Continuous twin-screw melt granulation of thermally labile drug – case study TONY L ISTRO, MS MBA

FOSTER DELIVERY SCIENCE

9 TH AMERICAN DRUG DELIVERY & FORMULATION SUMMIT

BOSTON, SEPTEMBER 9, 2019

1

Objectives Twin-screw melt granulation offers many advantages over roller compaction.

To investigate the effect of formulation and process variables on the physicochemical properties of granules • Binder type and binder particle size

• Screw design, barrel temperature, screw speed and feed rate

To understand the mechanisms and physicochemical changes during granulation • Dead-stop test

2

Presentation outline 1. Introduction of melt granulation and gabapentin (GABA), a

thermally labile drug with poor compaction property.

2. Selection of thermal binder and effect of thermal binder on the properties of GABA granules

3. Effect of processing conditions on the properties of GABA granules

4. Future studies and conclusions

3

Twin-Screw Melt Granulation

Granulation by TSE

Wet granulation

Melt granulation

• Continuous manufacturing • On-line and real time monitoring of product quality • Short granulation time and wider processing window • Reduction in binder (solution) level • Uniform distribution of formulation components • Less undesired physicochemical changes

• Use low-melting or thermoplastic materials as binders

• Energetic materials/explosives; powder metallurgy

• Improved flow and flow properties than roller-compacted granules

4

Nucleation mechanism of melt granulation: Depend on particle size and viscosity of binder

Distribution • Binder with low melt-viscosity • Molten binder is distributed onto the

surfaces of solid particles • Nuclei are formed by collision between

the wetted particle

Immersion • Thermoplastic binder with high melt

viscosity • Adhesion of solid particles onto the

surface of molten binder particles

James S, et al. Handbook of Pharmaceuitcal Granulation Technology. Taylor & Francis group LLC. 2005. 390-392. 5

Gabapentin (GABA) as a “Model drug” Goal of the study

• Identify formulation and process to (1) Improve compactability of gabapentin and (2) minimize processing-induced chemical degradation of gabapentin

Gabapentin as a “model drug” for melt granulation

• High-dose, poorly compressible drug

• Poor thermal stability

• Current commercial process: high-shear or fluidized-wet granulation. High impurity content of GABA tablets has been an real issue.

• During wet granulation, GABA is solubilized. The presence of polymeric binders prevent GABA from recrystallize during drying. The solubilized GABA undergoes significant degradation during the storage.

6

Properties of Gabapentin

Properties

Indication Anti epileptic

Description White to off-white, crystalline solid Form II, the most stable form, is used in this study

MW 171.24 g/mol

Melting point 162-166°C

pKa 3.7 (carboxylate), 10.7 (amine)

BCS class BCS class III (high solubility and low permeability)

Solubility pH-dependent solubility; soluble in water (100 mg/mL)

Particle size 6.1 μm (d10), 55.24 μm (d50), 215.64 μm (d90)

Others Crystalize rapidly, amorphous GABA could not be prepared

USP39 NF34 Gabapentin

https://pubchem.ncbi.nlm.nih.gov/compound/gabapentin#section=Top 7

Degradation pathway in solution & solid state: lactamization (GABA-L)

• Gabapentin degrades to a cyclic lactam via an intramolecular cyclization reaction triggered by a nucleophilic attack of the COOH group by the N of the amino group, followed by a dehydration reaction

• The degradation reaction is irreversible • USP specification of Gaba-lactam: NMT 0.4%

Zhizin Z, et al. The stabilizing effect of moisture on the solid-state degradation of gabapentin. AAPS PharmSciTech. 2011. 12(3):924-931. 8

GABA undergoes lactamization upon melting

Tm ~ 174°C -40

-30

-20

-10

0

0

20

40

60

80

100

50 100 150 200 250 300 350 400

% w

eigh

t

Temperature (°C)

Heat flo

w (W

/g)

Dehydration due to degradation (~10.5%w/w)

Overlap between melting and degradation

9

Melting and lactamization of GABA under hot-stage PLM

25°C 174°C 176°C (with bubble)

180°C 183°C 184°C

10

The experiment from DSC and Hot stage PLM confirm that Gabapentin is immiscible with binders

Preliminary study: binder selection Miscibility between GABA and binders

Hydrophilic binder

PEG 8000

Hydrophobic binder

Glycerol behenate (Compritol)

Thermoplastic polymer

HPC ELF (Klucel)

11

wt% hydroxypropyl groups: 53-81

Kittikunakorn N, Sun CC, Zhang F*. Effect of screw profile and processing conditions on physical transformation and chemical degradation of gabapentin during twin-screw melt granulation. Eur J Pharm Sci. 131:243-253 (2019).

