Opening The Medicine Cabinet

This spring, Pfizer opened the door to its medicine cabinet and invited outside scientists to take a peek. It signed a five-year, $22.5 million agreement with Washington University in St. Louis allowing faculty to access data on hundreds of Pfizer compounds that had either failed in the clinic or were in ongoing studies. The idea was that sharing the information would generate novel ideas for old and new drug candidates alike.

The pact inspired outside debate over Pfizer’s motives and the true worth of the compounds. But no one could deny it was intriguing. If done properly, the collaboration would be an innovative model and signal a newfound openness on the part of drug companies to share information they tend to keep under lock and key.

Indeed, Washington University is just the first in what will eventually be a network of academic collaborators who interact using Pfizer as a hub, says Rod MacKenzie, head of worldwide research for Pfizer PharmaTherapeutics. The strategy exemplifies the evolving relationship between industry and academia. Although Pfizer might be ahead of the pack, most drug companies are exploring enhanced academic partnerships. The new approach acknowledges that the old way of conducting R&D—guided by an inward focus and heavily guarded data—has deep flaws.

In the past, partnerships between drug companies and university researchers were fairly simple. The standard drill was for companies to fund individual scientists who, at some point down the road—sometimes a year, sometimes five—would report their findings. The model persisted, even though pharmaceutical makers didn’t always feel they were getting much out of it.

Today, with research budgets strained and productivity stalled, drug companies are looking for more collaboration with universities. Recent examples include AstraZeneca’s pact with Columbia University, which is centered on obesity and diabetes; Merck & Co.’s expansive relationship with Harvard University on osteoporosis; and a broad deal between Johnson & Johnson and Massachusetts Institute of Technology’s David H. Institute for Integrative Cancer Research for oncology research. Much more than in the funding relationships of old, these agreements encourage dialogue between corporate and academic scientists and even provide opportunities for working side by side in each other’s labs.

Pfizer’s relationship with Washington University has itself evolved amid the changing environment for pharmaceutical research, and it’s no coincidence that Pfizer chose the school as the site of its latest experiment in openness. The seeds of the agreement were planted back in 1982, when Monsanto and Washington University signed a five-year research pact worth $23.5 million.

The deal was controversial. Not only was a university accepting a large sum from a corporation, but the majority of the funds would be devoted to applied, rather than basic, science. The worry was that big money from industry would unduly influence the research going on at an academic institution.

Despite those concerns, the pact eventually became a model for industry-academia collaboration. In 2000, Monsanto was bought by Pharmacia, which continued the Washington University relationship even after spinning Monsanto back out just two years later. The pact remained intact after Pharmacia was acquired by Pfizer, also in 2002. Over the decades, the university received roughly $5 million in annual funding for the partnership.

An agreement signed in 2008 involving immunology and inflammation research deepened the ties. At the time, Washington University vowed to put up “no walls, no barriers” between itself and Pfizer as they pursued projects funded under the expanded deal. Scientists would make joint proposals and then work together at the university or at Pfizer’s St. Louis labs, with the freedom to use resources from either entity.

The end result of those decades of collaboration is “a legacy of trust that cannot be purchased,” says Jeff Gordon, director of Washington University’s Center for Genome Sciences.

But the relationship hit a snag a year ago when Pfizer, as part of its vast cost-cutting efforts, decided to eliminate 600 out of 1,000 jobs in St. Louis. The move included shifting inflammation research out of the area.

“We were left to ponder what sort of opportunities exist that could be predicated on the remarkable trust, interactions, and legacy of sharing that exist between Washington University and Pfizer,” Gordon recalls.

The answer, it turned out, lay in Pfizer’s two-year-old Indications Discovery Unit, or IDU, which was dedicated to finding new uses for compounds that had failed in clinical trials or were still in development. Don Frail, chief scientific officer of the St. Louis-based unit, and Karen Seibert, the vice president of research at Pfizer’s St. Louis site, who had overseen the 2008 pact, realized that Washington University’s resources could advance IDU’s mission.

They developed the idea of a virtual medicine cabinet that could be opened by more than just Pfizer scientists. Researchers at Washington University were intrigued, and a deal was hashed out.

When the collaboration was announced last May, it was greeted by industry observers with equal parts curiosity and skepticism. On the one hand, providing an academic institution with such open access to Pfizer’s molecular library was an intriguing model for collaboration. On the other, access wouldn’t mean much if the compounds in the library were not the crown jewels in the collection, but costume jewelry. Everyone wondered: Was Pfizer offering up compounds that it had abandoned for good reason?

Frail recalls seeing many blog posts suggesting the big pharma firm was merely handing over the dregs of its discovery engine. He could only chuckle. “I knew we weren’t,” he says, recalling the controversy the arrangement had caused inside the company.

