GCP meets GLP- The sponsor perspective

Peter van Amsterdam
Head Bioanalytics, Abbott Healthcare Products
DARQA themadag ‘When GCP meets GLP’, 30 Jan 2015, Utrecht

 The views and conclusions presented in this slide deck are
those of the author/presenter and do not necessarily reflect
the representative affiliation or company's position on the
subject.
 Exaggeration is often chosen as a style model to emphasize
the similarities or differences between various situations.
 The focus is on the bioanalytical arena
30-Jan-2015 GCP meets GLP @DARQA 2

1. Some history lessons
2. Outsourcing
3. Perspective / Perception
4. GCP in the BA lab
5. Anomalous results
6. Case studies

 1965: EEC 65/65 (reaction to Thalidomide)
o No real focus on bioanalysis
 1978: 21 CFR 58
 1982: OECD 1
o Both are General GLP guidelines (preclinical safety)
o Quality system ensure the uniformity, consistency, reliability, reproducibility,
quality, and integrity pre-clinical safety tests.
 Eighties (flowing over in the Nineties)
o Increased focus on Bioequivalence studies (including paragraphs on
bioanalytical methodology to be applied)
o EU, FDA, Australia, Canada to lead
 BMV workshop – (Crystal City-I)
o < 1990 = lack of uniformity in industry wrt validation bioanalytical methods
o Crystal City-I was first international conference with focus on Bioanalytical
method validation and sample analysis
o Resulted in Shah paper (Pharm Res. 1992;9:588-592).

Crystal City I
Shah
2001 FDA
Guidance
CC-I CC-II CC-III
Crystal City
Conferences
Conference papers
Regulatory
Guidance
Additional white
papers
CC-IV
(ISR)
CC-IV Fast
1990 2000 2010
CC III
Viswanathan
CC II
Shah (chrom.)
Miller (LBA)
DeSilva

20201960 1970 1980
1965:
65/65/EEC
1979:
US 58cfr21
1982:
OECD GLP
A B C
Anvisa RDC 899
HC removes ISR
1988:
Australian draft
CO6: 7581c
EMA draft
Anvisa update
Open letter to
FDA & EMA
GBC
formed
More countries or
regions likely to issue
Guidelines
D
E
1990 2000 2010
Thalidomide
A. scientist adopting home designed quality
systems
B. scientist shopping for inspiration in other areas
– peers, DIN, EPA,..
C. scientist regrouped around Shah paper
D. multiple countries issuing regulations of BA included in BE
guidelines
E. Industry increase meeting frequency (e.g. APA, EBF, CVG)
recommendation papers after (broad) internal discussions
EMA final
MHLW chrom
MHLW LBA
CFDA draft

A bucket full of other
adjacent regulations:
21 CFR 58 (GLP), OECD-
GLP, 21 CFR 11 (CSV),
ICH-S3A (TK), ICH-E6
(GCP), ICH M3 (MIST),
MHRA GCLP, EMA GCLP,
21 CFR 320.29 (BE), …,
…, …,

 1963 FDA drafting the GMP concept
 1964 Declaration of Helsinki
 1973 New Zealand draft GLP concept
 1978 FDA GLP final
 1978 FDA GMP final
 1981 OECD GLP
 1982 MHLW GLP
 1983 EPA GLP
 1989 Revised EPA GLP
 1989 ICH global standards for Clinical Research
 1997 21 CFR 11
 1997 Revised OECD GLP
 1997 ICH GCP
 2001 EU Clinical Trials directive
 2005 EU GCP directive
 2009 MHRA GCP for laboratories
 2009 WHO GCLP
 2012 EMA 'GCLP' reflection paper
Hmmm….
~1980 GLP
~2000 GCP
~2010 GCLP

 Within Pharma, bioanalytical departments are often part of
DMPK/Research.
o Basically a pre-clinical (GLP) area
o Science & technology driven
o High attrition rate
 Within CROs, bioanalytical departments are often core
business.
o Can cover all R to D area’s (GLP and)
o Science & technology driven
o Client focus, costs & timelines driven

2. Outsourcing
6. Case studies

By design, e.g.:
 Corporate strategy
o Science in house
o Operational external
 Flexible workforce
o Cope with peaks in workload
o No redundancies in slow periods
 Excess or shortage of money
o R&D expenditures as % of sales
Improvement of deliverables, e.g.:
 Better suited then own labs
o Better equipped
o GxP compliant
o Right location
 Faster then own labs
o Larger workforce
o More instruments
o Routine power
 Specific expertise

