In:
eLife, eLife Sciences Publications, Ltd, Vol. 3 ( 2014-11-21)
Abstract:
Cancer is not just one disease, but a collection of disorders; as such there is no single general treatment that is effective against all cancers. Different tissues and organs—including the lungs, skin, and kidneys—can get cancer, and each need different treatments. Even two patients with the same type of cancer might respond differently to the same treatment. Being able to distinguish between different cancer types would help doctors personalize a patient's cancer therapy—which would hopefully improve the outcome of the treatment. An important step in developing such personalized treatments is to find out how each type of cancer cell behaves and to see how this behavior differs both from normal, healthy cells and other types of cancer. Countless chemical reactions take place inside living cells, and these reactions essentially dictate how a cell will grow and behave. The chemical reactions occurring inside a cancerous cell can be described as its ‘metabolic phenotype’ and will likely be different to the chemical reactions occurring in a healthy cell. Now Yizhak, Gaude et al. have used a range of data, including gene expression data, to create computer models of the metabolic phenotypes of 60 different types of human cancer cell. The same approach was also used to create metabolic models of over 200 healthy human cells that were dividing normally. Yizhak, Gaude et al. used these metabolic models to predict how quickly the different types of cancer cell would divide and how the cells would respond to drug treatments. It may be possible to reduce the spread of all types of cancer—without also affecting healthy cells—by targeting proteins that help cancerous cells to proliferate. Yizhak, Gaude et al. used all of the models to search for genes that encode such proteins. One gene that was predicted to provide such a drug target encodes an enzyme that is needed to make and break down fatty acid molecules. Experiments confirmed that inhibiting this gene slowed the proliferation of both leukemia and kidney cancer cells, but had less of an effect on the growth of healthy bone marrow or kidney cells. Finally, Yizhak, Gaude et al. generated detailed metabolic profiles of cancer cells taken from over 700 breast and lung cancer patients and were able to use the models to successfully predict the outcome of the diseases in these patients. Yizhak, Gaude et al.'s findings might help future efforts aimed at developing and delivering personalized cancer therapies. The next challenge is to use additional data—such as gene sequencing data—to generate more detailed and more accurate metabolic models for many cancer patients, to both predict their individual responses to available drugs and identify new patient-specific treatments.
Type of Medium:
Online Resource
ISSN:
2050-084X
DOI:
10.7554/eLife.03641.001
DOI:
10.7554/eLife.03641.002
DOI:
10.7554/eLife.03641.003
DOI:
10.7554/eLife.03641.004
DOI:
10.7554/eLife.03641.005
DOI:
10.7554/eLife.03641.006
DOI:
10.7554/eLife.03641.007
DOI:
10.7554/eLife.03641.008
DOI:
10.7554/eLife.03641.009
DOI:
10.7554/eLife.03641.010
DOI:
10.7554/eLife.03641.011
DOI:
10.7554/eLife.03641.012
DOI:
10.7554/eLife.03641.013
DOI:
10.7554/eLife.03641.014
DOI:
10.7554/eLife.03641.015
DOI:
10.7554/eLife.03641.016
DOI:
10.7554/eLife.03641.017
DOI:
10.7554/eLife.03641.018
DOI:
10.7554/eLife.03641.019
DOI:
10.7554/eLife.03641.020
DOI:
10.7554/eLife.03641.021
Language:
English
Publisher:
eLife Sciences Publications, Ltd
Publication Date:
2014
detail.hit.zdb_id:
2687154-3
Bookmarklink