A novel transfer-learning based physician-level general and subtype classifier for non-small cell lung cancer.

Confirming histological patterns of lung carcinoma is important for determining the prognosis and the next steps of treatment for a patient. Confirming the histologic patterns (subtype) of lung adenocarcinoma is important for determining the prognosis and treatment options for a patient. The task is challenging, and often requires the input of experienced pathologists, who by themselves lack interobserver concordance. A computer-aided diagnosis holds the potential to accelerate the time to diagnosis. As many adenocarcinoma tissue samples contain multiple histologic patterns, accurate computer-aided diagnosis requires annotations manually labeled by pathologists. We propose a method that merges weak supervised learning and Integrated Learning using Transfer Learning using two datasets: The Cancer Genome Atlas (TCGA), and the Clinical Proteomic Tumor Analysis Consortium (CPTAC) to reduce the need for manual annotation by a pathologist while maintaining accuracy. Whole-slide images (WSI) are first determined to be either adenocarcinoma or squamous cell carcinoma, then further identify the subtypes by generating weak classifiers for each subtype, then using integrated learning to create a strong classifier. Our model was evaluated with independent datasets from the CPTAC dataset and a dataset from a private hospital. It can achieve AUC values of 0.86, 0.91, 0.82, 0.77, 0.96, 0.98 in Acinar, LPA, Micropapillary, Papillary, Solid, and Normal, respectively.

Highly Expressing SCARA5 Promotes Proliferation and Migration of Esophageal Squamous Cell Carcinoma.

Background: Thrombospondin type 1 domain-containing 7A (THSD7A) was reported to play a procancer role in esophageal squamous cell carcinoma (ESCC). The aim of the study was to screen the downstream functional genes of THSD7A and explore their functions in ESCC, based on the reported research into THSD7A function and on gene microarrays. Methods: We adopted quantitative reverse-transcription polymerase chain reaction (qRT-PCR) and Celigo high-content screening (HCS) technology to screen the downstream genes of THSD7A. The expression level of target genes was examined by PCR, western blot, and immunohistochemistry (IHC). The effects of these target genes on ESCC malignant biological behavior were performed in vivo and in vitro. The Kaplan-Meier (K-M) survival analysis and Cox regression were used to analyze the prognostic significance of target genes in ESCC patients. Experiments in the literature on liver cancer (LC) were repeated to verify the functions of these genes in different tumors. We further explored the cancer-promoting mechanism of target genes in ESCC by sequencing of the genes' exons. Results: Scavenger receptor class A member 5 (SCARA5) was proved to be the downstream driving gene of THSD7A. SCARA5 promoted cell proliferation and migration but inhibited apoptosis in ESCC. IHC results confirmed that SCARA5 expression in ESCC exceeded that in normal tissues. The K-M survival analysis indicated that SCARA5 expression quantity was not related to prognosis, but tumor volume and T classification were both the independent prognostic factors. Repetition of experiments in LC in the literature confirmed that SCARA5 had exactly opposite functions in EC and LC. Conclusion: SCARA5 was related to the development and occurrence of ESCC. Our findings suggested that it was a potentially diagnostic individualized therapeutic target for ESCC in the future and that its application could possibly be combined with that of upstream THSD7A gene.