Around two weeks ago, a Novartis official said that pre-clinical microdosing studies aimed at gauging the activity of investigational drugs in humans on a metabolic and pharmacokinetic level won’t predict whether a candidate will succeed or fail at higher doses and in broader populations [see PGx Reporter 03-14-07].
This week, the CEO of small British CRO Xceleron took issue with that notion, stating that microdosing studies, also known as Phase 0 studies, can help risk-averse, cash-strapped small- to mid-size biotechs pick molecules to advance into Phase I studies.
The Novartis official, Robert Schmouder, executive director of translational medicine, said that Phase 0 studies as outlined in a US Food and Drug Administration’s January 2006 guidance on “Exploratory Investigational New Drug Studies” are “not as useful as maybe we thought they were”
But according to Xceleron’s CEO Colin Garner, “where we would disagree with the gentleman from Novartis is that we see [microdosing studies] as part of the late-stage discovery process. It’s greatest usefulness is where you have several molecules coming out of discovery and you’re having difficulty deciding which one or ones to take into full scale Phase I studies.
“Currently, it’s small- to medium-sized biotech companies who seem to think they are useful,” Garner said. “Maybe they are the ones that have [fewer] resources than the Novartises of the world. So they can only take a few molecules to Phase I. But, obviously, Novartis can take many more molecules through to Phase I if they want to,” he said.
Xceleron, which specializes in microdosing studies, uses accelerator mass spectrometry to help pharma and biotech companies advance investigational agents early into humans to learn metabolism information, analyze human tissue and waste samples obtained from pharmacological doses of drugs, and measure biomarkers to gauge pharmacological effect.
Garner noted that while it is possible to do gene-based microdosing studies, it remains an unexplored area of research. “If you are trying to compare genotype with phenotype, then I think microdosing could play a role there,” he said. “But I’m not aware of any data to support that. From a scientific concept point of view it seems possible, but I don’t think anyone has actually done it.”
An FDA spokesperson told Pharmacogenomics Reporter last week that while gene-based microdosing studies are theoretically possible, the agency has not seen any such eIND studies as part of new drug application submissions.
Defending Microdosing
Several high-ranking FDA officials support microdosing studies as a tool to help industry modernize and accelerate drug development. According to the agency, a microdose exploratory IND study would expose a small number of patients to “1/100th of the dose of a test substance [needed] to yield a pharmacologic effect of the test substance with a maximum dose of <100 micrograms.”
However, there are many in industry who feel that eIND studies use far too small doses to predict whether a candidate will succeed or fail at higher doses and in broader populations, as Novartis’ Schmouder said at a recent Cambridge Healthtech Institute conference translational medicine.
As an alternative, UK-based Asterand provides a program it calls “Phase Zero” that uses human tissue samples rather than human subjects. This method might be more in line with big pharma’s efforts to increase confidence in making go/no go decisions in drug development [see PGx Reporter 03-14-07].
Asterand’s Phase Zero studies investigate a drug candidate’s efficacy, pharmacology, safety profiling, and metabolism by using a variety of techniques, including gene- and protein-expression profiling.
According to Asterand, most of the top 20 pharma companies are using its version of human tissue-based Phase Zero preclinical studies to gain more clarity in the drug development process.
Xceleron’s Garner defended microdosing studies by saying “the best model for man is man. If you don’t except that premise then obviously you won’t agree with the microdosing concept,” Garner said. “Most of the work people do in candidate selection out of discovery is using in vitro and animal models. Our view is that those have some flaws associated with them.”
According to Garner, in vitro models fall short because they only offer insight into how a drug metabolizes at a particular site in the body, not in the whole person.
“I think unfortunately today the whole person approach is the only approach to be taking,” he said. “The disadvantage to using human tissues as opposed to humans is that you have to look at every metabolic site in the body in order to try to do in vitro studies. So most of the in vitro studies tend to be done with liver tissue or liver cells. That’s only one of the sites of metabolism for the drug in the body.
