Meier Deutsch thought he was a healthy man. The Toronto software sales executive ran six kilometres at least three mornings a week, could race up a couple of flights of stairs without feeling winded and sometimes walked the 15 km home from work. He’d lost some extra weight, his cholesterol levels were good and his high blood pressure was well managed through medication.
There were no outward signs at all that Deutsch’s body was harbouring a cardiovascular time bomb.
But last October at his annual checkup, Deutsch’s family doctor heard a subtle whooshing sound in his chest and recommended he see a vascular surgeon. That specialist ordered an ultrasound, the imaging technique that uses high-frequency sound waves to visualize soft tissue and internal organs. The results were concerning enough to lead a week later to 3-D magnetic resonance imaging, an advanced type of MRI that creates detailed digital three-dimensional images using a combination of magnetic field and radio waves. Results confirmed that Deutsch’s left carotid artery, which supplies blood to the head, had a dangerous buildup of plaque that was threatening to rupture and lodge in his brain. In early December, Deutsch underwent a carotid endarterectomy, a surgical procedure to scrape out the plaque. “It was a damn good thing we went in when we did,” the surgeon later told Deutsch’s family. “He was a week or two away from a stroke.”
Deutsch, who is 63 and back to his regular routine, is now part of a U of T–affiliated study at Toronto’s Sunnybrook Health Sciences Centre that is using the latest imaging techniques to track his right carotid artery for changes. “Since I had none of the symptoms you would expect with artery disease, I’m proof positive that medical imaging works,” he says.
Medical imaging has come a long way from the basic X-ray technique, invented more than a century ago, which showed only bones and foreign objects lodged inside the body. Today’s imaging systems employ a variety of 3-D and digital technologies that can give detailed pictures of the inner workings of the body that previously couldn’t be seen without cutting the patient open. This has led to quicker diagnoses of disease, earlier interventions, better-targeted therapies and less invasive surgeries. “What’s guiding us today is no longer the X-ray film on the light box but the computers in the operating room,” says Dr. Mark Henkelman, a U of T professor of medical biophysics and a world-renowned imaging researcher. “Canada, especially Ontario, has had a disproportionately high involvement on the international scene in imaging research and development.” And U of T–affiliated researchers are having a particularly strong impact in the fields of cardiovascular disease, cancer and Alzheimer’s.
We used to think of blood vessels as merely the passive pipes of the body, much like a house’s plumbing system. It was believed that when the pipes became thick, thereby narrowing or even blocking the passageway, blood couldn’t get through and a heart attack or stroke could result. Conventional imaging techniques such as angiograms, which are X-rays of blood vessels, can show this narrowing. But that’s not the whole story, says Dr. Alan Moody, the newly appointed department chair of medical imaging at U of T. Sometimes patients whose angiograms look completely normal can have heart attacks and strokes. Other patients may show a narrowing, such as a 50 or 80 per cent blockage, but never have a cardiac event. “It suggested we were looking at the wrong thing,” says Moody, who is also radiologist-in-chief at Sunnybrook Health Sciences Centre.
The new focus, Moody says, is no longer just the degree of narrowing, but the type of plaque in the vessel wall itself. “Blood vessels are very active organs,” he says, “and the vessel walls themselves can become diseased.” Plaques, made up of substances such as cholesterol, fats or sugars, can grow into the vessel wall. Like mini-tumours, they develop their own little complicated system of fragile blood vessels. These are prone to leaking blood, causing inflammation and rupturing. So even if the vessel isn’t narrowed and the blood flow remains normal, the dislodged plaque can cause a blood clot that may move to a blood vessel near the heart or brain and cause a heart attack or stroke.
Moody helped develop an advanced type of MRI that uses high spatial resolution, multiple planes and a large number of images to visualize such a hemorrhage in a blood vessel wall. Called 3-D MRIPH, which stands for three-dimensional magnetic resonance of intraplaque hemorrhage, it shows the problem area as a bright white “hot spot” that jumps out from the grayscale image. “And with 3-D, we get a slab of imaging we can look at in any plane,” Moody says.
The technique is non-invasive and takes only slightly longer than a regular MRI. For instance, it can produce images from the aorta up to the brain, on both sides, in eight minutes. If disease is present, there are three possible therapies to lower the risk of a heart attack or stroke. Surgery can remove the plaque; a metal or plastic stent can support the artery; and intensive drug therapy, such as statins, can reduce inflammation.
Every clinical case investigated for carotid disease that goes through Sunnybrook now receives the imaging. But Moody says, “There’s evidence that there’s a whole tidal wave of people who could benefit from this.” One of Moody’s research studies includes diabetics, many of whom have had traditional ultrasounds that misleadingly show only minimal carotid disease, while the new MRIPH technology is revealing much more. “We’re already seeing really complicated disease in patients with no symptoms,” he says.
Radiation therapy is a proven treatment for many types of cancer, used on about half of all cancer patients. But targeting the tumour while sparing the normal tissue around it has always been a huge technical challenge. “For a long time, methods to localize radiation were pretty crude and produced toxicity in healthy tissues, which prevented us from treating some tumours,” says Dr. David Jaffray, a medical physicist, and professor and vice-chair in U of T’s department of radiation oncology. Areas of the body to receive radiation were traditionally marked on the patient’s skin, which ignored subtle changes that might be happening to the patient internally, both to the healthy tissue and to the tumour itself as the treatment proceeds. For instance, a bladder may fill, a lung inflate, a small bowel shift or a tumour alter, creating a moving target for the treatment.
