Linking Drug Stability to ManufacturingPhysical Chemical Foundations

Gabapentin

L. E. KirschStability team leader

Stability TeamGroup Team member

Minnesota Raj Suryanarayanan (Co-PI)

Aditya Kaushal (post-doc)

Kansas Eric Munson (Co-PI)

Dewey Barich (post-doc)

Elodie Dempah, Eric Gorman (grad. students)

Iowa Lee Kirsch (Co-PI)

Greg Huang (Analytical Chemist)

Salil Desai, Zhixin Zong, Tinmanee Radaduen, Hoa Nguyen, Jiang Qiu (grad students)

Duquesne(Unit-op team Interface)

Ira Buckner

3

Linking manufacturing to stability

API

API*

Degradant

PHYSICAL

TRANSFORMATION Chemica

l

trans

form

ation

(Stable form)

(Unstable form)

4

Gabapentin as a model drug substance

NH2

OH

O

gabapentin(Gaba)

• Multiple crystalline forms• Susceptible to stress-induced physical

transformations• Susceptible to chemical degradation

NH3+

O-

O

NH2

O-

O

NH3+

OH

O

pKa 3.7 pKa 10

KEY QUESTIONS1. Are physical and chemical instability

linked?2. How can manufacturing-induced stress be

incorporated in a quantitative chemical instability model?

Some Crystalline Forms of Gabapentin

5

API form Crystalline

I

II

III

IV

Ibers., Acta Cryst c57, 2001 and Reece and Levendis., Acta Cryst. c64 2008

Transition between forms by mechanical stress, humidity, and thermal stress

Hydrate

Stable polymorph (API)

Intramolecular H-bonding

4 6 8 10 12 14 16 18 20 22

2Theta

Physical transformation by Mechanical Stress

Form II

Form III

Milled Gabapentin

Physical transformation by Humidity

2theta

7

Inte

nsity

47 hrs in 40C 31 %RH29 hrs17 hrs 7 hrs 0 hr

Physical transformation by Thermal Stress

Kaushal and Suryanarayanan., Minnesota Univ. AAPS poster 20098

Aqueous degradation kinetics

OOH

NH2 NH

O

gabapentin lactam

Irreversible cyclization

+ H2O

toxicUSP limit: < 0.4%

Solid state degradation kinetics40 C 5% RH, milled gabapentin

0

1

2

3

4

5

6

7

0 100 200 300 400 500 600

Lact

am (m

ole

%)

hours

initial lactamrapid degradation of process-damaged gaba

autocatalytic lactam formation

11

Solid state Degradation Model

GABA (G)(stable form) LACTAM (L)

autocatalytic branching

spontaneous dehydration

branching termination

LDGk 1

Dk2GABA (D)(unstable form)

DGk3

Hypothesis:Manufacturing stress determines initial conditions (G0, D0 and L0)Environmental (storage) stress determines kinetics (k1, k2 and k3)

Building a quantitative degradation model

12

DrugStability

Compositional Factors

(e.g. excipients)

Environmental Stress

ManufacturingStress

13

Effects of Manufacturing Stress:Initial Lactam and Instability

0 5 10 15 20 25 300.0

0.5

1.0

1.5

2.0

2.5%

lact

am

time(days)

60 min milled

45 min milled

15 min milled

API as received

Thermal stressed at 50 °C, 5%RH

Lactam generated during milling(in-process lactam)

Milling caused faster degradation rate

14

Can Surface Area account for Lactamization Rate Changes upon

Mechanical Stess?

0 5 10 15 200

0.2

0.4

0.6

0.8

1

Specific Surface Area(m2/g)

Lact

amiza

tion

Rate

(mol

e%/d

ay 5

0 °C

)

Samples milled for different time

Sieved aliquots of 15min milled sample

Sieved aliquots of unmilled sample

NO, ALSO increased regions of crystal disorder caused by the mechanical stress.

Mechanical Stress Impact on Lactam Formation at 50 °C: No kinetic effects

15

TreatmentD0(%)

unstressed 0.027

15min milled 0.46

45min milled 0.92

60min milled 1.30

60min

0 200 400 600 800 1000 1200 14000.000

0.500

1.000

1.500

2.000

2.500

3.000

3.500

4.000

4.500

hours

Lact

am m

ole

%

45min

15min

0 min

Effects of Temperature:predicted values based on parameterization of

autocatalytic model

0 50 100 150 200 2500

5

10

15

20

25

time(hours)

Lact

am m

ole%

Gaba 40C 5%RH

Gaba 50C 5%RHGaba 60C 5%RH

17

Effects of Moisture

18

Why moisture appears to slow and shut down lactam formation?

• In general, effect of moisture is NOT to slow reaction rates

• Analytical issue?

• Reversible reaction?

• Formation of stable hydrate?No gabapentin formed from gaba-L in solution or solid state

No hydrate found from XRD patterns

Most gaba-L could be recovered from solid powder, only ignorable gaba-L was detected in saturated salt solution.

