Expanding Application of Unique Ultrasound Technique
Trahey and Nightingale
Pratt engineers evaluating a unique Acoustic Radiation Force Impulse (ARFI) ultrasound are now expanding the technique's usage beyond previous applications to breast cancers, to include other kinds of tumors, and more tissues. The sound waves produced by ARFI ultrasound "push" on tissues to help physicians diagnose abnormalities such as tumors.
"We have really advanced our technique," said Kathy Nightingale, an assistant professor of biomedical engineering who, together with her thesis adviser Gregg Trahey, pioneered ARFI imaging technology beginning in 1993.
"Our ARFI program has been very successful in attracting research support," said Trahey, the professor of biomedical engineering. "We began with small grants from the Department of Defense and the Whitaker foundation. We are currently funded by the National Institutes of Health (NIH) to develop this modality and investigate different clinical applications."
Both professors' research groups were just awarded NIH research grants of about $1.4 million for each team over five years for ARFI research. With their latest funding, Nightingale's investigators plan to evaluate ARFI's use for imaging colorectal cancers and associated lymph nodes.
Trahey's team will use theirs to study ARFI's potential for guiding a probe used in liver cancer treatment. With another NIH grant, Trahey's team will also evaluate whether ARFI can discriminate atherosclerotic arteries from healthy ones.
Ultrasound focuses high frequency sound waves into the body either to create images of internal tissues or to heat such tissues. ARFI is a special kind of ultrasound that employs two different sound pulses. One pulse type is a high-energy beam that pushes on tissues like sonic fingers. The other is a tracking beam that monitors the resulting tissue motion.
Those ARFI pulses pass into the body and are reflected from internal structures via a "transducer" the size of an iPod that is held next to the skin. A computer processes and records the returning information for display on a monitoring screen.
In the April 2002 issue of Ultrasound in Medicine and Biology, Nightingale and three other authors, including Trahey, published an initial report showing that ARFI was able to move various kinds of tissues in a way that provided potentially diagnostic information. Evaluating the method on human breast, bicep, thyroid and abdominal tissues, that study concluded that it "holds considerable clinical promise."
Nightingale's ultimate initial goal is to use the technology to distinguish benign from malignant lumps in the breast by the different ways they move in response to the pushing beams. Physicians try to do that by "palpating" breast surfaces with their fingers. But Nightingale's hope is that ARFI sound waves can probe deeper and provide more detailed information about the tumors.
Continuing research suggests that malignant breast tumors are stiffer than the tissues surrounding them. "We are also investigating the differences we have observed in the speed of responding tissues," she said. "Smaller, stiffer tumor sites respond more slowly than the surrounding tissue."
Early clinical breast studies with human patients are underway under supervision of Duke Medical Center radiologists Mary Scott Soo and Jay Baker, but results have not yet been published. Another study presented in May at the American Roentgen Ray Society's annual meeting found that a related ultrasound technology also developed by Nightingale and Trahey -- called "streaming detection" -- could better help physicians distinguish between benign fluid-filled breast cysts and solid masses.
Since the group's 2002 Ultrasound in Medicine and Biology report "we have really advanced our technology," she said. "Through mathematical modeling, experiments and simulations, we have come to better understand the complex physics underlying ARFI imaging. And we've utilized this knowledge to make more detailed, clinically relevant images." Trahey added that the researchers -- who work with modified commercial Siemens ultrasound equipment under an arrangement with that company -- now have more selectivity and control of their specialized ultrasound beams, and more knowledge of how to avoid problems.
With growing confidence and sharper pictures the ARFI investigators have taken on more graduate students and are extending their interests to other organs.
"We have found that stiffness imaging also does a good job of imaging different structures," Nightingale said. "It looks promising for differentiating layered tissues and determining the depth of tumor penetration into the layers."
With this new imaging promise in mind, under its new NIH grant the research team led by Nightingale is now evaluating ARFI to define the spread of malignant rectal tumors.
Depending on how deep the cancer has penetrated, Nightingale said physicians treat those in one of two ways. Deep-penetrating tumors require chemotherapy or radiation, followed by extensive surgery. Shallow-penetrating varieties may need only a less-aggressive form of surgery.
