Guillaume Etchepare
Research Subjects
Guillaume Etchepare has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Loïc Ferrer
UMR1219 INSERM Univ. Bordeaux ISPED F-33076 Bordeaux France
7 shared publications - Amanda Remorino
10AstraZeneca, Barcelona, Spain
6 shared publications - Ammar Ahmed Chaudhry
AstraZeneca, Gaithersburg, Maryland
6 shared publications - David Dellamonica
12AstraZeneca, Baar, Switzerland
6 shared publications - Edward Brian Garon
Department of Medicine, David Geffen School of Medicine at UCLA, Los Angeles, USA
6 shared publications - Ferdinandos Skoulidis
Department of Thoracic/Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
6 shared publications
Latest Publications
- Open Access
- 1 October 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S1 from Utilizing Machine Learning to Identify Multimodal Signatures for Patients Who Would Benefit from the Addition of Tremelimumab to Durvalumab and Chemotherapy (TRIDENT)
Supplementary Figure 1. Super T-learner architecture. The active arm refers to treatment with tremelimumab plus durvalumab and chemotherapy while the control arm refers to durvalumab plus chemotherapy
- 1 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S1 from Utilizing Machine Learning to Identify Multimodal Signatures for Patients Who Would Benefit from the Addition of Tremelimumab to Durvalumab and Chemotherapy (TRIDENT)
Supplementary Figure 1. Super T-learner architecture. The active arm refers to treatment with tremelimumab plus durvalumab and chemotherapy while the control arm refers to durvalumab plus chemotherapy
- Open Access
- 1 October 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S2 from Utilizing Machine Learning to Identify Multimodal Signatures for Patients Who Would Benefit from the Addition of Tremelimumab to Durvalumab and Chemotherapy (TRIDENT)
Supplementary Figure 2. Overall survival in patient subgroups who received tremelimumab plus durvalumab and chemotherapy or durvalumab plus chemotherapy (analysis set 1).
- 1 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S2 from Utilizing Machine Learning to Identify Multimodal Signatures for Patients Who Would Benefit from the Addition of Tremelimumab to Durvalumab and Chemotherapy (TRIDENT)
Supplementary Figure 2. Overall survival in patient subgroups who received tremelimumab plus durvalumab and chemotherapy or durvalumab plus chemotherapy (analysis set 1).
- Open Access
- 1 October 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S3 from Utilizing Machine Learning to Identify Multimodal Signatures for Patients Who Would Benefit from the Addition of Tremelimumab to Durvalumab and Chemotherapy (TRIDENT)
Supplementary Figure 3. Partial dependence plots (PDPs) for (A) Model 4, analysis set 2 (clinical and radiomics); (B) Model 7, analysis set 3 (clinical and genomics); (C) Model 19, analysis set 6 (clinical and genomics, non-squamous patients only). The PDPs show the predicted CATE against the selected feature, by averaging the model's predictions over all patients with that feature set to a specific value and all other features held constant. Panel A shows PDPs for the top 5 features in Figure 3A, panel 3. Panel B shows PDPs for the top 5 features in Figure 3B, panel 3. Panel C shows PDPs for the top 5 features in Figure 3C, panel 3.
- 1 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S3 from Utilizing Machine Learning to Identify Multimodal Signatures for Patients Who Would Benefit from the Addition of Tremelimumab to Durvalumab and Chemotherapy (TRIDENT)
Supplementary Figure 3. Partial dependence plots (PDPs) for (A) Model 4, analysis set 2 (clinical and radiomics); (B) Model 7, analysis set 3 (clinical and genomics); (C) Model 19, analysis set 6 (clinical and genomics, non-squamous patients only). The PDPs show the predicted CATE against the selected feature, by averaging the model's predictions over all patients with that feature set to a specific value and all other features held constant. Panel A shows PDPs for the top 5 features in Figure 3A, panel 3. Panel B shows PDPs for the top 5 features in Figure 3B, panel 3. Panel C shows PDPs for the top 5 features in Figure 3C, panel 3.
