(586am) Feasibility of Spray Dried Dispersion (SDD) Based Low Dose Tablet Formulation

BACKGROUND

The design and development
of low-dose solid drug products presents a unique set of challenges during drug
product development.  The key concern with a low-dose product is achieving and
maintaining acceptable content uniformity of the therapeutic active.   In
addition, frequently there is a need to utilize enabling technologies to increase
bioavailability of poorly soluble drugs.  Rendering the active therapeutic
moiety to be in an ?amorphous state' is a promising approach to obtain a
material with high kinetic solubility.  In pharmaceutical industry, the bulk
amorphous dispersions are often manufactured using the spray drying
technology.  The process involves dissolving the active drug and polymeric
carrier(s) in a solvent system, followed by evaporating the solvent in a spray
dryer.  Notably, the spray dried materials often have physical properties
(size, shape and flow) that are different than those of traditional solid dose
excipients.

This work assesses
the feasibility of developing a low dose SDD formulation with a representative material,
 composed of  30% API and 70% hydoxypropyl-methylcellulose acetyl succinate (HPMC-AS)
polymer.  To support clinical development a wide dose range is often necessary;
it is preferable to use a common SDD (same drug loading) to ensure consistent
performance and to minimize complexity with regard to SDD process development.   Hence
the provision of low dose tablets, for example as low as 0.1mg, using very
small quantities of SDD, necessitates high assurance of content uniformity.  The
manufacturing process described in this work follows a dry granulation process
which was selected to provide for acceptable physical stability, and to reduce
the risk of segregation and poor content uniformity.

Traditional models
for predicting content uniformity (Rohrs et al.) based on mean particle size
and distribution skew suggest that the risk of content uniformity failure for typical
SDD material (e.g. D50 40µm) in a 1mg formulation is low (P<0.1%), whilst for
a 0.1mg tablet the risk of failure rises particularly as the size distribution
becomes skewed, for D90/D50 ratio >2, P failure >8%.  The feasibility of
using low dose SDD based formulations therefore presents unique challenges
which to date have not been fully explored.

METHODS

Tablets were manufactured
at a 10 kg scale (0.1 mg and 1.0 mg potency) using 0.33 and 3.33% SDD (SDD
contained 30% w/w active in its amorphous state).  The SDD and excipients
(microcrystalline cellulose, anhydrous lactose, croscarmellose sodium and magnesium
stearate) were mixed in a low shear blender, followed by co-milling and further
blending, and subsequently roller compacted (Alexanderwerk WP120).  The resultant
ribbons were milled, lubricated and compressed on a rotary tablet machine to a
target press weight of 100 mg.

During compression
tablets were sampled over five regular intervals and analyzed for content
uniformity (total 5 x 10 tablets).  Each final blend was riffled and separated
into sieve fractions and the active content determined using HPLC for each
fraction (>425µm, 180-425 µm, 63-180 µm, <63 µm).  A modified Jenike
& Johanson segregation test was also employed to evaluate segregation
potential.  The test used four cycles through a standard hopper design and a
1.8m gravity drop, six 100 mg samples were analyzed for content at different
segments of the discharged blend.   The physical distribution of SDD in the
formulated tablet was also evaluated by FTIR imaging  (Thermo Raman DXR). SDD
particle size was determined using a Malvern Morphologi G3 and the morphology
was confirmed by SEM.

RESULTS

SDD particles
produced from the representative spray process were determined to have a mean
D50 and D90 of approximately 30-40 µm and 50-60µm, respectively, with a
characteristic collapsed sphere morphology.  The content uniformity for low
dose tablets gave consistently low variation throughout compression , for
example the %RSD  for 0.1mg tablets remained between 1.5%- 2.4%  for all five
segments.

The sieve cut
potency data, however, showed an uneven distribution of active through the
particle size range, with fines (<63um) being super-potent and at least 1.6x
nominal content. 
The data corroborated with the J&J
segregation test results which showed an elevated RSD and potency range for the
samples analyzed; suggesting a risk for segregation of the final blends
on scale up.

                                        *Based on the proposed
segregation risk classification system, Shah et. al.

CONCLUSIONS

Although the
theoretical content uniformity model predicted borderline uniformity for the
0.1mg strength product, the experimental data from the manufacture of low dose
SDD tablets at a ~10kg scale indicated good dose uniformity is feasible.  The
content uniformity model assumes that particles are solid, however for hollow
or collapsed SDD structures the particles will contain less mass per particle
than expected.  Therefore, the risk of content uniformity failure will be lower
than predicted, with the risk decreasing with lower solid volume fraction (wall
thickness).  Acceptable uniformity was supported by FTIR imaging showing even
SDD distribution on the tablet surface.  The granule sieve cut potency and the
J&J segregation test results indicate that the risk for segregation
remains, and needs to be addressed during process scale-up and bulk transfer
operations.

ACKNOWLEDGEMENT

The
technical contributions by the Bend Research Inc. team in developing the Spray
Dried Dispersions are hereby acknowledged.

REFERENCES

-       
Rohrs,
B.R., Amidon, G.E., Meury, R.H., Secreast, P.J., King, H.M., Skoug, C.J.
(2006). Particle size limits to meet USP content uniformity criteria for
tablets and capsules.  J. Pharm. Sci. 95(5), 1049?1059.

-       
Shah,
K.R et al. (2007)' Assessment of Segregation Potential of Powder Blends', The
Pharmaceutical development and technology', Informa healthcare, Volume
12, pp 457-462.