Abstract 2788
Background
Deep learning (DL) is one of the best approaches to predict nonlinear behaviors from high dimensional data. Nevertheless predicting the outcome of patients affected by cancers from transcriptomic data has shown limited performance, even with DL (C-index usually <0.65). Transfer learning is a DL two-step method where a model is pre-trained for a basic task on large amount of data, and then fine-tuned on the aimed task. We hypothesized that using TL with RNAseq may improve the performances of cancer patients’ outcome estimation.
Methods
The model was a Multi-Mayer Perceptron (MLP) with 22913 inputs corresponding to genes bulk tumor whole genome RNAseq expression analysis. An important restriction was applied to the number of units at second layer (N = 100), with further linear decrease across subsequent layers. Architecture of the model (number of layers, skip connections), L1 normalization value and learning rate were optimized by grid search on 30 parallel models. Training was performed using Keras package in R. Data were split into 70% training, 15% cross validation, 15% validation for each step, without contamination between the 2 transfer learning steps. The pre-training step consisted in predicting the organs of sample origin using 17.487 public RNAseq data of normal & cancer tissues (GTEX from gtexportal.org & TCGA from cBioportal.org). Fine-tuning on patients survival used 6401 training tumors. The model’s performance on survival prediction was evaluated by C-index and the area under the survival receiver-operating characteristic curve (AUROC).
Results
The pre-training using GTEx and TCGA reached very high performance with validation accuracy of 0.96 to predict organ of origins for the best model (all models had validation accuracy > 0.9). Fine-tuning on survival, the prognostic performance of the best model on the validation cohort was C-index=0.74 and AUROC= 0.81 (80% of models had a C-index > 0.6). The best model had 8 hidden layers and a small penalization value.
Conclusions
Thanks to this original transfer learning method, we achieved a high performance to estimate cancer patients’ prognostic from whole genome expression, a classically challenging task. Learning on public databases is a valuable method of DL for personalized cancer care.
Clinical trial identification
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
E. Angevin: Advisory / Consultancy: Amgen; Advisory / Consultancy: Astellas; Advisory / Consultancy: AstraZeneca; Advisory / Consultancy: Bayer; Advisory / Consultancy: BeiGene; Advisory / Consultancy: BMS; Advisory / Consultancy: Celgene; Advisory / Consultancy: DebioPharma; Advisory / Consultancy: Genentech; Advisory / Consultancy: Ipsen; Advisory / Consultancy: Janssen; Advisory / Consultancy: Lilly; Advisory / Consultancy: MedImmune; Advisory / Consultancy: Novartis; Advisory / Consultancy: Pfizer; Advisory / Consultancy: Roche; Advisory / Consultancy: Sanofi; Advisory / Consultancy: Orion. A. Hollebecque: Advisory / Consultancy: Amgen; Advisory / Consultancy: Spectrum Pharmaceuticals; Advisory / Consultancy: Lilly; Advisory / Consultancy: Debiopharm; Travel / Accommodation / Expenses: Servier; Travel / Accommodation / Expenses: Amgen; Travel / Accommodation / Expenses: Lilly; Travel / Accommodation / Expenses: Incyte; Travel / Accommodation / Expenses: Debiopharm. E. Deutsch: Advisory / Consultancy: Boehringer; Advisory / Consultancy: Medimune; Advisory / Consultancy: Amgen; Research grant / Funding (self): AstraZeneca; Research grant / Funding (self): biotrachea; Research grant / Funding (institution): BristolMyersSquidd; Research grant / Funding (self): Clevelex; Research grant / Funding (self): EDF; Research grant / Funding (self): Lilly; Research grant / Funding (self): GlaxoSmisthKline; Research grant / Funding (self): Merk; Research grant / Funding (self): Nanobiotix; Research grant / Funding (self): Oseo; Research grant / Funding (self): Ray Search Laboratory; Research grant / Funding (self): Roche; Research grant / Funding (self): Ipsen; Research grant / Funding (self): Servier; Research grant / Funding (self): Takeda. C. Massard: Advisory / Consultancy: Amgen; Advisory / Consultancy: Astellas; Advisory / Consultancy: AstraZeneca; Advisory / Consultancy: Bayer; Advisory / Consultancy: BeiGene; Advisory / Consultancy: BMS; Advisory / Consultancy: Celgene; Advisory / Consultancy: DebioPharma; Advisory / Consultancy: Genentech; Advisory / Consultancy: Ipsen; Advisory / Consultancy: Janssen; Advisory / Consultancy: Lilly; Advisory / Consultancy: MedImmune; Advisory / Consultancy: Novartis; Advisory / Consultancy: Pfizer; Advisory / Consultancy: Roche; Advisory / Consultancy: Sanofi; Advisory / Consultancy: Orion. L. Verlingue: Research grant / Funding (self): Bristol-Myers Squibb; Advisory / Consultancy: Pierre Fabre; Advisory / Consultancy: Adaptherapy. All other authors have declared no conflicts of interest.
