Mar 29, 2012
This is part II of a series of articles on innovation in Pharma based on an interview with Jose Carlos Gutierrez Ramos, the man in charge of revolutionizing drug discovery at Pfizer. In part I we established the difficulties facing Big Pharma as brand name drugs go off patent and pipelines run dry. Pfizer is trying to rewire the relationships between pharma, biotech, and academia to align incentives and take advantage of the strengths of each player. In part II, below, we explore new research initiatives between Pfizer and academia. In part III we discuss a new paradigm for pharma-biotech collaboration. In part IV we observe the effect these initiatives are having on Pfizer and on other pharma companies, and what a new, more innovative pharma sector might look like.
The relationship between academia and pharma is an old one and a common one, but often not a particularly fruitful one. In the status quo, says Jose Carlos, “money was going from pharma to academia and it was not really being used. I’ve done this myself, when I was on the faculty at Harvard. You get money [from pharma], and it’s like another R01 (a grant from the National Institutes of Health). You keep doing the same research as before. And therefore pharma was withdrawing. They’d say, ‘this is nice, but we don’t get anything out of it.'”
For Jose Carlos, the problem lies very much in the nature of academic versus industrial research. “As we discussed before,” he says, “from the moment you have a cluster of publications—ten, twenty—to the moment that a pharmaceutical company becomes interested, you could easily expect 5–10 years. Meanwhile, great research is being done around that pathway or phenomenon but it’s not directional, because that’s not the nature of academic research.”
Earlier in our conversation, Jose Carlos emphasized the need to align the goals and understand the incentives of different stakeholders. One thing pharma can provide to academic investigators is an opportunity to test their hypotheses in humans.
This is an unmet need, says Jose Carlos, since the physician/scientist who wants to commercialize a drug has three imperfect options. “The investigator may be so motivated that she starts the discovery program on her own. At Harvard and other institutions there are drug discovery facilities. These outfits are really nice to get in vitro drugs or pharmacological drugs, but very rarely do the drugs have the quality to make it in humans. Whether because of safety or because they don’t have the right pharmacokinetics, you cannot test the right hypotheses. That path is not the optimal one.”
“Another path is for the energetic investigator to get some friends in VC and go start a company. Perhaps she takes a year to go on the road to get funding, and that is great. But the problem is, once the money is in, the whole trajectory of how the idea is executed changes. I was in biotech and I love it, but I know that once the money is in, the focus becomes how to provide a return on investment in three years.
“And then the third avenue is the one we were discussing before, and that is wait 10 years until someone in pharma picks up the idea and runs with it.”
None of these options are particularly efficient ways to discover new medicines. In particular, new ideas and innovative science languish in academia, enveloped in an ever-swirling random walk of incremental discoveries. But what if a directional vector, pointing straight to clinical research, were overlaid upon this random walk?
To this end, Pfizer has begun a new initiative, called Centers for Therapeutic Innovation, or CTI. The new model is designed to focus academic research towards translational goals, and speed the introduction of new science into the pharma pipeline. A CTI is both a partnership between Pfizer and an academic institution, and a physical place. There are four CTIs in life science hubs around the country; the one in Boston opened last year near the Longwood medical campus.
Jose Carlos emphasizes that a CTI collaboration is designed to be unlike most current interactions between academic institutions and industry. For one, there is the physical proximity. “In the past,” he observes, “when collaborations between pharma and academia occurred, they were 5000 miles away. This meant there was no incentive for the investigator or the pharma project manager. Here we built the physical proximity, and we hired people in the CTIs who understand science and understand drug development, but who also understand the value and the drivers of the investigator—that publications are important, etc.—and they form one team to get that drug into humans and test the hypothesis.”
There are other innovative elements as well. Projects are selected based on scientific promise but must include a clear path to a proof-of-mechanism or proof-of-concept trial in humans. Publication restrictions are relaxed.
“We want to work with the best people,” says Jose Carlos. We want to see the best science. We don’t look at the best science and filter it through a Pfizer lens. Rather than call for specific areas of interest, the joint scientific committee made up of academic deans and Pfizer execs votes unanimously whether or not to fund a project. We select based on the science because market conditions change, reimbursements for patients change, our priorities change. Later on, if the drug works in humans and we are not interested in that space, then that’s fine, it goes back to the investigators and the institution with no strings attached.
“To this point we have received 300 proposals. We funded 16 in the first round. We’re trying to make this program like a short-term Howard Hughes grant, where if you are an investigator you get funding and the possibility to do human research. It’s very valuable. So for 24 of the projects that were not selected, we and academic medical centers provided peer review, such as ‘We love it, but in translatability it lacks tak-tak-tak.’
