Posted: March 22, 2011

Effectiveness of Therapeutic Inhibitors in Glioblastoma Multiforme is Dependent on a Specific Pattern of Gene Changes

Catherine Evans

A group of Harvard researchers showed in a June 22, 2010, Proceedings of the National Academy of Sciences (PNAS) paper that therapies targeting specific cancerous gene changes may be effective in only a subset of patients having a particular pattern of gene abnormalities. 

The researchers began by screening glioblastoma (GBM) cell lines for the genes that are most essential for the tumor’s growth. By systematically turning off 1,002 candidate genes, they narrowed down the list to 62 genes whose lack of activity resulted in slowed tumor growth. Then they compared this gene list to GBM data from The Cancer Genome Atlas (TCGA) to determine which of the 62 genes had extra copies, an indication of genetic dysfunction and a characteristic of oncogenes. Oncogenes, when activated via abnormal changes, enhance cancer cell growth. Eight of the 62 candidates met these criteria. 

CDK6 Emerges as Candidate for Targeted Treatment

Of these eight genes, the researchers focused on a gene called CDK6 because it is a participant in a cell growth pathway called the Retinoblastoma protein (Rb) pathway, which is dysfunctional in many GBM tumors. Genetic abnormalities in CDK6, Rb and other genes within the pathway occur in more than 78 percent of human GBM cases.

To confirm that CDK6 is actually an oncogene in GBM, the group expressed it in cell lines of normal brain cells. The cells showed enhanced growth and division, an indicator of cancerous growth. When the cells containing CDK6 were implanted into mice, most developed GBM tumors. CDK6, like its close relative CDK4, was confirmed as an oncogene.

Only a Subset of GBM Tumors Respond to CDK4/CDK6 Inhibitors

Since it appeared that CDK6 is an essential and commonly observed GBM oncogene, the researchers then turned to testing a potential CDK6 inhibitor. They focused on an inhibitor of CDK4 and CDK6, which is currently being tested in clinical trials. They treated 25 GBM cell lines with the inhibitor and measured cell growth. Surprisingly, the cell lines with the most slowed growth were not those containing the highest levels of CDK4 and CDK6. The most responsive cell lines contained deletions of the genes CDKN2A and CDKN2C. These genes code for proteins within the Rb pathway. They are tumor suppressors whose activity is lost in many GBM cases.

Using TCGA GBM data again, the group determined that 6 percent of TCGA GBMs contained deletions in both CDKN2A and CDKN2C. Given the study’s findings, GBM patients falling into this grouping are predicted to be the best responders to the CDK4/6 inhibitor currently being tested in clinical trials.

The study’s identification of a potential responder group should aid in the current development of the drug. Moreover, this work demonstrates how experimental studies and large scale genomic datasets can be used together to generate new knowledge about cancer genomics.


Wiedemeyer, W.R., Dunn, I.F., Quayle, S.N., Zhang, J., Chheda, M.G., Dunn, G.P., Zhuang, L., Rosenbluh, J., Chen, S., Xiao, Y., et al. (2010) Pattern of retinoblastoma pathway inactivation dictates response to CDK4/6 inhibition in GBM. Proc Natl Acad Sci USA. 107(25):11501-11506. Read the full article.