We subscribe to the HONcode principles.
Verify here.

Gene expression profiling -
predicting the clinical course of prostate cancer

Developing genetic "signatures" able to pick out recurrent versus nonrecurrent disease.

PSA Rising, March 29, 2004. Figuring out whether you, the cancer patient, will do well by choosing a certain primary treatment is not easy. Doctors who specialize in prostate cancer are developing genetic analysis of tumor samples to help with this challenge. The aim is to be able to predict much more precisely than today which individual patients who do in fact need to be treated in order to live a full lifespan and who will most likely be cured by a primary treatment without needing other treatments.

In this quest to match treatment to patient, a study in March 15 2004 issue of the Journal of Clinical Investigation (JCI) describes an exciting application of genomic technology. This particular study focuses on predicting the outcome of radical prostate surgery. In an accompanying commentary, James McKiernan and Mitchell C. Benson, of Columbia University look at strengths and weaknesses of the approach.

The leading author of the paper, Gennadi V. Glinsky is a named inventor on related patent applications.

How would it be used?

Since the spread of the prostate-specific antigen (PSA) test, more men with prostate cancer are diagnosed at early stages of the disease, with a higher likelihood of surgical cure. But side effects of surgical treatment affect many men, even though Dr. Patrick Walsh first began nerve-sparing radical prostatectomy in the 1980s.

To put this another way -- while negative side-effects from radical prostatectomy for clinically localized prostate cancer have decreased yet not been eliminated, many men who undergo readical surgery and are not cured (as well as many who are cured) are left with some urinary incontinence or, especially, erectile dysfunction.

At present, decisions about whether or how aggressively to treat prostate cancer are, as they note, "based on serum PSA levels, biopsy Gleason score, and clinical stage." Although leading centers use "powerful multifactorial nomograms" to help in the decision process" -- and now Veterans Affairs has developed a profiler -- the ability to predict individual patient outcome remains relatively limited.

Novel prognostic indicators

Today medical scientists are using genomics to try to figure out levels of risk of treatment failure for patients with localized prostate cancer undergoing radical prostatectomy. Recent advances in functional human genomics allows researchers to take samples of tumors and scan them for thousands of genes.

Dr. Gennadi V. Glinsky and colleagues analyzed the gene expression profiles of human prostate cancer samples in a total of 100 different tumors. Glinsky's team screened 21 patients for 12,625 gene transcripts to identify genes that would predict relapse-free survival following radical prostatectomy using the Affymetrix GeneChip system. Then, using a smaller set of five-gene clusters,they were able to predict clinical outcome in a larger validation set of 79 patients.

By comparison with other methods, Gennadi Glinsky's gene-expression-based prostate cancer recurrence predictor algorithm is highly accurate. Their microarray-based gene-expression profiling method was able to find molecular signatures that could distinguish subgroups of patients with different disease outcomes. Of those patients who ultimately had recurrence of prostate cancer, the algorithm correctly classified 88% of these patients into the poor-prognosis group. This represents a significant advance in ability to properly classify the clinical outcome of prostate cancer patients for use in determining appropriate treatment.

This study represents the largest clinical series of genomic classification in prostate cancer to date using a high-throughput functional human genomic technique. Microarray technology, which they used to pick out minute genetic differences in the biopsy samples, has the potential to revolutionize identification of high-risk groups of patients.

Some drawbacks to this study

The reviewers point out some weaknesses to this approach. Firstly, the biopsy samples were taken from the prostate after its removal during surgery. It is much easier (and may be more accurate) to take samples that way. But, obviously, if the procedure aims to help men and their physicians to decide which treatment is appropriate, the genetic information should be obtained (before anyone is cut open) from DNA analysis of pretreatment, transrectal ultrasound–guided biopsy material.

But this presents another problem -- as the reviewer says, analysis of a relatively few biopsy samples (rather than slices of the removed prostate prostate) "is subject to the unavoidable risk of biopsy sampling error due to the known heterogeneity of multifocal prostate cancer." The Gleason Score as such reflects this mixed state or variability of prostate tumors. Do the characteristic distinct foci, or patches, of cancer within a given tumor all have similar genetic profiles? So far, no one knows.

At most, then, the "postoperative risk stratification" utilized in the current study can add some pointers to which patients need extra therapy (such as radiotherapy) or closer follow-up.

Statistical analysis

Another major limitation, according to the reviewers, is that the final statistical analysis is flawed. "By separating patient groups into high Gleason score or high PSA or high stage and then allowing the cluster analysis to discriminate outcome, they have limited the power of these traditional risk factors,which should be considered simultaneously in each case."

The authors would have done better, in other words, to figure out a way to see if the gene cluster analysis in fact adds anything to the traditional model, which already includes stage, Gleason score, and PSA level. "Further stratification of a given patient’s risk of treatment failure is important, but the ultimate goal must remain accurate and reproducible individual prediction. Only then will such approaches accomplish more than altering informed consent in selection of therapy."

Despite these limitations the reviewers say that the presented data are a significant advance in the ability to leverage the awesome power of functional genomics in a clinically useful fashion. "It is this translational approach to risk stratification," they conclude, "that is most likely to lead to progress in our understanding of each individual prostate cancer patient’s risk of treatment failure and allow us to intelligently counsel patients regarding what today remains a complex and often confusing array of treatmentoptions."

Links and References

Gene expression profiling predicts clinical outcome of prostate cancer Gennadi V. Glinsky, Anna B. Glinskii, Andrew J. Stephenson, Robert M. Hoffman, and William L. Gerald, JCI 2004 113: 913-923. [Abstract] [Full Text/Free]

This page reported by J. Strax, last updated April 22, 2004

This page is based on an editorial review of an article in J. Clin. Invest. 113:913-923 (2004) titled: Gene expression profiling predicts clinical outcome of prostate cancer by Gennadi V. Glinsky et al. Both the article and the editorial are free online.

Technology Using gene chip arrays for gene expression analysis (Affymetrix)

Try the Prostate cancer survival calculator
prostatecalculator.org

Prostate Cancer Treatment Guidelines for Patients
Version III, October 2002 The national Comprehensive Cancer Network