Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Mini Oral session 2

193MO - Development of a Deep Learning model using a large real-world database to predict overall survival in patients with metastatic breast cancer (MBC)

Date

12 May 2023

Session

Mini Oral session 2

Topics

Tumour Site

Breast Cancer

Presenters

Laura Vuduc

Citation

Annals of Oncology (2023) 8 (1suppl_4): 101223-101223. 10.1016/esmoop/esmoop101223

Authors

L. Vuduc1, W. Jacot2, J. Frenel3, E.G.C. Brain4, V.C. Dieras5, T. Bachelot6, A. Mailliez7, F. Dalenc8, P.H. Cottu9, M. Arnedos10, C. Lefeuvre-Plesse11, A. Gonçalves12, T. Grinda13, A. Antoine6, M. Chevrot14, D. Perol6, P. Cournede15, S. Delaloge16

Author affiliations

  • 1 CentraleSupélec - Paris-Saclay campus, Gif sur Yvette/FR
  • 2 ICM - Institut du Cancer de Montpellier, Montpellier, Cedex/FR
  • 3 ICO Institut de Cancerologie de l'Ouest René Gauducheau, Saint-Herblain/FR
  • 4 Hopital René Huguenin - Institut Curie, Saint-Cloud/FR
  • 5 Centre Eugene - Marquis, Rennes/FR
  • 6 Centre Léon Bérard, Lyon/FR
  • 7 Centre Oscar Lambret, Lille/FR
  • 8 Centre Claudius-Regaud, 31059 - Toulouse/FR
  • 9 Institut Curie, Paris/FR
  • 10 Institute Bergonié, Bordeaux/FR
  • 11 Centre Eugène Marquis, 35042 - Rennes/FR
  • 12 Institut Paoli-Calmettes, Marseille/FR
  • 13 Institut Gustave Roussy, Villejuif, Cedex/FR
  • 14 Unicancer, 75654 - Paris/FR
  • 15 Université Paris-Saclay, CentraleSupélec, Gif sur Yvette/FR
  • 16 Institut Gustave Roussy, Villejuif/FR

Resources

This content is available to ESMO members and event participants.

Abstract 193MO

Background

Despite recent improvements in MBC management, sensitivity to treatment and disease outcomes remain heterogeneous. We developed a deep learning and user-friendly model to predict individual outcomes and select the most adequate treatments according to the course of the disease.

Methods

We used data from the Unicancer large national multicenter real-world Epidemiological Strategy and Medical Economics (ESME) database (NCT03275311) that included 27,855 women diagnosed between 2008 and 2020 with MBC. The primary outcome was overall survival (OS). To measure this outcome and follow the disease, treatment line (TL) initiation dates were used as essential time points, allowing us to build a dynamic deep learning survival model. The main structure of this model is the extension of a Long-Short Term Memory cell, a recurrent neural network designed to handle sequential data with irregular timing. The time-dependent concordance index (C-index) and the Integrated Brier Score (IBS) allowed the selection of the best models.

Results

The cohort included 4,857 patients with triple-negative (TN), 5,027 with HER2+ and 17,971 with hormonal receptor positive (HR+)/HER2- MBC. Median follow-up was 65.1 months (95%CI 63.8, 66.4). The areas under the receiving operator characteristics (AUCs) for 6-month and 1-year OS prediction, the sensitivity at 90% specificity for 6-month OS prediction, as well as the global C-index and IBS of the model per subtype, averaged across 5-fold cross-validation cohorts, are reported in the table for the three first TLs. Table: 193MO

Average metrics for our model per subtype for the three first TLs

Subtype Initial date: TL initiation AUC for 6-month OS Sensitivity at 90% specificity for 6-month OS AUC for 1-year OS Global C-index Global IBS
TN TL1 0.76 0.40 0.76 0.69 0.14
TN TL2 0.76 0.38 0.76 0.69 0.10
TN TL3 0.75 0.40 0.76 0.69 0.08
HER2+ TL1 0.80 0.44 0.79 0.72 0.16
HER2+ TL2 0.78 0.39 0.78 0.71 0.17
HER2+ TL3 0.77 0.39 0.76 0.70 0.16
HR+/HER2- TL1 0.81 0.48 0.79 0.70 0.16
HR+/HER2- TL2 0.78 0.42 0.77 0.70 0.16
HR+/HER2- TL3 0.78 0.42 0.77 0.71 0.14

