FDFrançoise Ducimetière

Françoise Ducimetière

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Françoise Ducimetière has a total of 331 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Insights into rhabdomyosarcoma in the adolescent and young adult population: a comprehensive clinical study from the French Society of Pediatric Oncology (SFCE) Malignant Mesenchymal Tumor (MMT) Committee and the French Sarcoma Group (GSF-GETO)

    by C. Khanfar, F. Sevrin, E. Bogart, M. Karanian, L. Fenwarth, A. Buisson, J.-Y. Blay, M. Cleirec, N. Corradini, E. Bompas, et al.

  • Open Access

    Linking the NETSARC+ national sarcoma database with the SNDS to evaluate adjuvant and/or neoadjuvant therapy: report on the linkage process and result (Health Data Hub’s DEEPSARC pilot project)

    by Erwan Drezen, André Happe, Vincent Thevenet, Nicolas Penel, François Gouin, François Le Loarer, Gonzague Du Bouexic De Pinieux, Hugo Crochet, Claire Chemin Airiau, Françoise Ducimetiere, et al.

    Background DEEPSARC, one of the first project running on the Health Data Hub aimed to identify real-life treatment regimens that could improve overall survival. The project is based on matching the national database of the sarcoma reference network with the SNDS. Objectives We aimed to report a transparent description of the linking process and its results. Methods The sarcoma database encompasses 33,548 patients matching the selection criteria divided in three subsets: 13,507 patients with a complete dataset gathering clinical and pathological data; 5,844 patients with clinical data alone; and 14,197 patients with pathological data alone. As no ICD-10 code reliably identifies patients with sarcoma the subpopulation extracted from the SNDS was extended to 3 million patients who underwent surgery for their cancer. An indirect record linkage process used a combination (called a signature) of so-called chaining variables to uniquely identify a pair of patients from each of the bases. Two metrics (signature robustness and overall quality) were calculated for ease of interpretation. Results The overall matching rate of 73.1% (24,539 pairs out of 33,548 observations), reaching 90.5% in the intersection of the sarcomas databases (with extended data, 12,225 pairs out of 13,507 observations). Conclusion Detailed reporting, along with dedicated metrics, contribute to the transparency of the process, as discussion and interpretation of the chaining results are crucial for the validity of the main results of the study.


  • Open Access

    Sclerosing Epithelioid Fibrosarcoma (SEF) versus Low Grade Fibromyxoid Sarcoma (LGFMS): Presentation and outcome in the nationwide NETSARC+ series of 330 patients over 13 years

    by Jy Blay, C. Tlemsani, M. Toulmonde, A. Italiano, M. Rios, E. Bompas, T. Valentin, F. Duffaud, Lr Le Nail, S. Watson, et al.


  • Open Access

    Evolution of Patterns of Care and Outcomes in the Real-Life Setting for Patients with Metastatic GIST Treated in Three French Expert Centers over Three Decades

    by Maud Toulmonde, Derek Dinart, Mehdi Brahmi, Benjamin Verret, Myriam Jean-Denis, Françoise Ducimetière, Gregoire Desolneux, Pierre Méeus, Jean Palussière, Xavier Buy, et al.

    Gastrointestinal stromal tumors (GIST) are rare mesenchymal tumors characterized by KIT or PDGFRA mutations. Over three decades, significant changes in drug discovery and loco-regional (LR) procedures have impacted treatment strategies. We assessed the evolution of treatment strategies for metastatic GIST patients treated in the three national coordinating centers of NetSarc, the French network of sarcoma referral centers endorsed by the National Institute of Cancers, from 1990 to 2018. The primary objective was to describe the clinical and biological profiles as well as the treatment modalities of patients with metastatic GIST in a real-life setting, including access to clinical trials and LR procedures in the metastatic setting. Secondary objectives were to assess (1) patients’ outcome in terms of time to next treatment (TNT) for each line of systemic treatment, (2) patients’ overall survival (OS), (3) evolution of patients’ treatment modalities and OS according to treatment access: p = 0.18). In multivariate analysis, age, AFIP Miettinen classification, mutational status, surgery of the primary tumor, participation in a clinical trial in the first line and LR procedure to metastatic sites were associated with longer TNT in the first line, whereas age, mitotic index, mutational status, surgery of the primary tumor and LR procedure to metastatic sites were associated with longer OS. This real-life study advocates for early reference of metastatic GIST patients to expert centers to orchestrate the best access to future innovative clinical trials together with LR strategies and further improve GIST patients’ survival.


