CAG EMEAI FY14 | Agilent Restricted | Page 2
Contents
• EU GMP – A Tale of Change (Vs. CFR)……………………………………….....
• FDA Warning Letters – Data Gathering…………………..…
• Instrument Qualification - GAMP / <1058> ……….…
• Quality by Design………………………………………………..……………….
• Data Integrity……………………………………………………..………………………
• FDA Warning Letters – a Deeper Dive………………………..
• How Do You Know ?..............................................................................
• Questions……………………………………………………………………………………….
• Appendix – More FDA Warning Letters………………………………………..
CAG EMEAI FY14 | Agilent Restricted | Page 3
EU GMP – Current Update Status (Vs. CFR Vs. FDA Guidance)
Section Title Issue Date Current Proposed
http://ec.europa.eu/health/documents/eudralex/vol-4/index_en.htm
FDA Warning Letters
CAG EMEAI FY14 | Agilent Restricted | Page 5
EudraGMDP – European Non-Compliance
http://eudragmdp.ema.europa.eu/inspections/gmpc/searchGMPNonCompliance.do
1 Listed for Spain:
CAG EMEAI FY14 | Agilent Restricted | Page 6
FDA Warning Letter Page – Recent Warning Letters
A B
C
D E
F G
H
I J
http://www.fda.gov/ICECI/EnforcementActions/WarningLetters/default.htm
Recent FDA Warning Letters
CAG EMEAI FY14 | Agilent Restricted | Page 7
What an FDA Warning Letter Might Contain
Unique Reference – e.g. “ucm1048179”
Audit Dates
Location
References to 21 CFR and / or USP requirements
Common Words:
Adulterated Misbranded Acknowledgement of Response:
We acknowledge receipt of your reply, but….. Lacks sufficient detail / corrective actions Was not considered as > 15 days from WL
Details of non-conformance
Expression of audit / response concerns
Expression of WL response requirements:
Risk-Assessment for US distributed products Copies of anything promised (procedures, validation) Counter arguments (if difference of opinion)
A “ucm” number is assigned to each FDA web page…. Google will “find” it, if “links broken”.
How many audit days (but not how many people), days to WL
Auditors Carry GPS
Look Up [+ EIR Data]
FD&C Adulterated Drug Definition (Chapter 9, section 351)
FD&C Misbranded Drug Definition (Chapter 9, section 352)
Often Basic Errors…….
Analysis of………
“Tone” of response……
Consequences of what they found, how you responded, what you promised…..
CAG EMEAI FY14 | Agilent Restricted | Page 8
Example Warning Letter – “ucm1048179”
“…a live rodent that you brushed off your Shoulder onto the floor, and then kicked under a pallet….”
Audit Days
[ 15 “Working” Days ] Days to Issue
[ 74 Days, 53 “Working Days” ]
CFR Look Up
[ “Google CFR” ]
“Tone”
Non-Conformance
CAG EMEAI FY14 | Agilent Restricted | Page 9
FDA Warning Letter Trends
725
535 538
471
445 474
673
1720
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
?
Trend Downwards – In Terms of Number of Warning Letters
Trend Upwards !
Margaret Hamburg Joined FDA
( 21st Commissioner, May 18th 2009 )
http://www.fda.gov/ICECI/EnforcementActions/ucm247813.htm
?
725 535 538 471 445 474 673
1720
4882
6760
0
1000
2000
3000
4000
5000
6000
7000
8000
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
FDA Audits: 97 % of US Facilities [pharmaceutical, every 2 years]
FDA Audits: 7 % of Non US Facilities [pharmaceutical every year]
[Source: Journal of Validation Technology; D. W. Vincent, Winter 2012 edition]
US Audits: Down 40 %
Overseas Audits: Up 30 %
Mutual Recognition Audits [Tougher “National” Audits]
Move From “Fixed” Time Triggers To RISK Based Triggers…… Whistle Blowers…. [other regulators – UK MHRA…… ]
976 [2013] down to 591 [2014]
604 [2013] up to 843 [2014]
Source: Pharmaceutical Manufacturing.com
> 10 Times Increase
CAG EMEAI FY14 | Agilent Restricted | Page 10
How Many ? How Many Times are Techniques Cited ?
