Market Insight - Microdosing reaching a tipping point?
This article was originally published in Scrip
Advocates of microdosing, a technique that involves testing very small doses of products in humans, think it will improve the efficiency of drug development by flagging at an early stage those product candidates that are likely to fail. So why has it yet to enter the mainstream, asks Phil Greenfield
Drug development is a lengthy, resource-intensive and expensive process, with a low yield of successful compounds. The costs are estimated to be more than $1 billion per registered drug. Much of this figure is associated with drugs that, after significant development expenses, ultimately do not make it to market. Inappropriate pharmacokinetic (PK) parameters lead to up to 40% of drug candidates failing in first-in-man studies.
Microdosing, a technique developed in the late 1990s, is designed to allow products to be de-selected at an earlier stage. It could finally be at the point of realising its potential in terms of saving time and resources that would otherwise be spent on the development of unsuccessful drug candidates.
Microdosing involves testing small (1/100th of the pharmacological dose or 100µg, whichever is lower) doses in humans. Even though such low doses are not pharmacologically active, they still allow the study of PK and certain pharmacodynamic (PD) parameters in humans up to one year earlier than is possible following the traditional Phase I approach.
Microdosing is now enshrined in guidance from both the European and US medicines regulators for the conduct of clinical trials in support of new drug approval applications. For example, the June 2009 ICH M3 guidance (Non-Clinical Safety Studies for the Conduct of Human Clinical Trials and Marketing Authorisation for Pharmaceuticals) recommends two types of microdosing studies where doses must not exceed 100µg.
A pioneer of microdosing, Professor Colin Garner, principal at Garner Consulting and former CEO of Xceleron, a UK provider of drug development services, thinks the technique could be near its "tipping point" in terms of being accepted into the drug development process. "Microdosing is now regarded as part of [the] exploratory clinical development of a drug. Plus there is regulatory guidance [for microdosing] in place, through [the] EMA and the [US] FDA. It is now down to pharma to adopt the approach."
first movers
Small and medium-sized biotech companies have adopted microdosing earlier than big pharma, due in part to the fact that human data (PK but not safety/efficacy data due to the small doses used) can be produced earlier than in conventional Phase I trials, enabling them to license their molecules sooner and at lower cost to themselves.
It appears that increasing numbers of larger companies are following in the footsteps of the early adopters, such as GlaxoSmithKline and Servier, in taking up the technique. These include Organon (Merck & Co), Sanofi-Aventis (both of which are clients of Xceleron) and Amgen.
Two key technologies are used in microdosing: accelerator mass spectrometry (AMS) and positron emission tomography (PET).
In the case of microdosing using AMS, human volunteers are given a carbon-14-labelled drug candidate at such a minute dose that safety is assured. Blood samples are taken from volunteers at specified intervals for analysis.
Since the technique can detect trace amounts of the isotope, concentrations of the bioactive ingredient and its metabolites are determined and used to obtain the absorption, metabolism and excretion rates. This information is essential for companies to determine the further development of candidate drugs. Therefore, microdosing will allow firms to focus their clinical development programmes on those candidates with the most suitable PK profiles in humans.
Using AMS, it is possible to conduct a full human metabolism study, measuring drug concentrations in biological fluids at doses down to 1,000 times smaller than those used in a Phase I study. Trials are usually split into intravenous/oral crossover groups to demonstrate bioavailability, as IV is 100% bioavailable and oral is unknown. Two 50µg doses are needed in crossover studies so that the maximum dose of 100µg is not exceeded. This is still well within the sensitivity range of AMS, which can measure blood drug concentration at human doses as low as 0.5µg.
Meanwhile, positron emission tomography (PET) imaging, using microdoses of radiolabelled drug tracers, is also gaining traction as a modern clinical drug development tool. This approach is unique in that it allows for the direct quantitative assessment of drug concentrations in the tissues targeted for treatment, thereby bridging the gap between pharmacokinetics and pharmacodynamics.
Current applications of PET include anticancer, anti-infective and CNS drug research. Situated at the interface of preclinical and clinical drug testing, PET microdosing can be used to produce PD data, such as receptor selectivity or occupancy profile by using short-half-life radioisotopes, including carbon-11 and fluorine-18. It is particularly useful in detecting whether CNS drugs pass through the blood-brain barrier.
evidence to support microdosing
The collaborative, industry-sponsored CREAM trial (Consortium for Resourcing and Evaluating AMS Microdosing) reported results in 2005. The trial, sponsored by pharma companies such as Lilly, Schering AG (Bayer Schering Pharma) and Roche, was set up as a rigorous test using compounds that were expected to strongly challenge the microdosing concept.
CREAM was undertaken using several drugs with known human PK characteristics at pharmacological dose levels. Each compound was administered at a microdose level and at a therapeutic-dose level to subjects in an appropriate crossover design.
The CREAM trial involved five drugs of which the microdose PK data were studied. One compound was Roche's Versed (midazolam), a sedative drug that is a selective substrate for CYP3A4. The results of the experiment showed a high first pass metabolism correlation of key PK parameters. The microdosing study carried out with midazolam gave excellent correlation with the pharmacological dose. According to Xceleron, many sceptics of the microdosing concept suggested that drugs with high first pass metabolism would not be predictive at microdoses.
At the time, Xceleron stated: "Currently, preclinical studies can take up to 18 months at a cost of $3-5 million. Microdosing techniques could reduce the time to four to six months and the costs to $0.35 million per new molecule." A typical microdosing study costs $250,000-400,000 per molecule, including the labelling, toxicity, clinical and data analyses.