Conclusions

Our models showed promising results in predicting OS for MBC patients of each subtype at different time points. The ability to anticipate early failure and death despite adjusted standard treatment is important for identifying patients requiring a more innovative investigation. The strong prognostic capacity of our deep learning survival model could inform treatment selection for MBC patients.

Clinical trial identification

NCT03275311.

Legal entity responsible for the study

Unicancer.

Funding

The used database is the ESME MBC database, which receives financial support from an industrial consortium (Roche, Pfizer, AstraZeneca, MSD, Eisai, and Daiichi Sankyo). Data collection, analysis, and publication are managed entirely by Unicancer independently of the industrial consortium. The first author received fundings by the doctoral school Interfaces of the Université Paris-Saclay.

Disclosure

W. Jacot: Financial Interests, Personal, Advisory Board: AstraZeneca, Eisai, Novartis, Roche, Pfizer, Eli Lilly, MSD, BMS, Chugai, Seagen, Daiichi Sankyo; Financial Interests, Personal, Invited Speaker: AstraZeneca, Pfizer, Seagen, Daiichi Sankyo; Financial Interests, Institutional, Research Grant: AstraZeneca, Daiichi Sankyo; Financial Interests, Invited Speaker: Roche, Novartis, Daiichi Sankyo, Daiichi Sankyo. J. Frenel: Financial Interests, Personal, Advisory Board: Pfizer, Novocure, Pierre Fabre, Eisai, Seagen, Gilead; Financial Interests, Personal, Invited Speaker: GSK, Amgen; Financial Interests, Institutional, Advisory Board: Exactscience, Lilly, Daiichi Sankyo, AstraZeneca, Clovis Oncology; Financial Interests, Institutional, Invited Speaker: Novartis, MSD; Financial Interests, Invited Speaker: AstraZeneca, Seagen, MSD, Daiichi; Non-Financial Interests, Principal Investigator: Novartis, Lilly, AstraZeneca, Pfizer, Daiichi, MSD. E.G.C. Brain: Financial Interests, Personal, Invited Speaker, Webinars, optimized endocrine therapy for older breast cancer patients: Eli Lilly; Financial Interests, Personal, Advisory Board, Palbociclib and older breast cancer patients: Pfizer; Financial Interests, Personal, Invited Speaker, Symposium HER2+ MAO conference 03/21: Seagen; Financial Interests, Personal, Invited Speaker, ELEVATE 10/2021 and ABC 11/2021 meeetings: Pfizer; Financial Interests, Personal, Other, IDMC DESTINY 05: DAIICHI; Financial Interests, Personal, Advisory Board, GCSF and FN in older patients: Sandoz; Financial Interests, Personal, Advisory Board, Underserved Patients Populations with breast cancer, advisory board: Pfizer; Financial Interests, Personal, Invited Speaker, Management of older patients with cancer, series of seminars in Canada for HCP: Pfizer; Financial Interests, Institutional, Invited Speaker, APPALACHES study EORTC 1745: Pfizer; Financial Interests, Institutional, Invited Speaker, TOUCH study (IBCSG 55/GERICO study): Pfizer; Financial Interests, Institutional, Invited Speaker, DEESTINY 09: DAIICHI; Financial Interests, Institutional, Invited Speaker, DESTINY 06: DAIICHI; Financial Interests, Institutional, Invited Speaker, SERENA 06: AstraZeneca; Financial Interests, Institutional, Invited Speaker, AMEERA 6: Sanofi. V.C. Dieras: Financial Interests, Personal, Advisory Board, National advisory board: Pierre Fabre Oncologie; Financial Interests, Personal, Advisory Board, Steering Committee, consultant, Symposium, travel expenses: Roche Genentech; Financial