Jaffray attacked this problem by integrating computerized tomography (CT) imaging into radiation therapy to give a more precise real-time picture. His technique, called cone-beam CT, uses cone-shaped beams to produce a more accurate picture of where the tumour is immediately before and during treatment. So the system, which emits radiation beams, at the same time is acquiring radiographs of the patient and feeding them into a computer. The generated images then determine how to precisely align the radiation beam for the most effective treatment for that patient for that particular day.
A treatment unit equipped with cone-beam CT is about 20 per cent more expensive than regular CT, but it delivers more precise X-ray radiation to the patient. “We’ve seen a reduction in toxicity with image-guided techniques,” Jaffray says. “We’re also hitting more targets that we wouldn’t have otherwise, like prostate, lung and spinal cancers.” First used on a patient in Canada in 2003, cone-beam CT is fast becoming standard, with more than 1,000 machines now in use worldwide. Jaffray’s team is also in the process of building a $15-million research project that integrates state-of-the-art MRI and radiation therapy technology with a robotic system that moves the patient, to increase precision even more.
Jaffray says that in the near future, we’ll see imaging techniques that will further personalize cancer medicine. His team is using a wide array of imaging technologies to measure the delivery of drugs and monitor how a cancer may change with therapy. For instance, if on the first day of treatment a tumour is short of oxygen – which makes it resistant to radiation – will it stay that way? “We’re cracking open the old-style treatment, which was the same for everybody, and looking at, ‘what is the best thing to do for this patient?’” Jaffray says. “The frontier is exciting.”
Alzheimer’s disease can be tricky to diagnose in the early stages, since many other conditions – low thyroid, depression, Lyme disease, vitamin B deficiency or a stroke – can cause symptoms such as memory loss and concentration problems that mimic those of early Alzheimer’s. In a patient exhibiting signs of dementia, a conventional CT scan or MRI can show if the brain has atrophied and shrunk, which may suggest a diagnosis of Alzheimer’s. But there’s a problem with that, says Dr. Paul Fraser, a U of T professor of medical biophysics and the Jeno Diener Chair in Neurodegenerative Diseases. “By the time a brain is losing tissue, it’s difficult to do much about it,” he says, adding that current drug therapies at that point can only temporarily slow the progression, at best. “So a lot of the push is to get early detection, to pick up people before nerve cells start to die.”
The disease may begin 10, 20 or even 40 years before a patient comes in with symptoms. Scientists believe that a protein called tau accumulates in the brain, and is involved in the death of nerve cells and the shrinkage of brain tissue. But long before that happens, it’s thought that fragments of another protein called amyloid start accumulating and forming plaques between nerve cells instead of being cleared out, as in a healthy brain. Amyloid may be linked to the formation of tau. So if we could see when someone’s brain is accumulating too much amyloid, we might be able to intervene before irreversible damage happens.
“The most important new development is that we can now do amyloid imaging,” says Dr. Sandra Black, the Brill Chair in Neurology in U of T’s department of medicine. Multimodal technology, which uses several different imaging techniques together, not only creates a pictorial map of the brain but can look at brain functioning, such as the way the brain accumulates amyloid or takes up glucose, the cells’ fuel. Functional MRIs follow blood flow to measure brain activity. SPECT, which stands for single photon emission computerized tomography, is a nuclear medicine technique that involves injecting the patient with a radiotracer that emits gamma rays, producing 3-D images.
Amyloid buildup alone doesn’t signify the presence of Alzheimer’s dementia, as 30 per cent of people with the biomarker don’t show any cognitive loss. But Black explains, “We can now say the person [with amyloid buildup] has Alzheimer’s pathology and may be at risk of getting dementia.” She’s also looking at the interaction between vascular disease – disease of the brain’s blood vessels – and Alzheimer’s. “The commonest cause of dementia is Alzheimer’s and vascular disease together,” she says.
As research director of the brain sciences program at Sunnybrook Research Institute, Black is leading a Canada-wide amyloid imaging project in people with Alzheimer’s disease and cerebral small-vessel disease. Black shares her findings with her patients. “We show them the images of their brain, and that’s often an incentive for patients to make changes,” she says. Many are motivated to get their cholesterol or diabetes under control, quit smoking, stop drinking to excess and start exercising. “A minimum of half an hour of aerobic exercise three times a week has shown to be brain-protective,” she says.
There may also be drug therapies available soon. The Toronto Dementia Research Alliance, a research collective, is following 7,000 patients a year in five U of T–affiliated hospitals that have memory clinics. “There have been a number of anti-amyloid therapies with thousands of patients in clinical trials, but it’s thought they haven’t worked because they’ve been started too late,” says Dr. Barry Greenberg, the alliance’s director of strategy and the director of neuroscience drug discovery and development at University Health Network. The therapies need to start sooner, he says, and that can happen now that imaging is able to pick up signs of Alzheimer’s much earlier than before. “Some drugs may be on the shelf already and others are in development, so there’s reason for optimism.”
By bringing artificial intelligence into chemistry, Prof. Aspuru-Guzik aims to vastly shrink the time it takes to develop new drugs – and almost everything else