Pioneering Prostate Imaging for More Accurate Cancer Diagnosis
January 04, 2017
Innovations in Cancer - Winter 2017
In a world where medicine is becoming increasingly precise, prostate cancer remains a stubborn outlier. Screening methods are bogged down in controversy, which is confusing for both physicians and their patients. For many physicians, diagnosing the condition can almost feel like a shot in the dark.
“The prostate is truly the last organ in the body that we are biopsying without hitting a visualized target,” says Lee Ponsky, MD, Chief of the Division of Urologic Oncology with University Hospitals Seidman Cancer Center and UH Urology Institute; Professor of Urology, Case Western Reserve University School of Medicine. “There have been some advances in how we treat certain cases, but prostate cancer has been relatively stagnant over the past 20 years.”
Impatient with the pace of progress, a team of physicians and scientists at UH is working to alter that reality – and not just with incremental improvements around the edges. The multidisciplinary team of urologists, radiologists, radiation oncologists, medical physicists, pathologists and biomedical engineers is well on its way to establishing a new, disruptive model of care.
“It’s a ‘leave your egos at the door’ approach,” says UH radiologist Vikas Gulani, MD, PhD, Director of Magnetic Resonance Imaging; UH Cleveland Medical Center; Associate Professor of Radiology, Case Western Reserve University School of Medicine. “Our goal is to form a multidisciplinary clinic in which the patient gets imaged and we come up with a diagnosis and treatment plan, ideally on the same day, eliminating the uncertainty that can carry on for weeks on end.”
“We have a clear vision together to make this happen, to make it clinically relevant and clinically impactful,” Dr. Ponsky adds. “It’s wonderful to discover new technologies. But we want to move the needle. We want to make a monumental impact, and that’s what we’re working to do.” Innovations in prostate imaging are at the heart of the approach.
“We’re using all the tools at our disposal to change how prostate imaging is done,” Dr. Gulani says. “That includes current state-of-the-art MRI, developing screening methods with MRI, developing MR fingerprinting technology and developing advanced image analytics with the help of biomedical engineering. We want to achieve the goal of reducing the uncertainty and reducing the number of unnecessary biopsies, procedures and treatments against low-grade cancer.”
The team has already developed a rapid, noncontrast screening exam that they have used in more than 100 research cases. This exam, which does not require an IV and is performed in under 15 minutes on the MR table, has an extremely high negative predictive value for prostate cancer in patients who ordinarily would have gone straight to non-targeted biopsy. In patients who have a suspicious focus on the MRI, a biopsy can be targeted at the suspicious area using one of several approaches available.
UH Seidman Cancer Center physicians are also among the first in the country to offer in-gantry, MRI-guided prostate biopsy.
“The in-gantry MRI-guided biopsy has the potential to revolutionize prostate cancer detection,” Dr. Gulani says. “The MRI reveals the specific characteristics that distinguish between normal and diseased tissue. Visualizing the tumor’s location within the prostate enables physicians to more accurately target and biopsy the lesion.”
Currently, transrectal ultrasound (TRUS) guided biopsy is still the standard test to diagnose prostate cancer. Ultrasound can show the size and shape of the prostate gland, but not the exact shape and location of a tumor. Thus physicians are “blindly” biopsying tissue in the gland without foreknowledge of the tumor. Fusing an MRI to the ultrasound is possible, and this is done at UH as well, but is an imperfect solution.
In addition to increased accuracy, the MRI technology may improve the safety of prostate biopsy. Due to its precise targeting, only two to four biopsy samples are needed for an accurate diagnosis, compared to 12 samples with the standard technique, thus reducing risk of infection, bleeding, pain and recovery time. Additionally, standard ultrasound biopsy has false-negative rates of up to 30 percent. There also can be a need for repeat biopsies when PSA continues to rise despite negative biopsies.
Patients who may benefit from an in-gantry MRI-guided biopsy include men who are suspected of having prostate cancer, but who may have inconclusive tests results and have experienced any of the following: persistent, unexplained elevated PSA, prior negative biopsy, or increased prostate cancer gene expression, or low-risk prostate cancer being monitored by active surveillance (or watchful waiting). The use of MRI-guided biopsy is just the latest innovation adopted by Drs. Gulani, Ponsky and a multidisciplinary team of urologists, radiologists, radiation oncologists, medical physicists, pathologists and biomedical engineers.