Physicians currently use conventional ultrasound imaging to evaluate penetration depth, the goal being to identify the bottom "margin" of tumor growth. "But it's only 70 percent accurate," Nightingale said. "If ARFI could do a better job of determining the depth of penetration, that could save some patients from the more aggressive form of treatment."
In initial ARFI screening of rectal tumors "we've seen delineations of bottom margins that are in some cases better than normal ultrasound, so we're encouraged," she said.
Its new grant will also allow her team to test ARFI imaging to show whether rectal tumors have spread to nearby lymph nodes. "Right now, our ability to determine malignant involvement of lymph nodes is not very good," she said. "Magnetic resonance imaging (MRI) computerized tomography (CT) and conventional ultrasound are all used, but none of them work very well."
Nightingale is working with Duke Medical Center pathologist Marcia Gottfried and surgeon Kirk Ludwig on this project.
Meanwhile, with its new grant. the research team led by Trahey is evaluating ARFI imaging to monitor the effectiveness of radiofrequency ablation heat treatments that attempt to destroy liver and kidney cancer cells.
That procedure uses ultrasound, CT or MRI guidance to help physicians aim a probe that heats tumor tissues to the point that it "denatures" protein as when eggs undergo cooking. That denaturing creates a lesion area of damaged tissues.
The challenge is to "cook" only the cancer cells while leaving adjoining cells unharmed, Trahey said. "That depends on how well you can see. A rigid electrode is directed through the abdomen into the tumor, you hit the button, and the machine delivers a fixed amount of current in a fixed amount of time.
"But presently, the procedure is performed in a largely blinded fashion. CT, MRI and regular ultrasound do not visualize the ablated tissue very well during that procedure. Our hypothesis, our hope, is that with ARFI we could actually see the perimeters of the cancer and preferentially cook the tissue to the tumor boundaries," he said.
Results of a preliminary study that used both heat and chemical treatments to induce lesion formation in animal liver samples were encouraging. A paper published in the March, 2004 in Ultrasound in Medicine and Biology found that, unlike conventional ultrasound, "ARFI imaging was capable of monitoring lesion size and boundaries."
For clinical evaluations of ARFI with radiofrequency ablation Trahey will work with Duke Medical Center radiologist Rendon Nelson. The technology has also been evaluated in animals to monitor the creation of radiofrequency ablation lesions in beating hearts -- used clinically in humans to correct some forms of cardiac arrhythmia.
According to a paper by Duke ARFI researchers in the April, 2005 issue of IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control, resulting images were able to distinguish differences between healthy and damaged tissue regions. Those differences were "not visualized well" by conventional ultrasound, the group reported.
Trahey's group's atherosclerosis studies ask a very different question: whether ARFI can look inside the body to sense whether some arteries are becoming clogged with a potentially dangerous variety of plaque that is vulnerable to rupture.
"We hypothesize that we can tell the difference between hard and soft plaques with ARFI, since ARFI is fundamentally a stiffness-imaging modality," said Trahey. "We also hypothesize that in healthy people we will detect soft vessels, and that in patients with known vascular disease we will see stiff vessels," Trahey said. "And our very preliminary data supports that."
That data, published in the September 2004 issue of Ultrasound in Medicine and Biology, focused ARFI pulses on arteries in amputated human limbs as well as on arteries in Trahey's and Nightingale's own necks and thighs.
That report concluded that ARFI images "demonstrated resolution and contrast on par with, and in some cases, better that that demonstrated in matched B-mode (normal ultrasound) images."
Other coauthors of that study were Duke Medical Center pathologist Rex Bentley and Nightingale's graduate student, Mark Palmeri. Palmeri "has been working with us since the inception of ARFI imaging, and has developed our modeling tools and studied the mechanical an thermal responses of tissues to ARFI pulses," she said. "The results of these studies have been used to design our experimental and clinical trials." He will complete his Ph.D. at Pratt this fall and then finish his M.D. training at the medical center.
Other graduate students on Nightingale's and Trahey's teams include Richard Bouchard, Doug Dumont, Brian Fahey, Kristin Frinkley, Molly Gregas, Stephen Hsu, Gianmarco Pinton and Liang Zhai.