A green, red, black, and white 3-D image of a patient’s entire body rotates on a computer screen before Gordon Harris and Wenli Cai. Harris points out a number of tumors, hallmarks of a genetic disease called neurofibromatosis (NF). Harris’s research group at Massachusetts General Hospital, where he is director of 3-D-imaging services, is preparing a clinical trial to test a technique for monitoring patients, like this one, affected by NF. The technique combines PET imaging and a relatively new imaging technology called whole-body MRI. [Disclaimer: Jason Pontin, the editor in chief and publisher of Technology Review, serves on the board of directors of the Children’s Tumor Foundation, which awarded a grant to Harris for the research described here.]
Most people have never heard of neurofibromatosis, though its three forms affect over 100,000 Americans–more than suffer from cystic fibrosis, Huntington’s disease, and Tay-Sachs disease combined. About one in 3,500 babies worldwide is born with the most common form, NF1. NF1 patients may have noncancerous tumors growing along the nerves and on the skin of any part of the body. These tumors can cause pain and disfigurement; more rarely, they turn cancerous.
Much is unknown about neurofibromatosis. There are genetic tests for two of its forms, but the causes of the third, which was characterized only recently, have yet to be determined. About 10 percent of people with NF1 will develop cancerous tumors. But doctors are unable to accurately predict which patients will have a mild form of the disease, which will develop severe complications, and which will get cancer. Detecting tumors early can be difficult. “Unless you can see a tumor pressing out, or it causes a symptom like shortness of breath, we don’t know it’s there,” says Scott Plotkin, a doctor specializing in cancers of the nervous system at Mass. General, where he collaborates with Harris. Currently, a doctor who suspects an internal tumor characteristic of NF will order a traditional MRI scan, which images only one part of the patient’s body–such as the brain, right arm, or left leg–and takes about 45 minutes. In contrast, whole-body MRI can scan the entire body in 45 minutes and should find all detectable tumors at once. Plotkin and Harris’s study, which starts early next year, will enroll about 250 patients and last four years. It will test whether whole-body MRI is able to image neurofibromatosis tumors as effectively as traditional MRI, and it will look at the effectiveness of combining whole-body MRI with PET images.
MRI provides detailed anatomical images that crisply map out the location and size of each tumor, but it doesn’t reveal anything about the activity of tumors and other tissue. PET scans show how quickly tissue burns through glucose, which gives physicians a measure of how metabolically active it is. Cancerous tumors have very high metabolic rates and, says Harris, are usually “hot” on PET scans. Benign tumors have low metabolic rates, consuming glucose much more slowly.
MGH's New Imaging Technology
If an NF patient’s tumor becomes cancerous, Plotkin and Harris will look back at previous images to see if there were any warning signs–a metabolic reading slightly hotter than normal. The normal range of metabolic activity for NF-related tumors is simply not known, so it can be difficult to distinguish between a tumor with a high but normal metabolic rate and one that is likely to turn malignant. Plotkin says the hindsight offered by the Mass. General trial will provide clearer guidelines for physicians about which tumors may be precancerous and should be biopsied.
Whole-body MRI is less than a decade old, and the technology is currently not part of routine patient care in the United States. But Ara Kassarjian, a Mass. General radiologist working on the NF trial, says the major manufacturers of the strongest MRI scanners used in hospitals (including Siemens, General Electric, and Philips) are now making the machines with whole-body imaging capability. The primary technological advances over traditional MRI are a table that moves a patient through the machine smoothly enough not to blur the image and software that can seamlessly weave together five or six sets of images. Researchers are testing the technique in a variety of diseases besides NF. For example, physicians hope clinical trials in progress will show whether whole-body MRI can detect the spread of cancer.
The Mass. General researchers’ trial will be run in full collaboration with a hospital in Hamburg, Germany, that also treats a large number of neurofibromatosis patients and performs whole-body MRI scans. To facilitate this long-distance collaboration, Harris has built a password-protected online database where all MRI and PET images from patients at each hospital will be stored. Rather than seeing only a radiologist’s report on images from the other hospital, the American and German researchers will be able to see original imaging scans from all patients. This will allow for more consistent analysis of both sets of images.
Plotkin hopes their study will take some of the guesswork out of diagnosing and managing neurofibromatosis. And in the future, he hopes to correlate the images with genetic tests, so he can better determine how gene abnormality affects the number and size of tumors in NF patients. “The ultimate goal is to understand why some patients are severely affected and others are not,” he says. “This is the first step on that path.”