The most comprehensive study of microdosing to date is the European Union Microdose Accelerator Mass Spectrometry Partnership Programme (EUMAPP), funded by the EU. The findings of this international, multicentre study were presented in 2009.
EUMAPP's objectives were:
- to assess if there was PK linearity following a microdose and a therapeutic dose for seven drugs (fexofenadine, paracetamol, phenobarbital, sumatriptan, propafenone, clarithromycin and S-19812). These drugs all represented situations where traditional pharmacokinetic predictive models (eg, in vitro and animal species) were problematic; and
- to compare the accuracy of the pharmacokinetic predictions made by microdosing to those made from physiologically-based pharmacokinetic (PB-PK) computer models.
For all of the drugs tested in EUMAPP, IV microdose data predicted half life, clearance and volume-distribution data well. Oral-dose data did not scale as well as the IV dose but, in general, the data obtained would have been useful in the selection of drug candidates for further development (or those to be dropped from the pipeline).
Where oral microdose data did not scale well, the reasons for this can be surmised from the known metabolic or chemical properties of the drug. Thus, EUMAPP has contributed to the scientific community's knowledge of the technique, including where it can be best applied to drug selection.
sceptics remain
Not all companies are converts to microdosing, though. The method has come under scrutiny, with particular concerns that it may not be able to predict the behaviour of clinical doses. "Companies have said PK values measured at these low doses (100µg) are not relevant compared to pharma doses. However, it is around 80% predictive and allows companies to reduce risk in the drug candidate selection process (and therefore increase the chance of a drug's success in Phase I trials) or, just as importantly kill off candidates before they reach Phase I," argues Professor Garner.
But the main concern is whether microdosing speeds up drug development. One researcher from Amgen comments: "Although this strategy is useful to characterise clinical PK with lower resource inputs, there is a significant time delay if you support a microdosing clinical trial and then conduct a standard Phase I programme. In effect, the microdosing clinical trial can slow your overall clinical programme."
Prof Garner thinks this is a somewhat simplistic view: "Microdosing does add a time delay if added onto Phase I trials, but it provides greater certainty that a candidate doesn't fail Phase I trials. It needs to be built into drug development strategies, rather than tagged on. For example, human in vitro data, in silico modelling and microdosing can replace many of the metabolic studies which currently require testing in five animal species."
Peter Gaskin, principal at Aptuit Consulting, agrees: "We are just about to start a microdosing programme in the US having completed the enabling non-clinical programme with a very small amount of API and in about 14 weeks. One of the key issues is bioanalytical sensitivity and in our case doubly so as we were working with a pro-drug. Whilst the overall timeline to complete the clinical programme is longer as you'll have to go back to Phase I enabling studies, if API synthesis is an issue, and [if] you have a number of interesting candidates to choose between, I think that microdosing offers some advantages worth careful consideration."
According to Professor Garner, microdosing is most appropriate where there are several molecules that look interesting from a pharmacological perspective and companies want to select which ones to take forward. But where to find four or five candidates to undertake a comparative microdosing study? Here Professor Garner suggests that it might be useful for pharma companies to go back to their compound libraries. Firms may have discarded potentially useful drug candidates on the basis of inappropriate animal PK when in reality these molecules might have good PK in humans. Microdose studies could be used to go back and re-examine some of these molecules.
growth drivers
Professor Garner says the recent announcement by GlaxoSmithKline to bolster malaria research by opening up its libraries is "a wonderful opportunity for microdosing". With microdosing, it is possible to measure drug concentrations in red blood cells, potentially enabling antimalarial drug candidates to be more quickly advanced to Phase I. GSK is also one of the few companies with in-house AMS facilities. AMS requires expensive and bulky equipment, taking up space equivalent to two tennis courts. The hope is that smaller and cheaper instrumentation will be developed in future.
Another driver behind the movement towards microdosing is the prospect of reducing reliance on animal studies. Indeed, this factor influenced TNO, a Dutch CRO, in its decision to purchase an AMS to support microdosing studies in humans. TNO is organising a microdosing symposium in 2010 with a number of Dutch institutions (Assuring Safety without Animal Testing, Top Institute Pharma, Food and Nutrition Delta, Netherlands' Biotech Industry Association, and Life Science and Health) whose aim is "reducing the resources spent on non-viable candidates and the amount of animal testing necessary to bring new drugs and foods to the market".
But there is still a great deal to be done to get microdosing into the drug development process for all companies. Smaller firms may even be able to educate the larger players on the advantages of microdosing. Two candidates for this job are Neurocrine Biosciences and Idenix Pharmaceuticals (which licensed a molecule to GSK). Neurocrine has used microdosing to evaluate the PK of five H
"One major UK pharma company is looking at microdosing studies for molecules for in-house licensing," says Professor Garner. When looking for potential partners the company plans to screen early-stage candidates using microdosing. Other firms may follow suit.
Also encouraging, in Professor Garner's view, is recognition of the microdosing technique's sensitivity and its potential use in paediatric studies. Most drugs used in infants are not tested in this patient group from a metabolism point of view; microdosing could enable this to be done safely. Moreover, companies can get a six-month patent extension for a drug in Europe and the US if the product comes with paediatric data, so the commercial benefits are also compelling.
References
1. British Journal of Clinical Pharmacology, Vol 67(3), pp288-298.
Phil Greenfield is a principal analyst for Scrip. Email: [email protected]. If you are a subscriber and have a question arising from our news and analysis coverage, or would like us to do some research for you, use the free Ask the Analyst service by emailing us at [email protected].