Issue No. 2012
Valerie Matthews Mehl
Date: December 20, 2011
Johns Hopkins Physicians, Scientists, and Engineers Working Together to Fight Cancer
Johns Hopkins Physicians, Scientists and Engineers Working Together to Fight Cancer
Add engineers to the list of specialists engaged in the fight against cancer. It is not unexpected. Johns Hopkins is revered for its brain trust and the willingness and desire of its leadership to apply knowledge to improving the health and wellbeing of humans. Johns Hopkins, after all, was the site of the first department of biomedical engineering—the very first organized effort that applied the tools of engineering to human medicine. It is a brilliantly simple concept. Gather together the smartest people, give them the freedom to pursue novel approaches, and good ideas will almost always result. This is perhaps the cornerstone of Johns Hopkins’ pioneering success in so many fields. With world-class programs in engineering and medicine and with some of the most talented scientists in both fields, why not bring them together.
“More interaction stimulates creative ideas,” says Kimmel Cancer Center Director William Nelson. “It leads to connecting two things together that have never been connected before. With more of this happening it is almost certain that something innovative will come about.”
The Cancer Center at Johns Hopkins was built on the premise of collaboration. The urgency of the problem demanded no less. Laboratories were located directly off of the patient wings to ensure the quick transfer of discoveries from scientists to clinicians.
Nowhere at Johns Hopkins was the physician-scientist, those who work both at the bench and the bedside, more visible than in the Cancer Center. Kimmel Cancer Center investigators and clinicians were doing “translational” medicine—transferring basic science discoveries to cancer medicine—long before there was a word for it.
Therefore, it is not surprising that a new and different type of collaboration has emerged. This way of working, one could say, is in our metaphorical “genes.”
Pioneering discoveries in genetics and epigenetics by Johns Hopkins Kimmel Cancer Center investigators have revealed the very origins of cancer and uncovered new targets for treatment and drugs that hit these targets. Engineers bring technologies that allow researchers to pick the best target, deliver drugs effectively to the target, and monitor whether or not it gets the job done. Cancer researchers made personalized cancer medicine possible. Engineers made it doable.
“We have to ask ourselves, ‘Why is Johns Hopkins so successful?’” says Dean of the Whiting School of Engineering Nicholas Jones, “The collaborations between investigators in the trenches all the way up through the Deans and Directors of departments surely must have something to do with it.”
Collaborations between oncologists and engineers are certainly not new to the Kimmel Cancer Center. Diseases like cancer demand forward-thinking study, and our investigators have always looked to innovation to develop new methods to control and cure cancer. Engineers and physicists, including John Wong, Justin Hanes, and Martin Pomper work in the cancer center with the sole purpose of using their skills to aid oncologists in diagnosing and treating cancer. Surgeons like Michael Choti have solicited the expertise of engineers for more than a decade to bring added precision through image guidance and robotics to cancer surgery.
Every instrument used and the process by which they operate were developed by engineers.
Under the leadership of Jones, the Whiting School of Engineering has infused considerable resources into the department to accommodate the growth in clinically relevant programs, such as nanotechnology, computer science, and robotics. In recent years, he has added a new engineering building and a computational and robotics facility. New programs, including the Engineering in Oncology Center, Institute for Nanobiotechnology, the Center for Cancer Nanotechnology Excellence, The Center for Nanomedicine, and Physical Science Oncology Center fund and foster research projects between Kimmel Cancer Center scientists and engineers.
The strength of these collaborations and their potential to improve healthcare has not gone unnoticed. If investments are a measure of our success, then the Kimmel Cancer Center and Whiting School have earned high marks. Both have recently received separate $30 million gifts. These funds will help support personalized approaches to treating cancer, stemming from Johns Hopkins research that distinguishes unique genetic and epigenetic differences among individual patients’ cancers, explaining, at least in part, why traditionally developed drugs help some people and not others.
The Kimmel Cancer Center received its gift, a grant from the Commonwealth Foundation for Cancer Research, to create a Center for Personalized Cancer Medicine. It supports collaborations and research that accelerate the pace of developing targeted therapies based on the distinctive cellular fingerprint of each individual patient’s cancer. The Whiting School of Engineering grant comes from Johns Hopkins alumnus John C. Malone and will be used to foster interdisciplinary research efforts where researchers will collaborate with colleagues from other Hopkins divisions to learn to tailor therapies for individual patients and devise systems-based approaches to medical problems, with an initial emphasis on cancer.
“This is a watershed moment,” says Nelson, M.D., Ph.D. “There are many opportunities for engineers and physicians to make cancer medicine better. Across a whole spectrum of components—molecular genetics, epigenetics, imaging, diagnostics, and treatment—we are doing pioneering work in engineering and in medicine. Hopkins is probably better positioned than anyone to make revolutionary advances in human diseases, like cancer, that could only be imagined a decade ago. It’s just a question of how we bring it all together.”
Genetics, Epigenetics, and Physics
Cancer is a genetic disease. That was proven by researcher Bert Vogelstein and team who first illustrated it in colon cancer, laying out an accumulation of genetic errors over time that cause cancer to arise and grow. Since that pioneering work in the 1980s, the same team has mapped the genetic blueprint of nearly 100 cancers. Other errors, called epigenetic events, that alter the DNA without mutating it were also proven by Kimmel Cancer Center researchers to contribute to the origination and progression of cancer.
Now with collaborations through the Engineering department’s Physical Science Oncology Center, Johns Hopkins teams are deciphering a third contributor called physical oncology. It is focused on solving the mystery of how cancer cells break away from the original tumor and form new tumors in other parts of the body. This little understood process is called cancer metastasis, and it is at the heart of what makes the disease so deadly. Cancers that spread are cancers that kill.
