May 1, 2005 (Vol. 25, No. 9)

Bridging Early Preclinical Drug Discovery With the Later Stages of Clinical Development

First comes genomics, then comes proteomics, then comesin vivo molecular imaging. In vivo molecular imaging is a relatively new, interdisciplinary field focused on the non-invasive measurement and characterization of biological, cellular, and molecular processes as they occur in living people and animals.

A recent Cambridge Healthtech Institute conference explored the burgeoning field, focusing particular attention on technology innovation and applications in preclinical drug development.

Clinical Modalities

In vivo imaging can be performed using a variety of modalities, and each has its own particular strengths and weaknesses. “The imaging modalities all have a lot to offer,” says Douglas A. Bakan, Ph.D., vp, business development, Alerion Biomedical (San Diego). You just “have to use them for the right applications.”

Dr. Bakan says he uses cost, spatial resolution, sensitivity, ability of depth probing, and quantification potential as discriminators to evaluate the different modalities and determine which is the best tool for specific situations.

The bulk of imaging in the clinical setting is carried out by a fairly small number of modalities, says Paul Picot, Ph.D., advanced applications manager, preclinical imaging, GE Healthcare (Chalfont St. Giles, U.K.).

“The workhorse is the CT scanner, which is wonderful, fast, and general in purpose. Its primary benefit is speed and cost, but it’s not good at soft tissue discrimination. To fill that niche, you have MRI.”

While CT and MRI are excellent for gathering anatomical information, the nuclear modalities like PET and SPECT are “the tools used to investigate the function of organs and molecular pathways,” says Dr. Picot.

One key example of functional imaging performed in the clinic using PET is the wide use of FDG, 2-fluorodeoxyglucose, as a surrogate marker for tumor metabolism. “It’s basically a heat seeking missile for tumors,” says Derek Maclean, Ph.D., senior scientist at KAI Pharmaceuticals (S. San Francisco).

More and more, companies are using imaging in their clinical trials to try to gain a better understanding of how their development candidates are working in patients. “We always try to incorporate noninvasive imaging into clinical trials hoping we can get prediction of efficacy relatively early,” says Peter Lassota, Ph.D., executive director, oncology, Novartis Pharmaceuticals (East Hanover, NJ).

Clinically-proven modalities like PET, SPECT, CT, MRI, ultrasound, and multi-modality or fusion approaches, like PET-CT, have been adapted for use in small animals and are now commercially available from companies such as GE Healthcare, among others.

Preclinical Imaging

The technologies allow imaging of drugs or contrast agents to be performed in animals prior to testing in humans, and also provide “more freedom on what agents can be used, since you don’t need FDA approval for the use of new experimental tracers in animals,” says Dr. Picot.

Some hurdles remain for use of the clinically-proven modalities in the preclinical setting. For PET, one of the hurdles is “the lack of availability of interesting compounds in radiolabeled form,” says Dr. Maclean. Another issue is expense. According to Dr. Lassota, the biggest hurdle may be lower throughput in animals.

Optical imaging has many advantages over the clinically-proven modalities in preclinical development, including throughput and cost, but one of its weaknesses is that the incorporation of genetic labels cannot easily be translated into use in humans.

“In general, the problem we are facing is that for the imaging modalities used in the clinic, the throughput and sometimes the resolution is too low for use in preclinical studies. For modalities that have good throughput in the preclinical setting you have to use labels, so you can’t use them in the clinic,” explains Dr. Lassota.

To merge preclinical and pre-clinical imaging, “we can use modalities like visible light to select compounds. Then once you have chosen the final compound, you can do a limited number of bridging studies using clinical imaging modalities like MRI, PET, or SPECT. This way you can have the best of both worlds,” suggests Dr. Lassota.

For broad use in preclinical drug discovery and development, though, Dr. Lassota believes, “it is difficult to imagine visible light imaging will be replaced by PET or CT in the next five years.”

Bioluminescence and Fluorescence

The two types of visible light imaging are bioluminescence and fluorescence. Xenogen (Alameda, CA) is one of the industry leaders in the usage of bioluminescent markers, while AntiCancer (San Diego) has pioneered the use of fluorescent markers like Green Fluorescence Protein (GFP). Cambridge Research & Instrumentation (CRI; Woburn, MA) develops and manufactures instrumentation systems for small animal fluorescence imaging.

“When I think about the activities that a pharmaceutical company is involved inpicking a disease mechanism to target, testing a drug against a particular target, and if you use a drug, what are the side effects and safety issues that arisewe can do all of that,” says Pamela R. Contag, Ph.D., president and co-founder of Xenogen.

In vivo imaging is playing an increasingly important role in bridging early preclinical drug discovery, which is often centered around in vitro and/or cellular assays, to later stages of preclinical and clinical development, which aim to elucidate and quantify a drug’s effects in living organisms.

Xenogen uses the light-emitting reporter luciferase to tag genes and proteins within living animals. They create transcriptional fusions, with a promoter of interest driving luciferase expression, or translational fusions to create protein/luciferase chimeras. The tools of molecular biology and the power of mouse genetics can be harnessed to allow tremendous flexibility in the kinds of experiments that can be performed.