Too good miscibility of GABA and binders is not desired!

Hold at 80°C Hold at 100°C Hold at 140°C

PEG (Tm ~60°C) Compritol (Tm ~70°C) HPC (Tg ~0°C, soften at 100-140°C)

12

Binders Zone 1 Zone 2 Zone 3

GB-PEG8000 80°C 80°C 40°C

GB-Compritol 90°C 90°C 60°C

GB-HPC ELF 120°C 120°C 70°C

GB-PEG8000 GB-Compritol 888 ATO GB-HPC ELF

80% GAGB and 20% binder; Feed rate 10 g/min, Screw speed 100 rpm

Leistritz nano 16 Open-end discharge

13

SEM Images of Granules

GAGB+HPC/1000 X

14

GABA+Compritol/1000 X GABA+PEG/1000 X

• Mill the granule and collect the granule between 20-60 mesh (250-850 μm)

mix with 1% Mg stearate compress into tablet

0.0

1.0

2.0

3.0

4.0

5.0

0.0 50.0 100.0 150.0

Ten

sile

str

engt

h (M

Pa)

Compression pressure (MPa)

Melt granulation(20%HPC ELF+GB)Direct compression(20%HPC ELF+GB)

0.0

1.0

2.0

3.0

4.0

5.0

0.0 50.0 100.0 150.0

Ten

sile

str

engt

h (M

Pa)

Compression pressure (MPa)

Melt granulation(20%PEG 8000+GB)Direct compression(20%PEG8000+GB)

0.0

1.0

2.0

3.0

4.0

5.0

0.0 50.0 100.0 150.0

Ten

sile

str

engt

h (M

Pa)

Compression pressure (MPa)

Melt granulation(20%Compritol+GB)

Direct compression(20%Compritol+GB)

Melt granulation significantly improves compaction properties. HPC is the most effective.

15

Degradation of GABA granules upon storage USP specification for GABA-L: NMT 0.4%

Induction-sealed HDPE bottles, desiccated

16

0.000

0.010

0.020

0.030

0.040

0.050

0.060

85 90 95 100 105 110

%Im

pu

rity

Barrel temperature (°C)

20%HPC ELF+GB

20%PEG8000+GB

20%Compritol+GB

Degradation of gabapentin USP specification for GABA-L: NMT 0.4%

• Higher barrel temperature led to higher level of degradant

• At the same temperature : HPC ELF-based granule shown higher % GABA-lactam than Compritol and PEG 8000-based granules

17

Particle size reduction during melt granulation

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

Den

sity

dis

trib

uti

on

q3

*

0.4 0.6 0.8 1.0 2 4 6 8 10 20 40 60 80 100

particle size / µm

x(10 %)

µm

21.72

7.57

4.69

1.57

x(50 %)

µm

63.12

42.88

21.41

10.45

x(90 %)

µm

116.13

96.49

49.06

41.90

Param. 3

Opt. concentration

%

33.47

25.69

26.51

30.75

Gabapentin drug substance

Compritol-GABA granules

HPC-GABA granules

PEG8000-GABA granules

GABA in HPC ELF based formulation has the smallest particle size high mechanical stress resulted in breakage of drug crystals and amorphization highest impurity

Acetone

Chloroform

18

Development of granule structure during the granulation along screw profile 20% HPC EXF + Gabapentin

70°C 120°C 120°C

19

Particle size of gabapentin along screw profile (EXF2-4)

• Sample the granules from each zone • Disperse in acetone in order to dissolve HPC • Measure the particle size of gabapentin

120°C 120°C 70°C

Feeding zone

Zone 1

Zone 2

Zone 3

Granules

Feeding zone Zone 1 Zone 2 Zone 3

Particle size of gabapentin decrease

20

PM Z1

Z2 Z3

PM

Z1

Z2

Z3

Total CH2N- C2H3O-

21

Binder Distribution on the surface of granules: Time-of-Flight Secondary Ion Mass Spectrometry (TOF-SIMS)

Melt rheology of binders

• Melt viscosity of HPC ELF (pseudoplastic) >> melt viscosity of PEG 8000 and Compritol (Newtonian fluid).

• The high viscosity of HPC

melt during granulation resulted in high shear stress that led to significant particle size reduction.