“Every tumor has mutations. But now, we are also looking at its structural differences,” says Kimmel Cancer Center investigator and pancreas cancer expert Anirban Maitra. “What are the physical consequences of the genetic and epigenetic events on the cell?
How does it change the cell structurally so that it can become metastatic? ” asks Maitra. Epigenetics expert Stephen Baylin agrees. “We know that cancer cells have the ability to reprogram to respond to and see cues in their environment,” says Baylin. “Understanding what these cues are would be very helpful.”
Maitra says genetic events happen first in cancers, but it’s what happens afterwards that scientists must begin to take a look at to get the full picture. We need only look to lower life forms like worms and plants to know there is more. “Worms have as many protein encoding genes as humans, yet humans are more evolved. There are layers of regulation and cellular characteristics beyond coding genes that we don’t yet fully understand and some of them likely sustain cancer” he says.
Maitra has teamed up with Denis Wirtz, a chemical and biomolecular engineer and director of the Physical Sciences Oncology Center to help unravel some of these mysteries. “If we all take pieces of the puzzle, we can figure this out,” says Maitra.
“Johns Hopkins is ahead of the game here because we do not work in isolation. Multidisciplinary collaborations aimed at moving research discoveries to patient care are the Hopkins model. Human tissue and clinical research are revered and encouraged here. We recognize that engineers are doing vibrant research that is directly relevant to our work, so it only makes sense that we engage in collaborations.”
Wirtz developed a technology that, in minutes, can take thousands of distinct structural measurements on tens of thousands of individual different cells. It’s not easy for a cancer cell to metastasize. “It’s like a steeple chase. The cancer cell has many obstacles to overcome,” says Wirtz. It must break away from the original tumor and the organ from which it arose, squeeze into narrow capillaries, make it through the sheer force of the bloodstream, and set up residence in a new and foreign environment.
Maitra and Wirtz are working to uncover the physical changes to a cell that help drive it away from its home and, against all odds, allow it to survive and thrive in a new and foreign place. Most cancers can be cured if they are detected and treated before they have begun to spread. If researchers can figure out how metastasis occurs, they could develop ways to prevent the deadliest event in cancer progression.
Wirtz likens a cancer cell that metastasizes to a decathlon athlete. A decathlon competition, whose winner is considered the world’s greatest athlete, features competitors who have physical attributes that make them stronger, faster, and simply more adept than other athletes. “Millions of cells are shed by tumors every day, but only one or two of them will have what it takes to become metastatic. These are the decathlon cells. We need to figure out what the physical properties are that give these cells an edge,” says Wirtz.
Maitra and Wirtz are now using his cellular measurement tool in the laboratory on human cancer cells to identify specific cellular characteristics that may determine how aggressive a cancer will be and how it will respond to treatment. Monitoring and measuring how cells change will allow scientists to predict which ones are likely to become the lethal, metastatic type.
“This kind of technology allows us to further personalize cancer medicine,” says Maitra. Within the next few years, Wirtz and Maitra hope to move the technology into clinical use to help select drugs that will work best against a cancer based on the genetic, epigenetic, and physical characteristics of tumor cells. “The engineers have the coolest tools around, and they are looking to us for the right biological uses,” says Maitra. “Now, we can use this new tool to correlate gene abnormalities with particular cellular characteristics and drug sensitivity so that we can get the most effective treatments to each patient.”
Engineering as Applied to Cancer Medicine
Denis Wirtz refers to Kimmel Cancer Center Deputy Director Stephen Baylin as Mr. Epigenetics. The title was bestowed for Baylin’s work in linking biochemical alterations in the DNA environment to the initiation and growth of cancer. These non-mutational alterations are categorized as epi—in addition to—genetic. The work of Baylin and his colleagues, clinician-scientists James Herman and Malcolm Brock, resulted in some of the first uses of epigenetic abnormalities as cancer biomarkers for diagnosis and to predict how a cancer would respond to treatment. Now Center for Cancer Nanotechnology Excellence engineer Jeff Wang is helping move these trailblazing therapeutic discoveries forward more rapidly.
In 2008, lung cancer research by Brock, Herman, and Baylin found epigenetic alterations to any or all of four genes caused cancers to be more aggressive. Depending upon the combination of altered genes, the recurrence rates of the lung cancers studied increased two to 25-fold. Based on these findings, clinician-scientists Ros Juergens, Charles Rudin, and Brock initiated a clinical study of epigenetic therapy for patients with advanced and heavily treated lung cancer, and a small percentage of patients experienced very robust and long-lasting responses.
With funding from Stand Up to Cancer (SU2C), Rudin, Brock and team are hoping to develop these findings into a personalized treatment approach for lung cancer, and Wang’s work in developing a nano-based assay to test for alterations in these and other genes is central to that.
SU2C support allowed the researchers to do an unusual thing—to go back and find out what happened to the patients they thought they didn’t help. They reviewed the records of patients who were participating in a clinical trial of epigenetic therapy because their lung cancers had not responded to standard therapy. Specifically, Rudin, Brock and team were interested in some 40 patients who were taken off of the epigenetic treatment trial because their tumors continued to grow. What they found surprised them all. Although many patients seemed to progress while on therapy, many, including those who had received just two or three treatments, had unprecedented and long lasting responses. The team poured over every scan and every clinical report, and even re-biopsied some tumors, and could come up with only two explanations. Either the epigenetic therapy sensitized their cancers to subsequent treatment with standard drugs or their improvement was a direct response to the targeted therapy.
