the pharmaceutical excipients database hosted on · pdf filethe pharmaceutical excipients...
TRANSCRIPT
The Pharmaceutical Excipients Database Hosted on PharmaHUB
2014 GPhA Fall Technical Conference
October 27-29, 2014 North Bethesda, MD , MD
Stephen W. Hoag, Ph.D. University of Maryland, Baltimore
School of Pharmacy Phone 410-706-6865
Email: [email protected]
Disclaimer
• This presentation contains a summary of the opinion and perspective from GPhA member industry representatives on the topic of The Pharmaceutical Excipients Database Hosted on PharmaHUB.
• This presentation does not necessarily represent the opinion of the presenter nor its employers.
Outline • Introduce NIPTE/FDA database project
– Identify Critical Material Attributes
– Cataloging Excipient Variability
• Database structure and usage – PharmaHub
– Database views
– Applications
• Future of excipient database – Form for collaboration
Dosage Form Variability
API
Dosage
Forms
Excipients
Manufacturing
Processing
Conditions
+
+
+
Spec. range
Excipient Properties Excipients are:
– Highly variable materials
– Have very different properties and performance attributes as compared to drug substances
– For example
Consider an excipient NF monographs vs API USP monograph
– To understand excipients, you need to think like an excipient
Key issues – Identifying Critical Material Attributes (CMA)
Most unit operations only empirically understood
– Understand or identify excipient property variability
If develop a product that is on the edge of failure
– Excipient variability → unexpected failure post approval
Database can help to solve these key issues
Example Ciprofloxacin in USP » Ciprofloxacin Hydrochloride
contains not less than 98.0 percent
and not more than 102.0 percent
of C17H18FN3O3·HCl, calculated
on the anhydrous basis.
Identification—
A: Infrared Absorption 197K.
B: Thin-layer chromatographic plate.
C: A solution of it responds to the tests for Chloride 191.
Chromatographic purity—
Calculate the percentage of each impurity peak in the chromatogram obtained
from the Assay preparation taken by the formula: 100(ri / rt)
in which ri is the response of each impurity peak; and rt is the sum of the
responses of all the peaks: not more than 0.2% of ciprofloxacin ethylenediamine
analog or of any other individual impurity peak is found; and the sum of all the
impurity peaks is not more than 0.5%.
Figure Source USP32-NF27
Excipients in NF Monographs » Anhydrous Lactose is:
O- -D-galactopyranosyl-(1 4)- -D-glucopyranose ( -
lactose) or a mixture of O- -d-galactopyranosyl-(1 4)-
-d-glucopyranose and O- -D-galactopyranosyl-(1 4)-
-D-glucopyranose ( -lactose).
i.e., - lactose or a mixture of - lactose and -lactose
Labeling—
Where the labeling indicates the relative quantities of
alpha and beta lactose
» Lactose Monohydrate is:
a natural disaccharide, obtained from milk, which consists of one glucose
and one galactose moiety. [note—Lactose Monohydrate may be modified
as to its physical characteristics. It may contain varying proportions of
amorphous lactose.]
Labeling—
Where the labeling states the particle size distribution, it also indicates the d10,
d50, and d90 values and the range for each. For modified Lactose
Monohydrate, also label it to indicate the method of modification. Figure Source USP32-NF27
Which Raw Material Inputs → CQA
Critical Quality Attributes
• Hardness
• Disintegration time
• Dissolution
• CU/mixing
• Stability
Excipient Properties
• Crystal form
• Particle size
• Bulk density
• Etc.
Relationships
Not well
understood
Raw Data Hidden Info &
Patterns
Data Mining
Data Mining
Raw
Data
Def: Discovery of previously unknown information and
patterns from set of data
Stage 1
Define
problem or
subject of
study
Stage 2
Select data
Visualize Data
Remove noise,
& Irrelevant
Data
Stage 3
Data
preprocessing
Stage 4
Data Analysis
Data Mining
Analysis of
Results
Assimilation
of Knowledge
Sources of Excipient Variability Intentional (designed into excipient)
– i.e., different grades and vendors
– Manufactures produce to improve performance
– Many examples
MCC, HPMC, lactose, et al.