In fact, it was the lessons learned in creating IDU that enabled the deal with Washington University. When IDU was proposed in the mid-2000s, some Pfizer project teams worried about other scientists handling compounds that were already in development—even if those scientists were Pfizer colleagues. IDU managers teamed with safety colleagues to develop a set of guidelines for using the compounds that allayed the concerns.

By the time Pfizer executives proposed expanding IDU’s network to include Washington University, scientists had grown more comfortable with opening their lab notebooks to others. “It’s kind of scary to let people outside the organization play around with molecules that may be in clinical development,” Pfizer’s MacKenzie acknowledges. “But we got over that.”

Another reason Pfizer scientists were game was their knowledge that target selection—the critical first step in the drug discovery process—would be next to impossible without the disease-pathway research going on in academia, MacKenzie says. And deep down, they recognized that the secretive nature of the industry might have become an impediment to research success.

Other drug companies are coming to similar conclusions. Although none of them is participating in data sharing to the extent that Pfizer is, some are taking baby steps toward opening up the compound vault to speed up the translation of ideas into drugs.

“Attitudes are really starting to change across the whole industry,” says Garry Neil, vice president of Johnson & Johnson’s corporate office of science and technology, which is tasked with enhancing J&J’s partnership network. Drug firms are starting to recognize that “we need to be open and share information,” Neil says. “We don’t compete in basic science; we compete in products in the marketplace.”

“We can’t be protective about everything,” agrees Håkan Wennbo, AstraZeneca’s head of project management for cardiovascular and gastrointestinal innovative medicine. “We need to take some risk exposing ourselves because, otherwise, we’ll sit isolated in our own facilities.”

AstraZeneca recently expanded its diabetes and obesity partnership with Columbia to include testing some of its drug candidates in the school’s preclinical disease models with the intention of finding new indications.

But wariness about broadening such partnerships doesn’t exist exclusively on the corporate side. Washington University, for example, took care to explain the new agreement with Pfizer to its faculty and address concerns about the arrangement. “Before any of this began, we had a series of town hall meetings describing what it was and answering any questions faculty might have,” Gordon recalls.

After Pfizer and Washington University managers made staff more comfortable with the deal, the realities of managing the collaboration set in. The first step was for Pfizer’s IDU to create a virtual medicine cabinet. The company chose 560 compounds, representing 228 mechanisms of action, for the database.

All faculty members at Washington University’s medical school, whether they are basic scientists or clinicians, can request access to the electronic portal. After verifying that they are indeed on the faculty, interested scientists are given a password allowing them to peruse the site at any time.

As might be expected, Pfizer doesn’t completely take the lid off the pill bottle until a researcher’s funding proposal has been approved. At first, Washington University scientists encounter a list of mechanisms of action, each of which translates into one or more compounds locked inside the virtual medicine cabinet. “They can interrogate the list and determine if that target, mechanism, or pathway intersects with their own research,” Frail explains.

The faculty member can also see how many compounds Pfizer has in development related to that particular mechanism and for what indications the compounds have been studied. The status of a given program is also revealed, so a researcher knows whether a compound is in clinical trials or, if the program has been discontinued, what phase of development it reached.

From there, the scientist can make a one- to two-page concept proposal and submit it on the spot. More often than not, Frail and Gordon say, the researcher needs assistance interpreting the data. Something akin to a help desk, manned by both a Pfizer scientist and a Washington University staff member, is available to help field questions and to connect researchers at the organizations.

IDU is located just a block away from Washington University’s medical school labs, and Pfizer often hosts lengthy sessions with university scientists to tease out ideas. Once a faculty member’s proposal is approved, access to the compound is granted, and the researcher learns more about Pfizer’s research and clinical data on the drug.

Importantly, Frail says, proposals are welcome at any time. And although some researchers might receive funding right away, others are asked to provide a “full” proposal, which must be coauthored by an IDU scientist. For the latter situation, “a dating process occurs to understand if the goals of the investigator mesh with the goals of Pfizer,” Frail notes. “The proposal can evolve quite a bit in that process.”

And therein lies the catch: Implicit in the pact is that Pfizer actually has an interest in the potential new indication a Washington University scientist proposes studying. Pfizer, like most big pharma companies in this age of R&D cutbacks, has become selective about the diseases it researches.

What happens if a Washington University faculty member has a great idea for a compound that is outside Pfizer’s research scope? The question is still being discussed. Both sides would like to have a mechanism by which the compound could be developed outside Pfizer’s labs, but the details have yet to be pounded out, Frail acknowledges.

He also points out that a formal proposal isn’t the only outlet for research ideas generated by the portal. The database is also leading to interactions outside the official agreement, in which Washington University researchers are launching independent research projects with Pfizer scientists.

