Stanford University Home

Stanford Report Online

Aggressive form of prostate cancer ID’d by test
Finding may lead to better diagnostic tools for disease


Prostate cancer is often a slow-grower that doesn’t spread. In some cases, however, a deadly migration of cancer cells invades other parts of the body. School of Medicine researchers have now found a way to distinguish between the two forms early, which may one day provide a test to determine when invasive treatment is the best option.

Urology postdoctoral scholar Jacques Lapointe, MD, PhD, working with senior study investigators Jonathan Pollack, MD, PhD, assistant professor of pathology, and James Brooks, MD, assistant professor of urology, scanned thousands of genes in normal and diseased prostates and discovered that as few as two specific genes appear to indicate which tumors will be aggressive and which will not. Their findings were published in the Jan. 20 issue of the Proceedings of the National Academy of Sciences.

“It is the first time we’ve been able to take tumors that generally appear the same under the microscope and say that we can see differences that appear useful in predicting how these patients will fare,” said Pollack. This work is “early translational research,” he emphasized, and will need to be validated by other groups and through prospective clinical trials.

Jonathan Pollack, MD, PhD (from left), James Brooks, MD, and Jacques Lapointe, MD, PhD, take a break from studying microarray data identifying genes that determine whether a case of prostate cancer will be slow-growing or aggressive. Photo: Mitzi Baker

Current techniques for estimating the severity of prostate cancer include measuring the level of prostate specific antigen, or PSA, and determining tumor stage and grade. These methods, said Lapointe, provide limited information about both the progression of the disease and the best course of action.

“We have the PSA blood test, which is not without faults,” said Brooks, who provided prostate samples from his patients who consented for the study. He explained that while the PSA test has proved to be a useful tool in screening men for prostate cancer, a number of less dangerous conditions can cause the level of PSA to rise, which has opened controversy over the test’s usefulness. An easily detectable indicator that not only unequivocally tells if cancer is present but also how aggressive it is would be ideal, he noted.

Using microarrays -- glass slides carrying more than 26,000 DNA spots, each representing a different gene -- the group profiled gene expression in 112 prostate samples to study the differences between normal tissue and tumor tissue. They discovered that the cancers fell into one of three subtypes based on which genes were turned on or off. Two of these subtypes were associated with more aggressive cancers.

Interested in streamlining the identification process, they tried using only a couple of the genes that appeared to indicate either a more or less aggressive type of tumor to use as markers. They chose the genes MUC1 and AZGP1 and used a simple antibody test to screen 225 archived prostate samples that had accompanying information about how each patient had fared on average eight years after their prostates were removed.

Although there are hundreds of genes that differentiate the subtypes on the molecular level, the group found that those two genes captured a significant amount of this information. “We have shown that these markers predict outcome very well,” said Pollack. “These markers improve our predictions for the recurrence of tumors months or years after surgery over and above anything available now.”

Since as few as two markers are effective at predicting how serious a tumor a man may have, the simple antibody test could someday be used clinically to screen for those markers. “It’s much easier to stain for two antibodies than it is to perform a DNA microarray,” said Pollack. “This test has the potential to provide clinicians with more information about how to treat tumors in their patients.”

Lapointe added that as they continue to screen more samples, they may find even more subtypes to further characterize and better predict the course of prostate cancer.

This research was made possible through National Institutes of Health and National Cancer Institute sponsorship, part of a larger research grant of biochemistry professor and Howard Hughes Medical Institute investigator Patrick Brown, MD, PhD. Brown developed microarrays at Stanford and collaborated on the study.

Seven other Stanford investigators from the departments of pathology, urology, health research and policy, statistics and genetics contributed to this project. This study was part of an international collaboration and included investigators from Johns Hopkins University, Louisiana State University and the Karolinska Institute in Sweden.

Researcher challenges value of widely used prostate cancer diagnostic tool (2/13/02)

Controversial faculty finding gets nationwide airing (10/9/02)