- Open Access
- 1 October 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Tables S1-S11 from Utilizing Machine Learning to Identify Multimodal Signatures for Patients Who Would Benefit from the Addition of Tremelimumab to Durvalumab and Chemotherapy (TRIDENT)
Supplementary Table 1. Time-to-event models’ hyperparameter space for Bayesian optimization. Supplementary Table 2. Baseline characteristics for patients included in analysis set 2. Supplementary Table 3. Baseline characteristics for patients included in analysis set 3. Supplementary Table 4. Baseline characteristics for patients included in analysis set 4. Supplementary Table 5. Baseline characteristics for patients included in analysis set 5. Supplementary Table 6. Baseline characteristics for patients included in analysis set 6. Supplementary Table 7. Distribution of missing values for all clinical features. Supplementary Table 8. Representativeness of study participants Supplementary Table 9. Integrated AUC between 0 and 24 months for all numerical features (analysis set 1) Supplementary Table 10. Wald test P-values for all categorical features, ranked according to P-values in the tremelimumab plus durvalumab and chemotherapy arm (analysis set 1) Supplementary Table 11. Group statistics for the 50% of patients with higher/lower treatment benefit (from addition of tremelimumab to durvalumab and chemotherapy) identified by the best performing model (Model 7) with deep dive in STK11, KRAS, and KEAP1 mutations.
- 1 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Tables S1-S11 from Utilizing Machine Learning to Identify Multimodal Signatures for Patients Who Would Benefit from the Addition of Tremelimumab to Durvalumab and Chemotherapy (TRIDENT)
Supplementary Table 1. Time-to-event models’ hyperparameter space for Bayesian optimization. Supplementary Table 2. Baseline characteristics for patients included in analysis set 2. Supplementary Table 3. Baseline characteristics for patients included in analysis set 3. Supplementary Table 4. Baseline characteristics for patients included in analysis set 4. Supplementary Table 5. Baseline characteristics for patients included in analysis set 5. Supplementary Table 6. Baseline characteristics for patients included in analysis set 6. Supplementary Table 7. Distribution of missing values for all clinical features. Supplementary Table 8. Representativeness of study participants Supplementary Table 9. Integrated AUC between 0 and 24 months for all numerical features (analysis set 1) Supplementary Table 10. Wald test P-values for all categorical features, ranked according to P-values in the tremelimumab plus durvalumab and chemotherapy arm (analysis set 1) Supplementary Table 11. Group statistics for the 50% of patients with higher/lower treatment benefit (from addition of tremelimumab to durvalumab and chemotherapy) identified by the best performing model (Model 7) with deep dive in STK11, KRAS, and KEAP1 mutations.
- Open Access
- 22 July 2026
- Published by American Association for Cancer Research (AACR) in Clinical Cancer Research
- Vol. 32 (19) , 4461-4473
Utilizing Machine Learning to Identify Multimodal Signatures for Patients Who Would Benefit from the Addition of Tremelimumab to Durvalumab and Chemotherapy (TRIDENT)
Purpose: POSEIDON (NCT03164616) was a randomized, open-label, multicenter phase III trial comparing first-line durvalumab with or without tremelimumab in combination with chemotherapy versus chemotherapy alone in patients with metastatic non–small cell lung cancer (NSCLC). Overall survival (OS) and progression-free survival were significantly increased in the tremelimumab plus durvalumab and chemotherapy arm. We conducted a post hoc analysis (TRIDENT) to identify patients who may receive greater OS benefit from the addition of tremelimumab to durvalumab and chemotherapy. Experimental Design: This analysis included clinical, genomic, and radiomic data from the POSEIDON trial (data cutoff March 12, 2021). Machine learning models leveraging multimodal data were trained to identify subpopulations of patients who benefit from the addition of tremelimumab to first-line durvalumab and chemotherapy. Results: Using clinical and genomic data, the model was able to predict treatment benefit from adding tremelimumab to first-line durvalumab and chemotherapy, with the top ranked 50% of patients with nonsquamous tumors achieving a hazard ratio of 0.56 (95% confidence interval, 0.33–0.97). EGFR wild type, FGFR3 wild type, CDKN2A wild type, KRAS mutations, and STK11 mutations were the factors most associated with higher OS benefit. Conclusions: By utilizing machine learning models to analyze POSEIDON data, we yielded genetic signatures identifying patients with nonsquamous metastatic NSCLC who may derive greater OS benefit from the addition of tremelimumab to first-line durvalumab and chemotherapy. Such approaches could be used in the future to enhance precision in tailoring therapies for individual patients.