CAG EMEAI FY14 | Agilent Restricted | Page 11
FDA Warning Letters – A Valuable Resource / Tool…
Country Number Austria 6 Belgium 15 China 143 Denmark 16 France 31 Germany 75 Hungary 2 India 48 Ireland 12 Israel 14 Italy 42 Netherlands 13 Spain 23 Switzerland 27 United Kingdom 58
By Country
Warning Letters by Country
Updated: 31/Mar./2014
0
20
40
60
80
100
120
140
160
6 15
143
16
31
75
2
48
12 14
42
13 23 28
58
CAG EMEAI FY14 | Agilent Restricted | Page 12
FDA Warning Letters – 5 Year Trend…
By Country
0
5
10
15
20
25
30
35
40
3
6 2
3
20132012201120102009
Warning Letters by Country – Last 5 Years
CAG EMEAI FY14 | Agilent Restricted | Page 13
2012 – Last Full Year, Warning Letters as % of FDA Drug Audits
0
10
20
30
40
50
60
70
0 0
61
42
23 19
33
3
14
0
21
50
40
8
24
Spain = 6 Warning Letters, 15 Drug Inspections (= 40 % !)
Drug Audits
How Good Are Pharmaceutical Companies in Spain ?
Instrument Qualification USP <1058> and GAMP
CAG EMEAI FY14 | Agilent Restricted | Page 15
All instruments contain some level of software (unless only electro-mechanical / mechanical)
All instruments and Software must be suitable for intended use – in GXP work
Qualify the Instrument
Validate the Software
• For Simple Instruments – these are combined
• For Complex Systems – these are independent
• For Simple Instruments – calibrate (e.g. pH Meter)
• For Complex Instruments– qualify (e.g. HPLC)
• For Complex Systems – validate (e.g. CDS)
Qualify the Instrument Can’t qualify without software control.
USP <1058>
Validate the Software Can’t validate the software without the instrument.
GAMP
Hardware & Software
CAG EMEAI FY14 | Agilent Restricted | Page 16
USP <1058> - Implemented in August 2008
Risk Based Categorisation: A, B, C:
A – Verify by Observation (e.g. Stirrer) - No calibration or Measurement
B – Verify by Calibration - Calibrate by SOP / Document (e.g. pH Meter)
C – Qualify (e.g. HPLC) - Complex Systems - Verify by Full Qualification
Strength of <1058>
Addressed in <1058> Update
CAG EMEAI FY14 | Agilent Restricted | Page 17
Summary of USP <1058>
Components of Data Quality / Integrity - Principles Apply to All Labs.
Definitions • Roles:
• User • QA • Supplier
• Responsibilities • DQ / IQ / OQ / PQ
Flexibility • What is done in OQ / PQ
Risk Management by • Categorisation: A / B / C
Areas of <1058>
Simplification of the Life Cycle Process:
A B C Examples…
Observe Calibrate Qualify
CAG EMEAI FY14 | Agilent Restricted | Page 18
GAMP – Software Categories
GAMP Defines Different Software Categories:
1
3
4
5
Infrastructure
From PIC/S Guidance
Non-Configured Software
Configured Software (COTS)
Bespoke Software
GAMP Good Practice Guide
A Risk-Based Approach to GxP Compliant Laboratory Computerized Systems
USP <1058> Analytical Instrument Qualification
Firmware – Group B
GAMP 5
CAG EMEAI FY14 | Agilent Restricted | Page 19
GAMP GPG – Risk-Based Approach to GXP Compliant Laboratory Computerized Systems
Appendices: 1. Determining System Impact 2. Testing Priorities 3. Supplier Assessment Scheme Glossary References
Appendices: 1. Categories of Software 2. System Description 3. Data Integrity 4. Simple Systems 5. Medium Systems 6. Complex Systems 7. System Interface Considerations 8. Robotic Systems 9. Defining Electronic Records and Raw Data 10. Security Management for Laboratory Computerized Systems 11. Supplier Documentation and Services
Aligned With GAMP 4 Risk Management by Instrument Categorisation.
