Carolyn Riley Chapman, PhD, Chapman is a freelance writer based in Harrison, N.Y.
Collaborative efforts of pharma-biotech, academia, and government agencies push advances in imaging technologies.
Because imaging can provide early indications of a drug’s efficacy (or ineffectiveness) and help determine optimal dosing and scheduling, it is increasingly being used to support decision making in drug development.
 |
Recently, pharmaceutical companies have been leveraging the power of imaging to advance cardiovascular disease, Alzheimer’s disease, and oncology drug development. Pharmaceutical companies and government agencies alike are making clear investments of time and money to develop and use a greater repertoire of imaging biomarkers.
GlaxoSmithKline, London, for example, partnered with Imperial College London in 2004 to build a £76 million ($144 million) clinical imaging research center with magnetic resonance imaging (MRI), positron emission transmission (PET), and optical imaging facilities at Hammersmith Hospital, UK. Last year, Pfizer, New York, opened a phase I clinical research unit in New Haven, Conn. The center was strategically located near Yale University School of Medicine, which has PET and MRI facilities. Pfizer will collaborate with Yale to include imaging in its clinical trials.
Cardiovascular disease:
atherosclerosis
Coronary intravascular ultrasonography (IVUS), which involves the insertion of a catheter into the coronary artery in order to measure the size of an atherosclerotic plaque, is the “imaging technique that is the hottest and most important right now in anti-atherosclerotic drug development,” says Steven Nissen, MD, interim chairman, Department of Cardiovascular Medicine, Cleveland Clinic Foundation. The technique is precise, sensitive, and specific, he says.
 |
Nissen has used IVUS to evaluate the efficacy of many different anti-atherosclerotic drugs. The technique is capable of supporting go/no go decisions in drug development; some trials have turned out positively and others have revealed drug ineffectiveness. For example, the ASTEROID trial, published in April 2006 in JAMA, demonstrated, for the first time, that statin therapy caused atherosclerotic plaque regression. In stark contrast, the ACTIVATE trial of pactimibe, an ACAT inhibitor that was under development by Daiichi Sankyo Company Limited, Tokyo, showed no difference between drug and placebo on the primary endpoint of percent atheroma volume [New England Journal of Medicine (March 2006)]. With drug-treated patients faring worse on secondary endpoints, development of the drug was halted.
Although clearly valuable, IVUS does have disadvantages, Nissen says. First, it is invasive. Also, the relationship between IVUS results and clinical events is not certain. Nailing down the link between atherosclerotic plaque regression and clinical outcomes like morbidity and mortality is difficult, Nissen explains, because IVUS trials tend to be modest in size and duration (roughly 300 to 1,000 patients with about a two-year follow-up).
However, historical evidence suggests that “the faster you’re progressing, the more likely you are to have a coronary event, and if you’re not progressing, if you’re regressing, the better you’re going to do,” Nissen says. In fact, imaging of atherosclerotic plaque may become an acceptable surrogate marker for cardiovascular disease, even for registration purposes. Nissen says that a number of drugs currently in clinical trials are aiming to use imaging as a primary endpoint for registration. Both carotid intima-media thickness (IMT) and IVUS studies are typically being performed for these purposes, he adds.
One drug currently undergoing evaluation in large imaging trials is Pfizer’s torcetrapib, a high-density lipoprotein elevator currently in phase III trials. According to Stephen Lederer, senior director of media relations, Pfizer Global R&D, Pfizer is hopeful that the US Food and Drug Administration (FDA) will accept imaging results as an endpoint, but there are also definitive mortality and morbidity trials in progress as well.
Li-ming Gan, MD, PhD, associate professor at Sahlgrenska University Hospital, Sweden, the Sahlgrenska Academy at Göteborg University, and research scientist with AstraZeneca, London, says pharmaceutical companies are searching for new, noninvasive techniques to measure coronary artery plaque with high sensitivity.
Gan uses a micro-ultrasound imaging system from VisualSonics Inc., Toronto, Canada, to obtain real-time, non-invasive, in vivo images of coronary artery plaque in mice. This is a difficult task because coronary arteries are small and move with the beating heart. With a resolution of 40 microns, the system produces images comparable to what can be obtained in humans, Gan says. According to him, the ultrasound technique is comparatively cheap, uncomplicated, and provides good throughput: about 15 mice can be scanned in three hours. In contrast, other techniques used for preclinical cardiovascular imaging, such as micro MRI, are more time consuming and do not enable real-time measurements, he says.
