By Turna Ray
Smaller diagnostics shops should not feel discouraged to bring their proposals to Pfizer because the pharma giant decided to partner with Abbott Molecular to develop a companion diagnostic for an investigational lung cancer drug, according to Hakan Sakul, the head of diagnostics at Pfizer's global R&D division.
Pfizer's recently announced deal with Abbott Molecular Diagnostics to develop a companion test for PF-02341066 — an EML4-ALK mutation-targeting non-small cell lung cancer drug — is one of several efforts underway at the company to develop personalized drugs.
In July, Pfizer invested $6 million in several Ontario, Canada-based research institutions for the discovery and development of genetic biomarkers to personalize new and existing colorectal cancer treatments in the pharmaceutical giant's pipeline [see PGx Reporter 07-29-2009].
The company is also working with Genomic Health to develop a companion diagnostic test for an unannounced renal cancer drug, and markets the HIV drug Selzentry, for which Monogram Biosciences, now part of Laboratory Corporation of America, has developed a companion test to establish patients' tropism status.
Historically, Pfizer has worked with comparatively smaller diagnostic shops, such as Monogram and Genomic Health, to develop tests for its pharmacogenomics-guided drugs. However, in an interview with Pharmacogenomics Reporter this week, Sakul noted that Pfizer picks its diagnostics partners on a case-by-case basis, evaluating the specific assay needs for a particular indication, commercialization aspects, and regulatory landscape, among other factors.
As a result of the Abbott deal, "we wouldn't want smaller companies out there to think Pfizer's not going to work with any small companies," Sakul said.
"I just wanted to put the message out that this particular announcement from Abbott Molecular should not discourage folks from approaching us and working with us, because we already have other relationships where we do work with much smaller companies as well."
During the interview, Sakul also provided a snapshot of Pfizer's efforts in personalized medicine, the company's stance on regulatory issues, and its overall focus on "omics"-guided medicine.
Below is an edited transcript of the interview.
When we spoke last, back in 2006, pharmacogenomics appeared to be mostly in the early research phase at the company. Now, it seems PGx-guided medicine is playing a role in Pfizer's late-stage and commercial products. Is this true?
I want to make a comment on the use of pharmacogenomics. We do use pharmacogenomics, but we also incorporate other 'omics' into our programs. We have a molecular medicine group, and I head up the companion diagnostics piece of that. We also have metabolomics, transcriptomics, as well as proteomics efforts, in addition to a pharmacogenomics effort at Pfizer. So, I'm going to refer to all these programs as the 'omics' and use 'personalized medicine' as a term to refer to all of them.
We've been active in pharmacogenomics, specifically, for quite some time, for over ten years now at Pfizer. Our activities are not limited to just research. This is really where the science is. We look for new targets, we become opportunistic and we look at what we can do with them. You know the story behind our drug Selzentry for HIV. That was developed as a result of findings around the CCR5 gene, which [were] published in Science in 1998, that initiated a program at Pfizer that eventually brought a drug to market. We continue to follow the science, not just research but what can be taken into development. Finding targets is really the first step, followed by proper validation of the targets, and giving us enough confidence in that target and in that mechanism of that target, so we can consider turning that into a program in personalized medicine that will look at particular segments of the patient population for that specific indication ...
When we look at personalized medicine … predominantly our programs are happening beyond the research stage, the programs are happening in the development and commercial stages right now.
In the research stage, the efforts are focused on identifying targets. We have announced collaborations … [for] biomarker development, where these biomarkers are the substrates for future diagnostics, but they are not diagnostics yet.
[ pagebreak ]
Many pharma companies, in their entry into the personalized medicine space, license gene targets or work to improve upon established gene targets, such as KRAS in colorectal cancer. In addition to advancing novel gene markers, is Pfizer planning to license established markers?
In our pipeline (available here), we have a program [for a drug] that targets KRAS. So, we have activity in that area. … In other areas, we have licensed in a compound for glioblastoma multiforme, which comes with a diagnostic assay that is also in the works. So, we do have several projects underway [in personalized medicine], not only in oncology but … we are also very active in the infectious disease space, in the neurosciences, particularly in Alzheimer's disease, in ophthalmology, and in metabolic disease. The difficulty, often times, is if you can't establish a cause-and-effect relationship between a set of genes and a clinical endpoint, and validate [that relationship], it is difficult to incorporate it into our programs. It is difficult to establish that cause-and-effect relationship and validate that to an extent that it becomes a no-brainer to pursue that program.
In terms of the regulatory environment, have the delays on the part of the US Food and Drug Administration to finalize its guidance on in vitro diagnostic multivariate index assays and issue guidelines on Rx/Dx co-development impacted Pfizer's ability to move ahead with its personalized medicine projects?
I don't believe anything of what the FDA is actually doing is delaying anything here at Pfizer. We're certainly impacted like everyone in this space, pharma as well as diagnostic companies. For the IVDMIAs, the final guidance documents haven't come out yet. We had some comments on the first draft of that document, which we had sent to the FDA. Obviously, IVDMIAs are only one part of the genetic test domain where the FDA has already exercised its regulatory authority.
