The aim of this study was to evaluate the incidence and predictive factors of Pelvic Insufficiency Fractures (PIFs) occurring after Intensity Modulated Radiation Therapy (IMRT) combined with chemotherapy for locally advanced cervical cancer (CC). Medical records of patients receiving radio-chemotherapy with IMRT between 2010 and 2020 for advanced CC were reviewed. PIFs were detected during follow-up on pelvic Magnetic Resonance Imaging (MRI) or Computed Tomography (CT). The cumulative incidence rate of PIFs and its confidence interval were calculated at 2 and 5 years of follow-up. Pre-therapeutic Bone Mineral Density (BMD) (g/cm3) was evaluated on CT simulation for sacrum and the fourth lumbar (L4) vertebrae. Sacrum dosimetric parameters (V30Gy, V40Gy, D50%, Dmean) were analyzed. 136 patients were included. The median follow-up was 4.4 years. Median dose of D50% and V40Gy sacrum were 35.2 Gy (20.6–46.4) and 32.2% (7.2–73.4) respectively. The 2-year and 5-year cumulative incidence rates were 15.7% (95% CI: 9.88–22.71) and 22% (95% CI: 14.58–30.45) respectively. Median time interval between RT completion and PIFs’ detection was 11.5 months (IQR: 7.4–22.3). Univariate analysis showed that older age (p < 0.01), postmenopausal status at baseline (p < 0.01), and lower sacral and spinal BMD at baseline (respectively p < 0.001 and p < 0.01) were significantly associated to all sites of PIFs, and lower sacral BMD with sacral fractures (p < 0.001). Post-IMRT PIFs were detected in 18.4% of patients with locally advanced CC. Individual predisposing factors as older age, postmenopausal status, decreased bone density on the CT simulation were mainly predictive.
Whether adjuvant chemotherapy (AC) improves overall survival (OS) after complete resection in localized soft tissue sarcoma (STS) remains unproven and optional in guidelines. We measured OS in patients with operated non-metastatic STS treated with or without AC in the nationwide exhaustive NETSARC+ registry. The DEEPSARC project linked the NETSARC+ registry and the French nationwide health data system (SNDS) into a retrospective cohort. Adults with non-metastatic, undergoing primary surgical resection between 2012 and 2017 were included, excluding those with incomplete linkage data or prior neoadjuvant chemotherapy. We analysed the entire series of 8331 patients with successful linkage and a complete-case subset of 4296 patients. Survival with or without AC were compared using crude and multivariate Cox proportional hazards regression models, including adjustment for confounders and inverse probability of treatment weighting (IPTW). Sensitivity analyses and subgroups analysis in high-risk patients were performed. In the complete-case series, the commonest histotypes were liposarcoma (30%), undifferentiated pleomorphic sarcoma (23%) and leiomyosarcoma (17%). Overall, 389 (9.1%) patients received AC and, compared with those not receiving AC, had fewer comorbidities and presented more high-risk tumour features. With 5.8 years median follow-up, 5-yr OS was 61% (95%CI: 60–63) in the entire series, and 64% (95%CI: 63–66) in the complete-case analysis. 5-yr OS was inferior in patients receiving AC, both in the entire series (50% [47%–54%] vs. 63% [61%–64%]) and in the complete-case analysis (53% [48–59%] vs. 65% [64–67%]). After adjustment (HR = 1.46 [1.12–1.93]) or IPTW (HR = 1.56 [1.17–2.08]), AC remained associated with inferior OS. Findings were consistent across sensitivity and subgroup analyses. In this real-world nationwide study, AC was not associated with superior OS, but rather with an inferior OS in patients operated of a localized STS and did not compensate for the adverse prognostic profile of treated patients. Future research should therefore focus on identifying predictive biomarkers in order to better identify patients who are likely to derive real benefit from systemic treatment. This work was supported by the Health Data Hub; the 10.13039/501100006364 Institut National Du Cancer and the French Ministry of Health (Direction Générale de l’Offre de Soins, DGOS) for the NETSARC, RRePS and RESOS networks; the INTERSARC+ program; the Agence Nationale de la Recherche; the LYriCAN+ program; the Ligue Nationale contre le Cancer; the Ligue contre le Cancer (Comité de l’Ain); the Fondation ARC; and the European Reference Network for Rare Adult Solid Cancers.
The COVID-19 pandemic highlighted the importance of reliable, real-time hospital forecasting. At Bordeaux University Hospital, we developed models to predict SARS-CoV-2-related hospitalizations 14 days in advance using integrated data sources. We identified six key lessons to guide future epidemic response: (1) Multimodal data improves accuracy; (2) Simple baseline models are essential for benchmarking and building trust; (3) Model and metric choices must align with decision goals which often means prioritizing absolute over relative metrics and beginning with simple models; (4) Prediction intervals should be provided to communicate the uncertainty associated with forecasts; (5) Real-world constraints such as computational cost, maintainability, and required expertise should guide model selection; (6) Forecasts must be contextualized and communicated carefully to policymakers. We advocate for a systems-level forecasting approach that balances accuracy, feasibility, and impact.