“People have endeavored to use in vitro systems to look at metabolism, to look at toxicity or mutagenicity, and it’s known that they are flawed in one way or another,” Garner added.
Xceleron has the FDA and the European Agency for the Evaluation of Medicinal Products behind it. In 2003, the EMEA implemented its Guideline on Microdosing. Three years later, shortly after issuing its draft guidance of exploratory IND studies, Janet Woodcock, deputy commissioner and chief medical officer of the FDA, told Pharmacogenomics Reporter that through Phase 0 studies the agency is encouraging industry’s move away from the traditional three-phase drug-development model toward more predictive strategies.
“Everyone wants to move from what has basically been, after the discovery phase, almost a trial-and-error approach, to a more mechanistic approach, where we have much more predictive information that we can generate at each step of the way,” Woodcock had said when the agency first introduced the draft version of its eIND guidance in 2005 [see PGx Reporter 05-12-05].
On its website, Xceleron quotes newly minted FDA Commissioner Andrew von Eschenbach stating: “Phase 0 studies are an extremely important mechanism to reduce risk for drug developers. I would encourage innovators to take more risk in exploring this approach. … I think the cancer community in particular should be adopting Phase 0 studies.”
|
“The best model for man is man. If you do not except that premise then obviously you will not agree with the microdosing concept.” |
Garner noted that in Xceleron’s experience, companies are increasingly using microdosing studies in the anti-infectives and CNS space to gauge pharmacokinetics and drug localization information.
Also, a May 2006 report by Cambridge Healthtech Associates on the pros and cons of microdosing showed that it costs between $1.5 million to $3 million and takes between 12 months and 18 months from the time preclinical candidates are selected to the time first-time-in-man studies are finished.
The microdosing technique could shrink the timeframe to between 5 months and 8 months and reduce the cost to between $300,000 and $500,000, according to the report.
According to CHA, although pharmaceutical companies have not warmed up to the microdosing approach, “by 2010 we believe that microdosing will already have gained a secure foothold at the interface between the preclinical and early clinical stages of drug development.”
Finally, Garner cited research conducted by Oxford University’s Saïd Business School that found that there is “no negative impact with microdosing and at its best … you could gain up to a 10-percent time and cost savings.”
Data Supporting Microdosing Needed
Xceleron conducted an industry-sponsored trial, “Consortium for Resourcing and Evaluating AMS Microdosing,” to investigate the correlation between microdose and pharmacological dose PK data for five drugs: the sedatives midazolam and diazepam, the anti-estrogen compound ZK 253, the anticoagulant warfarin, and the antibiotic erythromycin.
In the study, known as CREAM, microdosing predicted the pharmacokinetics at pharmacological doses of ZK 253, diazepam, and midazolam. ZK 253 – a candidate that Schering AG dropped after Phase I – showed extremely low bioavailability by microdosing. Although the study predicted slow metabolism for warfarin, microdosing did not predict the PK of warfarin. Erythromycin encountered problems in administration and was not tested.
According to Xceleron, the CREAM study showed approximately 70-percent correspondence between microdose and pharmacological dose PK.
“The results will represent a surprise for some in the pharmaceutical community as the concept works in circumstances that were cited as being unlikely for microdosing to succeed,” Graham Lappin, head of R&D at Xceleron, said in a statement.
“For example the microdosing study carried out with midazolam gave excellent correlation with the pharmacological dose that was seen as highly significant, as this is a well known substrate for CYP3A4,” he said. “Many skeptics of the microdosing concept suggested that drugs with high first-pass metabolism would not be predictive at microdose.”
According to Garner the current problem hindering industry adoption is the lack of peer-reviewed data regarding microdosing studies. “That’s a difficulty,” he said. ”We can talk forever about whether it’s predictive or not but until we have a body of data together then it’s impossible to come to any conclusion about” the benefits microdosing.
To address this, European Union is funding a $4 million Microdose AMS Partnership Program to investigate the reliability of microdosing to predict the PK of seven molecules at pharmacological doses. The study will be led by Xceleron and initial results are slated for release in April.