Under the direction of Christopher Sotak, WPI has become a world leader in the use of magnetic resonance imaging to study stroke and cancer and to evaluate new treatments for these deadly diseases.
A self-described "methods person,"
Christopher Sotak enjoys finding innovative applications for this MRI research instrument.
By Michael W. Dorsey
In the blink of an eye, a small blood clot breaks loose from the aorta of a 62-year-old man and tumbles into an artery in his neck. Approaching the brain, it is whisked into a smaller artery and quickly lodges in the narrowing vessel, blocking the flow of blood to a section of cerebral cortex about the size of a tennis ball.
Deprived of oxygen and nutrients, the brain cells nearest the artery anaerobically burn their small stores of glucose, releasing telltale traces of lactic acid. Then they starve. Without energy, the sodium pumps in their membranes shut down, letting sodium rush in, followed by water. As the cells balloon, they squeeze together, restricting the movement of water in the intercellular spaces.
The damaged cells release toxins, including glutamate that lets calcium flow freely across the membranes like an invading army. The calcium initiates a cascade of biochemical events that eventually kills the cells. Slowly, this sequence of deprivation, biochemical mayhem and death moves like a shock wave through the affected brain cells as the man becomes one of more than 700,000 people in the United States who will suffer a stroke this year.
Only recently has it been possible to see in such detail how stroke affects the brain. This new window on one of the nation's deadliest diseases has been opened by a remarkable medical diagnostic technique called Magnetic Resonance Imaging (MRI), which uses a combination of a strong magnetic field and radio waves to probe the body. Over the past decade, a handful of researchers have found new ways to use MRI to study how killers like stroke and cancer work, to see their effects on the body with stunning clarity, and to evaluate the effectiveness of new medications and therapies. Among these pioneers is Christopher H. Sotak, professor of biomedical engineering at WPI and newly named chairman of the Biomedical Engineering Department.
Since coming to WPI in 1988 to direct the MRI laboratory and research program that WPI operates jointly with the University of Massachusetts, Worcester, he has developed an intimate understanding of the technology and capabilities of MRI. He has put that knowledge to use to develop innovative methods that he has applied in a host of ways, from studying new treatments for tumors to creating new ways of probing systems as diverse as oil wells and tendons. The work has won him a number of awards and honors, including the Established Investigator Award from the American Heart Association and WPI's own Board of Trustees' Award for Outstanding Research and Creative Scholarship. It has also brought international recognition to WPI. Within the past decade, more than 10 visiting medical scientists from around the globe have collaborated on studies with Sotak and his team.
Much of that recognition has come from pioneering studies of stroke and of the effectiveness of pharmaceutical agents for treating stroke and its aftereffects. Working closely with Dr. Marc Fisher at UMass Memorial Hospital, Sotak and several graduate students have turned the WPI/UMass MRI lab into one of the global leaders in the use of the technology to evaluate stroke treatments. They have tested more new drugs than any other laboratory in the world and have produced more than 55 peer-reviewed articles in scientific journals and more than 130 presentations at conferences. The work has been supported by more than $2.5 million in grants and other funding from the American Heart Association, the National Institutes of Health, the National Science Foundation, major pharmaceutical companies and other organizations.
Sotak traces his involvement with MRI to the days before the first human scanners were marketed in the early 1980s. The clinical devices were based on nuclear magnetic resonance imaging (NMR) instruments that have long been used in chemistry labs to study the structure of organic molecules. In the late 1970s, a number of researchers, including Lawrence A. Minkoff '69, now president of M.R. Labs Inc. in Locust Valley, N.Y., found ways to scale the technology up to probe bodies instead of molecules.
While working toward his Ph.D. in chemistry at Syracuse University, Sotak used a large-bore chemistry NMR instrument to study metabolism in perfused rat hearts. After receiving his doctorate, he joined a company that was designing one of the earliest MRI instruments made specifically for animal research. Just before Sotak started work, the company was acquired by General Electric, a manufacturer of human clinical MRI instruments, and Sotak became a software engineer working on the design of the $650,000 device that now occupies about a quarter of his 1,600-square-foot laboratory at WPI.