Targeted therapies do not work like the old cytotoxic chemotherapies, which do not discriminate between cancer cells and normal cells. Instead they specifically seek out and reprogram the malfunctioning mechanism within cancer cells that is causing the tumors. As a result, Baylin says, targeted therapies shrink tumors more slowly over time as they make their repairs and genes are returned to normal function.
Jerry Morton, a 61-year-old retired firefighter, was building a sandbox for his grandson two years ago when he learned he had small cell lung cancer. The cancer had already spread throughout his lungs and to his liver. “I didn’t expect to live long enough to complete it,” says Morton. “Today, that sandbox is finished.” Morton’s tumors melted away on the epigenetic treatment. When he recently developed a new, small tumor of the same type Rudin, Brock, and team again tried epigenetic therapy, and Morton, once again, appears to be benefitting. “The team has evidence that the status of epigenetic alterations in these four genes, assayed in the blood by sensitive technologies that Dr. Wang is working to perfect, can predict the good responses in patients like Mr. Morton,” says Baylin. "It is a perfect illustration of the value of collaborative research and translational, personalized cancer medicine.”
With such promising results, Baylin and team are now planning another trial for newly diagnosed patients. If findings continue to show that the epigenetic-targeted therapy works, Wang figures prominently in the wide clinical application. What Brock, Herman, and Baylin do in their basic science laboratory to screen for the epigenetic alterations would be difficult to routinely do in a clinical lab, so they asked for help from engineers. Wang, who works in the Institute for Nanobiotechnology, is using nanotechnology, engineering at a molecular level, to develop a broad-based test that could be used routinely in any clinical laboratory to predict which patients would benefit from the targeted therapy. “This facet is critically important in taking these therapies beyond our own walls. Without the nano-tech approach, testing for the alterations might be too cumbersome for a clinical lab,” says Baylin. If ongoing trials confirm the earlier findings, Wang’s test can help ensure that the Johns Hopkins-based discovery becomes available to lung cancer patients everywhere.
Nanotechnology, in essence, refers to the field of engineering that focuses on ultra tiny functional devices and structures that are so minute they cannot be seen by the human eye. Think of it as engineering at the molecular level. Like their scientist counterparts in molecular genetics, engineers who specialize in nanotechnology work in the realm of invisibility and had to develop new technologies just to perform their craft.
Where nanotechnology intersects with medicine is the focus of the Johns Hopkins Institute for Nanobiotechnology. It began in 2006, bringing together scientists and students from engineering, physics, chemistry, biology, medicine, and public health to apply the technology to the treatment of diseases. It was the brainchild of Peter Searson and Denis Wirtz, both engineers, whose work was centered on the physical sciences. They began to wonder what Johns Hopkins engineers and physicists could contribute if they collaborated with medical scientists and directed their research toward health care and medicine.
Research teams comprise both engineers and medical research scientists, a point Searson says cannot be overstated. “The engineers attend lab meetings and visit the clinic and work directly with clinicians so that we understand the problems clinicians are facing,” says Searson. The model has been quite successful earning the Institute two large NIH center grants; one for the Physical Science Oncology Center and another for the Center for Cancer Nanotechnology Excellence (CCNE).
“Physicians and engineers approach problem solving in different ways,” says cancer immunology expert Hy Levitsky, who trained in bioengineering before going into medicine and is now working with engineers to improve cancer vaccines. “As an oncologist, I can point to a problem, but I don’t have the expertise to know the spectrum of possible solutions. Engineers can look at the problem from a very different perspective. When we bring the two perspectives together, we arrive at unique approaches we may not have otherwise found.”
A Trojan Horse for the Cancer War
Justin Hanes, a chemical and biomedical engineer specializing in nanotechnology for cancer treatment, explains that a nanoparticle is to a soccer ball what a soccer ball is to the planet earth. Suffice it to say that we’re talking very small.
As small as they are, part of their value in cancer medicine is that they are bigger than other things. Nanoparticles are much larger than small molecules—agents commonly used to treat cancer. Hanes says hundreds of thousands of small molecules can fit into one nanoparticle. As a result, they can be loaded up like a Trojan horse and sent out to deliver their cargo to tumors.
Hanes is a member of the CCNE, which has the specific aim of using nanotechnology to improve the treatment of cancer. He is excited about the prospects of nanotechnology to improve the delivery of anticancer drugs and combat toxicities. “Right now, we use systemic cancer therapies, which are incredibly useful, but typically a very small percentage of the chemotherapy ends up in the tumor. The rest of it poisons normal cells throughout the body,” explains Hanes. He uses nanotechnology in an attempt to reverse that ratio so that anticancer drugs go directly to tumors with very little injury to normal cells. “Think of a tumor as a weed in a prized rose garden. What we do now is spray the entire rose garden with weed killer,” says Hanes. “But, we can only spray so much poison or we risk killing the roses.” The project he is working on with lung cancer experts Charles Rudin and Craig Peacock is a nano-based treatment that seeks out and targets cancer cells. “Like a pre-addressed envelope, we are developing methods to direct them to where we want them to go,” says Hanes.
Hanes directs the Center for Nanomedicine where he and his colleagues are currently working on anticancer-drug-filled nanoparticles that can be inhaled into the lung. His team created the first nanoparticle system that can penetrate the mucus lining the lung airways. In the past, inhaled forms of nano-therapy have failed largely because the lungs’ natural mucus barrier, in place to keep out bacteria and other infectious invaders, swept the drugs away before they could go to work. When Hanes and team discovered that the mucus lining was more like a net than a blanket, they constructed nanoparticles small enough and slippery enough to get through the holes in the sticky mucus “net.”
