particle size analysis in the pharmaceutical industry ia: method validation *bell, dennis,...
TRANSCRIPT
© 2007 HORIBA, Ltd. All rights reserved.
Particle Size Analysisin the Pharmaceutical Industry
Mark Bumiller
© 2007 HORIBA, Ltd. All rights reserved.
Outline
Why is particle size important?Relationship between particle size and dissolution, absorption and content uniformity
Do I need to set a specification?Recent recommendations from PQRIParticle sizing techniques
Microscopy/image analysis, dynamic light scattering, laser diffraction, acoustic attenuation
– ISO standards & USP <429>
© 2007 HORIBA, Ltd. All rights reserved.
Monitor Particle Processes
Milling/size reductionMixing/blendingSeparationFiltrationGranulationHomogenizationCrystallization
Narrow particle size distributions minimize segregation problems during mixing – more homogeneous distribution of components in final product
© 2007 HORIBA, Ltd. All rights reserved.
Particle Size and Dissolution
David R. Friend, PhD; Gregory E. Parry, PhD; T. Francis, PhD; Gary Kupperblatt, PhD; Suggy S. Chrai, PhD; and Gerald Slack,Mathematical Modeling of a Novel Controlled-Release Dosage FormDrug Delivery Technology,
•
XS is the mass of solid drug (mg),t is time (minutes), D is the drug diffusivity (cm2/min), X0 is the initial drug mass (mg), r is the drug density (mg/mL), h is the diffusion layer thickness (cm), r0 is the initial particle radius (cm),CS is the drug solubility (mg/mL), Xd is the mass of dissolved drug (mg), V is the volume of dissolution media (mL).
© 2007 HORIBA, Ltd. All rights reserved.
Tablets SuspensionsSize of active ingredient effects content uniformitySize influences tablet hardnessSize and shape effects packingSize and shape effect powder flow
Dissolution and absorptionContent uniformityAbility to stay in suspensionFeel in mouth
© 2007 HORIBA, Ltd. All rights reserved.
Excipients
Particle size and physical characteristics critical in selection and performance
© 2007 HORIBA, Ltd. All rights reserved.
Not Just Theoretical
Included in the January FDA-483 report were observations alleging that the product quality review methods for unnamed drug delivery and particle size "were inadequate in that the methods exhibit various unidentified extraneous peaks."
PRODUCTUnnamed Drug Inhalation Aerosol with mouthpiece, Net Contents 14g (10 ml), Metered Dose Inhaler, 200 metered doses. Recall # D-840-2007CODELot Numbers: 050912W, expiration date Dec 07 and lot # 060359W, expiration date Jun 08.RECALLING FIRM/MANUFACTURERRecalling Firm: Unnamed Company, letter dated May 14, 2007. Manufacturer: Unnamed Company. Firm initiated recall is ongoing. REASON Product failed particle size distribution analysis during stability.VOLUME OF PRODUCT IN COMMERCE137,491 canisters
© 2007 HORIBA, Ltd. All rights reserved.
Preclinical and Phase I*
Scheme for outlining particle evaluation for preclinical studies.
Decision tree outlining particle evaluation for Phase I clinical studies*PQRI Recommendations on Particle-Size Analysis of Drug Substances Used in Oral Dosage FormsJOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 6, JUNE 2007
© 2007 HORIBA, Ltd. All rights reserved.
Phase II
Scheme for outlining particle evaluation for Phase III clinical studies
© 2007 HORIBA, Ltd. All rights reserved.
PQRI Recommendations: Techniques
PQRI Recommendations on Particle-Size Analysis of Drug Substances Used in Oral Dosage FormsJOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 6, JUNE 2007
© 2007 HORIBA, Ltd. All rights reserved.
Microscopy in the Pharm Industry
Microscopy used for many years– USP <776> “Optical Microscopy”
USP <788> for parenterals– Light obscuration or microscope
Home made, components since 1985Automated image analysis since ~1995New generations vastly superior
– Computer speed, cameras, software
© 2007 HORIBA, Ltd. All rights reserved.
USP <776> TermsAcicular— Slender, needle-like particle of similar width and thickness. Columnar— Long, thin particle with a width and thickness that are greater than those of an acicular particle. Flake— Thin, flat particle of similar length and width. Plate— Flat particles of similar length and width but with greater thickness than flakes. Lath— Long, thin, and blade-like particle. Equant— Particles of similar length, width, and thickness; both cubical and spherical particles are included.
© 2007 HORIBA, Ltd. All rights reserved.
Don’t use words to describe particle shape Quantify morphology through image analysis
Dynamic:particles flow past camera
Static:particles fixed on slide,stage moves slide
Two Approaches
© 2007 HORIBA, Ltd. All rights reserved.
