SCIENCE@home

Predicting ligand-dependent tumors from multi-dimensional signaling features

Helge Hass,Kristina Masson,Sibylle Wohlgemuth,V. Paragas,John E. Allen,M. Sevecka,E. Pace,J. Timmer,J. Stelling,G. MacBeath,B. Schoeberl,A. Raue

Published 2017 in npj Systems Biology and Applications

ABSTRACT

Targeted therapies have shown significant patient benefit in about 5–10% of solid tumors that are addicted to a single oncogene. Here, we explore the idea of ligand addiction as a driver of tumor growth. High ligand levels in tumors have been shown to be associated with impaired patient survival, but targeted therapies have not yet shown great benefit in unselected patient populations. Using an approach of applying Bagged Decision Trees (BDT) to high-dimensional signaling features derived from a computational model, we can predict ligand dependent proliferation across a set of 58 cell lines. This mechanistic, multi-pathway model that features receptor heterodimerization, was trained on seven cancer cell lines and can predict signaling across two independent cell lines by adjusting only the receptor expression levels for each cell line. Interestingly, for patient samples the predicted tumor growth response correlates with high growth factor expression in the tumor microenvironment, which argues for a co-evolution of both factors in vivo. The prediction of growth factor induced cancer cell growth was improved significantly by combining a signaling model with machine learning. A team led by Andreas Raue at Merrimack Pharmaceuticals, attempted to better understand growth factor-dependent tumors and their potential treatment with receptor-targeting antibodies. Interestingly, prediction of tumor response improved significantly by adding prior knowledge from a mechanistic signaling model. This conceptually new approach relies solely on publicly available gene expression data and can be readily applied in drug development and development of clinical trials. In patient data, correlation between growth factor expression in the tumor microenvironment and its predicted response were identified. This consolidates the belief of an addiction of tumors to growth factors abundant in the tumor microenvironment, and might enable a more robust patient stratification in the future.

PUBLICATION RECORD

CITATION MAP

EXTRACTION MAP

CLAIMS

  • No claims are published for this paper.

CONCEPTS

  • No concepts are published for this paper.

REFERENCES

Showing 1-83 of 83 references · Page 1 of 1

CITED BY

Showing 1-43 of 43 citing papers · Page 1 of 1