(605e) A Quality By Design Approach to Address the Challenges of Direct Compression Scale-up and Technology Transfer between Rotary Tablet Presses

A quality by design approach to address the
challenges of direct compression scale-up and technology transfer between
rotary tablet presses

Raghu V. G. Peddapatla¹, Gerard Sheridan²,
Conor Slevin², Shrikant Swaminathan³, Ivan Browning⁴,
Clare O’ Reilly⁴, David Egan²,Stephen Sheehan*⁴,
Abina M. Crean^1,5

¹School of Pharmacy,
University College Cork, Cork, Ireland

²Bernal Institute, University
of Limerick, Limerick, Ireland

³Alkermes Inc, Waltham, MA,
USA.

⁴Alkermes Pharma Ireland
Limited, Athlone, Ireland

⁵Synthesis and Solid State
Pharmaceutical Centre, Ireland

* Alkermes Pharma Ireland Limited, Athlone, Ireland, Tel: + 353906495412,

email: Stephen.Sheehan@alkermes.com

INTRODUCTION

The key outcome of
implementing quality by design (QbD) is attainment of process understanding, i.e.
understanding the relationships between product critical quality attributes
(CQAs), process parameters and formulation attributes.¹ During the process
development stage a degree of process understanding can be attained using pilot
scale equipment. However, differences in equipment designs utilised at pilot
scale and production scale makes extrapolation of process understanding from
pilot to production scale challenging. When scaling-up a process, or
transferring a process between equipment models, different processing
parameters can be encountered which need to be considered for thorough process
understanding and hence QbD.

The concept of a design
space is integral to QbD. Design space is a multidimensional space that
encompasses combinations of material attributes and processing parameters to
provide assurance of processing conditions that can achieve product CQAs.
However, due to the restricted processing parameters that can be explored on
certain pilot scale equipment, it is difficult to generate a single design
space that will encompass both pilot and production scale equipment during
development. This is particularly evident for tablet presses where feeder designs and press
speeds can vary greatly between equipment models.²
Therefore, to create an accurate design space for a product processed on a
particular production tablet press additional experimental runs are required
using that production press. Each design space generated is to some degree limited
by the equipment on which the experimental runs were conducted.

The aim of this study is
to develop a generic understanding of how direct compression formulations with
differing material attributes are compacted on selected pilot and production scale
tablet presses. Six placebo formulations with differing flow and compaction
properties were designed and compacted on one pilot scale and two production
scale presses. For each press an experimental design space was generated
encompassing both formulations attributes (e.g. flow and compaction behaviour).
The objectives of this study were two fold. Firstly, to enhance understanding
of the impact of transferring formulations between press models, specifically
how formulation attributes and press parameters influence compaction and flow
behaviour. Secondly, to benchmark the attributes of commercial formulations to the
placebo formulations and thereby predict their behaviour on different press
models. The overall goal of this work is to create a knowledge base on how to
optimize a future OSD active formulation on each tablet press. From a
commercial process development perspective this information will be vital to
reduce a new products time to market.

MATERIALS AND METHODS

Tablet presses

The following rotary
presses were utilised in this study KG RoTab (KG-Pharma, Scharbeutz, Germany),
Fette 1200i (Fette Compacting, Schwarzenbek Germany) and Modul^TM P
(GEA, Belgium).

Formulations

Six placebo formulations were used in
this study designed to achieve differing flow and compaction properties, shown
in Table 1.

Blend preparation and
characterisation

Formulation ingredients were
dispensed and fed through a 450µm Sweco sieve to remove any agglomerates.
Except lubricant, all other formulation components (diluents, flow aid and
disintegrant) were added into a 100L IBC and blended for 18 minutes at 18 rpm.
1%w/w of lubricant was also passed through the 425µm sieve before being added
to the other components in IBC. Blending was then continued for a further 3
minutes.

Design of Experiments (DoE)

To investigate the impact of formulation
on flow and compaction behaviour on each press, two separate DoEs were designed
using JMP statistical software. Based on process parameters investigated and levels,
a factorial experimental design was used with 4 centre points (Table 1).

