Joint Pharmaceutical Analysis Group
Analysis of modified release products
Vinod Shah, a consultant from the US, gave a thought-provoking presentation
on the future of dissolution testing or, more accurately, drug-release
testing. The dissolution test is well established, reliable and reproducible.
It is used as a process control test and also to assess batch-to-batch
quality. Increasingly in vitro dissolution testing is relied on to assure
product performance.
For medicinal products having a systemic action, bioequivalence testing
is usually based on a comparison of the plasma concentration profile
of the medicinal product at issue with the profile obtained with the
reference preparation. However, in some situations such a bioequivalence
study may be replaced by in vitro dissolution testing. When such a substitution
is allowed by registration authorities it is referred to as a biowaiver.
Several regulatory guidances have been developed to provide biowaivers
based on dissolution profile comparisons for lower strengths, for test
product based on the biopharmaceutics classification system, and for
certain scale-up and post-approval changes. Thus the dissolution test
has brought about significant changes in regulatory perspectives.
The concept of dissolution — or in vitro release — can be
easily extended to other dosage forms. Examples of these include in
vitro release from semisolids and transdermal patches. Development of such
performance tests for other dosage forms is under investigation. Dissolution
testing remains essential, and its future is strong, asserted Dr Shah.
A working definition of in vitro–in vivo correlation (IVIVC) is
that it is a predictive mathematical treatment describing the relationship
between an in vitro property of a dosage form, such as the rate or extent
of drug dissolution and a relevant in vivo response usually plasma drug
concentrations or amount of drug absorbed. Harald Rettig, of Biovista,
Switzerland, was a firm believer in such correlations, particularly for
modified-release products. If successfully established, such correlations
provide strong support for biowaivers, with considerable savings in time
and costs.
A correlation can usually be expected when drug release from the product
is the step governing the subsequent absorption step. Normally this is
an essential design element for a modified-release dosage form. For oral
dosage forms the in vitro drug release is routinely measured and characterised
as dissolution rate.
Some compound properties will prohibit the successful application of
IVIVC, such as those with a narrow therapeutic window, those with a variable
first-pass effect, endogenous compounds, prodrugs, or those with multiple
response populations.
The relationship between the in vitro and in vivo characteristics is
expressed by a linear or non-linear correlation. However, the plasma
concentration profiles cannot be related directly to the in vitro release
rate, and must be converted first to the underlying in vivo release or
absorption profile, either by pharmacokinetic compartment model analysis
or by a model independent treatment.
The IVIVC becomes more robust when two or more different formulations
are tested in the same in vivo study. Also, the release-controlling excipient
in the formulations should either be identical or very similar. The corresponding
in vitro dissolution data can be obtained with different test conditions.
The IVIVC with the best predictive power is then selected for further
use.
Modified release and PAT
Since the US Food and Drug Administration launched its process analytical
technology (PAT) initiative in 2003 there has been an increase in activities
to better understand pharmaceutical manufacturing processes. PAT has been
successfully applied to blending and granulation processes although other
parts of the manufacturing process, including the coating processes have
received significantly less attention. However, said Alex Pysik, of Pfizer,
Sandwich, Kent, it is widely recognised that coating processes are important
and in order to achieve “quality by design” for these manufacturing
steps it is necessary to identify and assess the key quality attributes
and key process parameters. There are many assessment tools available and
the outcome of such a risk assessment is used to decide on how to gain
a better process understanding. This can be achieved in a variety of different
ways, including the use or implementation of PAT technology.
Osmotic pump tablets and multiparticulates in capsules are common examples
of modified-release drug delivery systems. Both of these dosage forms can
incorporate functional and active coating components that are critical
to product performance. Different approaches to monitoring the coating
processes using near-infrared (NIR) spectroscopy have been tested and compared.
Good correlations between NIR reflectance spectra and weight gain, coat
thickness and drug release performance have been demonstrated.
Therefore, it has been shown that application of PAT provides significant
benefit in enhancing monitoring, improving end-point determination and
understanding of the associated coating processes. Furthermore, this can
help understand and reduce intra-batch variability in product drug release
performance and potency.
The physiological environment
In an overview of gastrointestinal physiology, illustrated by the view
of a camera-in-a-pill on its journey throughout the entire gastrointestinal
tract, Abdul Basit, of the University of London School of Pharmacy, emphasised
that the environment was wholly unlike the environment of medical devices
being tested in dissolution testing. For example, the volume of fluid in
the fasting stomach was approximately 45ml, a far cry from the litre volume
specified in dissolution tests, and the buffering capacity of the intestinal
fluids is relatively weak compared with phosphate buffers. The varying
conditions in vivo mean a more acceptable way of assessing the behaviour
of a dosage form is by direct observation in situ. Gamma-scintigraphy is
a well established, non-invasive imaging technique for disease diagnosis
in many areas and has been adapted for following the transit and transformation
of orally administered medicines and devices. A radionuclide is incorporated
into the formulation and a gamma camera used to track its location. Thus
a liquid formulation can be seen to be relatively
smoothly removed from the stomach compared with a pellet formulation.
The difference in performance (based on drug plasma concentrations) of
an osmotic pump extended release tablet in two healthy volunteers could
be dramatically demonstrated as due to different residence times at the
main release site (the colon). Dr Basit used the example to point out that
single component devices such as this were susceptible to catastrophic
removal from the proposed absorption site, explaining the variability in
performance.