Hanes’ particles are composed of two parts made of molecules routinely used in existing medications. An inner core traps therapeutic agents inside while a dense outer coating allows a particle to slip through mucus nearly as easily as if it were moving through water. The nanoparticles cause no immune or inflammatory response so drugs travel freely through the lungs, including areas scientists believed were impenetrable, releasing their anticancer cargo over an extended period of time.
Hanes’ nano-based approached could be applied to cancers of any mucus-lined organ, including the colon and cervix. He said such treatments could be ready for clinical trials in as little as five years and cost about the same as traditional chemotherapy.
On Target Molecular Imaging
Central to targeted therapies is the ability to verify that an agent is hitting its intended target and having an impact against the cancer. Targeted agents work by killing cancer cells outright or, in some cases, by reprogramming broken mechanisms within the DNA to kill cancer cells more slowly over time. Traditional radiology, like CT, is focused on anatomy and explores the structure of organs and tumors and primarily measures reduction or growth in the size of tumor over a period of months. These types of scans cannot see or measure what is going on within the DNA of cancer cells.
This shortfall has led to the field of cancer molecular imaging and the research of forward- thinking scientists like Martin Pomper, a radiology expert who came to Johns Hopkins in the early 1990s because it was one of the few institutions where molecular imaging research was being done. Pomper has collaborated with chemical and biomolecular engineers and other scientists to adapt radiology to the molecular level.
Over the last two decades with advances in genetics and epigenetics leading to new therapies aimed at interfering with these cancer-causing alterations to the DNA, cancer imaging has grown from a developing program of the Kimmel Cancer Center to a formal program vital to its ability to move these new treatments from test tubes to patients.
Pomper says molecular radiology has the ability to image inside the cell and reveal to clinicians when a cancer cell is poised to metastasize, permitting a therapeutic strike before the cancer makes this critical and often deadly transformation.
More recently, Pomper has been developing agents that make cancer cells fluoresce so that they can be tracked and monitored—tracked so that scientists know where new targeted agents are going, and monitored so they know if the agents are having any benefit. “This helps us to see quickly if drugs are hitting their targets and doing what we expected them to,” says Pomper. “We can then personalize treatments to individual patients. We can give one dose of a drug. If cells light up, we know we have the right drug. If not, we know to move onto something new, rather than waiting three months for a scan to show that the tumor has grown.” Currently, this type of molecular imaging is done primarily in research studies and in conjunction with pathology and scans to validate its capabilities. Once proven, however, Pomper says, it has the potential to complement, if not replace, CTs and some types of pathology obtained in biopsy as the primary method for detecting and monitoring cancers.
Light-emitting nanoparticles known as quantum dots are another molecular imaging technique Pomper is exploring with Institute for Nanobiotechnology (INBT) director Peter Searson. These cancer-specific quantum dots are what investigators call theranostics (therapeutic and diagnostic) for their ability to aid both in the detection and treatment of cancer. Pomper says they can be engineered using a special coating that makes them seek out and bind only to cancer cells, and they can also be loaded with drugs to carry to tumors. In collaboration with Searson and Anirban Maitra, Pomper is attaching targeting molecules to quantum dots that are attracted to proteins over-expressed in pancreas cancer and working on methods to make the quantum dots circulate throughout the bloodstream to target tumor cells. “These would be like radioactive warheads that can travel throughout the body and not cause any damage until they come into contact with a cancer cell,” he says.
Imaging in Cancer Vaccines
In another example of targeting and monitoring treatments, cancer immunology expert Hy Levitsky is working with engineer Jeff Bulte to improve cancer vaccines. Bulte, a radiologist and biomedical, chemical, and biomolecular engineer, has devised a way for Levitsky to monitor and track the activity of immune-boosting cells in response to cancer vaccines.
Several key biological events must occur after injection with a cancer vaccine to get the immune system to kill cancer cells.. Levitsky says the antigens contained in the vaccine must move from the injection site into the lymph nodes, where immunity is initiated. Cells, called antigen-presenting cells, have the pivotal job of capturing the antigens and delivering them to the lymph nodes. This critical step of antigen transport is often the place where vaccine therapy breaks down. “If we don’t get this step to happen efficiently, there is no hope that the vaccine will work,” says Levitsky. As a result, clinicians almost always use adjuvants—additional treatments given in conjunction with cancer vaccines to help antigen-presenting cells make the journey to the lymph nodes.
Bulte and Levitsky recruited M.D., Ph.D.-student Chris Long to work jointly between the two laboratories to devise a method to quantify this process of antigen delivery. Long developed a technology that uses MRI (magnetic resonance imaging) and iron nanoparticles to track these vaccine-transporting cells. The magnetic iron particles become visible when exposed to a strong magnetic field, such as that used in MRI. Long was the first scientist to show that antigen presenting cells can capture iron-particle labeled antigens and be tracked inside the body. Before this work, cells could only be labeled through contrast agents administered outside of the body. The method allows Levitsky to know exactly how many antigen-presenting cells capture vaccine antigen and make it to the lymph nodes and also to test how well vaccine adjuvants are improving this transport.
“Within three to five days of vaccination, we can see if the vaccine is working as it should,” says Levitsky. “Adjuvants are almost always given with vaccine therapy, so this technology aids us in a critical area of cancer research.” More important, Levitsky says it could be moved to humans quickly. “All pieces of this project—the adjuvants, MRI, and contrast—have already been used safely in humans, just not in this way,” he says.
