Faced with a diagnosis of prostate cancer, a man and his doctor today must rely on an analysis called a Gleason grade. It’s performed on a surgically removed prostate gland to determine aggressiveness and best options.

Yet, neither Gleason grades nor the precursor PSA (prostate-specific antigen) — a protein marker blood test — are “slam dunks.” They leave an all-too-vast diagnostic wilderness. Some tumors need only “watchful waiting,” requiring little more than follow-up PSA screenings and physician exams. But the nagging nightmare is that the tumor might be more aggressive. If surgery is performed, incontinence and impotence are just two possible side effects.

“There is a big controversy with PSA screening today,” says George Klee, M.D., Ph.D., Mayo Clinic cancer researcher. “Some large studies suggest that PSA-based prostate cancer screening may not add to the life span or quality of life of the patient. Better tests are needed.”

A Mayo team led by Dr. Klee and a group from the University of Minnesota led by Donald Connelly, Ph.D., joined forces to find such a test. They assembled 20 investigators and obtained one of the first research awards from the Minnesota Partnership for Biotechnology and Medical Genomics. Begun in 2004, the Minnesota Partnership is an initiative among Mayo, the university and the state of Minnesota to fund promising research projects that are too preliminary to garner federal support. That initial grant was the beginning.

Likely Suspects

The Minnesota Partnership team analyzed frozen prostate tissue samples from 100 volunteer patients with either benign prostates, or glands with early-stage or aggressive cancers. When patients’ cells mutate to form cancer, molecular changes occur in the messenger RNA. mRNA is the template outside the cell’s nucleus used to make amino acids into proteins. Understanding these changes, says Dr. Klee, should provide insight into understanding changes that lead to prostate cancer.

The team used laser capture microdissection to select only the prostate gland epithelial cells thought to be crucial. At Mayo, John Cheville, M.D., a prostate cancer pathologist, helped in tissue selection and marker identification; and George Vasmatzis, Ph.D., and his team performed the laser microdissection and molecular amplification. Bioinformatics specialists at both the University of Minnesota and Mayo Clinic identified multiple tissue and blood biomarker candidates.

Spawning New Studies

This initial Minnesota Partnership prostate project has led to five follow-up projects. Each has become part of the Mayo Clinic Specialized Program Oncology Research Excellence (SPORE) program, headed by Donald Tindall, Ph.D. Two projects are developing panels of gene markers to predict who will advance to more serious stages of cancer and how quickly. Mayo’s Robert Jenkins, M.D., Ph.D., developed a 17-gene prognostic panel in tissue and Dr. Vasmatzis developed a three-gene prognostic marker panel. A third project, by Farhad Kosari, Ph.D., of Mayo, and Michael Wilson, Ph.D., at the university, is looking at the “field effects” of gene expression in the tissue adjacent to biopsy areas showing cancer - to identify men who would be misdiagnosed if the biopsy needle did not pass through their tumor. A fourth project, by Krishna Donkena, Ph.D., and Charles Young, Ph.D., both of Mayo, is exploring the prognostic effects of DNA methylation in tissue and blood. The fifth project is targeting proteins in blood that correspond to the differentially expressed tissue genes, with the hope of developing a noninvasive blood test to identify aggressive prostate cancer. For this challenging project, Dr. Klee teamed with his son, Eric Klee, Ph.D., who added expertise in biomedical informatics with novel blood measurement techniques.

Dr. George Klee says, “The amount of proteins specifically coming out from that relatively little prostate gland (it’s the size of a walnut) into the large volume of circulating blood is very low.”

Narrowing the field was formidable. The researchers were looking at 1,000,000-to-1 to 1,000,000,000-to-1 ratios of other common circulating proteins and comparing them to predicted candidate proteins. Using the original Minnesota Partnership data — 22,000 genes — they whittled out 36 novel candidate genetic biomarkers. Advanced study of these markers, a validation of their presence in blood serum and correlation with different stages of prostate cancer, could provide the answers to save lives.

Using computational tools, from the mRNA they could predict characteristics of the protein and how they could slice it up to find the most viable peptide, a tiny piece of protein between 12 and 20 amino acids that serves as the candidate marker. “We came up with methods to measure the peptides on the mass spectrometer, so the focus of the project was a ‘mass spec measurement’ of peptides in blood,” says Dr. Eric Klee. “The hurdle was the concentration that we were predicting to be in blood was far lower than the detection limit of the mass spec.”

To finally net their markers, the researchers used a technique based on that pioneered by N. Leigh Anderson, Ph.D., of the Plasma Proteome Institute, Washington, D.C. It enabled them to enrich low abundance proteins with special antibodies. The antibodies captured peptides and through mass spectrometry and complementary tests they determined how many markers appeared.

Analyzing prostate tissue and blood samples from the consenting men, they found that seven of the candidate markers had significantly increased concentrations in protein cancer tissue compared to matched benign tissue. Ten of the proposed markers were more elevated in the blood of some men with advanced prostate cancer compared to men with less advanced cancer and controls, but none of the markers were elevated in all of the men with advanced prostate cancer.

They did not find a “home run” marker, one that was expressed in every case of aggressive cancer. But Dr. George Klee says, “We have a number of pieces that we hope to weave together with other things such as PSA, and potentially the new gene panels, to make a more efficient procedure.”.

“What we were able to do that’s very exciting is find these novel markers,” says Dr. Eric Klee. “Very little was known about them. If history of the past 10 years or so tells us anything, it’s that a few of these 36 may be good but many are likely to be reflecting some other condition or molecular process. We’d be very happy if we came up with just one or two markers that give validated signals and will augment detection of prostate cancer.”

Next Steps

Going forward, these teams will be analyzing specimens from volunteers suspected of having aggressive prostate cancer to see if they can distinguish the men who really have it from those who don’t. The mass spectrometry peptide measurement technology developed in the blood marker project also has helped Mayo obtain a new federal grant from the National Institute of Standards and Technology for development of reference methods for measuring PSA and activities of sex steroids.

Enrollment in the SPORE Project 1 first clinical trial is just beginning. Results aren’t expected for three to four years. Dr. Jenkins’ retrospective study may have results earlier than that, estimates Dr. Klee. Three patent applications have been filed. Not only does this study improve care for future prostate patients, it also showcases how research collaborations and seed funding can make a difference.

— Tony Fitzpatrick, March 2010