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
4506 - Single intravenous preoperative administration of the oncolytic virus Pexa-Vec to prime anti-tumor immunity
Presenter: Adel Samson
Session: Poster Display session 3
Resources:
Abstract
1631 - Randomized phase 2 clinical trial of NY-ESO-1 protein vaccine combined with cholesteryl pullulan (CHP-NY-ESO-1) in resected esophageal cancer patients
Presenter: Shinichi Kageyama
Session: Poster Display session 3
Resources:
Abstract
4244 - T cell repertoire sequencing reveals dynamics of response to dendritic cell vaccine plus dasatinib for checkpoint blockade resistant metastatic melanoma
Presenter: Luca Quagliata
Session: Poster Display session 3
Resources:
Abstract
5791 - Ixovex, a novel oncolytic E1B-mutated adenovirus
Presenter: Mohiemen Anwar
Session: Poster Display session 3
Resources:
Abstract
4170 - Anti-CSPG4 DNA vaccination as a promising strategy for the treatment of CSPG4+ tumors: a comparative oncology trial
Presenter: Federica Riccardo
Session: Poster Display session 3
Resources:
Abstract
5780 - Antitumor activity, immunogenicity and safety of a novel PD-1 vaccine in combination with two chimeric HER-2 peptide vaccine in syngeneic Balb/c, C57Bl/6 models and in beagle dogs
Presenter: Pravin Kaumaya
Session: Poster Display session 3
Resources:
Abstract
5860 - Maternal immunization against ALK as a weapon to fight neuroblastoma
Presenter: Giuseppina Barutello
Session: Poster Display session 3
Resources:
Abstract
4720 - Phase 1 study evaluating safety, pharmacokinetics (PK), pharmacodynamics, and preliminary efficacy of ABBV-428, first-in-class mesothelin (MSLN)-CD40 bispecific, in patients (pts) with advanced solid tumors
Presenter: Jason Luke
Session: Poster Display session 3
Resources:
Abstract
5717 - Anti-PD-L1/IL-15 fusion protein generates robust adaptive immune gene signatures in tumors leading to tumor inhibition and memory responses
Presenter: Stella Martomo
Session: Poster Display session 3
Resources:
Abstract
1802 - Evaluation of the anti-tumor efficacy and immune effects of N-809, a novel IL-15 superagonist/anti-PD-L1 bispecific agent
Presenter: Kristin Hicks
Session: Poster Display session 3
Resources:
Abstract