“At steady state our goal is to have 20–25 of the projects running in CTIs across the country. Of those, we’ll have 5 in first-in-human trials at any given time, which represents a huge acceleration not only to the way we were operating, but also in how ideas are translated to products. Even if the mechanisms don’t always pan out in humans and we have to go back, at least the starting point for drug discovery will have changed.”
In addition to the type of projects that are selected in a CTI, the way they are funded is innovative. Money is provided in increments based on research progress, which is highly unusual in academia. “A full project is up to $8 million,” explains Jose Carlos, “but the funding is based on milestones that are shared and agreed upon by investigators and us. That’s how we work in biotech. You get your funding, series A or B, but you don’t get it in your bank. You get $20 million in the series but it’s all tranched.”
For example, a project may aim to humanize a mouse antibody that shows promise in animal models. An initial grant would fund the work to change the original amino acids to reflect a more human sequence. The investigator would then have to show that the binding and biophysical properties of the molecule are unaffected to unlock more funding. A further milestone might be to demonstrate that the mechanism of action is the same in human cell lines as it was in mouse cells. This type of results-based payment is a sharp departure from typical partnership arrangements, and it aims to provide a clear focus for the project and cut the time to clinic down to 18 months.
Not all projects succeed, and those that fail to meet translational objectives are cut without remorse. “Of the 16 funded projects,” notes Jose Carlos, “we just terminated one. After eight months. It turns out that the epitope we thought was going to be internalized in humans wasn’t. This is the risk you take when you go to very early science: Will the phenomenon be robust and repeatable? So together with the investigator and the joint steering committee we terminated the project. It was a great try. And this is going to be happening all the time, and that’s the value of the CTI.”
I ask Jose Carlos about the overall vision. What happens to the molecule that meets all of its milestones but doesn’t meet Pfizer’s current development goals? Where does it go? And what about the ones that do fit in? How are they integrated into the Pfizer pipeline?
He shakes his head. “Two years ago we were seeing it in the binary way you describe, that either we would take it or we wouldn’t take it, but not anymore.”
Jose Carlos describes three likely outcomes for a molecule from a completed CTI project. “One, if the data is negative, then that’s how science is and that’s a significant risk you take when you go into early science. Maybe the data is negative because you tested the wrong hypothesis, but now you have GMP material to test another hypothesis, and the investigator is completely incentivized to take that next step.”
“Two, if the signal in humans is interesting but maybe not commercially interesting to us, or the risk is too high, the program stays with the investigator and is equivalent to a series A funding in a biotech. The investigator now has $8 million invested, an IND open, intellectual property around the drug, and half a kilo of GMP antibody. They can go to the lupus foundation or leukemia foundation or company X or venture capital firm Y. It’s equivalent to seed funding or more.
“Three, if we take the molecule fully forward, then there are pre-negotiated payments—between $2–4 million at proof of mechanism in humans and $15–25 million for proof of concept in humans—that are paid to the institution and the investigator.
“But something different, which I think will happen more and more, is that the signal is good but the science has evolved. Perhaps there are more experiments to do. What we’re studying now is how to build a hybrid model of funding with investigators, with venture groups, with biotech, to allow the program to go forward. In this scenario, maybe Pfizer doesn’t take on 100% of the risk, and neither the investigator nor the institution nor the venture group takes on all the risk.
“What I’m trying to put on the table applies not only to CTI but to all the innovation we’re building here. What is critical is that this cannot be binary for us or any other party because too many opportunities are left on the table.”
The CTIs were founded about a year ago. The projects are up and running, though the initiative is not expected to reach steady state until 2014. The milestone payments are a sharp departure from the “take the money and run” attitude pervasive in most academic-pharma collaborations. The clear focus on translatability, and the willingness to slash projects that don’t meet their marks, represent an industrialization of academia that, while perhaps not welcome everywhere, is crucial to accelerate the translation of innovative academic discovery into actual drugs.
Academic centers have signed on to this vision. Pfizer is all in, and other Big Pharma companies are taking notice. In March 2012, Merck announced the creation of a new drug discovery institute in San Diego to help academic investigators develop ideas into products for industrial development. The California Institute for Biomedical Research shares many objectives with the Centers for Therapeutic Innovation.
It remains to be seen if more new drugs will come out of these projects. What is true, however is that it marks a pronounced shift in the way pharma interacts with academia. CTIs align incentives and harness, rather than subsume, discovery, one of the core competencies of academic labs. Pharma gets access to innovative science, validated in humans, faster than in the current system. Academic investigators and their institutions get money, resources, and the once-in-a-lifetime chance to ask the truly important question about their discoveries: Does it work in people?