Download PDF

Computational And Structural Biotechnology Journal

Publication date: 2019-01-01
Volume: 17 Pages: 537 - 560
Publisher: Elsevier

Author:

Meshcheryakova, Anastasia
Svoboda, Martin ; Jaritz, Markus ; Mungenast, Felicitas ; Salzmann, Martina ; Pils, Dietmar ; Castillo-Tong, Dan Cacsire ; Hager, Gudrun ; Wolf, Andrea ; Braicu, Elena Ioana ; Sehouli, Jalid ; Lambrechts, Sandrina ; Vergote, Ignace ; Mahner, Sven ; Birner, Peter ; Zimmermann, Philip ; Brindley, David N ; Heinze, Georg ; Zeillinger, Robert ; Mechtcheriakova, Diana

Keywords:

Science & Technology, Life Sciences & Biomedicine, Biochemistry & Molecular Biology, Biotechnology & Applied Microbiology, Sphingolipid/lysophosphatidate system, On-site immune response, Patient-specific expression data sets, Integrative analysis algorithm, Patient stratification, From systems biology to systems medicine, SPHINGOSINE 1-PHOSPHATE, SPHINGOSINE-1-PHOSPHATE, ACID, EXPRESSION, CERAMIDE, GROWTH, CELLS, SPHINGOMYELINASE, AUTOTAXIN, ROLES, 0103 Numerical and Computational Mathematics, 0802 Computation Theory and Mathematics, 3101 Biochemistry and cell biology, 4601 Applied computing

Abstract:

The sphingolipid and lysophosphatidate regulatory networks impact diverse mechanisms attributed to cancer cells and the tumor immune microenvironment. Deciphering the complexity demands implementation of a holistic approach combined with higher-resolution techniques. We implemented a multi-modular integrative approach consolidating the latest accomplishments in gene expression profiling, prognostic/predictive modeling, next generation digital pathology, and systems biology for epithelial ovarian cancer. We assessed patient-specific transcriptional profiles using the sphingolipid/lysophosphatidate/immune-associated signature. This revealed novel sphingolipid/lysophosphatidate-immune gene-gene associations and distinguished tumor subtypes with immune high/low context. These were characterized by robust differences in sphingolipid-/lysophosphatidate-related checkpoints and the drug response. The analysis also nominates novel survival models for stratification of patients with CD68, LPAR3, SMPD1, PPAP2B, and SMPD2 emerging as the most prognostically important genes. Alignment of proprietary data with curated transcriptomic data from public databases across a variety of malignancies (over 600 categories; over 21,000 arrays) showed specificity for ovarian carcinoma. Our systems approach identified novel sphingolipid-lysophosphatidate-immune checkpoints and networks underlying tumor immune heterogeneity and disease outcomes. This holds great promise for delivering novel stratifying and targeting strategies.