Aligned With GAMP 5 Risk Assessment...... No Formal Categorisation ! (Fixed Categories VS. Risk Based Thinking)
Expansion and Examples....
7 Categories Examples:
• A - Sonicator • B - pH Meter • C - Key Pad HPLC • D - PC HPLC • E - NMR • F - Spread sheet • G - Bespoke
Available From ISPE
Contents: 19 Sections - Process Based Approach (Table of Contents Available on Line)
Contents: 7 Sections - Risk Based Model (Table of Contents Available on Line)
GPG Edition 2
GPG Edition 1
CAG EMEAI FY14 | Agilent Restricted | Page 20
Common Goals Good Evolution
• GAMP 5 for Lab.
• Worked Examples
• Inclusion of <1058> Principles
• Life Cycle Mgt.
• Harmonisation
• Draft New <1058> With USP Panel
• Integrates <1058> & GAMP
GAMP 5 - 2008
AIQ <1058> - 2008
ISPE
USP
May 1987
FDA Process
Validation Guide Lines
• Simplify Process
• Risk by Categorisation - Software (GAMP) - Hardware (<1058>)
• Roles and Responsibilities
• Maximise use of Supplier Documentation
• 5 pages !
• Easy to Follow
• A and B Simple
• Flexible
• Very Comprehensive
• Great For Custom Software
GAMP 4 - 2001
GAMP 5 Good Practice
Guide
AIQ <1058> Draft 2013
GAMP and USP <1058>
CAG EMEAI FY14 | Agilent Restricted | Page 21
Pharmaceutical Engineering Publication
CAG EMEAI FY14 | Agilent Restricted | Page 22
Proposed <1058> Classification – Awaiting Feedback
Bob McDowall & Chris Burgess Publication
Quality by Design
CAG EMEAI FY14 | Agilent Restricted | Page 24
Example – Laboratory Lean Sigma Application How Many Days to Test
Product Batches of Material Before
Mean Workflow Optimisation
After
Mean
Impressive Results:
• 50 % Reduction In Tests Times are “Common” • Improved Testing Control
Moderately Robust HPLC Method
• Sample Based Testing
• Product Team [analysts perform all tests]
• Optimised for Control / Speed
Less-Robust HPLC Method
• Test Based Testing
• HPLC Team [HPLC “specialists”]
• Designed to Reduce Problems
CAG EMEAI FY14 | Agilent Restricted | Page 25
What is Quality By Design ?
Pharmaceutical Manufacturing
Process
Inputs
Variable Quality Raw Materials
• Variable Process • Validated Design • Registered
Fixed Output
QbD
Design Space
Historical Pharmaceutical
Process Validation
Quality by Design Pharmaceutical
Process Validation
Inputs
Pharmaceutical Manufacturing
Process
Output
• Fixed Process • Validated • Registered • Can’t be Changed
Variable Quality Raw Materials
Variable Quality Drugs
Specification
Pharmaceutical Manufacturing 2.5
• Can Change – Within Design Space
CAG EMEAI FY14 | Agilent Restricted | Page 26
Wider Application of QbD Principles…..
Application of QbD To Analytical Method Development & Validation
Pharmacopeia Forum
<1224> Analytical Method Transfer <1225> Analytical Method Validation <1226> Analytical Method Verification
<220> Basic Requirements for Validation
<1220> Lifecycle Management of Analytical Procedures
CAG EMEAI FY14 | Agilent Restricted | Page 27
Example – Why QbD is Needed…
Significant QC Problems With Method Variable “Resolution”
CAG EMEAI FY14 | Agilent Restricted | Page 28
3D Graph Called Response Surface
Example - Why Analytical QbD is Needed….. Method Conditions Mobile Phase pH 4.5 Column Oven Temp. 35oC
For This Method: • Temperature – Minimal Effect • pH – Critical Effect – on Resolution
Change Method Conditions pH & Temperature
Measure Effect
Here QbD used in Problem Solving If Can’t Change Method CONTROL pH
With QbD Design A DIFFERENT pH Would Have been Used pH – 3 [flat part of response]
CAG EMEAI FY14 | Agilent Restricted | Page 29
Range of Use…………(Best Practice to Qualify Range of Use)
Wavelength (nm)
Oven Temperature (oC)
Flow Rate (mL/Min.)
