Name: Richard Kivel
Position: CEO, TheraGenetics
Background: CEO/President, BioBehavioral Diagnostics; Trinity PharmaSolutions; MolecularWare.
Education: M.S., Boston College; B.A., American International College
TheraGenetics, a new company spun out of the Institute of Psychiatry at King’s College in the UK in April, recently announced the completion of a $6 million venture funding round.
The company evolved out of a team that was formed 15 years ago at the institute researching the genetics of various psychiatric disorders, and how genes are related to response and side effects associated with psychiatric medications.
After £10 million of investment into these initial research efforts, in 2005 researchers at the institute created what TheraGenetics is touting as the first diagnostic predictive test for clozapine, a widely prescribed schizophrenia drug associated with severe side effects.
Based on the success of this initial test, several investors and King’s College decided to create a company around the technology of identifying novel biomarkers and predictive algorithms to determine response and side effects associated with CNS drugs.
TheraGenetics CEO Richard Kivel, who joined the company in April, spoke to Pharmacogenomics Reporter this week about the company’s products under development and its strategy for entering the CNS diagnostics space.
What kind to technology is the clozapine test based on and what other tests are currently under development at the company?
We use a series of different approaches to develop a prediction test. Certainly we use general candidate gene approaches, using automated genotyping and things of that nature. So, for example, we look at genetically determined drug metabolic status, so we looked at SNPs, UGTs and other metabolic enzymes. We look at mutations associated with side effect and response. And we often use candidate gene approaches with technology such as the ABI 3100 or SNP Plex and so forth.
Another approach we have underway is sort of a genome-wide approach, which is most often used in conjunction with Affymetrix technologies. With a combination of those technologies, a team of geneticists, and statistical geneticists, what we do is we identify specifically SNPs that may be relative to a response or a side effect. We test those SNPs against large candidate populations looking for genetic diversity within populations to determine the accuracy of those SNPs.
Right now we have two pharmacogenetic tests in late-stage development. Those two tests right now are being commercialized some time in the next 24 months. One of them is for a drug called olanzapine [Eli Lilly’s Zyprexa], and the other one is for a drug called risperidone [J&J’s Risperdal]. Both of these drugs are on patent and generating tremendous revenue. Olanzapine does approximately $4 billion a year in revenue and risperidone does approximately $3 billion a year in revenue. We filed two provisional patents around the prediction of response to both olanzapine and risperidone last December.
We’re also looking at a series of other drugs within schizophrenia, such as quetiapine [AstraZeneca’s Seroquel] and aripiprazol [Bristol-Myers Squibb’s Abilify]. We’re also right now doing a series of projects related to depression and bipolar disease.
For these projects are you developing the diagnostics in collaboration with the drug companies?
We are actually not working with Lilly and Janssen on olanzapine and risperidone. It’s sort of an interesting model. For drugs that are already commercialized and achieving tremendous amounts of revenue, and not at the end of their patent, it is much more advantageous for us to work independently on those drugs, because there is not a big advantage for those drug companies to work with a diagnostics company. However, we are regularly approached and working with other pharmaceutical companies and biotechs around drugs that are in late stage clinical trials. We are working on one project right now around a drug that was actually abandoned at one point during the clinical trials and we’re working with [the pharma company] to use genetics to bring that drug back to the market.
I guess there is a three-pronged approach. For large scale drugs, billion dollar drugs on the market, you’re primarily creating a pharmacogenetic test, getting clinician buy in, reimbursement, and getting it to the market, and you can do that completely autonomously. The two other models we work with are to work with pharma and biotech companies on rescuing drugs that were in late stage clinical trials but that perhaps did not achieve the volume of success that they wanted through the clinical trial, and now they want to adopt genetics to repurpose that drug or look at that drug again. Or we simply help pharma or biotech companies better select patients for clinical trials with genetics.
Ultimately those [last two] models are very advantageous to us and the drug companies because it not only helps them move more quickly through the clinical trial process but it might actually turn into a drug-diagnostic combination similar to Herceptin. … That’s when it’s really important to work with the pharma companies because they have a tremendous vested interest in getting that drug to the market. They want it to be more efficacious than the other drugs that are out on the market. They want to demonstrate it has [fewer] side effects. One way to do that other than modifying the chemical entity is to better choose the patient population.
Can you talk about the project you are working on to rescue the drug candidate using genetics?
Unfortunately, we’re under a CDA with [the drug companies]. It was a pharma company that was working on this project with another pharma company, so we’re kind of locked in by two different groups. I would say that probably in the six months we would be able to discuss some of those external projects.
Now that our funding is in place it has really put us in a really wonderful position where we’re really able to further the research and development around the olanzapine and risperidone. We’re looking at other areas like depression and bipolar and really pursuing these other projects. We’re bringing aboard more senior level staff.
How many staff members do you have now?
We have a dozen people with the company right now, and three senior level openings that we are just now opening up. We are hiring a new head of R&D, someone that really has tremendous experience in the diagnostics space, a business development director, as well as a chief operations officer. Those three senior level positions are open, and we always have lower level R&D scientific roles available. The R&D director and the scientific roles will be based in the UK, and the business development role will most likely be based in the US.
Do you have plans to commercialize these tests in the US?
We view the US market as the biggest market and probably the most important one for us. The tests that are under development in the UK are actually representative of an international population. Being that we’re in the genetics space, when we first create the test, in the very earliest stages of development, we actually take an approach where we look at a very, very homogenous population. So, we’ve often collected samples from Northern Spain or areas where there is lot less genetic noise. We identify the algorithms and the SNPs, and we begin to test those hypotheses and scientific findings in larger, noisier populations. We’re right now collecting samples from Asia, throughout the US, Canada, and elsewhere. That way when we bring the test to market, it’s clinically available and valuable in multiple large populations. So the US will be part of our initial launch.
Will you pursue FDA clearance for your tests?
Yes. We’re going to be going for a combination. We’ll be looking for a CE Mark, initially, and moving most likely to a 510(k) approval in the US.
How difficult is it to garner reimbursement for tests in the CNS space, and how do you plan to overcome any physician reluctance to adopting this technology into their practice?
One of the things that was very important in the earliest days of this company’s launch was that the clinical staff interviewed hundreds of clinicians in the US and Europe, specifically around the probability that they would use predictive tests to help drive prescribing decisions. Based on the fact that approximately 50 percent of the prescribing decisions they make are wrong right now, the clinician population feedback as a whole was very clear that if we could provide a prescribing application that would better drive their prescribing, both helping them determine both response to a particular drug but also to protect their patients from unnecessary side effects, that was something they warmly embraced. We really believe that the clinician population will embrace this and I guess you can say that we’ve tested that.
The reimbursement piece of it is the other side of the equation and that’s a much longer road. But what we’re doing is in a very methodical way, as we build our board of advisors and directors, we’re getting the key opinion leaders to embrace the technology we have, to begin to use it in a very early alpha or beta form. That is really helping our reimbursement story. From a health economics perspective it’s a very compelling story. When you consider the very adverse side effects of antipsychotics, the 50 percent probability of response based on present prescribing behavior, it’s a great story for the health care world, especially health care providers in the US.
How would you characterize the competition in this space? And what are your advantages in comparison?
Without naming names of specific companies that are in this space, I’d say they typically fall into a couple of categories. Most companies chasing personalized medicine or using pharmacogenomics or genetics for prediction tests are not in the CNS space. They are in oncology, they are in infectious disease. They are in these big blockbuster areas that get all the attention. CNS has been widely ignored. With that said, there are some small players that are doing things in the CNS space. But most of what they’re doing seems to be around identification of side effects in genetic populations.
But nobody is as advanced as TheraGenetics from the standpoint of total body of patient samples collected. The rate-limiting step to success in this field is being able to collect large quantities of well-documented and defined genetic and clinical samples. So we not only need the blood samples, the DNA from the patient that has a specific disorder, but we need the clinical notes with that DNA sample in order to properly build our algorithms. And nobody has access to that kind of data. We have access to thousands and thousands of samples, not only through our own collaborations but also through the Institute of Psychiatry. And that’s pretty much the rate-limiting step for everyone in the industry.
What do you perceive to be some of the challenges in this space in developing diagnostics?
The diagnostics industry as a whole, having followed it for a decade, it’s amazing how it has changed. Diagnostics really was associated with instrumentation and very, very low-cost tests. … What we’re seeing is a tremendous shift. And companies that are generating tests like the Herceptin test and others are really showing us that the test price, the quality and the quantity of tests, are all going up at the same time because of the use of really high quality genetic information. We do see that as one of the challenges.
One of the big challenges is getting market acceptance around the use of genetic testing, and then getting the reimbursement companies to say we’ll pay for this, and then using that information to make prescribing decisions. I think that’s going to be a challenge for the coming years as this new technology is adopted. Every time you see a new pharmacogenomic or genetic diagnostic test in the market, I think it lowers that barrier a little bit more.