Moisture-facilitated termination of branching

19

Effect of Moisture:Shut down Lactam Formation

0

1

2

3

4

0 20 40 60 80 100

Gaba

-L C

once

ntra

tion

(Mol

e %

)

Hours

Pretreated at 5% RH 25°C for 24 hours before thermal stress

Pretreated at 81% RH 25°C for 24 hours before thermal stress

Thermal stress: 50°C 5%RH

0.001

0.01

0.1

1

10

100

1000

104

105

0 10 20 30 40 50 60

Estim

ate

cons

tant

s (r

ate

x 10

6)

Relative Humidity (%)

Humidity effects (40 °C)

20

0.000

1.000

2.000

3.000

4.000

5.000

6.000

7.000

8.000

9.000

10.000

0 100 200 300 400 500 600 700 800

Lact

am m

ole%

hours

5% RH

11% RH

30% RH50% RH

k3 termination

k1 branching

lactam time profiles rate constants vs RH

k2 cyclization

– Mixtures of gabapentin & excipients– Co-milled– Storage conditions: 5 to 50% RH at 50 ˚C

• Excipients (50% w/w)– CaHPO4.2H20 (Emcompress)– Corn starch– Microcrystalline cellulose (Avicel PH101)– HPMC 4000– Colloidal SiO2 (Cab-O-Sil)– Talc (Mg silicate)– HPC (6.5% w/w)

Evaluation of the role of excipients in gabapentin SS degradation

Saturated solution 50˚C

0

10

20

30

40

50

0 100 200 300 400 500

5RH 4:47:40 AM 10/22/2010

gabaAviHPMCCabTalcHPC Calccorngaba obscalc obscorn obsAvi obsHPMC obsCab obsTalc obsHPC obs

Gaba

Starch

CaHPO4SiO2

HPC AvicelHPMC

Talc

Lact

am m

ole

%Time (hr)

Excipient Effectscontrolled temperature (40-60 C) and humidity (5-50% RH)

0 100 200 300 400 500 600 700 8000.000

2.000

4.000

6.000

8.000

10.000

hours

Lact

am m

ole%

0 100 200 300 400 500 600 700 8000.000

4.000

8.000

12.000

16.000

20.000

hours

lact

am m

ole%

No excipient

6.5% HPC

• Crystal damage (D0) during milling• Kinetics of branching(k1) and

termination(k3)

Effect of co-milled excipients on crystal damage during milling

0 5 10 15 20 25

Si02

CaHPO4

Talc

HPMC

HPC (6.5%)

Starch

None

% Unstable Gabapentin (D0)

Excipients50% w/w

Moisture and excipient effectsNo excipient Co-milled excipient (SiO2)

5 %RH 11 %RH30 %RH

50 %RH

11 %RH

30 %RH

50 %RH

5 %RH

24

0

10

20

30

40

50

0 100 200 300 400 500 600

Data 10

BDFH

Lact

am m

ole

%

Time (hr)

0

5

10

15

20

0 100 200 300 400 500 600 700 800

moisture effect gaba50RH

0RH11RH30RH50RH

Effect Moisture on Lactamization Kineticsfor gabapentin/HPC (6.5%) mixtures: blue: HPC and red: no HPC

0.01

0.1

1

10

100

1000

104

105

10 20 30 40 50 60

Estim

ated

con

stan

ts (k

1 106 ,

k 3107 )

Relative Humidity (%)

K (k1/k3)

k1 branching

k3 termination

Linking Stability in Design SpaceManuf.Design SpaceModel

L0D0

Post-Manuf.

Degradation

Model

LtEndof

Expiry

• Key Research Findings• Manufacturing Stress impacts drug stability upon storage:

L0 (in-process lactam) D0 (unstable gabapentin)

• Predictive model for drug stability includes:• Environment factor: temperature () & humidity ()• Compositional factors: both kinetic and initial condition

effects• Manufacturing factors: L0 and D0

• Model validation: completion of long term stability

Measuring the manufacturing stress effects• Physical methods

– Raj Suryanarayanan (University of Minnesota) – Eric Munson (University of Kentucky)

• Chemical and kinetic measurements– Lee Kirsch (University of Iowa

Solid State NMR KansasRaman spectroscopy MinnesotaPowder x-ray diffraction (XRD) MinnesotaDSC/TGA All Water vapor sorption MinnesotaHPLC Iowa

Chromatographic Approach for Manufacturing Stability Measurement

Minutes

1 2 3 4 5 6 7 8 9 10

mAU

0.00

0.25

0.50

0.75

1.00

1.25

1.50

mAU

0.00

0.25

0.50

0.75

1.00

1.25

1.50

4093

741

Gab

apen

tin

3.65

8

2388

5.39

0

4339

La

ctam

7.28

8

3853

9.11

7

Detector 1-210nmhydBt0H

AreaNameRetention Time

Detector 1-210nmhydBt24H

Comparison of HPLC chromatograms before (black) and after (red) thermal stress:

∆ lactam = 0.004%.

Minutes

1 2 3 4 5 6 7 8 9 10

mAU

0

1

2

3

4

mAU

0

1

2

3

4

2801

635

Gab

apen

tin

3.66

8

8278

La

ctam

7.30

7

Detector 1-210nmhydAt0H

AreaNameRetention Time

Detector 1-210nmhydAt24H

Comparison of HPLC chromatograms before (black) and after (red) thermal stress:

∆ lactam = 0.059%.

Minutes

1 2 3 4 5 6 7 8 9 10

mAU

0

5

10

15

20

mAU

0

5

10

15

20

4635

741

Gab

apen

tin

3.61

8

2878

43

7.57

2

(Lactam)

Detector 1-210nmlotAH

AreaNameRetention Time

Detector 1-210nmlotAHbefore

Comparison of HPLC chromatograms before (black) and after (red) thermal stress:

∆ lactam = 0.174%.

Manufacturing-stability measurements

• In process lactam (L0)– Change in lactam levels during specific treatment or unit

operation in % lactam/gabapentin on molar basis• Initial Rate of Lactam Formation (V0 or STS)

– Daily rate of lactam formation upon thermal stress at 50°C under low humidity

• D0 from Chemical Analysis

dayCk

kVD

DkV

o %/37.0)50(2

2

00

020

Insert Sury

Insert Eric

Applied Manufacturing-stability Measurements to Design Space and Risk Assessment

• Laboratory scale stability design space• Pilot scale stability design space• Risk assessment using Manufacturing-

stability Measurements