Multi-particulates, Dynamic IA
Inert CoreDrug + BinderPolymer CoatingDrug release
Sugar Sphere Plus Drug Layer Plus Polymer Layer
Ghost After Dissolution,
Expansion Noticeable
Sugar Sphere Plus Drug Layer
Sugar Sphere
Heinicke et al., Pharm Dev Tech, 10, 85, 2005 G. Heinicke©
© 2007 HORIBA, Ltd. All rights reserved.
Drug Layering*13%Weight gain
15%Weight gain
Sphericity = 0.96
G. Heinicke©
*Heinicke et al., Pharm Dev Tech, 10, 85, 2005
© 2007 HORIBA, Ltd. All rights reserved.
Drug Layering*
Heinicke et al., Pharm Dev Tech, 10, 85, 2005G. Heinicke©
© 2007 HORIBA, Ltd. All rights reserved.
Pharmaceutical ApplicationsSugar Spheres Globules
Coating thickness
20% < 0.98 Roundness1% < 0.98 Roundness
© 2007 HORIBA, Ltd. All rights reserved.
Static Image Analysis: PSA300
•Release vacuum•No point of impact•Software controlled•Break up agglomerates•Don’t break fragile particles•Even distribution on slide
undispersed dispersed
© 2007 HORIBA, Ltd. All rights reserved.
Image Processing
Erosion: remove
Dilation: add
Closing: bothsmaller valleysfilled
Delineation:increasecontrast at edges
Tool box to improve imagesbefore assigning size &shape values
© 2007 HORIBA, Ltd. All rights reserved.
Image Processing
Multi-layer grab
Contrast Thresholding
Convex hull
© 2007 HORIBA, Ltd. All rights reserved.
Fiber Separation
Acilcular SeparationEach fiber has its own bitplaneActually measure longestdiameter
© 2007 HORIBA, Ltd. All rights reserved.
Active Ingredients
150.050.020.010.05.02.01.00.3CircDiam (µm)
0.0
5.0
10.0
15.0
20.0
25.0
Sph
Vol (
%)
0
20
40
60
80
100
Cum
ulat
ive (%
)
StatisticsMinimum: 0.3 µmMaximum: 117.7 µmD[4,3] 52.3 µmStd Dev.: 34.6 µmSum: 553229657.7 µmCount: 11208Under: 0Over: 0Accepted: 100.0 %Field Count: 400Field Area: 183702.5 µm²Total Area: 73480997.2 µm²D10: 15.9 µmD50: 40.1 µmD90: 100.0 µm
1.00.50.20.10.1 0.0Roundness
0
5
10
15
20
25
Cou
nt (%
)
0
20
40
60
80
100
Cum
ulat
ive (%
)
Mean 0.6Std Dev 0.2Count 11208Field Count 400
20.010.05.02.01.0AspRatio
0
5
10
15
20
25C
ount
(%)
0
20
40
60
80
100
Cum
ulat
ive (%
)
Mean 1.9Std Dev 1.0Min 1.0Max 17.6Count 11208Field Count 400
© 2007 HORIBA, Ltd. All rights reserved.
Static IA: Method Validation
*Bell, Dennis, Hendriksen, North, Sherwood, Position Paper on Particle Sizing: Sample Preparation, Method Validation and Data Presentation, Pharmaceutical Technology Europe, November 1999
RecommendationsOf the possible validation headings discussed above, the following items are considered significant and should be included in a particle sizing validation report:• the procedure (or reference to it)• precision• range (suitability assessment including microscopic comparison)• robustness.
© 2007 HORIBA, Ltd. All rights reserved.
Submicron by PCS: Brownian motion: 1 nm – 1 µm
Particles in suspension undergo Brownianmotion due to bombardment by solventmolecules in random thermal motion.
Nanoparticles attachedto red blood cells
© 2007 HORIBA, Ltd. All rights reserved.Explore the Future Seminar Series 2009 Horiba Particle Characterization
Dynamic Light Scattering: 1 nm – 1 µm
Particles in suspension undergo Brownian motion due to solventmolecule bombardment in random thermal motion.