Dr Basit concluded that pharmacokinetic assessment of modified-release
formulations provides an incomplete and often misleading picture of gastrointestinal
performance and the use of visualisation techniques, not just scintigraphy,
can help to explain in vivo variability and aid formulation development.
Characterisation and performance of parenterals
Brian Clark, of AstraZeneca Macclesfield Cheshire, described the characterisation
and performance of parenterals based on poly(lactide-co-glycolide), a
polymer often used as a release-controlling excipient in sustained release
parenterals for subcutaneous or intramuscular administration.
Duration of release is determined by several factors, including the characteristics
of the active substance, drug loading, size,
shape and morphology of the delivery system, polymer characteristics including
molecular weight, lactide:glycolide ratio and other parameters, such as
residual solvent
levels.
Because the efficacy of the product is largely dependent upon the rate
and extent of drug release in vivo, the development of an in vitro drug
release test which is relevant to the physiological situation is important.
Based on an understanding of the factors that influence release, a factorial
experimental design approach was used to maximise discrimination in vitro
for batches known to perform differently in in vivo animal models. This
approach was used to establish an in vivo–in vitro relationship for
an experimental microsphere formulation.
Parenteral preparations are often intended to release drug over a period
of up to six months. It is therefore also desirable to develop a valid
accelerated drug release test, particularly at the formulation design stage
where the researcher needs to evaluate
dose dumping, assess the duration of
action, minimise the number of animals required, evaluate product stability,
and
establish the effect of changes in processing conditions.
Drug release can be accelerated by use of elevated temperature, extreme
pH, reduction in ionic strength or the use of a catalyst to increase the
rate of polymer degradation. Above all, however, the accelerated test must
not distort the physiological environment. Although accelerated testing
is feasible, it may be relatively insensitive to the morphological factors
which mediate the initial “burst” release often observed with
such formulations, concluded Mr Clark.
Jayne Lawrence, of King’s College London, also tackled the problem
of modified dosage forms for parenterals, concentrating on microemulsions
and vesicles. Microemulsions are thermodynamically stable, transparent
dispersions of oil and water, stabilised by a surfactant often in combination
with a cosurfactant, such as butanol. Vesicles are spherical structures
consisting of one or more bilayers entrapping a central aqueous core and
with a layer of water trapped in each bilayer.
Both microemulsions and vesicles are used as vehicles for drug delivery,
and a variety of physicochemical techniques have been applied to characterise
them. This is necessary to understand their structure and their capabilities
so as to aid in design of drug delivery systems. For example, phospholipid
vesicles are rapidly removed from circulating blood by a process of opsonisation
(coating with protein). To extend the plasma half-life of such vesicles,
they can be coated with poly(oxyethylene) chains, said Professor Lawrence.
Understanding the physical chemistry of modified-release systems
The properties of water-soluble polymers are of considerable importance
in industrial and biotechnology processes, said Simon Ross-Murphy, of King’s
College London. Their dissolution — which may be defined as an increase
in viscosity — is complex and depends on temperature, concentration,
molecular weight and particle size.
Professor Ross-Murphy focused on investigations on the effect of particle
size on the hydration rate of macromolecules over a wide range of particle
sizes. The model system studied was guar gum, used in pharmaceutics to
control release in the gastrointestinal tract or as a bulk-forming laxative.
Unsurprisingly, small particles are hydrated faster than large particles,
as shown by simple plots of viscosity versus time. Attempts to establish
a more precise relationship, however, proved problematic as larger particles
take such a long time to achieve even 80 per cent of their final hydration
level. Log-log plots did not seem to throw any further light until a shift
factor was introduced for the time axis and superimposable curves were
obtained. Although this shift factor is, at present, an arbitrary figure,
its value (somewhat closer to 2 than 3) suggests it may be related to surface
area, rather than volume. Establishing an accurate figure for the shift
factor would allow prediction of hydration time for any other particle
size of interest and the model allows some new physical insights to be
developed. Further work is needed on different molecular weight samples
of controlled particle size, concluded Professor Ross-Murphy.
Temperature and pH-responsive polymers offer a number of important advantages
for controlled release systems compared with conventional release systems,
said Brian Saunders (School of Materials, University of Manchester). They
have enabled new designs for temporal and distribution-controlled release
devices. Microgel particles are crosslinked colloidal particles that swell
under appropriate conditions. Poly(N-isopropylacrylamide) is a temperature-responsive
polymer with a lower critical solution temperature of about 32C in water,
meaning that the microgel will swell and collapse, respectively, when the
temperature is below and above this temperature. Poly(diethylamino-ethyl
methacrylate) is a pH-responsive polymer with a pKa of about 6.8. By combining
either type with a poly(ethylene oxide) macromonomer it is straightforward
to prepare a range of systems with varying characteristics. Architectural
control during construction of the devices provides considerable flexibility
in tailoring the release properties of delivery vehicles. Through strategic
use of the polymers it is possible to design delivery systems that behave
in different ways in vivo even though responsiveness originates from the
same polymer. Model polymers demonstrate principles that also apply for
biocompatible responsive copolymers, said Dr Saunders. |
The
Pharmaceutical Journal