22

0.00

0.01

0.01

0.02

0.02

1 2 3 4

%G

AB

A-L

WEEK

40C, 10%RH

40C, 30%RH

40C, 75%RH

When granules were stored in open containers, slower degradation at higher humidity – due to crystallization of amorphous GABA

0.10

0.15

0.20

0.25

0.30

0 1 2 3 4

%G

AB

A-L

WEEK

40°C, 10%RH

40°C, 30%RH

40°C, 75%RH

GABA drug substance GB-HPC ELF Granules

23

“The stabilizing effect of moisture on the solid-state degradation of gabapentin”, Z. Zong, AAPS PharmSciTech, 12(3) 925-31 (2011)

Studying HPC of different particle size

• HPC ELF : D50 ~ 160 μm

• HPC EXF : D50 ~ 50 μm

• Spray-dried HPC : D50 ~ 10 μm

80% Gabapentin + 20% Binders

24

80% GAGB and 20% binder Feed rate 10 g/min, Screw speed 100 rpm

Effect of HPC particle size on GABA granule size

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

HPC ELF HPC EXF HPC SD

%w

/w

Granules size (μm)

>1180

850-1180

600-850

425-600

250-425

150-250

<150

25

Effect of HPC particle size on compaction profiles of granules

• Physical mixture : Small particle size of binder improve the compressibility of drug

• Melt granules : binder particle size does not have effect on the compressibility of drug

0.0

1.0

2.0

3.0

4.0

0 20 40 60 80 100 120 140

TEN

SILE

STR

ENG

TH (

MP

A)

COMPRESSION PRESSURE (MPA)

90°C 20% HPC ELF + GB (at 90°C)20% HPC EXF + GB (at 90°C)20% HPC SD + GB (at 90°C)

0.0

1.0

2.0

3.0

4.0

0 20 40 60 80 100 120 140

TEN

SILE

STR

ENG

HT

(MP

A)

COMPRESSION PRESSURE (MPA)

100°C

20% HPC ELF + GB(at 100°C)

0.0

1.0

2.0

3.0

4.0

0 20 40 60 80 100 120 140

TEN

SILE

STR

ENG

TH (

MP

A)

COMPRESSION PRESSURE (MPA)

110°C 20% HPC ELF + GB(at 110°C)

0.0

1.0

2.0

3.0

4.0

0 20 40 60 80 100 120 140

TEN

SILE

STR

ENG

TH (

MP

A)

COMPRESSION PRESSURE (MPA)

120°C

20% HPC ELF + GB (at120°C)

0.0

1.0

2.0

3.0

4.0

0 20 40 60 80 100 120 140

TEN

SILE

STR

ENG

TH (

MP

A)

COMPRESSION PRESSURE (MPA)

PHYSICAL MIXTURE

GABA-HPC ELF (PM)

GABA-HPC EXF (PM)

GABA-SD HPC (PM)

26

Effect of HPC particle size on the degradation of GABA

Smaller particle size of HPC more degradation

0.000

0.050

0.100

0.150

0.200

0.250

0.300

85 95 105 115 125

% G

AB

A-L

PROCESSING TEMPERATURE (°C)

GB+20%ELF granules

GB+20%EXF granules

GB+20%SD HPC granules

Gabapentin

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

Densi

ty d

istr

ibuti

on q

3*

0.4 0.6 0.8 1.0 2 4 6 8 10 20 40 60 80 100

particle size / µm

x(10 %)

µm

21.72

1.57

1.46

1.34

x(50 %)

µm

63.12

10.45

10.77

8.83

x(90 %)

µm

116.13

41.90

43.71

47.16

Param. 3

Opt. concentration

%

33.47

30.75

34.95

29.85

Gabapentin

20% HPC ELF + GB

20% HPC SD + GB

20% HPC EXF + GB

No significant difference in particle size reduction after melt granulation

27

Effect of processing variables: screw speed and feed rate

70°C 110°C 110°C

• Move kneading element further down stream • Remove some narrow pitch conveying element to lower the torque

vent

28

5 g/min

7.5 g/min

10 g/min

100 rpm 150 rpm 200 rpm 300 rpm

29

The effect of screw speed and feed rate on GABA extrudate size

Effect of screw speed and feed rate on the level of GABA-L

• Degradant content increases with increasing feed rate and decreasing the screw speed (increasing specific rate)

0.000

0.020

0.040

0.060

0.080

0.100

0.120

0.140

100 150 200 250 300 350

% G

AB

A-L

SCREW SPEED (RPM)

10 g/min

7.5 g/min

5 g/min

30

Impurity increases as degree of fill of conveying elements prior to kneading elements increases

y = 0.0042x + 0.0263 R² = 0.7966

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.00 5.00 10.00 15.00 20.00 25.00