Gan believes the VisualSonics machine may even prove useful as a noninvasive method for measuring coronary artery atherosclerosis in humans: “We are trying to use the mouse ultrasound biomicroscopy machine to measure intima-thickness in humans as well.” According to Nissen, other approaches in development for noninvasive measurement of human plaque include computed tomography (CT) angiography, and MRI.
Alzheimer’s disease
GE Healthcare, Chalfont St. Giles, UK, is currently developing diagnostic imaging agents for Alzheimer’s disease. Imaging agents that diagnose the presence of beta-amyloid plaque before disease symptoms manifest could potentially enable earlier treatment of the disease with better outcomes. They could also be used to facilitate development of Alzheimer’s drugs.
“The whole idea is, if we can measure plaque, then not only can we identify Alzheimer’s patients, but we can also measure the effect of drugs, assuming that the drug effect is intended to reduce plaque,” explains Richard Frank, MD, PhD, vice president of medical and clinical strategy at GE Healthcare. Although drugs would need to demonstrate improvement of Alzheimer’s symptoms to be approved, the ability to see them affect plaque levels would improve decision-making and enable “labeling to include the additional claim of disease modification,” Frank says.
Validated Alzheimer’s imaging agents could also be used to improve lead identification and optimization during preclinical development, Frank points out. In the last few years, PET scanners for small animals have become available, he says, which greatly facilitates translation of results from animals to humans.
 |
Last year, the company announced that it will collaborate with Roche, Basel, Switzerland, in clinical trials of Roche’s anti-amyloid drug. Patients will be monitored with GE’s PET imaging agent, Pittsburgh Compound-B (PIB), Frank says. PIB was licensed by GE Healthcare from the University of Pittsburgh in 2003. The collaboration aims to validate the efficacy of both Roche’s anti-amyloid drug and GE’s diagnostic agent.
In a separate collaboration, GE Healthcare is working with Eli Lilly, Indianapolis, to further expand GE’s portfolio of beta-amyloid diagnostic agents for Alzheimer’s disease. GE will have access to Lilly’s libraries, and Lilly will be able to utilize diagnostic agents that GE identifies through the partnership. By collaborating on the development of diagnostics and therapeutics for the disease at the same time, the companies hope to realize synergies on both fronts.
Earlier this year, the National Institutes of Health launched the Alzheimer’s Disease Neuroimaging Initiative (ADNI), a $60 million, five-year study aiming to attract 800 participants through 58 trial sites located in the U.S. and Canada. “The primary goal is to try to develop and determine the best biomarkers of disease progression in Alzheimer’s disease,” says Susan Molchan, MD, program director for the ADNI project at the National Institute on Aging (NIA). The study will monitor patients using imaging technology, as well as blood, cerebrospinal fluid, and urine sample analysis.
Molchan says that all participants in the study will receive serial anatomical MRI scans over a two- to three-year period. About half the participants will also receive FDG-PET scans to measure cerebral glucose metabolism. According to Molchan, prior studies have linked changes in MRI and FDG-PET scans with Alzheimer’s disease progression and risk factors.
Initiated by the NIA, the ADNI study involves many federal agencies, pharmaceutical companies, and private non-profit organizations.
In oncology, imaging already plays a large role in clinical development as surrogate endpoints for registration include reduction in tumor burden or progression, typically measured by CT or MRI. But tumor size may not tell the whole story, particularly with the new generation of anti-angiogenic and cytostatic drugs.
Anti-angiogenesis drugs decrease the rate that new blood vessels are recruited, but because they do not kill cells directly, they may not cause tumor shrinkage, explains Homer H. Pien, PhD, managing director, Center for Biomarkers in Imaging, Massachusetts General Hospital Dept. Radiology, Chelsea, Mass. Because many of the new cancer drugs that are hitting the market or that are in the pipeline rely on this mechanism, the big question is how to assess their efficacy without relying on mortality/morbidity data, he says.