– Important to understand for:
Optimizing a formulation performance, in vitro and during production
Make regulatory decisions, e.g., excipient substitution and change control
Random
– Lot-to-lot variation
Introduced from all factors disused previously
– Within a lot variation (container-to-container)
During production process parameters can drift
– Especially continuous process where a lot maybe several
days of production
– Important to understand: so can produce robust formulations
PH-200
PH-102 SCG
PH-102
PH-101
PH-105
PH-103 PH-113
PH-112
Decrease moisture content
NMT
1.5%
NMT
2%
NMT
3%
NMT
5%
Particle size
180
um
150
um
100
um
50
um
20
um
PH-302
PH-301
0.3g/cc
0.4g/cc
Loose bulk density
NMT= not more than
FMC
Intentional Variation—Grades of MCC
Intentional Variability– Manufacturers of MCC
Manufactures Grades Particle Size, µm Moisture, % Loose Bulk Density,
g/cc
FMC Avicel PH101 50 3.0-5.0 0.26-0.31
JRS Vivapur 101
65 -- 0.26-0.31
Emcocel 50M 0.25-0.37
AKC
PH-101
50 2.0-6.0
0.22
UF-711 0.21
KG-802 0.12
KG-1000 0.29
FMC Avicel PH-102 100 3.0-5.0 0.28-0.33
JRS Vivapur 102
100 -- 0.28-0.33
Emcocel 90M 0.25-0.37
AKC PH-102 90 2.0-6.0 0.30
FMC = FMC BIOPOLYMERS
AKC = Asahi Kasei Corporation
JRS = J Rettenmaier & Söhne GmbH and Co.KG
Tracking of Random Variability Track random variability overtime and plot in histogram
Excipients Database
Where to Host the Database ?
pharmaHUB: a cyber infrastructure developed at Purdue University
to support digital scholarship, dissemination, collaboration and outreach for the pharmaceutical engineering community.
Analysis Tools make decisions and solve problems
Derivations & Plots of Measurements histograms, yield loci, powder flow function,
frequency & cumulative distributions
Measurements raw data, chemical and test descriptions
Catalogs excipients, vendors, products, lots, test methods,
equipment, properties
Web Interface
Database Structure
Excipients
- MCC
Products
- PH101
Lots
- Lot # 1
Properties
- particle size
Test method
- Light scattering
Equipment
- Malvern
← USP compendial category
← Actual product on the market
A Publicly Available Database
Excipients
Database
A Publicly Available Database
Excipients
Database
A Publicly Available Database
Excipients
Database
A Publicly Available Database
Excipients
Database
The Mechanical Property Group
• Particle Size Distribution
• True Density
• Bulk Density
• Cohesiveness
• Elastic Modulus
• Compactibility
• Brittleness
21
12
Effective Yield Locus’
’
’
12
Effective Yield Locus’
’
’
D
P0
s
Slip Region
h
Nip Regionx
D
P0
s
Slip Region
h
Nip Regionx
-50
-30
-10
10
30
50
70
90
110
130
150
-25.0 -20.0 -15.0 -10.0 -5.0 0.0
( ° )
(M
Pa
)
= 40 ° D = 8 inches
45 ° s/D = 0.029
K = 10 P0 = 0.1 psi
h = 45 ° 0 = 0.28 psi
w = 4 inches = 11.5 °
Calculated
Johanson
Calculated Schönert
=0.3
= 40 ° D = 8 inches
45 ° s/D = 0.029
K = 10 P0 = 0.1 psi
h = 45 ° 0 = 0.28 psi
w = 4 inches = 11.5 °
Calculated
Johanson
Calculated Schönert
=0.3
log
log
1
2
K
1log
log
1
2
K
1
A.Hlinak, NIPTE-NIST Meeting, April, 2005
Development Lots
-2
-1
0
1
2
3
4
5
-5 -4 -3 -2 -1 0 1 2 3
Prin
cip
al C
om
po
ne
nt 5
Principal Component 3SIMCA-P 11 - 4/14/2010 10:17:54 AM
Domain of Prior
Experience
7143C
7545C
7513C
• Based upon random selection of lots:
• Domain of prior Experience: typical production lots
• 3 lots used for development can tell if develop formulation using
typical material
Source: Kindly supplied by Joe Kushner
Inactive Ingredient Database
Provide little ability to interact with data
Supplier’s Data
User’s Data
Data Mining Database
User Friendly
Web based
Interface
Targeted
Experiments to
Support Database
Relational
Modeling Literature Data
Database
Management
Tools
Acknowledgements
FDA & NIPTE for
funding
Ting Wang
Ann Christine Catlin
Sumudinie Fernando
Sudheera R. Fernando
Carl Wassgren
Kristine Alston
Linas Mockus
Paribir Basu
Individuals with
invaluable advice
FMC, Brian Carlin
Pfizer, Bruno Hancock
Roquette, Leon Zhou
Colorcon Dave Schoneker
IPEC
And many others