Interests, Personal, Advisory Board, + Symposia and travel expenses: Novartis; Financial Interests, Personal, Advisory Board, Advisory boards, symposia, travel expenses: Pfizer; Financial Interests, Personal, Advisory Board, Symposia, travel expenses: Lilly, AstraZeneca, MSD; Financial Interests, Personal, Advisory Board, Symposia,travel expenses: Daiichi Sankyo; Financial Interests, Personal, Advisory Board, symposia,travel expenses: Seagen, Gilead; Financial Interests, Personal, Advisory Board, Steering Committee: AbbVie; Financial Interests, Personal, Advisory Board: Eisai; Financial Interests, Personal, Other, IDMC: Sanofi; Financial Interests, Personal and Institutional, Other, IDMC: Sanofi; Financial Interests, Institutional, Invited Speaker: Roche Genentech, AstraZeneca; Financial Interests, Institutional, Invited Speaker, Steering Committee: Lilly; Financial Interests, Institutional, Invited Speaker, + IDMC: Daiichi Sankyo; Financial Interests, Institutional, Invited Speaker, PI: Seagen. T. Bachelot: Financial Interests, Personal, Advisory Board: Roche, Novartis, AstraZeneca, Pfizer, Seagen, Daiichi Sankyo; Financial Interests, Institutional, Research Grant: Novartis, Roche, AstraZeneca, Seagen, Pfizer; Financial Interests, Personal, Invited Speaker: Roche; Financial Interests, Institutional, Invited Speaker: AstraZeneca. A. Mailliez: Financial Interests, Institutional, Invited Speaker: AstraZeneca, MSD, Novartis, Roche; Financial Interests, Personal, Invited Speaker: Pierre Fabre, Oseus, Seagen, Pfizer, Daiichi Sankyo; Financial Interests, Personal, Expert Testimony: GSK. F. Dalenc: Non-Financial Interests, Principal Investigator: Roche, AstraZeneca, Gilead, Novartis. P.H. Cottu: Financial Interests, Personal, Advisory Board: Pfizer; Financial Interests, Personal, Invited Speaker: Pfizer, Lilly; Financial Interests, Personal, Expert Testimony: Roche; Financial Interests, Institutional, Invited Speaker: Daiichi, Lilly, Gilead; Financial Interests, Institutional, Funding: Novartis. A. Antoine: Financial Interests, Personal, Research Grant: Roche. M. Chevrot: Financial Interests, Institutional, Funding: Roche, Pfizer, AstraZeneca, MSD, Eisai, Daiichi Sankyo. D. Perol: Financial Interests, Personal, Training: Roche, AstraZeneca, Bayer, Boehringher-Ingelheim, Bristol Myers Squibb, Daiichi Sankyo, Eli Lilly, Ipsen, Novartis, Merck sharp and Dohme, Pfizer; Financial Interests, Personal, Advisory Role: Takeda. S. Delaloge: Financial Interests, Institutional, Advisory Board: AstraZeneca, Novartis, Pierre Fabre, Orion, Sanofi, Rappta, Cellectis, Isis/servier; Financial Interests, Institutional, Invited Speaker: Exact Sciences, Pfizer, Seagen, Lilly, AstraZeneca, MSD, Roche Genentech, BMS, Puma, AstraZeneca, Orion, Sanofi, Pfizer; Financial Interests, Institutional, Advisory Board, ad board: Besins Healthcare; Financial Interests, Institutional, Invited Speaker, ESMO symposium: Gilead; Financial Interests, Institutional, Advisory Board, scientific board: Elsan; Financial Interests, Institutional, Funding: GE; Financial Interests, Institutional, Invited Speaker, clinical research funding to my institution: Taiho; Non-Financial Interests, Invited Speaker, Société Française de Sénologie et Pathologie Mammaire: SFSPM; Non-Financial Interests, Principal Investigator, H2020 funding: European Commission. All other authors have declared no conflicts of interest.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.