Cancer Imaging Devices
Scientists in the Department of Biomedical Engineering make image-guided devices for diagnosis and treatment of diseases, including cancer.
Devices that produce remarkably clear and detailed images of cells are aiding oncologists, radiation oncologists, and surgeons in the detection and treatment of cancer. Xingde Li adapted a device he developed to look at the walls of blood vessels and arteries for use in cancer. Li’s optical imaging device can be placed in a patient’s esophagus where it produces super-high resolution images of the esophageal wall. The images are so clear clinicians are able to decipher normal cells from a benign inflammatory condition called Barrett’s esophagus that can predispose those affected to developing esophageal cancer.
A similar device is in development to assist neurosurgeon Alfredo Quiñones-Hinojosa in delicate brain cancer surgery. When removing a tumor from the brain, the command center for all that is human, surgeons work diligently to protect brain tissue not invaded by the tumor. Surgeons walk a clinical tightrope—they must remove as many tumor cells and as few normal cells as possible. Leave tumor cells behind and the tumor may return, but harm normal brain cells, and the patient could lose important mental and physical functions. Typically, surgeons will start at the border of the tumor, painstakingly shaving off very small sections, piece by piece. Figuring out when they are finished can be a difficult problem, so Li is working on a microscopic device that surgeons can place against cells to quickly distinguish tumor cells from normal cells.
Department chairman Elliot McVeigh and team also are working to make lung cancer detection safer and more accurate. The quality of CT scans has improved so greatly that it can detect tiny tumors in the lung no larger than a few grains of rice. This type of early detection, occurring before a cancer produces any symptomatic evidence of its existence, could allow for earlier intervention. To intervene, particularly with targeted therapies personalized to the specific DNA alterations in a tumor, clinicians must be able to get an accurate biopsy of the tumor. Usually this is done through bronchoscopy, but it is not an exact method. “I’ve watched many bronchoscopies, and sometimes it can take 10 to 20 attempts to get the right tumor sample,” says McVeigh. “This can be very traumatic for the lung and the patient. We want to help the physician get the important cells while causing the least amount of tissue damage.” His team has developed a targeted system that can be used in combination with CT images, the bronchoscope, and fluoroscopy to provide real-time moving images inside the lung. It transforms the tumor into an easily visible target that makes even tiny tumors easier to see and biopsy.
Image-guided methods for localized delivery of chemotherapy are another focus for McVeigh. Imagine a patient in an MR (magnetic resonance) scanner, and her physician is using the images it displays to clearly see and direct a needle or catheter directly to her tumor, injecting high doses of anticancer drugs into and around the tumor. In some cases, physicians could potentially place a needle into a blood vessel feeding the tumor and inject an agent that would harden and starve a tumor to death. The precise hardening agents to be used are still being studied. “Imagine going down the vessel tree, delivering a chemotherapy-filled liquid that gets drugs to the tumor and then polymerizes [undergoes a chemical reaction] to block off further blood supply to the tumor as you watch it all unfold under image guidance,” says McVeigh. In essence, this method gets the cancer drug directly to the tumor and then blocks the exit, keeping the anticancer drugs in and the tumor-nourishing blood supply out.
“Our clinicians are wonderful about embracing engineers,” says McVeigh. “One of the best things about Johns Hopkins is that so many are hugely famous for their work, but you wouldn’t know it. You sit at a table and talk to them about a project, and you go back and read about them and learn they are ridiculously famous. People here are great to work with. They are not protective about their ideas. They share them. It’s in the culture of Johns Hopkins. It’s who we hire.” McVeigh says he finds that scientists and clinicians at Johns Hopkins are focused on the overall mission rather than their own, individual accomplishments. “I can show up at any lab meeting anywhere in this institution and be welcomed. This is the perfect research environment.”
Making Sense of it All
In terms of engineering, perhaps one of the greatest contributions to cancer research is in the computer engineering and computational science that has allowed researchers to make sense of the volumes of data generated by genetic and epigenetic discovery. “Engineering is not only responsible for instruments we use in sequencing the genomes of cancer but also in helping us interpret the raw data we get from it,” says leading cancer genetics expert Bert Vogelstein. “It is much like hieroglyphics went it comes out. It is computer scientists and engineers that turn it into meaningful data.”
Their laboratories are not like those we typically associate with medical research. Take for example the laboratory of Rachel Karchin, a computer scientist in the department of biomedical engineering. Instead of benches, there are desks with computers. Just beyond the computer lab is a large, dark glass-enclosed room filled nearly floor to ceiling with computers. Green lights that glow and blink as if communicating in some futuristic code are the only outward sign of the complex calculations occurring within.
Karchin is working with Kimmel Cancer Center investigators Bert Vogelstein, Kenneth Kinzler, Victor Velculescu, and Nickolas Papadopoulos to help them sift through the vast genetic blueprints of cancer to identify mutations that warrant further study. “Our role it to try to distinguish which of these mutations are worth investing lab resources, money, and time in following up,” says Karchin.
The mutational landscape of tumors is quite diverse, characterized by alterations in a variety of genes which differ from patient to patient. Karchin is helping them separate the wheat from the chaff, deciphering the mutations that are actually driving the cancer from those that are just along for the ride. A cancer cell acquires thousands of different alterations, but only a few of these changes will lead to cancer. “We make it tractable,” says Karchin. “We offer a way to prioritize large amounts of data.” Karchin is well schooled in this type of work. She received her doctorate in computer science, with a focus on computational biology, from University of California, Santa Cruz, where the human genome was assembled in 2000. She came to Johns Hopkins University in 2006 to join its Institute for Computational Medicine. The opportunity to work with researchers of the magnitude of Vogelstein drew her here, she says. “Johns Hopkins makes it possible to forge collaborations with people across many disciplines, and these collaborations are speeding the pace of discovery,” says Karchin.