245 nm
45 oC 1 mL/min.
Min. Flow
Max. Flow
Low
est
Wav
elen
gth
Hig
hest
W
avel
engt
h
Highest Temp.
Lowest Temp.
Range of Use
Qualification should “bracket” THE RANGE OF USE (good science & easier to “defend” users may have to do extra qualification……)
Why Range of Use is Important ?
1 ml/min. (Pump)
45 oC (Oven)
245 nm (Detector)
Customer HPLC Method
Qualification Space
CAG EMEAI FY14 | Agilent Restricted | Page 30
Design Space = Qualification Space
EQP (Equipment Qualification Plan)
Instrument Qualification
Analytical Instrument
(Range of Use)
QTPP = Quality Target Product Profile ATP = Analytical Target Profile MODR = Method Operable Design Region EQP = Equipment Qualification Plan EQR = Equipment Qualification Report
Pharmaceutical QbD
Analytical QbD
(Process Operability)
Sharmista Chatterjee Jan. 2013
CAG EMEAI FY14 | Agilent Restricted | Page 31
Example – Range of Use Warning Letter The qualification – MUST BE – representative, of how the instrument is used…..
Failure of your quality control unit/laboratory to ensure that analytical instrumentation and test equipment used to assure the quality of your APIs has been appropriately qualified and calibrated for their intended use.
Specifically, your firm has failed to conduct adequate qualifications of your analytical instruments and test equipment. For example, the residual solvent method used to test (b)(4) API has an initial starting gas chromatograph (GC) oven temperature of (b)(4). Your firm’s current qualification of the GC oven temperature does not include temperatures below 100˚C. http://www.fda.gov/ICECI/EnforcementActions/WarningLetters/2010/ucm236841.htm
Qualify the Range of Use
Data Integrity
CAG EMEAI FY14 | Agilent Restricted | Page 33
What Triggered The Current FDA Focus ?
• Scientific “Validity” • ICH Q9, Q10, Q11 support
• FDA Part 11 “Add On” Audits • Jan. 2011
• Data Integrity has always been an FDA focus.
• 1997 Implementation • 2003 Guidance
Science Based GMP Decisions (e.g. cGMP for 21st Century)
Fundamental Data Integrity
21 CFR Part 11
Part 11 Add on Audits
FDA DATA
INTEGRITY
FOCUS • PAI
• Consent Decree • Data Integrity Training…
CAG EMEAI FY14 | Agilent Restricted | Page 34
So, What is Data Integrity ? “The Degree to which a collection of data is complete, consistent and accurate” * *(FDA Glossary of Computer System Software Development Terminology (9/95)
Data: (Collins Complete and Unabridged English Dictionary)
PLURAL NOUN, a series of observations, measurements, facts, information.
Integrity: (Collins Complete and Unabridged English Dictionary)
NOUN, adherence to moral principles, honesty; or the quality of being unimpaired; soundness, or unity; wholeness
The definition can influence the SCOPE that people Interpret Data Integrity:
Narrow:- Specific to 21 CFR Part 11
Wide:- Anything that causes a concern with the validity of the data
This means there is risk of confusion and poor preparation for defending YOUR Data Integrity in an audit…… (…effects your audit preparation)
THE FDA will base their audit on what they find – NOT YOUR DEFINITION !
CAG EMEAI FY14 | Agilent Restricted | Page 35
One Slide History – Data Integrity FDA Warning Letters: HPLC & GC warning letters show high level of
Data Integrity related issues (so not new)
History of Generic Drug Issues
2 Key Areas to Note: Generic Drug scandal in the late 80’s, early 90’s (falsified data, bribes to FDA .. Etc.).
Food Drug & Cosmetics Act Amended by: - Generic Drug Enforcement Act (GDEA)
Debarment – of Individuals and Companies
Clinical ANDA Data – Clinical Data Investigations……
ISPE Conference on Risk-Based Systems Validation - New Jersey, USA, 12-13th September 2011
FDA Data Integrity Auditing Training…… (AGILENT) - Dublin, Ireland, 9-10th July 2012 Train the Trainer Data Integrity Auditing - Dublin, Ireland, 14-15th May 2014
Monica Cahilly (Now Trained ALL FDA Auditors – in Data Investigations)
Reference Publication [Hyperlink]
CPG 7346.832 [Hyperlink]
FDA PAI – 3 Focus Areas: - Objective 3, Data Integrity Audit
CAG EMEAI FY14 | Agilent Restricted | Page 36
Is Paper Good Enough ?