Certainly, FDA is still looking at this space. There are currently two regulatory pathways to market: one goes through the CLIA process and the other through the FDA. The question is, how will this all end up? [Will] the FDA have oversight of everything or not?
Pfizer is a member of the California Healthcare Institute. I represent Pfizer on the diagnostic working group on CHI, which supported a submission in April to the FDA docket in response to Genentech's Citizen Petition, [which urged FDA to regulate all predictive laboratory-developed tests] … CHI supports regulatory reform harmonizing the analytical as well as clinical requirements for diagnostic tests. CHI also advocates for clear oversight of genetic testing.
There are lots of genetic tests out on the market that are not FDA approved and are being used for clinical decision-making for treatments. The question is how the FDA will actually use its authority: whether they will enforce it on everything, whether they will go back and look at some of these tests that are on the market being used for decision making, or whether they will grandfather those in. Those are obviously unknown. What is known is [that] as of 2005, FDA put out a white paper on drug/diagnostic co-development … The FDA has been having conversations with various constituencies, including the Personalized Medicine Coalition, of which Pfizer is a member. We had that conversation with the FDA a few months ago … From what I understand, they plan on putting out a few white papers on different parts of that document, then they will follow up with guidelines.
How [the Rx/Dx co-development guidelines are written] is going to impact us, as well as diagnostics companies we work with. We don't have a diagnostics standalone business at Pfizer. So, we look to other diagnostics companies to develop companion diagnostics for products we intend to put on the market. So, how it impacts them also impacts us and we have a vested interest in following that area.
[ pagebreak ]
In choosing which companies you partner with to develop companion tests, historically, Pfizer has worked with smaller firms like Monogram. However, your most recent companion Dx partnership was with a larger firm, Abbott Molecular. Are you shifting your strategy to partner with larger Dx shops that have more regulatory experience and capabilities in terms of commercialization?
We are driven by the kinds of assays we need first. Not everybody has the kinds of assays that we might need in our clinical programs. It's also very important to work with companies … that can also be around to help us with the commercialization of those assays. Developing the assays and using them in our clinical trials is really not the end of the story. We still need them approved by regulatory authorities, and commercialized globally, given that we are a global company and we market our drugs globally.
The short answer is, we work with all kinds of companies. We don't have a strategy to limit ourselves to large companies. In the particular case with Abbott Molecular, we did use, in Phase I studies. their probes … We did look at other competing companies. In the end, we made the decision to go with a company that had the IP, the capacity to commercialize, the ability to receive approvals, not just from the FDA but in Europe [because] we'll need a CE-mark.
Some of this may be difficult for a smaller company, but it certainly may be possible for [smaller] companies to do that. In cases where an assay is only available at a small company, and that company doesn't have all the support they would need to be able to commercialize [it] properly, then we'll consider what to do to ensure those capacities are in place. It could be through collaborations, three-way collaborations, other creative choices and decisions.
We wouldn't want smaller companies out there to think Pfizer's not going to work with any small companies.
We have two types of efforts [in companion diagnostics]. One is program-specific, looking at certain targets, and developing certain diagnostic assays for those targets. And another is a platform-development effort that cuts across therapeutic areas. One of those is looking at the point-of-care diagnostics space for new patient testing, where we would end up with kind of fully automated sample-to-result-type platforms that are amenable to use in physicians' offices, clinical trial sites, research hospitals, or whatever the case may be. That is another area we are actively engaged in.
When you look at all these different areas, there are a lot of companies we work with, small and big. We have unannounced collaborations with smaller companies. I just wanted to put the message out that this particular announcement from Abbott Molecular should not discourage folks from approaching us and working with us because we already have other relationships where we do work with much smaller companies as well.
What's important is to end up in a situation where innovation continues, and that regulatory issues and partnership considerations don't actually leave out the smaller companies. There's a lot of creativity in those companies, as well.
As a follow up, is how you choose a companion diagnostic partner dictated by the drug's patient population and market size?
Typically, we start with the medical unmet need. It's not really driven by how big the population will be in the end … [With regard to the c-Met NSCLC drug], on average, between 4 percent and 6 percent of NSCLC patients [have the gene rearrangements] that make them appropriate for the treatment. I believe about 40,000 people worldwide and 6,000 people in the US have this specific form of NSCLC. So, the number may not be huge but the medical unmet need is big.
If we have a drug that can address a specific need for a segment for that there is a much better chance for that drug to work, and to benefit the patients, then that's our approach. I don't know how the c-Met drug's numbers compare to the HIV drug, but I'd venture to say [the population for the NSCLC drug] is [smaller] than [Selzentry's market size].
Have you made use of FDA's Voluntary Genomic Data Submission program to get the agency's input on any of your personalized medicine projects?
When that system was being developed at the FDA, Pfizer shared some data with the agency. I believe part of that was on major depression where we had generated a lot of data. It wasn't just limited [to] that, however.