 CRO selection
 Oversight
 Monitoring

Sponsor
 Clinical
 QA
 Stats
 Bioanalytical
 Operations
 Trial managers
 CRAs
 Supplies
 Contracting
 Legal
 Data management
 Regulatory
 Marketing
CRO
 Project manager
 Recruitment
 CRAs
 Data management
 Pharmacokinetics
 Statistics
 Legal
 Regulatory
 QA
 Medical writers
 Archives, TMF


Central Lab
 Project manager
 Logistics
o Kits
o Samples
 Kit building
 Safety analysis
 Data basing
 Reconciliation
 QA
 Reporting
Bioanalytical Lab(s)
 Study lead
 QA
 Management
 Sample receipt
 Archives

Other
 Ethical Committee
 Safety Review Board
 Health Authorities

And
 Public disclosure
 Share holders
 Senior management
 Financial analists

Clinical sites
 Investigators
 Hospital staff
 Recruitment
 Inclusion / Exclusion
 Medical history
 Dosing
 Observations
 Sampling
 Pharmacists

2. Outsourcing
6. Case studies

In the Bioanalytical (‘GLP’) Environment
BioAnalysis Man
Knows BMV Guidelines
Knows (OECD) GLP
Knows (a bit) CSV
Knows the (relevant) SOPs
Knows his science
Can handle equipment
Can perform validations
Can run samples
Can report the results
23GCP meets GLP @DARQA30-Jan-2015

In the Research (GLP) environment
BioAnalysis Man
Knows bioanalysis
Knows (OECD) GLP
Knows a bit about DMPK
Knows a bit about the study
Takes care of the PK/PD samples
Part of team
Report appendix
24GCP meets GLP @DARQA30-Jan-2015

In the Clinical (GCP) environment
2530-Jan-2015 GCP meets GLP @DARQA
BioAnalysis Man
Knows bioanalysis
Knows some GCP
Knows a bit about DMPK
Knows a bit about the study
Takes care of the PK/PD samples
May be part of sub-team
Report appendix / part of the reported data

 In the BA lab, bioanalytical objectives
o “You can do anything you want“
• Change method, amend something, redo an experiment, ….
• (as long as you follow some SOPs and QA and management rules)
 In a GLP study
o “You can mostly do what you want“
• Change method, amend something, redo an experiment, BUT …
• (as long as you follow the SOPs, protocol, SD input/agreement and QA and
management rules)
 In a GCP study
o “You can hardly do what you want“
• NO to change of method, amend something, redo an experiment, UNLESS ..
• (and follow the SOPs, protocol, sponsor input/agreement, patient rights,
and QA and management rules, ….)

2. Outsourcing
6. Case studies

From: Principles and Practice of ICH GLP
Rebecca Luk, Pfizer (2012)

EMA:
EMA BMV: The validation of bioanalytical methods and the analysis of study samples for clinical
trials in humans should be performed following the principles of GCP. Further can be found in
the “Reflection Paper for Laboratories That Perform The Analysis Or Evaluation Of Clinical Trial
Samples.” (EMA/INS/GCP/532137/2010).
EMA BMV: The validation of bioanalytical methods used in non-clinical pharmaco toxicological
studies that are carried out in conformity with the provisions related to Good Laboratory Practice
should be performed following the Principles of GLP.
EMA BE: The bioanalytical part of bioequivalence trials should be performed in accordance with
the principles of GLP. However, as human bioanalytical studies fall outside the scope of GLP, the
sites conducting the studies are not required to be monitored as part of a national GLP
compliance programme.
Note: OECD GLP
Draft FDA: Pre-clinical adhere to GLP (21 CFR 58), Clinical adhere to 21 CFR 320.29
ANVISA: Not addressed
MHLW: Not clearly addressed

 Availability of the clinical protocol
 Central lab between clinic and BA lab
 Lab manual
 Sample roster ≠ sample tube label
 Data base (LIMS) requirements
 Timelines
 Reporting final data before issuing final report
 Who to report what to (communication plan)
 Patients and IC
 CRFs

From: Unexpected Results in a Bioanalytical Laboratory – a Safety and Compliance Issue?
Silke Luedtke, Boehringer Ingelheim (2011)

Excerpt from the MHRA "Guidance for the notification of serious breaches of
GCP or the trial protocol", released 2006, update Jan 2013
 non-compliance with GCP or the protocol
If this non-compliance has a significant impact on the integrity of trial
subjects in the UK or on the scientific value of the trial, this will constitute a
serious breach. For example, widespread and uncontrolled use of protocol
waivers affecting eligibility criteria, which leads to harm to trial subjects in
the UK or which has a significant impact on the scientific value of the trial.
Another example would be an investigator repeatedly failing to reduce or
stop the dose of an IMP in response to a trigger defined in the protocol (for
example, abnormal laboratory results).
 Examples of Serious Breaches Notified to MHRA
Investigator site failed to reduce or stop trial medication, in response to
certain laboratory parameters, as required by the protocol. This occurred
with several subjects over a one year period, despite identification by the
monitor of the first two occasions. Subjects were exposed to an increased
risk of thrombosis.