Reference Publication [Hyperlink]
http://www.pharmtech.com/pharmtech/IT/Data-Integrity-in-the-Analytical-Laboratory/ArticleStandard/Article/detail/842036
• Paper • Hybrid • Electronic
Need to implement data integrity improvements strategically, as many laboratories:
• Still define paper as raw data • Use Excel to calculate results • May have “imbedded” compliance risks – Data Mapping / Flow / Fish Bone
CAG EMEAI FY14 | Agilent Restricted | Page 37
Who Next ?
The UK MHRA is setting an expectation that pharmaceutical manufacturers, importers and contract laboratories, as part of their self-inspection programme must review the effectiveness of their governance systems to ensure data integrity and traceability.
This aspect will be covered during inspections from the start of 2014, when reviewing the adequacy of self inspection programmes in accordance with Chapter 9 of EU GMP.
It is also expected that in addition to having their own governance systems, companies outsourcing activities should verify the adequacy of comparable systems at the contract acceptor.
The MHRA invites companies that identify data integrity issues to contact: [email protected]
16 December 2013
Unknown – the situation in other EU inspectorates – Spain ?
FDA Warning Letters HPLC and GC
CAG EMEAI FY14 | Agilent Restricted | Page 39
Examples – Search on “HPLC” - 68
A B C D
1. Enter Search Criteria (e.g. HPLC)
2. Select Sort By (Default is Alphabetical)
3. Select Sort Criteria (e.g. Descending)
4. Select Go
5. Select ALL
6. Select Warning Letter (to open)
Use “Ctrl F” to FIND KEY WORDS
(Copy & Paste into Excel)
CAG EMEAI FY14 | Agilent Restricted | Page 40
FDA “HPLC” Warning Letters…
0
1
2
3
4
5
6
7
8
9
10
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
6
10
6
9
5 5
4
3
2
3
4
5 5
13 years on FDA Warning Letter Page: 46 % - last 4 Years
Key Points: • HPLC has the Highest FDA Focus of ANY Analytical Technique • High focus area in recent years
CAG EMEAI FY14 | Agilent Restricted | Page 41
FDA – Takes 10 x Longer
To Issue Warning Letters
Than Companies Have to Respond
FDA Warning Letters That Mention HPLC or GC Days to Issue Warning Letter [ 84 ]
Days to Issue Warning Letter [HPLC Only]
Audit Duration [ (HPLC & GC) = 84 ]
CAG EMEAI FY14 | Agilent Restricted | Page 42
0 1 2 3 4 5 6 7
Data Integrity
Calibration
Method
N/A
No GC
7
4
2
3
1
Review of FDA Warning Letters About GC Most of these categories cause concern over the validity of the analytical results….
Read Evaluate Categorise Knowledge
GC Methods
Calibration
Data Integrity: Now Biggest Category
Capability of the Method
Would trust this data ?
GC in Name, GC in SOP…..
GC Warning Letters – Most are relatively Recent !
CAG EMEAI FY14 | Agilent Restricted | Page 43
Primary Cause – HPLC Warning Letters
Calibration: Biggest Single Category
Decisions & Justification
Using the Method
Data Integrity
Capability of the Method Poor HPLC Use / Training
HPLC Methods
Would trust this data ?
0 2 4 6 8 10 12
Calibration Defeciency
No Calibration
N/A
OOS Problems
Problems With System Suitability
Chromatography - Unknown Peak
HPLC Method Validation
Defecient HPLC Method
Data Integrity - Data Deletion
Data Integrity - Incomplete Data
Data Integrity - Security
Data Integrity - Fraud
Stability Testing
Sterlity
Impact of Instrument Repair / Failure
9
8
8
7
2
2
11
5
1
3
4
3
1
1
1
HPLC Methods
Would trust this data ?