In a typical experiment, “you essentially set up a model, a mouse line, that represents something that occurs in a disease state. Once you have the model established, you can run a series of compounds against the model,” Dr. Contag says. One example is the development of models for testing cancer compounds.

“Using our imaging system, you can watch the tumor cells emit light, grow, and calculate the rate of growth. Then you can administer drug and watch the light go away. Since it’s nondestructive, you can watch the animal for a long period of time. If the tumor cells form resistance to the drug, you can see the light come back,” she adds.

Fluorescent markers and imaging systems can also be employed for a variety of purposes, including tumor models. Both bioluminescence and fluorescence have explicit advantages over the measurement of tumor growth in animals using palpation, calipers, and/or histology.

“Each mouse has only one timepoint, and the caliper isn’t reliable. You have to sacrifice lots and lots of animals,” says Richard Levenson, M.D., director of R&D, biomedical systems.

“With imaging, you can get a full curve from one injection. You don’t have to sacrifice animals, and the data is better quality, since each animal serves as its own control. It’s not quicker or cheaper, but it’s competitive in its cost effectiveness, and it’s more ethical,” explains Dr. Maclean.

According to Dr. Lassota, visible light is particularly useful for reporter assays in preclinical development. “I think we’re making steady progress in reporter assays that can be used in vivo, and that at some point, these can replace efficacy assays.

“With reporter assays, you can get a mechanistic readout pretty quickly in mice. You don’t have to wait for the tumor to shrink. You can effectively screen compounds with reporter assays, but they take a lot of work upfront.”

CRI’s fluorescence imaging system uses multispectral technology to reduce autofluorescence and improve sensitivity, allowing the detection of smaller and fainter targets.

“Our system enables the imaging of low level fluorescence in vivo. Before, it was completely obscured by background autofluorescence except for very bright signals,” says Cliff Hoyt, vp and CTO of CRI.

“Fluorescence-based imaging also allows multiplexing,” adds Dr. Levenson. “You can look at two, three, four things at once by taking advantage of different wavelengths. You unmix each from autofluorescence but also each from each other.”

One of the problems with visible light imaging is that deep imaging is more difficult due to the scattering and absorbance of light. “The best results are for things that are more superficial,” says Dr. Levenson, who adds that, by moving into the near-infrared range (750900 nm), signals can be detected from all the way through a mouse.

One of the hurdles can be finding appropriate molecular probes. “The imaging side is easier than the biology and the strategy. There can be problems with specificity and in where they gofor example, they can all end up in the liver and bladder,” says Dr. Levenson.

Imaging Agents, Probes, and Therapeutics

“Custom molecular probes are probably going to be a hugely important tool in the future,” says Dr. Picot. Many companies are focused on the development of molecular probes as imaging agents which can be used in the diagnosis and detection of disease or as surrogate markers for therapy success. The same formulations may be adapted to specifically deliver therapeutics.

For example, Kereos (St. Louis) uses its technology platform to create 250-nm emulsion “particles” which can serve as targeted imaging agents or targeted therapeutics. The formulations contain a perfluorochemical core, which is “extremely inert biologically and chemically,” explains Timothy J. Pelura, Ph.D., CTO.

On the surface, Kereos displays a targeting molecule, such as a monoclonal antibody, small molecule ligand, or peptide, to specifically direct the emulsion particles. The surface also incorporates an imaging effector molecule.

Kereos’ first lead candidate is an MRI imaging agent that targets avb3 integrin, which is specifically expressed by angiogenic blood vessels that feed tumors. The technology allows clinicians to “see tumors down to 1 mm in size, compared with 1 cm with PET, which is the gold standard,” says Pelura. Another product in its pipeline also targets avb3 integrin, but delivers a potent dose of chemotherapeutic instead of an imaging agent.

Kereos’ technology can be adapted to different imaging modalities and therapeutic areas. “The formulations are subtly different, but they are basically plug and play. It’s easy to swap out different effector molecules, or the targeting ligand,” says Pelura.

The company hopes that its technology will allow smaller amounts of drug to be delivered more effectively to the right patients, those who are known to express the disease-associated protein target.

Alerion is another company developing imaging agents and therapeutics. Alerion launched its first two products for experimental research in animals using CT and microCT imaging. Using its lipid emulsion (LE) technology, the company has “exploited normal lipid metabolism to serve as a drug delivery system,” says Dr. Bakan.

Alerion’s Fenestra LC mimics the actions of the chylomicron remnant, which serves to deliver lipids selectively to hepatocytes of the liver through recognition by the ApoE receptor. The agent provides anatomical and functional information in the liver and billiary system.

Another version, Fenestra VC, has a modified surface such that uptake by the ApoE receptor is significantly reduced. Because of this difference, Fenestra VC can be used as “a long lasting blood pool imaging agent. This is important because with microCT, scans still take up to twenty minutes,” explains Dr. Bakan. Both versions of Fenestra allow for “a very prolonged window of opportunity.”

“The ultimate goal is to translate from animal studies to clinical studies, to check up on clinical trials,” says Dr. Bakan.

Previous articleBiotechnology in Vancouver
Next articleDatabase of Macromolecular Movements