Although it’s not yet clear whether the Pfizer-Washington University partnership will yield results, scientists at the university are certainly interested in it. Since the virtual medicine cabinet went live at the end of May, more than 130 faculty members have explored the database, and dozens have made proposals. “Each person brings their own unique set of lenses to this medicine cabinet—that’s of course the most precious asset of all,” Gordon says.

Both Pfizer and Washington University seem happy with the early interest in the experiment. When the database opened, “we didn’t know if it would be a flood or a trickle,” Frail says. “It’s actually been a nice steady state.”

The long-term goal is to move compounds into late-stage development. The ultimate validation for the Pfizer-Washington University pact would be to find a new mechanism for a Pfizer molecule on the basis of a hypothesis coming out of the academic institution, MacKenzie says. Because early clinical data already exist for these compounds, a promising theory could swiftly lead to a midstage trial.

Companies that are now several years into deep relationships with universities say they are reaping results they could not have achieved on their own. Merck, among the first drug firms to forge a broad scientific agreement with an academic institution, is now two years along in a collaboration with Harvard immunologist Laurie Glimcher. Under the pact, Merck and Harvard scientists are working side by side at the university to translate Schnurri-3, a protein discovered in Glimcher’s lab that controls adult bone formation, into drug targets and candidates.

So far, the researchers have elucidated the bone-building pathway regulated by Schnurri-3 and found several potential drug targets. Moreover, they have already validated the targets using multiple techniques and, in the process, have found small molecules that block them. Merck is now identifying lead compounds to push into clinical development. “We’ve done this in two years, which is remarkable,” says Alan Ezekowitz, head of bone, muscle, immunology, respiratory, and endocrine R&D at Merck Research Laboratories.

Meanwhile, in the two years since forming its partnership with Columbia, AstraZeneca has generated new hypotheses around drug targets for metabolic diseases and has quickly been able to validate them in the clinic. “It would have been impossible for AstraZeneca to do this work by ourselves because we don’t have that expertise and background,” Wennbo says. He points to the new funding for the collaboration as further evidence of its success.

For Pfizer, the partnership with Washington University is just one model for how big pharma can work with academia, MacKenzie says. “There will be others coming along shortly,” he says. In the next few months, “you’ll see us forming a core set of relationships.”

Ultimately, the goal is to create a network of partners that connect through Pfizer to tackle problems together. “If you think of a brain, you’ve got individual neurons firing on certain pathways,” Mac­Kenzie says. “But it’s only when you get all of them firing together that you get higher intelligence.”

In practice, this means connecting partners with expertise across the scientific spectrum—say, a geneticist to a systems biologist to a clinician, all at different academic centers or at Pfizer—and then assigning the group a “grand challenge” that could encompass biomarkers, diagnostics, and medicines. MacKenzie points to Alzheimer’s disease as ripe for this type of project.

“What I’m describing here is nothing short of a different R&D model for the future,” MacKenzie says.

Other pharmaceutical executives agree that becoming more innovative at partnering is essential for accelerating drug development. If the industry is going to get better at finding novel drug targets and the compounds that can modulate them, “we need to be interacting with people that do the best science,” AstraZeneca’s Wennbo says. “That really means we need to be much more open with universities in these types of collaborations.”

Still, forming them is no easy feat. The challenge is to ensure that both partners benefit. For the company, a successful partnership means tangible results; for the academic lab, it primarily means advancing scientific knowledge and furthering the education and careers of students and postdocs in the lab. All the companies involved with these broader pacts stress that the goals of the organizations must be clearly aligned.

“It has to be the right relationship. It has to be managed, and it takes a lot of work to manage these things correctly,” Merck’s Ezekowitz says.

Given how time-consuming project management can be, Ezekowitz cautions against undertaking too many broad collaborations. “From a Merck perspective, we’re still being highly selective about choosing these opportunities, and they are fairly heavily scrutinized internally,” he says.

MacKenzie concedes that some cultural barriers need to fall before big pharma widely embraces relationships with academic centers. One parameter he finds particularly limiting is a university’s overhead. When any type of funding comes in to an academic lab, the university takes a cut to support its infrastructure. Those overhead costs can be high and can generate a roadblock to the creation of a true network of academic partners.

“It seems crazy to me that I’m paying and supporting academic infrastructure at the same time as I’m making hard decisions about our own footprint,” MacKenzie says. “In an ideal world, these relationships would be cash-flow neutral.”

He also notes that working with industry is not every academic scientist’s cup of tea and that it quickly becomes apparent which faculty members are comfortable with this new model. Several pharma executives say young investigators seem most interested in working with companies to translate their research into medicines.

Companies that have forged broader pacts with universities argue that the benefits are already outweighing the challenges.

For Pfizer, the relationship with Washington University is helping to change the mind-set about how research is accomplished. Ultimately, MacKenzie says, such agreements will ensure that Pfizer “never again gets introverted or obsessed with our own ideas of the world.”