INTRODUCING TEMPUS NEXT: AI-ENABLED CARE PATHWAY INTELLIGENCE /// EXPLORE NOW INTRODUCING TEMPUS NEXT: AI-ENABLED CARE PATHWAY INTELLIGENCE ///

RNA Pathway Enrichment in Serum-Based Mass Spectroscopy Prognostic Analyses

PUBLICATIONS

05/13/2020

ASCO Annual Meeting 2020, Tempus-acknowledged —

Background: Veristrat (VS) good and poor labels were found to be prognostic in the front-line NSCLC setting with or without immune checkpoint inhibitors. Little is known about RNA expression data corresponding to these good and poor prognostic labels.

Methods: Raw RNA seq expression data was obtained from TEMPUS XP RNA analysis. Sequences were filtered for allowing only genes that had an overall count of at least 20 across all samples. Samples were then processed through the EdgeR analysis in the R programming environment. Along with a Z-scored heatmap, PCA plot was also generated. Differential abundance analysis was done further filter out the genes. Benjamini-Hochberg correction was done and only genes with corrected p-value of < 0.05 were selected for further pathway analysis. 242 genes with corrected p-value < 0.05 were used. Pathway analysis was performed using the Broad Institute GSEA (Gene Set Enrichment Analysis) and DAVID (Database for Annotation, Visualization and Integrated Discovery). The gene list was ran using KEGG database in GSEA and using KEGG, BioCarta and Reactome pathway databases. Functional analyses against the KEGG and REACTOME database were reported out to determine the enriched pathways in the VS poor and VS good groups. Kaplan-Meier and Logrank were used to compare overall survival in the poor vs. good label cohorts of patients.

Results: 26 patients had VS label and RNA expression data available. Nine patients had VS poor label and 17 had veristrat good label. 6(66%) VS poor 13(76%) VS good patients received anti PD1 agents during the course of their therapy. Overall survival was not significantly different between good and poor label groups in this small series. 242 genes had significantly different abundance in the VS poor vs good patients with the majority (211) over expressed in the poor label patients. Significant pathways enriched in the VS poor group included the intrinsic pathway of fibrin clot formation, platelet degranulation (DAVID), steroid hormone biosynthesis and dilated cardiomyopathy (KEGG). Systemic lupus erythematous pathway was enriched in the VS good group (KEGG).

Conclusions: In this small set of front-line patients with RNA pathway analyses and known VS poor or good labels, differential expression of acute phase reactants was found in VS poor patients. DAVID and KEGG pathways were differentially enriched and may reveal potential targets to mitigate prognosis in poor risk NSCLC.

View the full publication here.

Authors: Benjamin D German, Jeffrey Allen Borgia, Ankur D Naqib, Philip D. Bonomi, Marta Batus, Sanjib Basu, Mary J. Fidler