Prediction of Drug Approval After Phase I Clinical Trials in Oncology -RESOLVED2 - Institut Gustave Roussy Accéder directement au contenu
Article Dans Une Revue JCO Clinical Cancer Informatics Année : 2019

Prediction of Drug Approval After Phase I Clinical Trials in Oncology -RESOLVED2

Résumé

Research support for the study: none. Training fees (VT, VC) were covered by the DITEP department. Part of this study will be presented (poster communication) at the AACR 2019 Congress. Disclaimers: none. Key Points *Key objective: we aimed to evaluate the feasibility and utility of a machine-learning recommender system to predict drug development outcome in Oncology and therefore to support early go/no-go decision as soon as phase I trials completion. *Knowledge generated: RESOLVED2 is a Lasso-penalized Cox regression model. To train RESOLVED2, we developed a new metric, namely Food and Drug Administration approval free survival (FDA-aFS), defined by the time between publication of the first early clinical trial (ECT) reporting clinical effect of a drug, and FDA approval, censored by date of last news. From simple pharmacological data and ECT's PubMed abstract, RESOLVED2 can accurately predict the time to FDA approval of a new antineoplastic agent. *Relevance: our work demonstrates machine learning approaches can enhance drug development in Oncology by supporting early go/no-go decisions.
Fichier principal
Vignette du fichier
RESOLVED2 - 08 - MANUSCRIPT_VF_tracked changes_V2.pdf (486.99 Ko) Télécharger le fichier
RESOLVED2 - 11 - Table 1 - drugs & pubmed data.docx (13.87 Ko) Télécharger le fichier
RESOLVED2 - 11 - Table 2 - prediction on examples.pdf (12.62 Ko) Télécharger le fichier
RESOLVED2 - 13 - Figure 2 - FDA_FS~prediction on test & training setV2.jpg (1.81 Mo) Télécharger le fichier
RESOLVED2 - Figure 1 - FLOW CHART DRUGS.pdf (181.23 Ko) Télécharger le fichier
RESOLVED2 - Figure 2 - model.png (51.95 Ko) Télécharger le fichier
RESOLVED2 - Figure 3 - FDA_AFS.png (43.31 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-02894327 , version 1 (08-07-2020)

Identifiants

Citer

Guillaume Beinse, Virgile Tellier, Valentin Charvet, Eric Deutsch, Isabelle Borget, et al.. Prediction of Drug Approval After Phase I Clinical Trials in Oncology -RESOLVED2. JCO Clinical Cancer Informatics, 2019, 3, pp.1-10. ⟨10.1200/CCI.19.00023⟩. ⟨hal-02894327⟩
70 Consultations
878 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More