Abstract 3607
Background
We aimed to establish a prognostic model based on magnetic resonance imaging using deep learning to predict disease-free survival in patients with non-metastatic nasopharyngeal carcinoma.
Methods
In this retrospective, cohort study, we included 1636 patients who were diagnosed with non-metastatic nasopharyngeal carcinoma and underwent radical treatment at the Sun Yat-sen University Cancer Center. Patients from October 2010 to March 2015 were randomly divided into training cohort (n = 878) and validation cohort (n = 376); 382 patients from April 2015 to September 2015 were separated as test cohort. 3D DenseNet models learned deep representations of pre-treatment MRI and risk scores were extracted to predict PFS in the training cohort. We evaluated the accuracy of the prognostic model in validation and test cohorts. The primary endpoint was DFS, and the secondary endpoint was distant metastasis-free survival (DMFS).
Results
A series of risk scores for each patient were extracted from 3D DenseNet models, and an optimal cut-off value of risk scores was generated to classify patients into low-risk and high-risk group in the training cohort. Patients with low-risk scores had better DFS (hazard ratio [HR] 0.62, 95% CI 0.55 -0.70; p < 0.0001) and DMFS (HR 0.62, 95% CI 0.48 -0.81; p < 0.0003) than patients with low-risk scores. And we validated the prognostic accuracy of risk scores in the validation and test cohorts. In addition, patients who received concurrent chemotherapy had a poorer DFS (hazard ratio [HR] 7.79, 95% CI 1.08 -56.00; p < 0.041) compared with those who did not receive concurrent chemotherapy in low-risk group, meanwhile, patients with or without concurrent chemotherapy had similar outcomes in the high-risk group (HR 2.39, 95% CI 0.59 -9.62; p = 0.22). We also developed a nomogram based on risk scores and several clinical factors that predicted an individual’s risk of DFS.
Conclusions
MRI-based 3D DenseNet models are effective tools to learn deep representations and extract risk scores of DFS. Risk scores can be reliable prognostic factors to select which patients benefit from concurrent chemotherapy.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
The National Natural Science Foundation of China.
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
2316 - A 3D co-culture platform of breast cancer and patient derived immune cells to analyse the response to chemotherapy and immunotherapies
Presenter: Diana Saraiva
Session: Poster Display session 3
Resources:
Abstract
4290 - Characterization of the mechanism of action and efficacy of MEN1611 (PA799), a novel PI3K inhibitor, in breast cancer preclinical models.
Presenter: Alessio Fiascarelli
Session: Poster Display session 3
Resources:
Abstract
2167 - Neat-1: culprit lnRNA tying PIG-C, MSLN, CD80 in TNBC
Presenter: Nada Hussein
Session: Poster Display session 3
Resources:
Abstract
1829 - A novel RAF/MEK inhibitor CH5126766 in phase 1 clinical trial has an effectiveness in the combination with eribulin for the treatment of triple negative breast cancer
Presenter: Hisako Ono
Session: Poster Display session 3
Resources:
Abstract
4357 - Identification of a stemness-related gene panel associated with BET inhibition in triple negative breast cancer
Presenter: Eva Galan-Moya
Session: Poster Display session 3
Resources:
Abstract
5163 - Preclinical Evaluation targeting both IGF1R and IR in Triple Negative Breast Cancer
Presenter: Alex Eustace
Session: Poster Display session 3
Resources:
Abstract
832 - Monospecific antibody targeting of CDH11 inhibits epithelial-to-mesenchymal transition and represses cancer stem cell-like phenotype by up-regulating miR-335 in metastatic breast cancer, in vitro and in vivo.
Presenter: Jia-Hong Chen
Session: Poster Display session 3
Resources:
Abstract
3781 - Pharmacological screening with Chk1 inhibitors identify synergistic agents to overcome resistance to platinums in basal breast and ovarian cancer
Presenter: Ana Lucia Sanabria
Session: Poster Display session 3
Resources:
Abstract
3275 - Comparison of 11 circulating miRNAs and CA125 kinetics in ovarian cancer during first line treatment: data from the randomized CHIVA trial (a GINECO-GCIG study)
Presenter: Patrick Robelin
Session: Poster Display session 3
Resources:
Abstract
3391 - Inhibiting Ehmt2 and Ezh2 histone methyltransferases alters the immune microenvironment in a Trp53-/- murine ovarian cancer model
Presenter: Pavlina Spiliopoulou
Session: Poster Display session 3
Resources:
Abstract