  • Open Access

    Deep learning predicts patients outcome and mutations from digitized histology slides in gastrointestinal stromal tumor

    by Yu Fu, Marie Karanian, Raul Perret, Axel Camara, François Le Loarer, Myriam Jean-Denis, Isabelle Hostein, Audrey Michot, Françoise Ducimetiere, Antoine Giraud, et al.

    Risk assessment of gastrointestinal stromal tumor (GIST) according to the AFIP/Miettinen classification and mutational profiling are major tools for patient management. However, the AFIP/Miettinen classification depends heavily on mitotic counts, which is laborious and sometimes inconsistent between pathologists. It has also been shown to be imperfect in stratifying patients. Molecular testing is costly and time-consuming, therefore, not systematically performed in all countries. New methods to improve risk and molecular predictions are hence crucial to improve the tailoring of adjuvant therapy. We have built deep learning (DL) models on digitized HES-stained whole slide images (WSI) to predict patients’ outcome and mutations. Models were trained with a cohort of 1233 GIST and validated on an independent cohort of 286 GIST. DL models yielded comparable results to the Miettinen classification for relapse-free-survival prediction in localized GIST without adjuvant Imatinib (C-index=0.83 in cross-validation and 0.72 for independent testing). DL splitted Miettinen intermediate risk GIST into high/low-risk groups (p value = 0.002 in the training set and p value = 0.29 in the testing set). DL models achieved an area under the receiver operating characteristic curve (AUC) of 0.81, 0.91, and 0.71 for predicting mutations in KIT, PDGFRA and wild type, respectively, in cross-validation and 0.76, 0.90, and 0.55 in independent testing. Notably, PDGFRA exon18 D842V mutation, which is resistant to Imatinib, was predicted with an AUC of 0.87 and 0.90 in cross-validation and independent testing, respectively. Additionally, novel histological criteria predictive of patients’ outcome and mutations were identified by reviewing the tiles selected by the models. As a proof of concept, our study showed the possibility of implementing DL with digitized WSI and may represent a reproducible way to improve tailoring therapy and precision medicine for patients with GIST.


  • Open Access

    Comprehensive Immune Profiling Unveils a Subset of Leiomyosarcoma with “Hot” Tumor Immune Microenvironment

    by Xiaolan Feng, Laurie Tonon, Haocheng Li, Elodie Darbo, Erin Pleasance, Nicolas Macagno, Armelle Dufresne, Mehdi Brahmi, Julien Bollard, Francoise Ducimetière, et al.

    Purpose: To investigate the immune biomarker in Leiomyosarcoma (LMS), which is rare and recognized as an immune cold cancer showing a poor response rate (<10%) to immune checkpoint inhibitors (ICIs). However, durable response and clinical benefit to ICIs has been observed in a few cases of LMS, including, but not only, LMS with tertiary lymphoid structure (TLS) structures. Patients and methods: We used comprehensive transcriptomic profiling and a deconvolution method extracted from RNA-sequencing gene expression data in two independent LMS cohorts, the International Cancer Genome Consortium (ICGC, N = 146) and The Cancer Genome Atlas (TCGA, N = 75), to explore tumor immune microenvironment (TIME) in LMS. Results: Unsupervised clustering analysis using the previously validated two methods, 90-gene signature and Cell-type Identification by Estimating Relative Subsets of RNA Transcripts (CIBERSORT), identified immune hot (I-H) and immune high (I-Hi) LMS, respectively, in the ICGC cohort. Similarly, immune active groups (T-H, T-Hi) were identified in the TCGA cohort using these two methods. These immune active (“hot”) clusters were significantly associated, but not completely overlapping, with several validated immune signatures such as sarcoma immune class (SIC) classification and TLS score, T cell inflamed signature (TIS) score, immune infiltration score (IIS), and macrophage score (M1/M2), with more patients identified by our clustering as potentially immune hot. Conclusions: Comprehensive immune profiling revealed a subset of LMS with a distinct active (“hot”) TIME, consistently associated with several validated immune signatures in other cancers. This suggests that the methodologies that we used in this study warrant further validation and development, which can potentially help refine our current immune biomarkers to select the right LMS patients for ICIs in clinical trials.


