysis. Data from patients with SCCDS diagnosed between 2000 and 2018 were extracted from the SEER database. After applying the inclusion and exclusion criteria, 1600 eligible patients were included. Patients were categorized into surgery and non-surgery groups. Overall survival (OS) and cancer-specific survival (CSS) were assessed using Kaplan-Meier analysis and Cox proportional hazards models. To further reduce treatment selection bias, propensity score-based inverse probability of treatment weighting (IPTW) was performed as a sensitivity analysis. A random forest algorithm was used to identify important prognostic factors. Subgroup analyses were conducted to explore the association between surgery and survival across different patient subsets. The surgery group (n = 400) showed better unadjusted 1-year OS (52.69% vs. 32.82%) and CSS (55.40% vs. 35.12%) than the non-surgery group (n = 1200; both P < 0.001). Multivariable Cox regression identified age, sex, tumor stage, hepatic metastasis, radiotherapy, chemotherapy, and surgery as independent prognostic factors for OS and CSS (all P < 0.05). In IPTW analyses, surgery remained independently associated with better OS and CSS. Machine learning analysis further identified surgery, tumor stage, and hepatic metastasis as major contributors to survival prediction. Subgroup analysis suggested that the association between surgery and survival may be less evident in patients with esophageal or retroperitoneal tumors and in Asian or Pacific Islander populations. Surgery was independently associated with better OS and CSS in patients with SCCDS. However, given the retrospective design and the potential for residual confounding, these findings should be interpreted cautiously. The survival association of surgery may be less pronounced in certain subgroups, highlighting the importance of careful patient selection.