By Kevin Davies
May 12, 2005 | For Millennium Pharmaceuticals, it was a very Good Friday. That afternoon, the company learned that its prize cancer drug, Velcade, had won a second indication from the FDA for the treatment of multiple myeloma patients.
Velcade sales topped $140 million in 2004, arguably the most visible sign of Millennium’s maturation from genomics pioneer to profitable biopharma. But to flourish, the company has to ensure the success of its expanding drug development pipeline.
A critical platform in that endeavor is Millennium’s new Imaging Center. According to Craig Muir, vice president for platform technology, the center is a “formal, integrated multimodal facility that signals our maturation as a pharma company. It’s one of the only facilities of its type.”
The center, which opened in 2004, is little more than a concentration of high-tech imaging instruments housed in a warren of non-descript labs. But it is this close proximity and space optimization that makes the center almost unique. In most other research departments or clinical settings, such an array of imaging technologies would be scattered across different buildings.
The 7-Tesla MRI instrument
at Millennium's Imaging Center.
For Millennium, the capability to employ an assortment of computer tomography and imaging technologies - known in the trade as CT, PET, SPECT, MRI — or all of the above, enables scientists to probe cells and tissues in live animals, whether it’s the degeneration of bone in arthritis, lung fibrosis in chronic pulmonary disease, or the runaway growth of a prostate tumor.
Published reports put the cost of the Center at $3 million. Muir hedges, but says the investment is within the scope of other technologies the company uses, such as mass spectrometry and X-ray crystallography. “Imaging is fundamental to us being more effective and accelerating what we’d like to do — increase our productivity with greater ROI,” says Muir. In fact, he adds, “The ROI may be higher than any other platform enhancement we’ve ever considered. Imaging delivers at the interfaces of drug discovery, preclinical development and clinical practice”
Scanning the Horizon
The ability to study drug effects non-invasively in animals has important implications for extrapolating the results into clinical situations. “These modalities are all used in clinical medicine. The ability to apply imaging in animal models will refine the use of these technologies in the clinic and vice-versa,” says Muir. Moreover, with the Velcade indication hot off the press, Muir says: “It isn’t just about getting things to the market, imaging can also help expand and support indications for previously approved drugs.”
Conceiving and building the Imaging Center was largely a ‘bottom-up’ process, says Muir. Throughout the ‘90s, Millennium harnessed and pushed the latest in gene and drug discovery technology, including genomic sequencing, microarrays, and proteomics. About five years ago, Muir explains, things changed: “DNA sequencing was less key to people. We said, ‘Imaging could be the next major thing to implement for our interests, and to refresh the technology solutions business.” Millennium has since dropped that part of its business model. “We are more of a pharma than five years ago,” says Muir.
The head of the Imaging Center is Sudeep Chandra, who was recruited from GlaxoSmithKline. Originally an engineer, Chandra trained at the University of Illinois with Paul Lauterbur, who went onto share the 2003 Nobel Prize for Physiology or Medicine for developing magnetic resonance imaging (MRI). “Sudeep is one of the best in the business, the perfect person to build and lead this center,” says Muir.
“We have to have platforms match up to the clinic,” says the softly spoken but fast talking Chandra. “We need to provide a platform to innovate in the preclinical domain, building paradigms, establishing pathways, and build confidence in taking molecules into man. In terms of assay capabilities, many of the tools matured in the last 3-5 years are analogues of tools in the clinic. We can make measurements in animal that are similar to what would be used in man.”
Chandra’s group benefits from a multidisciplinary band of group of software programmers, bioengineers, and molecular biologists. For example, the engineers have developed a computer controlled rotating drum that can automatically remove animals’ food during the night prior to an imaging test. The IT infrastructure has to handle large quantities of data. The team uses a 5-terabyte Dell server, which cost $30,000 to configure. “We’ve written image analysis software to go beyond what is provided by the instrument vendors and other groups in the field,” says Muir. (See Sidebar, “The IT perspective”).
Using a micro-PET (positron emission tomography) machine from Concord Microsystems, Millennium has produced dramatic evidence of Velcade’s efficacy in curtailing cancer growth. Fast-growing tumors take up more glucose than normal tissues. By administering FDG — glucose tagged with a potent, short half-life fluoride isotope (F-18) — the tumors are easily tracked. Within the space of about a week, a response to Velcade is demonstrated by reduced glucose uptake and growth of these tumors.
Last fall, Chandra and colleagues published the first results from the center in the journal Molecular Imaging, using micro-CT (computer tomography) to produce a three-dimensioned X-ray image of bones in rats with collagen-induced arthritis. The technique is similar to PET, but has a longer image half-life owing to the use of different radioactive isotopes, including Barium and Iodide.
Programmers in the Platform Technology group developed special software to not only measure bone volume, but also bone roughness and morphometry. ‘[A]dvanced imaging methods and analysis may provide additional preclinical data to assist the advancement of an experimental drug” such as the new candidate, MLN922, the authors write. “Ultimately, imaging models may contribute to the goal of providing a more personalized standard of clinical care based on the disease presentation of the individual patient.”
The juxtaposition of so many imaging techniques facilitates their use in tandem. For example, MLN2704 is an antibody that targets the PSMA (prostate-specific membrane antigen) on prostate cancer cells, and delivers a lethal chemotherapeutic drug, maytansinoid. Using the Center’s prized Varian 7-Tesla MRI machine for the soft tissue and circulation, and CT for the bone, the drug demonstrates a 90 percent reduction in tumor volume and significant preservation of normal bone anatomy. MRI is also being developed for functional lung imaging using a technique called 3-Helium diffusion imaging in collaboration with Harvard-Smithsonian physicists.
There are several other imaging technologies on hand. Variations of single photon emission computerized tomography (mSPECT) are helping Millennium investigate immune cell migration in models of inflammation. A Xenogen optical imaging system, measuring the fluorescence or luciferase and other “reporter genes,” and infrared imaging facilities are available. An exciting new technology called fluorescent molecular tomography (FMT), developed by a company called VisEn, delivers precise 3D fluorescent imaging of animal tissues and some experimental clinical settings. “We want to be able to deliver quantitative functional and anatomic information — it’s key to making better decisions about pre-clinical compounds and improving clinical results,” says Muir.
All this begs Chandra’s rhetorical question: “Why isn’t everyone doing it?” The answer, he says, is, “Skill sets. One of the big rate-limiting steps is imaging scientists. We’ve recruited a very talented group of people, tasking them with individual goals of contributing to the pipeline. From molecular biology to physical chemistry to cancer pharmacology to bioengineering — all under one roof, talking about compounds.”
And, of course, the “patients” — Chandra’s pet name for his rodent subjects. “We reduce the number of animals used by following them over time. We treat them as patients coming in for regular scans.”