Computational Biology
Microarrays Prove Reliable in Cross-platform Tests

Immunology
T Cell Misfits May Spell Autoimmunity

Pathology
Prostate Tumor Chemistry Reveals Early Disease

Medical Education
Dienstag Named Medical Education Dean

Student Research Takes Stage at Soma Weiss Day

Minorities Appear to Use CAM Less Than Whites

Blueprint Drawn for Ebola Infection

Director Appointed for Research Compliance at HMS

Abelardo Morell Photograph Donated

Stem Cell Institute Awards 12 Seed Grants

Connors Gift Renames MGH Building After Thiers

Zelen Award and Talk Announced

Rabkin Fellows Chosen for ’05–’06

Honors and Advances

In Memoriam

Surgery Center Joins Push for Quality Improvement

Front Page

Prostate Tumor Chemistry Reveals Early Disease

Profiling Metabolites Through Magnetic Resonance Spectroscopy Could Signal Cancer Course

Prostate cancer defies easy definitions—lethal for some, but benign for others. It is one of the few cancers that may grow so slowly it never kills, though it is also the second deadliest cancer in men behind lung cancer. When tumors do metastasize, there is no cure. The best method of characterizing the danger of a tumor is by how it looks: cancerous tissue taken from biopsies is graded according to its histopathology using the Gleason score. About 70 percent of newly diagnosed prostate tumors receive a Gleason score of 6 or 7 on a scale of one to 10, yet the outcomes of these tumors vary widely. Scientists would like to find a better way of identifying the tumors most likely to cause disease, to construct a better profile of a killer.

Photo by Steve Gilbert

A study on prostate tumors tracks cancer progression through the patterns of chemical changes in cells. Study authors include (clockwise from left) Wenlei He, Elkan Halpern, W. Scott McDougal, Leo Cheng, and Chin-Lee Wu.

A study led by Leo Cheng, HMS assistant professor of radiology (pathology) at Massachusetts General Hospital, suggests that tumors could be more accurately profiled by measuring their metabolites. The findings are part of the emerging field of metabolomics, which looks at patterns in the entire collection of small molecules involved in cellular energy turnover in cells. Just as genomics has been working to develop a genetic profile of cancer cells, metabolomics seeks to identify cells based on their overall chemistry. In the April 15 Cancer Research, Cheng’s team used magnetic resonance spectroscopy to create a profile of all the metabolites in samples of human prostate cancer. The method was able to sort out cancerous and benign tissue with 98 percent accuracy, and certain metabolic profiles correlated with tumor size and aggressiveness, even when taken from benign tissue elsewhere in the prostate.

Invisible Pathology
Even though the prostate-specific antigen test has made chemistry a factor in cancer detection, the only way to know for sure is to take a look at the tissue. “The standard is whether a pathologist sees transformed cells or not under the microscope,” Cheng said. But the chemistry inside cells may change before cancer ever becomes visible. “Before pathologists can see a tumor at the microscope, a lot of things have happened. With chemical information, maybe we can give a second opinion to the clinician.”

“Before pathologists can see a tumor at the microscope, a lot of things have happened. With chemical information, maybe we can give a second opinion to the clinician.”

Cheng worked with Chin-Lee Wu, HMS assistant professor of pathology, and W. Scott McDougal, the Walter S. Kerr Jr. professor of urology, both at MGH, to obtain nearly 200 tissue samples from prostates that had been removed from 82 men with prostate cancer. They used MR spectroscopy to generate profiles of nearly 40 different metabolites for each sample. Analyzing solid tissue with MR is challenging because the bonds between molecules and their haphazard organization create a much less defined spectrum than the same molecules neatly dissolved in solution. But when the tissue sample is spun at high rates at the so-called “magic angle” of 54.7 degrees from the magnetic field, the effect of these confounders is reduced, and the chemicals in the sample give off clearly defined spectra as they would if they were in solution.

Magic-angle spinning has been used for decades, though only recently was it applied to cancer tissue. Cheng helped pioneer a method of analyzing the chemical composition of cancer tissue samples using the technique, while still preserving the samples so they can be analyzed later under the microscope. With the assistance of the Wu lab, they examined the samples using quantitative pathology, which yields a more comprehensive inventory of the tissue than traditional pathology. The samples are sliced and each section carefully analyzed for the proportion of cancerous and noncancerous tissue in order to create a basis for comparing the metabolic profiles.

Chemical Trails
The MR analysis created an unwieldy amount of data. To make sense of it, the team looked for differences in the entire roster of 36 cell metabolites that could account for the differences between cancerous and benign tissue. A statistical computer program used a technique called principal component analysis to identify patterns of change across the panel of metabolites that might reflect major functional changes in the cell.

Certain patterns could account for metabolic changes between different kinds of tissue, including cancer cells. The metabolites that were drastically altered from one to the other included choline and phosphocholine, which have been identified previously as components of cancer cell metabolism. Using this metabolic information, Cheng’s team was able to distinguish all but one of the 26 samples positive for cancer, which Cheng said helps to validate the method.

Beyond separating cancer from other tissue, the approach showed promise in characterizing tumor aggressiveness. Two of the metabolic profiles could distinguish slow-growing tumors from ones that have spread to most of the prostate or escaped to other tissues—even when measured in tissue elsewhere in the prostate. Though the correlation is not sensitive enough to accurately predict the stage for any individual case, the approach might help detect early chemical changes that biopsies miss. When applied to both Gleason score and tumor stage, the analysis could isolate the group with the least aggressive tumors, a group that might benefit more from watchful waiting.

Zaver Bhujwalla, associate professor of oncology at Johns Hopkins University School of Medicine, said, “The nice thing about this study is that they have carefully done the pathology and related it to the patterns they are seeing.” She added that metabolomics can offer information about the function of cells that genomics cannot. “Potentially it can be very useful in clinical assessment of the tumor,” she said.

Bhujwalla, who was not an author on the paper, said that it will be important to know how the prognoses for these patients bear out over the long term. Because it is impossible to obtain biopsies from healthy controls, the study could only use the histopathology of tissue from cancer patients as a basis for comparison. To boost its predictive power, the approach will need to be tested in a larger sample that includes patients who receive biopsies but remain free of cancer for several years.

—Courtney Humphries

top