Karchin and team have helped Kimmel Cancer Center researchers predict which changes in the genetic code may cause pancreas cells to turn cancerous. Cancer researchers sequencing the DNA from pancreas tumors and comparing it to normal tissue, found nearly 1,000 alterations unique to the pancreas tumors where just one letter in the ATCG chemical alphabet was changed. They turned this sequence data over to Karchin’s team for assessment.
The researchers developed a computer program that listed all of the individual genetic changes suspected of causing cancer and those highly unlikely to cause cancer. It employed 70 different predictive features for each change to distinguish characteristics of the so-called driver mutations—those alterations that contribute directly to the cancer development—from other genetic changes. They used the program to assess the 1,000 pancreas cancer associated alterations, scoring them with numbers between zero and one. “Our results can help cancer researchers set experiments to see how important these changes are in pancreas cancer and whether or not they are good targets for potential drug treatments,” says Karchin. “Researchers may want to make a mouse model of a mutation that could be important in cancer so that they can test an inhibitor drug. We help them figure out which mutation to pick.”
Decoding the Data
Like Karchin, Steven Salzberg is focused on helping investigators interpret the billions of data points generated by human genome sequencing. Researchers sequencing the genomes of a variety of human diseases have called upon his expertise. He and his team have developed computer software that has been used around the world to whittle the data down to the few mutations that could potentially be targeted clinically. As Johns Hopkins uses its genetic discoveries to personalize medicine to the unique cellular characteristics of each individual’s disease, scientists like Salzberg play an essential role.
Salzberg, a biologist and computer scientist, came to Johns Hopkins to focus more intensely on human genomics. Until now, his work has primarily focused on bacterial, viral, animal, and plant genomes. In 2001, he was called in to help sequence the anthrax sent through the mail in the 2001 terrorist attacks, and is now part of the Johns Hopkins research teams helping cancer centers make sense of complex cancer genomes.
Salzberg and colleagues identified mutations that helped the FBI trace the anthrax to a single vial at Ft. Detrick in Frederick, Md. Cancer investigators are hopeful he can again help zero in on a killer. This time the target is genes that cause cancer.
Salzberg’s team has developed one of the first off-the-shelf programs that systematically and accurately find fusion genes. These are, as the name implies, two pieces of different chromosomes that become fused together in error. It turns out that fusions rarely happen in anything but cancer so they have become a major focus in cancer research. This jumbled up DNA is not a therapeutic target, but because it is so specific to cancer, it can serve as a biomarker for early detection of tumor cells before they are visible and as a way to monitor whether a treatment is working. If cancer cells are disappearing then so should the fusion genes.
Until recently, cancer researchers did not appreciate the importance of these types of alterations or how common they were in cancer, says bioinformatics expert Sarah Wheelan. Genetics and epigenetics expert Victor Velculescu came upon similar gene rearrangements, almost by accident, when he was looking for gene mutations that could be potential therapeutic targets. Functionally, he wasn’t sure what the fused genes did but he quickly recognized that they were unique to cancer cells and could be used for individualized cancer screening and monitoring. Wheelan says that alterations such as this can only be found by aligning the cancer genome back to the normal genome and looking for differences. “Steve’s programs do this and may find things we have been missing,’ she says.
Salzberg and colleague Mihaela Pertea also have written a program to find mutations in the breast cancer-associated BRCA1 and BRCA2 genes. With only raw data from a gene sequencer, their program quickly tests a sample for any one of about 70 known mutations to these genes. “We’re trying to make this easier for people to do,” says Salzberg. “The software is very efficient, and it can be run on a standard desktop so that anyone with modest computing experience can do it.” Personalized cancer medicine is only possible if clinicians can rapidly access and interpret the individual information on each patient’s cancer, and this work is proof-of- concept that it can be done. While the initial work was done for breast cancer mutations, Salzberg says his program, which contains a library of mutations that can be added to, was written so that it could be used to find any known mutation in essentially any cancer. “We are still a few years away, but it demonstrates that we can do this in the physician’s office,” says Salzberg. The cost—about $5,000 to sequence a genome—may be the only major limiting factor, but he and other scientists say these costs are declining.
Helping Salzberg and Wheelan toward these personalized medicine goals is Alexander Szalay, an astrophysicist and acclaimed computer science visionary. Szalay, the director of the Johns Hopkins Institute for Data-Intensive Engineering and Science, designed new computer architecture that the Space Telescope Institute uses to measure the three dimensional positioning in space of over 200 million galaxies. His technology is now being used to pour through huge amounts of data related to cancer DNA to draw clinically relevant conclusions. In fact, while Salzberg’s office is located on the Johns Hopkins medical campus, he has his computer servers in Szalay’s Physics department computing facility on the Homewood campus a few miles across town.
Szalay is developing algorithms and databases that store genetic information and can connect it to patient-specific cancer problems. It has the capability to store data on 14,000 individual patient genomes per year and tie them securely to patient records. His algorithms allow researchers to do in real time what would take a basic science laboratory a year to accomplish. Szalay’s technology allows them to quickly query information on large groups of cancer patients and find out what they have in common. For example, they could find similarities among patients who were treated for a particular cancer and relapsed. “These pieces are essential to being able to assess genetic information on tumors better and faster, and they are necessary to providing personalized therapies,” says Wheelan.
