Survival estimation in patients with stomach and esophageal carcinoma using miRNA expression profiles

Srinivasulu Yerukala Sathipati; Ming-Ju Tsai; Tonia Carter; Patrick Allaire; Sanjay K Shukla; Afshin Beheshti; Shinn-Ying Ho

doi:10.1016/j.csbj.2022.08.025

Survival estimation in patients with stomach and esophageal carcinoma using miRNA expression profiles

Comput Struct Biotechnol J. 2022 Aug 13:20:4490-4500. doi: 10.1016/j.csbj.2022.08.025. eCollection 2022.

Authors

Srinivasulu Yerukala Sathipati¹, Ming-Ju Tsai^{2

3}, Tonia Carter¹, Patrick Allaire¹, Sanjay K Shukla¹, Afshin Beheshti^{4

5}, Shinn-Ying Ho^{6

7

8}

Affiliations

¹ Center for Precision Medicine Research, Marshfield Clinic Research Institute, Marshfield, WI 54449, USA.
² Hinda and Arthur Marcus Institute for Aging Research at Hebrew Senior Life, Boston, MA, USA.
³ Department of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA, USA.
⁴ KBR, Space Biosciences Division, NASA Ames Research Center, Moffett Field, CA 94035, USA.
⁵ Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA.
⁶ Institute of Bioinformatics and Systems Biology, National Yang Ming Chiao Tung University, Hsinchu, Taiwan.
⁷ College of Health Sciences, Kaohsiung Medical University, Kaohsiung, Taiwan.
⁸ Biomedical Engineering, National Yang Ming Chiao Tung University, Hsinchu, Taiwan.

Abstract

Identifying a miRNA signature associated with survival will open a new window for developing miRNA-targeted treatment strategies in stomach and esophageal cancers (STEC). Here, using data from The Cancer Genome Atlas on 516 patients with STEC, we developed a Genetic Algorithm-based Survival Estimation method, GASE, to identify a miRNA signature that could estimate survival in patients with STEC. GASE identified 27 miRNAs as a survival miRNA signature and estimated the survival time with a mean squared correlation coefficient of 0.80 ± 0.01 and a mean absolute error of 0.44 ± 0.25 years between actual and estimated survival times, and showed a good estimation capability on an independent test cohort. The miRNAs of the signature were prioritized and analyzed to explore their roles in STEC. The diagnostic ability of the identified miRNA signature was analyzed, and identified some critical miRNAs in STEC. Further, miRNA-gene target enrichment analysis revealed the involvement of these miRNAs in various pathways, including the somatotrophic axis in mammals that involves the growth hormone and transforming growth factor beta signaling pathways, and gene ontology annotations. The identified miRNA signature provides evidence for survival-related miRNAs and their involvement in STEC, which would aid in developing miRNA-target based therapeutics.

Keywords: Machine learning; Stomach and esophageal carcinoma; Survival estimation; miRNA signature.