Poor HPLC Use / Training
CAG EMEAI FY14 | Agilent Restricted | Page 44
Your Equipment - How DO You Know ? How Do You Know If Your Equipment Is Set up Correctly ?
How Do You Know If Your Equipment Is Working Correctly ?
Measure How it Is Setup……
Mechanical Qualification
USP Performance Verification
Test
ASTM – E2503-07 FDA – DPA-LOP.002
USP – <711>
Use a method sensitive to Variation in the instrument performance……
Dissolution Problems are one of the Top 10 reasons for Pharmaceutical Product Recall ! (Cindy Bhuse, FDA, AAPS, 2011)
CAG EMEAI FY14 | Agilent Restricted | Page 45
Dissolution and Hydrodynamics
Dissolution Vessel (In the….)
Dissolution Medium (Which contains….)
Paddle (Rotates)
Drug Product (Particles)
Drug Particle
Dissolution Medium - Sweeping past particles
Anything that influences the flow in the vessel will Change the results !
• Scratches – on paddle or vessel • Seams – if not certified vessel • Vibration • Air bubbles (if not degassed) • Bent shaft / Wobble…….. • …….Etc.
To PASS the PVT – The Instrument Must be Working Well !
CAG EMEAI FY14 | Agilent Restricted | Page 46
Dissolution “Snapshot”….
Dissolution Qualification
49% - USP PVT
29 % - Both (PVT + MQ)
20 % - MQ Only (FDA / ASTM)
2 % - In House Methods
http://www.dissolutiontech.com/DTresour/201105Articles/DT201105_A06.pdf May 2011
CAG EMEAI FY14 | Agilent Restricted | Page 47
Questions ?
Appendix
Additional Information
Word Warning - Deeper Dive
Additional Information
CAG EMEAI FY14 | Agilent Restricted | Page 50
HPLC Analysis – % Included in Warning Letter
71 % of HPLC letters Include Reference to Stability Testing
0 10 20 30 40 50 60 70 80
Calibration
Training
OOS
SSC
Method Validation
Defecient Methods
Data Integrity - Security
Data Integrity - Data Deletion
Data Integrity - Incomplete Records
Reaccurance
External Consultant Required
Data Review
Complete Data
Poor Response / CAPA
Stability
41
27
51
27
43
17
27
13
43
30
16
27
43
43
71
CAG EMEAI FY14 | Agilent Restricted | Page 51
Detailed Analysis – HPLC Warning Letters - Stability
Solution Stability
Not Stability Indicating
Storage Temperature
Storage Excursion
Validation of
Deficiencies
Deviations
Specification Failure
Storage Not Supported
Calculation of Results
Stability Testing
Facilities
Protocol
Stability Results
37 %
16 %
38 %
38 %
17 %
78 %
50 %
50 %
40 %
67 %
21 %
54 %
29 %
21 %
71 % of HPLC warning letters
Include Reference to Stability
CAG EMEAI FY14 | Agilent Restricted | Page 52
GC Analysis – % Included in Warning Letter
79 % of GC letters Include Reference to aspect of Data Integrity
0 10 20 30 40 50 60 70 80
Calibration
Training
OOS
SSC
Methods
Data Integrity
Reaccurance
External Consultant Required
Complete Data
Poor Response / CAPA
Stability
50
21
36
29
43
79
30
7
36
79
36
0 20 40 60 80
Data Integrity
Security
Manufacturing
Data Change
Data Deletion
Incomplete Records
Audit Trail
79
21
21
7
7
50
14
GC Data Integrity Review (%)
CAG EMEAI FY14 | Agilent Restricted | Page 53
Proposed <1058> Sub Classification (McDowall & Burgess)
Group A: Apparatus Group B: Instruments • Type 1: Firmware
• Type 2: Firmware plus built in calculations
• Type 3: Firmware plus user defined programs
Group C: Systems • Type 1: Low complexity system - instrument plus non-configured software
• Type 2: Medium complexity system - instrument plus configured software
• Type 3: High complexity system - Instrument plus configured software & macros