Salzberg is hoping to collaborate with Gary Rosner, the Kimmel Cancer Center’s director of quantitative sciences, to better understand how cancers spread. By comparing tumor samples from several sites of metastasis to samples of normal tissue, Salzberg hopes to quantify the accumulating genetic mutations and determine the order in which the tumors occurred, providing a schematic for how a particular cancer progressed. “Perhaps we can help cancer researchers get a better handle on the course of metastasis,” says Salzberg.
Salzberg perhaps most succinctly explains the value of the engineering and cancer researcher collaboration. Engineers are the technology experts, he says, and physicians and scientists are the disease experts. Both are needed to conquer difficult problems like cancer. “Cancer is complicated,” he says, “but together we might figure it out.”
After working for 19 years as a research manager at IBM, engineer Russell Taylor became interested in the role of computers and robots in surgery and medicine. He recognized that the same combination of innovation, computers, and technology that revolutionized the electronics industry had the potential to do the same for health care.
Taylor decided that if he wanted to pursue medical applications, he ought to be in the same place as his customers. He came to Johns Hopkins in 1995 and focused his efforts on building an engineering research laboratory that used computers, robots, and information-based technologies to help plan, carry out, and assess new treatments for a variety of diseases, including cancer. The concept earned him $33 million in seed funding from the National Science Foundation to begin the Center for Computer-Integrated Surgical Systems and Technology, which opened in 1997.
Taylor’s philosophy combines practical business sense with forward-thinking science. “Start with everything you know about the patient, including images, lab results, and clinical data, and add in what you know in general about people and diseases based on statistics and anatomy,” says Taylor. “Combine all of this to make a computer representation, and use it to plan an intervention. Then use computer-based technology to help a surgeon carry out this plan and to assess the results.” His years in electronic manufacturing gave him the industry contacts and knowledge to develop and outfit a unique laboratory—one that could not only create innovative approaches for challenging medical problems, but carry them through from inception to production.
His years in industry also taught Taylor that information is never wasted. “In manufacturing, we save all of the data. When you have a problem, you go back and figure out how a process variation can fix the problem,” says Taylor. His goal is to bring this kind of data mining to medicine.
Radiation oncology physicist John Wong was of a similar mind. Taylor was one of the first people he sought out when he came to Johns Hopkins seven years ago. The two agreed that delivery of medical care should be more information driven, and Radiation Oncology, with computers intrinsic to all aspects of the complex planning and treatment, seemed a natural place to start.
Wong wanted to develop a system that would store data from clinical trials in a way that the information did not just help measure the usefulness of potential new therapies being studied, but could also be queried at any time in the future to improve the care of all patients. They looked to the data archive of Alexander Szalay as the model. The system Szalay and team created not only stored data but was also able to perform interactive, on-the-fly analyses. Wong, fellow Radiation Oncology physicist Todd McNutt, and Taylor envisioned a similar system that could be applied to radiation therapy.
“When we participate in a multicenter clinical trial, we send the data off, and it stays there,” says Wong. “We move on to the next trial and the next patient and start from scratch,” says Wong. They envisioned a system that stored data on all previous patients and used advanced computer technology to cull information and apply it to the planning and treatment of each individual patient. Instead of relying only on their specific patient or trial experiences, radiation oncologists could now use data from all those patients to improve the treatment of new patients.
Most adults make better decisions than children, largely because they have years of life experience to call upon—memories of things they’ve done; what worked, and what did not. In many ways, the system that Taylor, Wong, and McNutt have devised is simply a grander, technology-based form of what our brain helps us do everyday. Now, however, physicians not only benefit from their own experiences with patients but also with the experiences and results of countless physicians and patients, avoiding what does not work and zeroing in on what does.
“Rare cases, a doctor may see every two years,” says Wong. “But now he or she can go back and look at certain attributes of prior similar patients and make a better informed treatment plan.” Keeping data allows care to more quickly reflect advances in science.
“Let’s say there is a discovery in the future that completely changes how a cancer is staged. If that’s the case, then all of my previous study of that stage won’t apply anymore,” says Wong. “But, if you keep the data, when a new discovery happens, you simply go back to the original data and recalibrate it and make it useful again.”
They have piloted their concept, called Oncospace, in head and neck cancers which McNutt says are very geometrically complex and among the most difficult cancers for radiation physicists and oncologists to plan, often requiring 10 to 20 plan revisions. There is a lot of anatomy in the area of these tumors. Critical structures, such as the voice box and salivary glands must be spared radiation to prevent harmful and unpleasant side effects to patients. In their new data-mining approach, the team analyzes the geometry of the patient and tumor. Then they go to a database of all patients who should have been as hard or harder to treat because of tumor proximity to critical structures and other factors, to come up with an optimal treatment plan for each individual patient. In early studies, the method considerably improved treatment plan quality and spared critical organs. As a result, studies in pancreas, prostate, and other cancers are being planned.
Wong, who has worked both in medical physics and research and in the clinical setting feels strongly that advances developed in the academic medical setting should be put into practice everywhere. Data mining like this, he says, allows doctors in a community where they may not see as many patients as a Johns Hopkins doctor, to call upon our data to develop better treatment plans for their patients.
Computer-based efforts such as this are considered infrastructure, not research, and so their work is not supported by public funding. While the National Cancer Institute has encouraged their efforts, it will not fund it, so forward progress is often dependent on the ability to find private support.