• Brownian Motion– Random– Related to Size– Related to viscosity– Related to temperature
© 2007 HORIBA, Ltd. All rights reserved.Explore the Future Seminar Series 2009 Horiba Particle Characterization
Diffusion
Particle is randomly diffusingLarger particles will diffuse more slowlyLarger particles have more Inertia
Scatter light off this diffusing particleMeasure the Frequency Shift of the signal
Laser
Dete
ctor
Frequency Shifted Signal
© 2007 HORIBA, Ltd. All rights reserved.Explore the Future Seminar Series 2009 Horiba Particle Characterization
Stokes-Einstein
Dynamic Light Scattering
Measured frequency-intensity distribution (power spectrum)
Power spectrum takes form of Lorentz distribution, whose half-value width can be expressed as 2Dq²
All parameters in the half-width are known or measured
The Diffusion Coefficient D is related to the Particle Size
Frequency Shifted Signal
Frequency-IntensityDistribution
Iterative Calculation
© 2007 HORIBA, Ltd. All rights reserved.Explore the Future Seminar Series 2009 Horiba Particle Characterization
Liposomes
Liposomes to target tumor growthSize is critical to how the liposome
Encapsulates proteinFunctions within bodyRemains stable over timeDelivers the protein
© 2007 HORIBA, Ltd. All rights reserved.Explore the Future Seminar Series 2009 Horiba Particle Characterization
2mg/mL filtered BSA
BSA- well characterized protein DLS – Can be used to determine the aggregation state of the protein
© 2007 HORIBA, Ltd. All rights reserved.Explore the Future Seminar Series 2009 Horiba Particle Characterization
Protein Aggregation Time Study*
Unstabilized 10mg/ml lysozyme at pH 2
*Lisa Cole and Ben Burnett at the Florida Institute of Technology
© 2007 HORIBA, Ltd. All rights reserved.Explore the Future Seminar Series 2009 Horiba Particle Characterization
Time EvolutionStability is influenced by:
TemperatureProtein concentrationpHIonic strength
Aggregation is influenced by:Freezing Exposure to airInteractions with metal surfaces
© 2007 HORIBA, Ltd. All rights reserved.
ADVANTAGES:Can accommodate high sample concentrationsNo need to consider complex optical properties
Acoustic Spectroscopy
ApplicationsEmulsions & suspensionsDispersion stability with zeta potential
DetectorSignalsource
Signal output
Pulsed electric field applied to sampleParticles vibrateSound waves converted to
size
© 2007 HORIBA, Ltd. All rights reserved.
Dispersion Stability
10 -2 10 -1 10 0 10 1
Diameter [um]
85nm suspension
aggregates at IEP
© 2007 HORIBA, Ltd. All rights reserved.
Laser DiffractionParticle size 0.01 – 3000 µm
•Converts scattered light to particle size distribution
•Quick, repeatable•Most common technique•Suspensions or powders
© 2007 HORIBA, Ltd. All rights reserved.
Low End SensitivitySensitivity: small particle detection30 nm silica 40 nm latex
© 2007 HORIBA, Ltd. All rights reserved.
Low End Sensitivity: CosmeticsSome (unfounded?) concernswith particles <100nmLA-950 good at determining sub 100nm particlesSoftware set to display % underany given sizeData shown left is for skin creamand TiO2 suspensionOther systems capable of thesemeasurements:
-DT-1201 acoustic spectrometer-LB-550 DLS system
© 2007 HORIBA, Ltd. All rights reserved.
Small Sample Volume (MiniFlow)
Colloidal Silica (weak scatterer)Median (D50): 35 nmSample Amount: 132 mg
Magnesium StearateMedian (D50): 9.33 μmSample Amount: 0.165 mg
Bio-degradable PolymerMedian (D50): 114 μmSample Amount: 1.29 mg
© 2007 HORIBA, Ltd. All rights reserved.
Reproducibility: Dry Powder Feeder
Direct flow ofpowder down tocell rather thanturn 90o, thenaround plastic tube
© 2007 HORIBA, Ltd. All rights reserved.
Reproducibility: Dry Powder Feeder
Automatic control of sample feed rateLA-950 monitors amount of sample supplied by the vibratory feeder. Automatic feed back control keeps constant mass flow rate of powder during measurementThis is CRITICAL
– More reproducible, robust– No ghost peaks– No cutting off results
© 2007 HORIBA, Ltd. All rights reserved.
USP<429> Software
ISO13320-1-1999 “Particle size analysis — Laser diffraction methods — Part 1: General principlesEP 2.9.31 “Laser Diffraction Measurement of Particle Size”; Lead for this monographAppearance in Pharmacopeial Forum 28, Number 3 2002Now in USP 28, NF25
– in Stage 4 of the harmonization process with the EP and the JP
© 2007 HORIBA, Ltd. All rights reserved.
USP<429> Software
USP <429>: ReplicatesMay call this reproducibility
At least three different representative samples from the same batch
CV < 10% at median d 50CV < 15% at d10 & d 90
Can double when below 10 µm
© 2007 HORIBA, Ltd. All rights reserved.
Qualification: Accuracy and Repeatability
Three independent measurements, calculate meanX50 <3% “certified range of values”X10 & X90 < 5% “certified range of values”Also check repeatabilityCOV X50 < 3%COV X10 & X90 < 5%
© 2007 HORIBA, Ltd. All rights reserved.
ISO 13320
ISO 13320 has samerequirements
Just different specs
EVERY user shouldUse these features
© 2007 HORIBA, Ltd. All rights reserved.
Conclusions
Particle size and shape are critical physical characteristics for pharmaceutical powders & suspensionsLaser diffraction most popular technique
USP<429> may drive method developmentImage analysis becoming more popular
PQRI recommendations suggest useAcoustic spectroscopy w/zeta potential useful in formulation