The team has also used magnetic resonance fingerprinting (MRF) in about 150 prostate cancer patients to date. This new technology, developed at Case Western Reserve University and UH and reported in the journal Nature in 2013, adds a quantifiable, reproducible aspect to traditional MR. It uses highly unusual and novel MR signal acquisitions to generate simultaneous measurements of multiple tissue properties for each pixel in the image, yielding quantitative maps of these tissue properties. These maps are used to noninvasively and definitively characterize tissue – both normal and abnormal
“If you look at a typical MR image, it is a map of the anatomy and that is it,” Dr. Gulani says. “The maps of the anatomy are beautiful, but they are not quantitative. What that means is that day to day, scanner to scanner, site to site, it is very difficult to compare results.”
Dr. Gulani worked on this problem off and on for about 10 years, together with Mark Griswold, MD, Professor of Radiology at Case Western Reserve University School of Medicine, and other colleagues. Just when the group reached its lowest point and was about to give up, Dr. Griswold posed an important question: What would happen if they didn’t care what the image looked like?
“We decided to only go after a map of tissue properties that we care about,” Dr. Gulani says. “As soon as you stop trying to collect a traditional image and go instead after the properties, then you can let your signal vary in ways that you never thought possible.”
The result of this insight is magnetic resonance fingerprinting (MRF) – a new imaging technology that can generate immediate and simultaneous values for T1 and T2 for each pixel in the image. These tissue characteristics describe the environment of each particular proton, such as fat or water. These values are then matched against a “dictionary” of all possible signals, even when the values are “noisy.”
Dr. Gulani likens MRF to distinguishing the letter “A” from the letter “B” on a page, even when parts of the letter are faint or missing, or to being able to correctly identify “The Star-Spangled Banner” on a radio station with a lot of static, just based on the words “say” and “see” and a fragment of melody.
“The pattern-matching algorithm that is your brain knows what it is,” he says. “MRF works the same way.” The application to prostate cancer is one of the earliest clinical uses of this technology. The hope is to provide a quantitative separation of prostate cancer from normal prostatic tissue, and perhaps even provide an estimate of the aggressiveness of the cancer. Dr. Gulani is pleased to see the pioneering work in MR physics start to move to the clinical setting.
“We develop technology for the sake of helping patients, and I am really excited to move our work to this all-important stage of application,” he says.
Case Western Reserve University, UH Cleveland Medical Center and Siemens Healthcare recently announced an exclusive research partnership to further refine and develop MRF. But already, the approach is yielding dividends for prostate cancer patients.
Beyond MRF, advances in engineering technology are also part of the approach. Anant Madabhushi, PhD, Professor of Biomedical Engineering at Case Western Reserve University, has worked to develop technology that aligns MRI and transrectal ultrasound-guided (TRUS) imagery without manual intervention. This MAPPER Multiattribute Probabilistic Prostate Elastic Registration) technology has been shown to increase the yield of cancer-positive biopsies. Dr. Madabhushi is also collaborating with the clinicians to overlay pathological information from prostate tumors with data acquired using MRF.
“The idea is that computers can analyze MRI images and extract quantitative information about the structures in the images. Then we can input the pathological data from the patient’s tumor after surgery, including measurements of the texture of the cancer,” Dr. Ponsky says. “By fusing this information, the computer can then learn to recognize cancer. This is revolutionary.”
Another goal of the prostate group at UH is exploring cost-effectiveness, especially when it comes to MRI.
“Diagnostic MRI and MRI-guided biopsy have been shown to be effective in detecting clinically significant prostate cancer,” Dr. Gulani says. “However, despite the advantages, there is reluctance to incorporate MRI into standard practice because it is perceived to be expensive.”
New data from the UH-Case Western Reserve School of Medicine prostate team challenges this perception. The group compared cognitive MRI-guided biopsy, ultrasound-MR fusion or in-gantry MRI with standard prostate biopsy, presenting its findings at the recent meeting of the American Urological Association.
“Contrary to the knee-jerk reaction many people have, we found through careful analysis that MR prior to biopsy is actually cost-effective in the patient’s care,” Dr. Gulani says. “This is a very important fact, given the present state of discussion of our health care system. Whatever the costs of prostate cancer care are today, they are set to just skyrocket in the next 10 years. To be successful requires paying attention overdiagnosis and overtreatment.”
For their part, Dr. Gulani and Dr. Ponsky say they are committed to disrupting the current model of care for prostate cancer, attacking its shortcomings from every angle.
“It’s a commitment to conferencing every week, reviewing our cases, talking with radiologists and urologists, and trying to improve our ability to interpret every day,” Dr. Ponsky adds. “We’re coming at this from all aspects and building a huge foundation that will allow us to change how we practice for prostate cancer. We think we have the tools to do it. “