A 2⁴ factorial DoE was used to study the compaction
behaviour for formulations 1-3 (Table 1). The effect of formulation,
pre-compression force, main compression force and press speed on the tablet
CQA’s (weight variability, thickness, breaking force, porosity, friability and
disintegration time (DT)) was investigated.

A 2³ factorial DoE was used to study the
flow behaviour for formulations 4-6 (Table 1). The effect of formulation,
feeder speed and press speed on tablet CQA’s.

Table 1 Formulations with different compaction and flow
behaviours used in the study

Blend characterisation

Yield pressure and elastic recovery
were used to differentiate the different compaction mechanisms within
formulations 1-3. Freeman FT4 powder rheometry was used to assess the flow
behaviours of formulations 4-6.

Blend compaction

All tablet blend formulations were compressed at target compression
forces of 1.2 to 21 kN, which corresponded to compression pressures of 20 MPa
to 360 MPa using the shield shape punches on all three tablet presses. Compression
profile was further used to optimise the main compression forces for
compression and flow DoE.

RESULTS AND DISCUSSION

Formulation characterisation

Yield pressures and tablet elastic
recovery were determined to characterise the compaction mechanism (plastic or
brittle or elastic) in formulations 1-3. Table 2 shows the yield pressures of
the formulations on two tablet presses. On both tablet presses, formulation 1
showed a lower yield pressure, indicating higher plasticity compared to
formulations 2 and 3. Tablets 24 hr thickness recovery, showed higher elastic
recovery in formulations 2 (Table 2). Formulation 3 contained a higher
percentage of an excipient with known brittle behaviour.

Table 2 Yield pressures and elastic recovery of
formulations 1-3 compacted on Modul P and KG RoTab.

FT4 flow analysis of
formulations 4-6 showed that formulation 4 is good flowing compared to
formulation 5 (medium flow) and formulation 6 (passable flow), shown in
Figure-1.

Figure
1 Flow function of formulations 4-6 analysed by FT4 powder flow rheometer.

Breaking force and
disintegration time variation by model factors on Modul P

For each tablet press and each
formulation a model was developed using JMP software from which a design space
could be constructed. A representative output of the model is shown in Figure 2.
The effect of
process parameters on tablet breaking force and DT was assessed by model
factors for formulations 1-3. Press speed, formulation type and main
compression force showed a significant effect on the tablet breaking force and
DT for the formulations, shown in Figure 2A-B. Pre-compression force did not show
significant effect on breaking force and DT.

Figure 2 Prediction profiler for mean
breaking force and disintegration time for formulation 1-3 on Modul P. (A) Mean
breaking force (B) Disintegration time

This model can be used to predict the
quality attribute of tablet for specific type of formulation, on specific
tablet press using prediction profiler (Figure 2). For example, at 14.2kN
compression force and 93000 tablets per hour (TPH) press speed, the expected
quality for formulations can be predicted. At these settings formulation-1
produced tablets with greater breaking force and DT (Figure 2B) compared to
formulation 2 and 3, where they produced tablets with similar breaking force
(Figure 2A).

Similar models will be
presented for the KG RoTab and Fette 1200i presses.  Examples will be presented
illustrating how these models can be applied to support scale-up and technology
transfer of formulations between rotary tablet presses.

CONCLUSIONS

Experimental designs were
executed on one pilot-scale and two production scale rotary tablet presses
studying the impact of both process parameters and formulation behaviours (compaction
and flow) on tablet quality attributes. A predictive model was developed to identify
significant factors impacting on tablet attributes to create a process design
space for each tablet press. For future OSD active formulations these
predictive models will allow for a seamless transition in scale-up and/or
technology transfer between each tablet press, thus reducing a new product’s time
to market.

BIBLIOGRAPHY

1.        FDA,
2009. Guidance for Industry: Q10 Pharmaceutical Quality System. (Accessed on April
04, 2019) https://www.fda.gov/downloads/Drugs/
GuidanceComplianceRegulatoryInformation/Guidances/UCM073517.pdf.

2.        Narang
AS, Rao VM, Guo H, Lu J, Desai DS. Effect of force feeder on tablet strength
during compression. Int J Pharm. 2010;401(1-2):7-15.
doi:10.1016/j.ijpharm.2010.08.027