Magnetics And Metabolites
In late October 2005, Leo L. Cheng stood before an audience at the National Cancer Institute (NCI) in Rockville, MD. This assistant professor of radiology at Harvard Medical School described how magnetic resonance spectroscopy (MRS), a sophisticated imaging technique that analyzes the biochemistry rather than the structure of tissues, could be used to differentiate benign tissue from cancerous tissue in the prostate.
After his talk, Cheng took a seat in the audience, when something happened he will never forget. “A gentleman sitting behind me tapped on my shoulder,” recalls Cheng. The man said, “I’m sorry, Dr. Cheng, but I have to say I really don’t understand what you’re talking about. The only reason I’m here is my boss at NCI asked me to find out whether we can do metabolomic imaging.” That is, the researcher wanted to know if MRS could image the entire ensemble of prostate metabolites, or small organic compounds that are reactants in or the end products of metabolism.
Immediately, Cheng replied, “No, we cannot.” But a few seconds later, he looked at the man again and said, “Yes, we can do it.” In that moment, Cheng realized that he could define a combination of many measurable metabolites as one abstract value, then overlay the values on anatomic images, creating a three-dimensional disease map—profoundly altering the way diagnostic radiology looks at cancer.
In the January 27, 2010, online edition of Science Translational Medicine, Cheng and his colleagues at Massachusetts General Hospital demonstrated a 93–97 percent overall accuracy in detecting prostate cancer with this new imaging approach.
Prostate cancer is currently the second leading cause of cancer-related deaths among American men. Moreover, prostate cancer is increasing worldwide, primarily in Western countries. The current approach to diagnosis measures the prostate-specific antigen (PSA). If the PSA level surpasses a threshold—typically 4.0 nanograms per milliliter—then a physician usually recommends a tissue biopsy. In some studies, however, PSA measurements detected as few as 20 percent of the cases of prostate cancer. In addition, although a positive tissue biopsy is currently the only method of definitively diagnosing prostate cancer, ultrasound-guided prostate biopsies gather tissue at random, potentially leaving a malignant tumor undetected.
In sharp contrast, Cheng’s imaging approach could ultimately permit a highly accurate and routine test to identify suspicious regions within the prostate, guide biopsies directly to those regions and identify a tumor’s aggressiveness.
The implications of such imaging are far-reaching. “Eventually,” says Cheng, “we hope to produce radiological images that reveal the disease process itself.” In the near term, after further studies verify their results, the researchers hope to move into clinical trials for detecting prostate cancer within a year or two.