%G

AB

A-L

%DEGREE OF FILL OF CONVEYING ELEMENTS

IMPURITY VS DEGREE OF FILL

10 g/min7.5 g/min5 g/min

𝐾𝑊 (𝑎𝑝𝑝𝑙𝑖𝑒𝑑) =𝐾𝑊 𝑚𝑜𝑡𝑜𝑟 𝑟𝑎𝑡𝑖𝑛𝑔 𝑥 %𝑡𝑜𝑟𝑞𝑢𝑒 𝑥 𝑟𝑝𝑚 𝑥 0.97

𝑀𝑎𝑥. 𝑟𝑝𝑚

𝑆𝑝𝑒𝑐𝑖𝑓𝑖𝑐 𝑒𝑛𝑒𝑟𝑔𝑦 =𝐾𝑊 𝑎𝑝𝑝𝑙𝑖𝑒𝑑

𝐹𝑒𝑒𝑑 𝑟𝑎𝑡𝑒 (𝑘𝑔ℎ𝑟

)

%𝐹𝑖𝑙𝑙 = 𝑭𝒆𝒆𝒅 𝒓𝒂𝒕𝒆 𝑥 100

(𝐶𝑟𝑜𝑠𝑠 𝑠𝑒𝑐𝑡𝑖𝑜𝑛 𝑎𝑟𝑒𝑎 𝑥 𝑃𝑖𝑡𝑐ℎ 𝑙𝑒𝑛𝑔𝑡ℎ 𝑥 𝒓𝒑𝒎 𝑥 𝐷𝑒𝑛𝑠𝑖𝑡𝑦)/2

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.10 0.11 0.12 0.13 0.14 0.15 0.16 0.17 0.18

%G

AB

A-L

SPECIFIC MECHANICAL ENERGY (KW)

IMPURITY VS SPECIFIC MECHANICAL ENERGY

10 g/min

7.5 g/min

5 g/min

31 Kittikunakorn N, Koleng JJ III, Listro T, Calvin Sun C, Zhang F*. Effects of thermal binders on chemical stabilities and tabletability of gabapentin granules prepared by twin-screw melt granulation. Int J Pharm. 559:37-47 (2019).

GABA particle size of GABA in extrudates: highest vs. lowest degree of fill

Granule, 5 g/min, 300 rpm

Granule, 10 g/min, 100 rpm

GABA drug substance

High degree of fill

Low degree of fill

32

0.000

0.020

0.040

0.060

0.080

0.100

0.120

% G

AB

A-L

AC

TAM

% GABA-L in GABA granules along screw profile

5 g/min and 100 rpm

33

10 20 30 40

Two-Theta (deg)

x103

2.0

4.0

6.0

8.0

10.0

12.0

14.0

Inte

nsity(C

ounts

)[MG Run 2-10.raw] MG Run 2-10 (repeat)

[MG Run 2-7.raw] MG Run 2-7

10 g/min, 100 rpm High degree of fill

5 g/min, 300 rpm Low degree of fill

XRD profiles of GABA granules indicate higher amorphous content at high degree of fill

34

More compressible GABA granules at higher degree of fill

0.0

1.0

2.0

3.0

4.0

0 20 40 60 80 100 120 140

TEN

SILE

STR

ENG

TH (

MPA

)

COMPRESSION PRESSURE (MPA)

10 g/min at 100 rpm

5 g/min at 300 rpm

Physical mixture

35

Ongoing studies

36

1. Evaluate split feeding to minimize drug degradation while improving the compaction properties of GABA granules • Option 1: feed molten HPC into the primary extruder

• Option 2: use the primary extruder to melt HPC and side-stuff GABA

2. Quantify the thermal and mechanical stress during melt extrusion

3. More advanced technique to characterize binder distribution

Conclusions • From improving the compaction properties perspective, hydroxypropyl

cellulose, a thermoplastic polymer, is more effective than low melting point waxes such as PEG 8000 and Compritol.

• High melt viscosity of HPC resulted in more chemical degradation during processing and upon storage.

• Both the size of the granules coming off the extruder and the impurity of GABA correlate better with the degree of fill (or specific rate) than the specific mechanical energy.

• Processing parameters (screw speed and feed rate) should be optimized to achieve the balance between improving GABA compressibility but also minimizing GABA degradation.

37

Acknowledgements

38

Feng Zhang, PhD Nada Kittikunakorn

Charlie Martin Augie Machado Brian Haight

Larry Acquarulo