In the operating room, Taylor also is using technology to improve treatment. Basically, Taylor says, they put a computer between the patient and the surgeon and use technology to enhance his or her capabilities, mainly by giving the surgeon more information. To do this, engineers use a variety of technologies. They use computers, robots, sensors, imaging devices, and guidance systems to interface with clinicians so that they can maneuver inside the body without opening patients up and still clearly see and precisely manipulate their anatomical targets.
With an impressive resume, which includes work on technology used in the da Vinci robotic surgical system for minimally-invasive surgery, Taylor is piloting new advanced technologies that can be used in laparoscopic surgery for liver cancer, kidney cancer, prostate cancer treatment, and other malignancies. Among his inventions are high-dexterity robotic devices that can, for example, get into a bone where a cancer has spread and clean out tumor cells. The snake-like device is deployed through a tiny hollow pipe that moves around the cavity. Other appliances, including water jets and brushes, can be used in conjunction to help remove cells. The void in the bone is then filled with a special cement to maintain bone strength and stave off treatment-related fractures. These nimble, image-guided robotic devices also are being explored for head and neck surgery where surgeons must maneuver with little visibility below the vocal cords and other areas.
Cancer surgeon Michael Choti was one of Taylor’s first collaborators. A recent project involving Choti, radiologist Emad Boctor, and others from the center uses ultrasound elastography to help Choti improve treatment for patients with advanced liver cancer. For many patients, Choti can avoid cutting these tumors out by using special devices to burn or ablate lesions in the liver. The problem for Choti and other surgeons is that there is no real quantitative measurement to let them know that the tumor, and an adequate margin around it, is completely destroyed. They must rely on visual observations and generalized time recommendations. Surgeons want to heat just what they need to—no more, no less, so Choti and the other team members are piloting the elastography tool that acts as a gauge for the firmness of tissue being heated. “Cooked” tissue will be stiff, so the new technology allows surgeons to direct the right amount of heat to cancerous lesions and monitor the surrounding normal tissue to prevent damage.
Taylor and team also are working with Choti and other Johns Hopkins surgeons on improving technologies to advance minimally invasive surgery. These approaches require precise and clear image guidance because surgeons have to be able to see their target without actually opening the patient up. “Targeting and monitoring has been suboptimal and a major roadblock to advances,” says Choti. “Our goal is to do these procedures the best way they can be done, and our collaborations with engineers have been invaluable in making this happen,” says Choti. “It brings innovation to the bedside and helps to keep it cost effective.”
Big Shiny Object Syndrome
Radiation oncology is unique among other cancer-related specialties in that it is completely dependent upon technologically-advanced equipment, such as linear accelerators, to deliver the precisely targeted X-ray beams that cut through and destroy tumor cells.
Patients hear advertisements about one machine or another, and they think if a hospital doesn’t have a particular one, they’re not practicing good medicine. In many instances, Wong says, the information is marketing-driven, not fact-driven, and it is confusing patients. All too frequently claims of clinical benefit are made with no scientific evidence to back it up.
In this new age of medicine driven by marketing, it is Wong’s job to help Radiation Oncology and Molecular Radiation Sciences director Ted DeWeese determine which advanced technologies are needed to improve patient care and which are simply a product of aggressive advertising by industry and for profit health care practices. “Here with Dr. DeWeese, we do it right,” says Wong. “We don’t believe in purchasing big, expensive pieces of equipment simply because it’s the newest thing out there.” The truth is many of the machines actually do the same thing. The only people really benefitting are the people making and selling them. We’ll explore these technologies because we want to be sure Hopkins tells people how to do it right,” says Wong.
Radiation treatment, Wong says, is in essence a form of minimally invasive surgery. Instead of scalpels, radiation oncologists use precisely targeted beams of radiation to cut through tissue and tumors. The main reason there is no scientific evidence to support many of the industry claims is that, for radiation oncology, there was no way to do the research. “There was no laboratory counterpart for what we do in the clinic,” says Wong. So, he called in Taylor to build a downsized version of a human machine for mice. It is radiosurgery for mice. With unprecedented precision, Wong uses the machine to perform human-quality radiosurgery on mice so that he can have a realistic model to study what the doctors do in treatment. “In radiation oncology, we didn’t have the means to study mechanisms in animals, and this machine helps us do that,” says Wong. The miniature machine allowed Wong’s medical collaborators to uncover the causes of—and how to avoid—radiation treatment-induced bone fractures. In radiation oncology there is nearly equal interest in how to destroy tumors as there is in how to prevent damage to normal tissue, so being able to study human therapies in animal models is paramount to developing more effective and safer ways to treat patients.
While fancy machines are often the focus of attention, Wong says, the biggest problem in radiation oncology often is not the treatment technology but rather the ability to characterize the tumor. “If I don’t know where the disease is, it doesn’t matter how great the equipment is, I can’t treat it,” says Wong. “People at Hopkins ask, ‘What is the problem? What is the solution?’” This, he says, is where the collaboration between engineers and Cancer Center experts comes into play. Cancer imaging experts like Martin Pomper are helping make cancers more visible by engineering light-emitting pharmaceuticals that are specific to cancer cells to make them glow. “Now, I can see the cancer, and not just the more obvious tumor but other cells that may be hiding,” says Wong. Taylor is building robotic devices that integrate image guidance and combines technologies, such as ultrasound and CT, to provide clearer and more detailed pictures of tumors and anatomy.
“These types of advances are unique to Johns Hopkins,” says Wong, “and they are a direct result of the Kimmel Cancer Center and Engineering collaborations.”