  • Open Access

    Structural and Socio-Spatial Determinants Influencing Care and Survival of Patients with a Pancreatic Adenocarcinoma: Results of the PANDAURA Cohort

    by Gaël S. Roth, Yohan Fayet, Sakina Benmameche-Medjahed, Françoise Ducimetière, Amandine Charreton, Claire Cropet, Sylvie Chabaud, Anne-Marie Marion-Audibert, Olivier Berthelet, Thomas Walter, et al.

    Background and aims: Pancreatic cancer is highly lethal and often diagnosed at an advanced stage. This cohort study analyzes the impact of care pathways, delays, and socio-spatial determinants on pancreatic cancer patients’ diagnosis, treatment, and prognosis. Method: Patients with pancreatic adenocarcinoma newly diagnosed at all stages between January and June 2016 in the AuRA French region were included. The influence on survival of delays of care, healthcare centers’ expertise, and socio-spatial determinants was evaluated. Results: Here, 538 patients were included in 76 centers including 116 patients (21.8%) with resectable, 64 (12.0%) borderline-resectable, 147 (27.6%) locally-advanced tumors, and 205 (38.5%) with metastatic disease. A delay between first symptoms and CT scans did not statistically influence overall survival (OS). In resected patients, OS was significantly higher in centers with more than 20 surgeries (HR = 2.236 and HR5-20 surgeries/year = 1.215 versus centers with > 20 surgeries/year p = 0.0081). Regarding socio-spatial determinants, patients living in municipalities with greater access to a general practitioner (HR = 1.673, p = 0.0153) or with a population density below 795.1 people/km2 (HR = 1.881, p = 0.0057) were significantly more often resectable. Conclusion: This cohort study supports the pivotal role of general practitioner in cancer care and the importance of the centralization of pancreatic surgery to optimize pancreatic cancer patients’ care and outcomes. However, delays of care did not impact patient survival.


  • Open Access

    Refining Prognosis in Localized Gastrointestinal Stromal Tumor: Clinical Significance of Phosphatase and Tensin Homolog Low Expression and Gene Loss

    by Xiaolan Feng, Haocheng Li, Joanna Fourquet, Mehdi Brahmi, Armelle Dufresne, Alexandra Meurgey, Isabelle Ray-Coquard, Qing Wang, Julien Bollard, Francoise Ducimetiere, et al.

    PURPOSE: To investigate the use of PTEN biomarker to improve prognostic stratification in patients with localized gastrointestinal stromal tumor (GIST). METHODS: PTEN expression and genomic analysis were performed on two independent GIST-60 (n = 60) and GIST-100 (n = 100) cohorts, respectively. RESULTS: PTEN expression was significantly lower in patients with local and metastatic recurrent tumor compared with those with no recurrence ( P = .004). PTEN low expression was significantly associated with poor disease-free survival (DFS) compared with PTEN high expression (43.73 v 117.95 months; P = .0084) and distant metastatic-free survival (DMFS; 57.95 v 117.95 months; P = .0032). PTEN heterozygous loss was observed in approximately 10% of the patients in each cohort and was associated with poor DFS compared with patients with PTEN normal status (27.56 months v not reached [NR]; P < .001) and DMFS (27.56 months v NR; P < .001). Multivariate analysis revealed that PTEN expression was an independent clinical prognosis factor besides tumor size, mitosis index, and location (hazard ratio for DFS: 3.8; P = .033; hazard ratio for DMFS 5.7, P = .01). Furthermore, PTEN low expression was independently associated with poor DMFS in clinically high-risk patients (mDMFS: 42.28 v 65.61 months; P = .0166). In addition, PTEN heterozygous loss was independently associated with poor DMFS in patients at either low/intermediate risk (mDMFS: 18.05 months for PTEN loss v NR for PTEN normal status; P < .001) or at high risk (mDMFS: 27.19 months for PTEN loss v 105.36 months for PTEN normal status; P = .044). CONCLUSION: PTEN low expression/gene loss is an independent significant prognostic factor and a promising component to strengthen the clinical prognostic tools in patients with localized GIST.


  • Open Access

    Abstract 1939: Multimodal prediction of metastatic relapse using federated deep learning outperforms state-of-the-art methods in soft-tissue sarcoma

    by Charles Maussion, Jean-Michel Coindre, Jean-Yves Blay, Kathryn Schutte, Axel Camara, François Le Loarer, Raul Perret, Myriam Jean-Denis, Alexandra Meurgey, Françoise Ducimetiere, et al.