After two years of writing software, he became an applications scientist, a job that was part salesman and part researcher. He demonstrated the instrument to potential customers, mostly physicians and scientists at teaching hospitals and universities. He also traveled to sites where the instruments had been installed and helped researchers get their experiments to run properly on the complicated devices. Finally, to demonstrate what the devices could do, he conducted original research. "Nearly everything I did then was breaking new ground and was publishable," he says. "It was very exciting. It was the first time that I came up with some experiments of my own. I realized that I probably knew enough about MRI to be an academic."
His chance to join the academic world came in 1988. One of GE's research instruments was sold to the UMass Medical Center and was installed in a laboratory at the Central Massachusetts Magnetic Imaging Center, a clinical MRI facility operated by several area hospitals at the nearby Massachusetts Biotechnology Research Park. The UMass Department of Radiology entered into an agreement with WPI to operate a collaborative research program in MRI, with UMass providing the research instrument and WPI providing a director for the research center.
On a visit to the new laboratory, Sotak learned that the post of center director (a position that also included an appointment as assistant professor of biomedical engineering at WPI) was vacant. He says his growing reputation in the young MRI field and his established research program helped him land the job. At WPI, Sotak continued some of the work he had started at GE. One such project focused on the development of new methods for studying the effectiveness of cancer-killing drugs.
How effective radiation treatments are in destroying cancer cells depends, to a large extent, on how much oxygen those cells receive. Radiation and many chemotherapy agents work best when the cells are well oxygenated, but the rapid growth of tumors often outpaces the growth of blood vessels, leaving cells in the center of a tumor too far from the blood supply to receive adequate oxygen. These hypoxic, or oxygen-starved, cells can be missed by cancer killers, only to become seeds for renewed outbreaks of cancers after the treatment ends. To overcome this problem, radiation is often administered in stages so that cells that are not killed by one dose will be caught by later doses, after oxygen has found its way to them.
|"Nearly everything I did then was breaking new ground and was publishable. It was very exciting.... I realized that I probably knew enough about MRI to be an academic."|
Currently, physicians use their experience and judgement in planning the course of radiation treatments. Sotak would like to provide them a tool that would take some of the guesswork out of the process. The tool is based on a group of chemicals called perfluorinated hydrocarbons (perfluorocarbons). When injected into the body, these compounds are captured by macrophages, tiny cells that congregate in areas of infection and in tumors. Perfluorocarbons make tumor cells stand out in MRI images, but they also have measurable properties that are directly proportional to the amount of oxygen in the tissue surrounding them. By measuring the perfluorocarbon signal from the treated tumor cells before and after radiation treatments, physicians may one day be able to carefully adjust the timing of their treatments to achieve the optimal destruction of cancer cells.
Like other kinds of cells, tumor cells deprived of oxygen must metabolize glucose anaerobically, which generates lactic acid as a by-product. "This is what happens if you exercise and you're not in shape," Sotak says. "There's not enough oxygen to meet the demands that you're placing on your muscle tissue, but your muscles don't stop. Instead, they go into anaerobic glycolosis and produce lactic acid as a by-product. It's the lactic acid that makes your muscles hurt."
While at GE, Sotak earned two patents for new techniques he developed to selectively locate lactic acid using NMR spectroscopy. The research was made possible, in part, by newly developed gradient coils that greatly increased the sensitivity of the GE machines. "One of the advantages of working for a company is that you have access to technology that doesn't yet exist in the academic environment," he says. "That gives you an incredible competitive advantage. You can do experiments that no one else can do."
Once at WPI, he secured equipment grants that enabled him to install the same gradient coils on the machine there. He then continued his research aimed at using lactic acid imaging to map tumors and to evaluate the effectiveness of chemicals called adjuvants, which increase the flow of oxygen to hypoxic cancer cells, making them more susceptible to radiation and chemical treatments.
Sotak says it became clear early on that lactic acid imaging could also be an effective method for spotting cells damaged or compromised by strokes. Lactic acid released by oxygen-starved brain cells can be detected by Sotak's imaging techniques. But before he could design experiments to more fully develop this diagnostic technique, his interest in stroke would take a quite different course.
In 1990, Sotak began collaborating with Marc Fisher, then of the Medical Center of Central Massachusetts, now chief of neurology at UMass Memorial Health Care and professor of neurology and radiology at the UMass Medical School. A team, which also included scientists from Japan and UMass, set out to develop new, more accurate and useful methods for visualizing the progress of stroke and evaluating potential treatments.
When the research project began, doctors had no tools that would enable them to clearly see which sections of the brain were being damaged or killed by a stroke, or how extensive the damage was. Even worse, there were no treatments for the disorder, which is the third leading cause of death in the United States and the leading cause of adult disability. It would be another six years before the FDA approved the first emergency stroke treatment, the clot-busting drug TPA (tissue plasminogen activator). Sotak's work with MRI would help overcome both of those shortcomings.
Sotak's laboratory was one of the first to use a type of MRI called diffusion-weighted imaging to look at stroke damage. The technique focuses not at the brain cells themselves, but at the movement of water molecules within and between them. The diffusion coefficient of water, a measure of the speed of water movement in tissue, has proven to be a highly sensitive indicator of ischemia, the condition caused by obstruction of blood flow to the brain. Within the first few hours after the onset of stroke, brain cells are compromised and the diffusion coefficient in their vicinity declines. The change is probably due to the swelling of injured cells, which makes it more difficult for water molecules to move freely.
In his studies, Sotak augmented diffusion-weighted imaging with perfusion-weighted imaging, which provides a measure of how much blood is flowing through tissue. Perfusion imaging produces a picture of all tissue that is in danger of serious damage, some of which might be saved with prompt treatment. Diffusion imaging shows the tissue that is probably irreversibly damaged. When combined, the techniques create a striking portrait of the progress of a stroke and insight into how to best treat it. Unlike x-rays or CAT scans, which can fail to detect the damage from stroke even many hours after it begins, MRI can provide useful information within minutes of the obstruction of blood flow.
"Everything we've developed is directly applicable to clinical use," says Sotak, who notes that many MRI instruments in hospitals and clinics are now capable of performing perfusion and diffusion imaging. Some of these machines are now being used in clinical trials of new stroke drugs.
"In practice, clinical machines are not as sophisticated as the research machines, and they don't need to be," Sotak says. "In an emergency, physicians need to get a visual estimate of the damage. They need to be able to quickly compare the area shown in the diffusion and perfusion images, because the difference is potentially salvageable tissue. After treatment, physicians want to see if the size of the lesion has been reduced. This is not as accurate a measure as we make with our machine, but it is a great deal more information than physicians had even two years ago for making therapeutic decisions."
|"This laboratory is one of the few (if not the only one) in the United States with a research MRI instrument that is fully dedicated to undergraduate, graduate and postdoctoral research and education in biomedical engineering."|
Working with graduate students in his laboratory, Sotak spends part of his time refining the perfusion and diffusion imaging methods to increase their sensitivity. Another large chunk of time is spent evaluating potential new treatments for stroke. Over the past decade, the laboratory has become one of the world's leading centers for MRI studies on stroke drugs.
Nearly every new medication introduced to treat stroke--developed by drug companies large and small--has come to the lab; in fact, Sotak's team has evaluated far more drugs than any other lab in the world. They include a number of medications that are now in clinical trials, as well as several others that are still wending their way through the long and costly path to FDA approval. Based on WPI's capabilities in MRI research, AstraZeneca Pharmaceutical Corp. recently awarded WPI a three-year, $530,000 grant to use MRI to evaluate new therapeutic interventions. "It looks like this could turn into a long-term relationship," Sotak says.
Drugs for stroke generally fall into three categories. The first are agents that dissolve blood clots, restoring the flow of life-giving oxygen to brain cells. TPA is the only clot-buster currently approved for stroke treatment, but a number of others are in development. The next group, neuroprotective agents, shield surviving nerve cells from chemical changes that can lead to cell death. Many block the rush of calcium into weakened cells. The third group attempt to counteract the damage that can be done as blood rushes back into ischemic tissue, something known as reperfusion injury.
"In humans, the window for successfully treating stroke is potentially large. It may be possible to save affected brain cells perhaps six to eight hours after a stroke," Sotak says. "In practice, though, you can only administer clot dissolvers during the first three hours after the onset of stroke because the risk of death from reperfusion injury becomes too great after that. As fresh, oxygenated blood flows into tissue that is highly ischemic, it will generate oxygen free-radicals that attack the blood vessel walls, causing them to burst. In this case, mortality from reperfusion injury is potentially greater than the death rate from stroke alone in patients who receive no treatment at all."
Sotak with graduate student Matthew Silva.
In some ways," Sotak says, "our work on stroke drugs has become routine. Our methods are well established, though we continue to refine them all the time. Still, this work helps support the lab and enables us to get into new areas of cutting-edge research." One of those new areas may have startling implications for the course of stroke research and treatment.
"In our previous research," Sotak says, "we had done studies in which we occluded blood flow to the brain for 25 to 30 minutes and then restored it. When you stop the flow, you see a change with diffusion-weighted imaging that appears to be completely reversed once the blood returns. We had always assumed that the affected tissue completely recovers, but in more recent studies we've found out that this is not true."
When Sotak and his team of graduate researchers looked at the same brains one to two days later, they were in for a surprise. Tissue that had looked normal an hour after the restoration of blood flow now appeared injured, and the injury intensified with time. It is thought that the initial insult triggers a genetic change in brain cells that sets off a phenomenon known as apoptosis, or programmed cell death. The apoptotic process is a sort of biological clock that leads cells to commit suicide after a certain period of time. (Sotak says this theory is somewhat controversial, and some scientists believe that the delayed injury in brain cells may be the result of damage to the cells' mitochondria--the organelles that generate energy to keep the cells alive.)
If ischemia can set off apoptosis, then drugs being developed for stroke today may have to be augmented by medications that target the biological events responsible for cell death. "Today's drugs are designed to address the metabolic effects of an interruption of blood flow," Sotak says. "There is a complex metabolic pathway, and each drug interrupts it at a particular point to arrest or even reverse the damage. But the mechanistic pathway in a genetic process would be completely different. The effects of these drugs may prove to be temporary, as the genetic process eventually dominates the whole process."
Top (section labeled Treatment Animal): a clot buster drug increases blood flow to the area of the brain compromised by stroke (reduction in the dark area under CBFi) and reduces the apparent cell damage (the dark area under ). Bottom: damage caused by a 30-minute interruption of blood flow appears to vanish, only to reappear within 12 hours and steadily worsen. A 10-minute interruption results in no reappearance of tissue damage.
Sotak says this work may lead researchers to take a closer look at the long-term consequences of another medical condition called transient ischemic attacks. "These are temporary reductions in blood flow in the brain that may be caused by blood clots that spontaneously dissolve. It's now thought that these events are not harmful in and of themselves, but if they are actually triggering programmed cell death, they may have more serious consequences."
Over the past several years, the intensity of the research on stroke pushed some other research areas into the background--including research on the imaging of tumors. But that work is beginning to return to the forefront as more recent studies have revealed that the same diffusion MRI techniques perfected for stroke studies may be highly effective in monitoring changes in tumors. "It turns out that when you treat a tumor, the diffusion coefficient changes by a factor of two--in other words, you get a 100 percent change, which for MRI is spectacular."
A recently completed Ph.D. dissertation by Michael Mgiler used diffusion-weighted MRI to evaluate the effectiveness of the cancer drug 5-fluorouracil. The techniques made it possible to see, with exceptional spatial resolution, which sections of a tumor were killed after a treatment and which sections contained hot spots of living cells. The same techniques could be used in clinical MRI instruments to monitor how well patients respond to treatment. "Because it's now required for stroke interventions, diffusion-weighted imaging has been added to many clinical instruments," Sotak says. "This now standard technique could also be used to routinely monitor and customize chemotherapy.
"It would be possible to check on the progress of a treatment after a day or so to see if anything is really happening," Sotak says. "These treatments are so toxic and make people so sick, and you are frequently working against time, so if something isn't working you want to be able to switch to an alternative treatment as soon as you can. Diffusion-weighted MRI might make that possible."
The unique applications of MRI pioneered in Sotak's laboratory are only part of what has set the facility apart in the field of magnetic resonance imaging. The laboratory's emphasis on education is also unusual--if not unique. "This laboratory is one of the few (if not the only one) in the United States with a research MRI instrument that is fully dedicated to undergraduate, graduate and postdoctoral research and education in biomedical engineering," Sotak says.
Over the past decade, Sotak has worked with two postdoctoral students and advised eight Ph.D. dissertations (five more are in the works), three master's projects and a handful of undergraduate Major Qualifying and Interactive Qualifying projects. He says the fact that the lab has hosted so few master's candidates is a reflection of the daunting complexity of the theory and technology of MRI. "It typically takes a graduate student two years to really get the hang of this stuff. The instrument is very difficult to operate properly. It's also true that career opportunities in the field are open primarily to people with doctorates."
The challenges involved in operating the MRI instrument have also placed limits on the kinds of projects undergraduates can complete in this discipline, Sotak says. In projects undertaken to date, students in various engineering disciplines have computerized physiological instruments used in MRI studies to make them more versatile and easier to use, they have done studies of fluid flow in hollow-fiber bioreactors, and they have even examined the internal anatomy of birds.
With the recent launch of a new undergraduate major in biomedical engineering at WPI, Sotak says he foresees more interest in MRI among undergrads and more opportunities for projects in which the MRI instrument, operated by a graduate student, can be used as a tool for studying various areas of biomedical engineering, especially biomechanics. Some of these projects, he notes, may be in the area of soft-tissue mechanics, the subject of a new research collaboration he has forged with Allen Hoffman, professor of mechanical engineering, and Peter Grigg, professor of physiology at UMass Worcester. Hoffman and Grigg have spent many years studying the biomechanics of tendons using the conventional tools of the mechanical engineer. Sotak plans to see if similar information can be obtained noninvasively using diffusion-weighted MRI.
"Tendons are bundles of long fibers," he says. "Structures comprising those fibers should be impediments to the movement of water molecules. If you place a load on the tendon and cause those fibers to move closer together, the result might be a change in the diffusion coefficient, which we can measure with MRI. Then you would have a noninvasive MRI measurement that you might be able to relate to a mechanical property. We're in the early stages of this work and I'm not sure what's going to happen, but it should be interesting."
For Sotak, who describes himself as a "methods person," someone who enjoys developing and perfecting the techniques of MRI, finding novel ways to apply those methods is the challenge that makes his work fulfilling. "I know all of the history of these methods and have a lot of experience that I can bring to bear on solving problems," he says. "You develop a method for something over here and then you think, 'Yes, that might work over there--maybe there's something about that disease process that would manifest itself in that kind of measurement. If you understand the measurement and its physical underpinnings, you can see how it can be adapted to study other diseases or physical processes. That's why it's helpful to work on a lot of different types of applications. That's also what keeps this job interesting."
Last Updated: 7/7/99