Molecular signatures and prediction of poor prostate cancer outcomes over time

An article by Markert et al., just published on line in the Proceedings of the National Academy of Sciences, suggests that microarray analysis of messenger ribonucleic acid (mRNA) expression in prostate cancer tumors may be an independent indicator of the aggressiveness of prostate cancers in individual patients.

This is a highly technical paper based on analysis of clinical data and genetic information derived from stored tissue specimens  from two groups of patients:

  • A set of 281 Swedish patients recruited between 1977 and 1999 and followed on a watchful waiting protocol.
  • A somewhat younger set of 150 patients all treated by radical prostatectomy at Memorial Sloan-Kettering Cancer Center (MSKCC) in New York between 2000 and 2006.

The study allowed the authors to do a number of things. Specifically, they were able to show that: 

  • Gene expression data (molecular “signatures”) could be used to classify the men’s tumors into three subsets:
    • Those with “stem cell-like” signatures as well as P53 and PTEN inactivation. (The men with these tumors had very poor prostate cancer-specific survival over time.)
    • Those expressing the TMPRSS2-ERG fusion gene. (These men had intermediate prostate cancer-specific survival outcomes over time.)
    • Three other tunor groups whose carriers all had relatively “benign” outcomes over time.
  • The molecular signatures were independently predictive of risk for prostate cancer-specific mortality. (In other words, they were not specifically correlated with the Gleason scores of the patients.)
  • The molecular signatures appeared to be able to predict risk for prostate cancer-specific mortality in some men with Gleason 6 and Gleason 7 cancers.
  • The initial predictive information identified through use of data from the Swedish patients cohort was largely (if not perfectly) confirmed by the data from the MSKCC patinet cohort.
  • The “stem-cell-like” molecular signatures allow identification of two subtypes of prostate cancer that carry a 3.2-fold increased risk of prostate cancer-specific mortality.

On the one hand, this study is important because it validates the idea that we may be able to use genetic signatures to differentiate between high-risk and lower-risk forms of prostate cancer at the time of diagnosis. On the other hand, the increase in risk identified by these gene signatures appears to be relatively small. What we need are tools that can differentiate with a much higher level of probability between men at risk for metastatic disease and prostate cancer-specific death and those who have relatively benign forms of prostate cancer. To be really useful, such tests may need to have the capacity to demonstrate a 20-fold or higher risk for prostate cancer-specific mortality at the time of diagnosis.

Clearly data like these can be combined with Gleason scores and other data to better predict survival. Whether such data combinations can be highly accurate in prediction of risk for clinically significant prostate cancer is going to take a while to work out.

According to a report on, the senior author of this paper, Dr. Arnold Levine, is quoted as saying that these results “will need to be validated in a larger prospective trial,” and that “Early data may be available as soon as 5 years from now, and survival data in 10 to 15 years.”

For those interested in the full technical details, the full text of the paper is freely available on line.


2 Responses

  1. A very interesting article. As a Gleason 9 patient, I’m always trying to better understand what makes my cancer so much more aggressive. This is the best paper I’ve read so far about how this happens.



  2. Hi, everybody! I think that genetic science can help a lot in defining the type of the disease. Moreover, I believe that in future we will be able to cure the most serious diseases with the help of genetic modifications.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: