By Turna Ray
Using pharmacogenomics to resuscitate a drug that has previously failed clinical trials due to safety issues is one strategy no large pharmaceutical firm has yet had the guts to attempt in the US. However, this approach could reap benefits for smaller firms willing to license big pharma’s failed drugs, a speaker at a recent conference suggested.
Pfizer last year began funding a pharmacogenomic analysis of torcetrapib — the infamous hypercholesterolemia drug for which the drug giant torched development plans three years ago due to an increased risk of death associated with the compound.
While Pfizer has not announced any plans to reintroduce the drug using pharmacogenomic strategies, a drug, such as torcetrapib, could present an attractive licensing option for smaller firms. In fact, treatments deemed unsafe or ineffective in a large pharmaceutical firm's clinical program may turn out to be the lead product in a smaller personalized medicine company's pipeline using pharmacogenomic strategies, according to Michael Phillips, director of pharmacogenomics at Genome Quebec.
Phillips and Jean-Claude Tardif of the Montreal Heart Institute’s Pharmacogenomics Centre are leading a Pfizer-sponsored study to genotype 2,000 patients who were given torcetrapib and low-dose atorvastatin (Lipitor) and compare the genotypes of patients with good responses and poor responses from the perspective of HDL and LDL cholesterol levels, blood pressure, electrolyte levels, and cardiovascular outcomes [see PGx Reporter 11-12-2008].
Phillips described the PGx study at a meeting last week hosted by the Ohio State University Center for Personalized Healthcare, noting that "while it's an open question" as to "whether [torcetrapib] can or cannot be resuscitated," the pharmacogenomics data being generated by Genome Quebec and Montreal Heart Institute "might possibly be used to de-risk other drugs" in Pfizer's pipeline or present an opportunity for smaller companies wishing to advance the commercialization of such a drug.
Ripe for Licensing
Phillips noted that licensing a drug that a large pharmaceutical firm has abandoned due to low efficacy or heightened safety risks may be a viable business strategy for a smaller personalized medicine firm.
"Is this [resuscitating failed drugs with PGx] going to happen at a big pharmaceutical company? They may sell these kinds of drugs to a smaller firm," he suggested.
He gave the example of VIA-2291, a 5-lipoxygenase inhibitor being developed by Via Pharmaceutical as a treatment for atherosclerosis. Via Pharmaceuticals licensed VIA-2291 in 2005 from Abbott after the company abandoned development plans for the drug as an asthma treatment and after FDA approved a similar drug.
Genome Quebec is conducting a pharmacogenomics sub-study for VIA-2291 with the goal of developing a 5-LO genotyping panel for the drug and an ADME/tox panel that could be applied to the development of VIA-2291, as well as future clinical trials for other companies [see PGx Reporter 10-17-2007].
Clinical Data is one personalized medicine company that is employing this strategy. The company's lead drug product is the antidepressant vilazodone, a selective serotonin reuptake inhibitor and a partial agonist of the 5-hydroxytryptamine 1a receptor.
Originally discovered by Merck, vilazodone moved under a license agreement to GlaxoSmithKline in 2001 after Merck abandoned Phase II trials due to poor efficacy. Glaxo gave up on the drug in April 2003, and Merck decided to go through Genaissance in another attempt to get it to market. In 2005, Clinical Data acquired Gennaissance.
[ pagebreak ]
"The purchase of the vilazodone drug development program was a notable factor in the Genaissance acquisition," Theresa McNeely, VP of corporate communications at Clinical Data, told Pharmacogenomoics Reporter this week.
"The economics were great for us" in developing vilazodone, McNeely said. She added that in the case of vilazodone, Clinical Data has so far paid for two Phase III trials, both of which had positive results, and spent between $60 million and $70 million in R&D over the last three years, excluding non-cash items. This is significantly less than the $800 million to $1 billion it costs to take a drug from the discovery phase to market launch.
"There was a lot of clinical trial work done before the drug got to us,” McNeely said. “We were able to use that data and make decisions about trial design and dosing, among other things."
Although Clinical Data reported this summer that vilazodone affected a statistically significant improvement in the symptoms of anxiety associated with major depressive disorder, the company was unable to replicate findings for a biomarker that it had previously associated with increased response in a subpopulation.
Identifying a subpopulation that experienced better response to vilazodone would have allowed the company to price the drug higher for those patients and earn greater dividends. Clinical Data has said it will continue to evaluate other response gene markers. The Newton, Mass.-based firm plans to submit a new drug application for vilazodone in the first quarter of 2010.
Exiting the CVD Space?
According to Phillips, the torcetrapib PGx study has a dual objective: to identify genetic predictors of response to the torcetrapib/atorvastatin combination treatment and to gather data that can be used to de-risk follow-on compounds.
The researchers are conducing non-hypothesis-based genome-wide analysis on Illumina 610-Quad BeadChips and are performing candidate gene analysis using Illumina's custom platform iSelect in order to evaluate drug response and adverse reactions to the combination treatment.
However, all this data is going to be of little use to the pharma giant's internal development plans — particularly since the company announced last year that it plans to exit the cardiovascular disease space.
The decision to do so may or may not have been precipitated by torcetrapib's costly and highly publicized late-stage failure. Certainly, an important consideration was that the company's mega-blockbuster Lipitor is losing patent protection in 2011. Exiting the CVD space means Pfizer will not be committing R&D bucks to develop statins for hyperlipidemia/atherosclerosis, heart failure, obesity, and peripheral arterial disease.
Of course, the drug giant can ultimately change its mind if the torcetrapib PGx study yields particularly promising data. Torcetrapib's market potential in the general population was projected to be in the range of several billion dollars annually. Depending on the results of the PGx study, the company may be able to develop a follow-on to torcetrapib that avoids its safety concerns when administered with a companion genetic test.
Still another option for Pfizer, as suggested by Phillips, would be to license torcetrapib to a smaller firm, willing to develop a companion test for the drug.
Since Pfizer has paid for torcetrapib's clinical development, including the PGx studies, a smaller company looking to license the compound would essentially get a nearly market-ready product on the cheap. Such a scheme would also allow Pfizer to distance itself from the stigma of relaunching a drug that has failed due to safety concerns, while still netting a portion of its revenues.