    Soft Tissue sarcomas (STS) are a group of heterogeneous and complex diseases. Being able to predict the appearance of metastases is key to inform clinical decisions, especially the prescription of perioperative chemotherapy. While multiple methods including multivariable nomograms have been developed to tackle this issue, we leveraged artificial intelligence (AI) to extract information from raw images with Deep-Learning. In this study, we developed SarcNet: a multimodal deep-learning algorithm based on histological slides and clinical variables to predict metastatic relapse in patients affected with limbs and trunk wall STS. Two independent series were investigated simultaneously: Centre Léon Bérard, n=221 and Institut Bergonié, n=390. This study was able to gather a sufficient number of patients in this rare disease by training all the algorithms in a multicentric fashion using Federated Learning (FL). It is one of the first real-world applications of FL in Healthcare which allows to keep patients’ data behind hospital firewalls. The SarcNet algorithm achieved unprecedented performance. By leveraging an AI-driven analysis of digitized whole slide images in addition to clinical variables, our multimodal model achieved a 0.80 AUC to predict 5-year MFS (metastatic free survival) compared to 0.77 for Sarculator (current state-of-the-art nomogram). It also greatly outperformed current standard of care practices like FNCLCC grading that reached an AUC of 0.70. Interpretability investigations of the SarcNet model highlighted histological patterns that drive the prediction of metastatic relapse such as atypia, tumoral cellularity and mitosis. This model could lead to more informed clinical decisions by identifying high-risk patients that could benefit from perioperative chemotherapy. Finally, deeper investigations of the model and the highlighted biomarkers could lead to a better understanding of metastatic relapse in STS Performance of models to predict 5y MFS (AUC evaluated in 5 repeated 5-fold cross validation) Methods Whole cohort (n=611) IB (n=390) CLB (n=221) SarcNet 0.800, 95% CI [0.786-0.814] 0.770, 95% CI [0.750-0.791] 0.842, 95% CI [0.824-0.859] Sarculator 0.767, 95% CI [0.752-0.783] 0.736, 95% CI [0.718-0.755] 0.808, 95% CI [0.785-0.831] FNCLCC (French Federation of Centers for the Fight against Cancer) grading 0.699, 95% CI [0.684-0.714] 0.723, 95% CI [0.708-0.738] 0.637, 95% CI [0.612-0.663] Citation Format: Charles Maussion, Jean-Michel Coindre, Jean-Yves Blay, Kathryn Schutte, Axel Camara, François Le Loarer, Raul Perret, Myriam Jean-Denis, Alexandra Meurgey, Françoise Ducimetiere, Antoine Giraud, Jean-Baptiste Courrèges, Clément Gautier, Etienne Bendjebbar, Jean Du Terrail, Antoine Italiano, Marie Karanian. Multimodal prediction of metastatic relapse using federated deep learning outperforms state-of-the-art methods in soft-tissue sarcoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1939.


  • Open Access

    No Geographical Inequalities in Survival for Sarcoma Patients in France: A Reference Networks’ Outcome?

    by Yohan Fayet, Christine Chevreau, Gauthier Decanter, Cécile Dalban, Pierre Meeus, Sébastien Carrère, Leila Haddag-Miliani, François Le Loarer, Sylvain Causeret, Daniel Orbach, et al.

    The national reference network NETSARC+ provides remote access to specialized diagnosis and the Multidisciplinary Tumour Board (MTB) to improve the management and survival of sarcoma patients in France. The IGéAS research program aims to assess the potential of this innovative organization to address geographical inequalities in cancer management. Using the IGéAS cohort built from the nationwide NETSARC+ database, the individual, clinical, and geographical determinants of the 3-year overall survival of sarcoma patients in France were analyzed. The survival analysis was focused on patients diagnosed in 2013 (n = 2281) to ensure sufficient hindsight to collect patient follow-up. Our study included patients with bone (16.8%), soft-tissue (69%), and visceral (14.2%) sarcomas, with a median age of 61.8 years. The overall survival was not associated with geographical variables after adjustment for individual and clinical factors. The lower survival in precarious population districts [HR 1.23, 95% CI 1.02 to 1.48] in comparison to wealthy metropolitan areas (HR = 1) found in univariable analysis was due to the worst clinical presentation at diagnosis of patients. The place of residence had no impact on sarcoma patients’ survival, in the context of the national organization driven by the reference network. Following previous findings, this suggests the ability of this organization to go through geographical barriers usually impeding the optimal management of cancer patients.