AUTHOR=Zhao Xuefei , Xia Xia , Wang Xinyue , Bai Mingze , Zhan Dongdong , Shu Kunxian 

TITLE=Deep Learning-Based Protein Features Predict Overall Survival and Chemotherapy Benefit in Gastric Cancer

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.847706

DOI=10.3389/fonc.2022.847706

ISSN=2234-943X

ABSTRACT=Gastric cancer (GC) is one of the most common malignant tumors with a high mortality rate worldwide, and lacks effective methods for prognosis prediction. Postoperative adjuvant chemotherapy is first-line treatment for advanced gastric cancer, but only a subgroup of patients benefits from it. Here, we used 833 formalin-fixed, paraffin-embedded (FFPE) surgically resected tumor samples from patients with TNM stage II/III GC, and established a proteomic subtyping workflow using 100 deep-learned features. Two proteomic subtypes (S-I and S-II) with overall survival (OS) differences were identified. S-I has a better survival, and is sensitive to chemotherapy. Patients in the S-I who received adjuvant chemotherapy had a significant improvement in 5-year OS rate compared with patients who received surgery alone (65.3% vs 52.6%; log-rank P=0.014), but no improvement was observed in S-II (54% vs 51%; log-rank P=0.96). These results were verified in an independent validation set. Further, we also evaluated the superiority and scalability of the deep learning (DL)-based workflow in cancer molecular subtyping, exhibiting its great utility and potential in prognosis prediction and therapeutic decision making.