- 22 July 2026
- Published by American Association for Cancer Research (AACR) in Clinical Cancer Research
- Vol. 32 (19) , 4461-4473
Open AccessUtilizing Machine Learning to Identify Multimodal Signatures for Patients Who Would Benefit from the Addition of Tremelimumab to Durvalumab and Chemotherapy (TRIDENT)
Purpose: POSEIDON (NCT03164616) was a randomized, open-label, multicenter phase III trial comparing first-line durvalumab with or without tremelimumab in combination with chemotherapy versus chemotherapy alone in patients with metastatic non–small cell lung cancer (NSCLC). Overall survival (OS) and progression-free survival were significantly increased in the tremelimumab plus durvalumab and chemotherapy arm. We conducted a post hoc analysis (TRIDENT) to identify patients who may receive greater OS benefit from the addition of tremelimumab to durvalumab and chemotherapy. Experimental Design: This analysis included clinical, genomic, and radiomic data from the POSEIDON trial (data cutoff March 12, 2021). Machine learning models leveraging multimodal data were trained to identify subpopulations of patients who benefit from the addition of tremelimumab to first-line durvalumab and chemotherapy. Results: Using clinical and genomic data, the model was able to predict treatment benefit from adding tremelimumab to first-line durvalumab and chemotherapy, with the top ranked 50% of patients with nonsquamous tumors achieving a hazard ratio of 0.56 (95% confidence interval, 0.33–0.97). EGFR wild type, FGFR3 wild type, CDKN2A wild type, KRAS mutations, and STK11 mutations were the factors most associated with higher OS benefit. Conclusions: By utilizing machine learning models to analyze POSEIDON data, we yielded genetic signatures identifying patients with nonsquamous metastatic NSCLC who may derive greater OS benefit from the addition of tremelimumab to first-line durvalumab and chemotherapy. Such approaches could be used in the future to enhance precision in tailoring therapies for individual patients.
- Open Access
- 23 February 2024
- Published by Springer Nature in npj Precision Oncology
- Vol. 8 (1) , 1-10
UroPredict: Machine learning model on real-world data for prediction of kidney cancer recurrence (UroCCR-120)
Renal cell carcinoma (RCC) is most often diagnosed at a localized stage, where surgery is the standard of care. Existing prognostic scores provide moderate predictive performance, leading to challenges in establishing follow-up recommendations after surgery and in selecting patients who could benefit from adjuvant therapy. In this study, we developed a model for individual postoperative disease-free survival (DFS) prediction using machine learning (ML) on real-world prospective data. Using the French kidney cancer research network database, UroCCR, we analyzed a cohort of surgically treated RCC patients. Participating sites were randomly assigned to either the training or testing cohort, and several ML models were trained on the training dataset. The predictive performance of the best ML model was then evaluated on the test dataset and compared with the usual risk scores. In total, 3372 patients were included, with a median follow-up of 30 months. The best results in predicting DFS were achieved using Cox PH models that included 24 variables, resulting in an iAUC of 0.81 [IC95% 0.77–0.85]. The ML model surpassed the predictive performance of the most commonly used risk scores while handling incomplete data in predictors. Lastly, patients were stratified into four prognostic groups with good discrimination (iAUC = 0.79 [IC95% 0.74–0.83]). Our study suggests that applying ML to real-world prospective data from patients undergoing surgery for localized or locally advanced RCC can provide accurate individual DFS prediction, outperforming traditional prognostic scores.
- 23 February 2024
- Published by Springer Nature in npj Precision Oncology
- Vol. 8 (1) , 1-10
Open AccessUroPredict: Machine learning model on real-world data for prediction of kidney cancer recurrence (UroCCR-120)
Renal cell carcinoma (RCC) is most often diagnosed at a localized stage, where surgery is the standard of care. Existing prognostic scores provide moderate predictive performance, leading to challenges in establishing follow-up recommendations after surgery and in selecting patients who could benefit from adjuvant therapy. In this study, we developed a model for individual postoperative disease-free survival (DFS) prediction using machine learning (ML) on real-world prospective data. Using the French kidney cancer research network database, UroCCR, we analyzed a cohort of surgically treated RCC patients. Participating sites were randomly assigned to either the training or testing cohort, and several ML models were trained on the training dataset. The predictive performance of the best ML model was then evaluated on the test dataset and compared with the usual risk scores. In total, 3372 patients were included, with a median follow-up of 30 months. The best results in predicting DFS were achieved using Cox PH models that included 24 variables, resulting in an iAUC of 0.81 [IC95% 0.77–0.85]. The ML model surpassed the predictive performance of the most commonly used risk scores while handling incomplete data in predictors. Lastly, patients were stratified into four prognostic groups with good discrimination (iAUC = 0.79 [IC95% 0.74–0.83]). Our study suggests that applying ML to real-world prospective data from patients undergoing surgery for localized or locally advanced RCC can provide accurate individual DFS prediction, outperforming traditional prognostic scores.