According to Pien, there are two big trends in oncology imaging. First, more companies are turning to functional perfusion MR and CT imaging. Perfusion imaging involves taking repeated images of the tumor region after administration of a contrast agent. By measuring the rate the agent enters and exits the tumor, blood volume and flow can be determined. “If I am using an anti-angiogenic drug to try to starve the blood supply to a tumor, then you would think that the amount of blood that’s going in, or the blood volume, or conversely, the rate at which blood is going in, the blood flow, they would both decrease after the drug, and that’s exactly what we see [with] many of the anti-angiogenic drugs,” says Pien, noting that CT perfusion imaging was performed in Avastin rectal cancer trials (Genentech, S. San Francisco).
The second trend is FDG-PET, which Pien says is “incredibly important” for companies making go/no go decisions in clinical development. One case in point is Pfizer’s Sutent, approved in January 2006 for the treatment of gastrointestinal stromal tumors, a rare stomach cancer, and advanced kidney cancer. Sutent was approved under accepted criteria for reduction of tumor size. But FDG-PEG imaging also played a significant role in its trials. Tim McCarthy, PhD, director and platform technology leader for molecular imaging, Global Clinical Technology, Pfizer Global R&D, says that FDG-PET studies showed significant metabolic shutdown in patients ahead of tumor burden reduction or clinical benefit. The ability to see such an early response to the drug gave the organization confidence to push Sutent’s development forward, McCarthy says.
The striking FDG-PET results for both Gleevec (Novartis, Basel, Switzerland) and Sutent have certainly garnered industry attention. Stemming from the Critical Path Initiative, announced by the FDA in March 2004, there is a big push to facilitate the development of new imaging agents and protocols for use during drug development, says Barbara Croft, PhD, program director, Cancer Imaging Program, National Cancer Institute (NCI), Bethesda, Md. One example is the Oncology Biomarker Qualification Initiative (OBQI), which is a January 2006 agreement between the FDA, the NCI, and the Centers for Medicare & Medicaid Services to collaborate on oncology biomarker development. The first projects of the OBQI will be to see whether FDG-PET predicts tumor response for lymphoma and lung cancer. Croft says one of the critical factors for adopting imaging biomarkers is standardizing software and protocols. “We have got to, for example, be able to show that if you did a PET study in one institution, and you were able, which you typically don’t because of the radiation dose, to take that same patient across the street to another institution . . . and do the study again, you would get the same answer. We don’t have that assurance yet, and we do need to get that.”
While FDG-PET is promising, Croft points out that other markers are needed in oncology drug development. “FDG is not the perfect probe for slowly growing cancers that don’t use glucose as their primary energy source,” she says. So there is a demand for other imaging agents such as apoptosis and proliferation markers, which at this point are not as well-developed, she says.
Imaging Agents
VisEn Medical Inc., Woburn, Mass., founded in 2000 by Ralph Weissleder, MD, PhD, professor of radiology and director of the Center for Molecular Imaging Research at Massachusetts General Hospital, and Kirtland Poss, now president and CEO, is one company that develops and sells novel imaging agents that provide windows into biological processes.
According to Poss, VisEn’s probes all have at least two of three components: 1) a near-infrared biocompatible fluorochrome reporter, 2) a biocompatible carrier or backbone that allows the probe to circulate within the body and reach the appropriate place, and 3) either an affinity ligand or substrate that links it up to or is activated by a particular type of enzyme. VisEn’s probe portfolio includes ProSense, a protease-activatable agent that can be turned on by disease-associated proteases including cathepsin B, L, S, and plasmin, and AngioSense, a fluorescent macromolecule that enables imaging of blood vessels and angiogenesis since it remains in the vessels for a relatively long time. In March, the company announced plans to launch another line of probes, including AminoSpark, a platform that enables fluorescent labeling of peptides, antibodies, or small molecules of interest.
Although VisEn’s exogenous fluorescent bioprobes are currently only used in animals, Poss says VisEn’s probes are “fundamentally capable” of translation into clinical use. According to him, VisEn’s customers, which include Merck, Whitehouse Station, N.J.; Lilly, Novartis and Amgen, Thousand Oaks, Calif., are particularly interested in the ability to bridge their preclinical studies into clinical programs. As a potential answer to their wishes, VisEn is planning to bring ProSense into phase I sometime in 2007, Poss says.
This article was published in Drug Discovery & Development magazine: Vol. 9, No. 10, October, 2006, pp. 20-28.
Filed Under: Drug Discovery