Resources from the same session
2600 - Atezolizumab (atezo) vs chemotherapy (chemo) in patients (pts) with platinum-treated locally advanced or metastatic urothelial carcinoma (mUC): a long-term overall survival (OS) and safety update from the Phase III IMvigor211 study
Presenter: Michiel Van der Heijden
Session: Poster Display session 3
Resources:
Abstract
3598 - Three-Year Follow-Up From the Phase 3 KEYNOTE-045 Trial: Pembrolizumab (Pembro) Versus Investigator’s Choice (Paclitaxel, Docetaxel, or Vinflunine) in Recurrent, Advanced Urothelial Cancer (UC)
Presenter: Andrea Necchi
Session: Poster Display session 3
Resources:
Abstract
2382 - First Report of Efficacy and Safety From a Phase 2 Trial of Tislelizumab, an Anti-PD-1 Antibody, for the Treatment of PD-L1+ Locally Advanced or Metastatic Urothelial Carcinoma (UC) in Asian Patients
Presenter: Dingwei Ye
Session: Poster Display session 3
Resources:
Abstract
2388 - Quality of Life of Metastatic Urothelial Cancer (mUC) Patients Treated with Enfortumab Vedotin (EV) Following Platinum-Containing Chemotherapy and a Checkpoint Inhibitor (CPI): Data from EV-201 Cohort 1
Presenter: Bradley McGregor
Session: Poster Display session 3
Resources:
Abstract
3748 - Safety and efficacy of atezolizumab (atezo) in patients (pts) with autoimmune disease (AID): subgroup analysis of the SAUL study in locally advanced/metastatic urinary tract carcinoma
Presenter: Yohann Loriot
Session: Poster Display session 3
Resources:
Abstract
1126 - Validation of the VIO prognostic index in patients with metastatic urothelial carcinoma treated with immune-checkpoint inhibitors
Presenter: Rafael Morales Barrera
Session: Poster Display session 3
Resources:
Abstract
3693 - Pathologic outcomes after neoadjuvant chemotherapy for high-risk muscle invasive bladder cancer
Presenter: Justin Matulay
Session: Poster Display session 3
Resources:
Abstract
4840 - Analysis of response to prior therapies and therapies after treatment with erdafitinib in fibroblast growth factor receptor (FGFR)-positive patients (pts) with metastatic urothelial carcinoma (mUC)
Presenter: Arlene Siefker-Radtke
Session: Poster Display session 3
Resources:
Abstract
1221 - Clinical outcomes by sex with atezolizumab (atezo) monotherapy in patients (pts) with locally advanced/metastatic urothelial carcinoma (mUC)
Presenter: Jean Hoffman-censits
Session: Poster Display session 3
Resources:
Abstract
1715 - National Small Cell Bladder Cancer Audit: Results from 26 UK institutions
Presenter: Caroline Chau
Session: Poster Display session 3
Resources:
Abstract