2. Outsourcing
6. Case studies

 Is the bioanalytical laboratory (CRO) able to:
o Identify significant deviations from the protocol?
o Identify misdosing?
• Overdosing
• Underdosing
• Wrong medication
o Identify differences in drug clearance (a subjects ability to
metabolise or excrete an investigational drug)?
o Identify sloppy conduct at a clinical site?
o Report in a timely manner?
o Report to the right parties involved?

 Yes, e.g. in a non-placebo controlled trial with cross-over
design some subjects had normal PK profiles for treatment
1 but hardly any profile for treatment 2
 subjects did not take medication in treatment cycle 2
 clinic did not notice as e.g. no vomiting etc was reported
 Could be a compliance issue due to misconduct

 Yes, e.g. placebo or reference drug and active treatment
mixed-up
 may not be a safety issue for subject if the dose was
approved for the trial
 Compliance issue as it demonstrates that procedures at
the clinic are prone to error

 Yes, e.g. some or all subjects show much higher plasma
levels than expected (e.g. from sponsor info or literature or
comparison to historic data)
 may be a safety issue for the subject(s)
 Could be a compliance and safety issue as it indicates
that the study is not executed according to protocol and
subjects may have been overdosed

 Yes, e.g. some or all subjects show much lower plasma
levels than expected (e.g. from sponsor info or literature or
comparison to historic data)
 may be a safety issue for the subject(s): undertreatment
of patients
 Could be a compliance and safety issue as it indicates
that the study is not executed according to protocol and
patients may have been underdosed

 Yes, but may need help from a DMPK scientist (from the
sponsor)
e.g. profile differs significantly from others, no hint for
analytical problem like matrix effect
 Could be a safety issue

 Yes, e.g. sample mislabeling, wrong collection tubes, wrong
matrix, large volume variations, strongly hemolyzed, …
 may not be a safety issue
 sloppiness/errors in sample collection and handling can
be indicative of bigger compliance issues
 Compliance issue as it demonstrates that procedures at
the clinic are prone to error

 Not always but very often, i.e. for clinical trials were the
bioanalyst is unblinded and sample analysis is performed
while the trial is still ongoing

 Yes, if there is a communication plan
 Maybe yes, with known contact details of key (sponsor)
people
 Likely not in cases of ‘thrown over the fence’

2. Outsourcing
6. Case studies

In een multiple ascending dose study met een NCE worden plasma
monsters direct na de 1e en na de 7e dag geanalyseerd door een
CRO BA-lab. Gebaseerd op de PK (plasma spiegels) en de klinische
observaties wordt door de sponsor de beslissing genomen wat de
volgende (hogere) dosis gaat worden of dat de studie gestopt moet
worden.
Wanneer de sponsor de resultaten van de eerste (laagste) dosering
krijgt valt op dat er 1 subject is met 5-10x hogere plasma spiegels
dan de overige 5 subjects op drug.
 Hebben we een probleem?
 Wat had het BA lab (eerder) moeten doen?
 Wat moet de sponsor doen?
 Kan de volgende hogere dosering volgens plan plaatsvinden?

Om een op tijd belangrijke beslissing te kunnen nemen in een
fase III programma is het gewenst dat de resultaten van een
zeer grote fase I dubbelblinde DDI studie halverwege de
bioanalyse pharmacokinetisch verwerkt en geëvalueerd
worden. Hiervoor wordt een derde onafhankelijk partij
gecontracteerd zodat de sponsor en de onderzoeker en
biometrie van de CRO blind kunnen blijven. Het lab stuurt de
data en randomisatielijst naar de 3e partij en de sponsor …
 Wat moet het BA lab doen?
 Wat moet de 3e partij doen?

Bij de pharmacokinetische verwerking en statische analyse
van een dubbelblinde PK studie blijken 3 subjects zeer
vreemde PK curves te laten zien. De sponsor wordt gevraagd
wat met deze subjects te doen. De bioanalyticus van de
sponsor ontdekt dat er bij deze 3 subjects zeer waarschijnlijk
de data van de QCs tussen de monster data is gekomen
waardoor ‘alles is opgeschoven’
Curve 1: QC-HI ipv t=0, QC-ME ipv t=1, QC-LO ipv t=2, t=0 ipv t=3 …
Curve 2: t=n-21 ipv t=02, t=n-11 ipv t=12, t1=n1 ipv t=22, t=02 ipv t=32 …
 Wat had het BA lab (eerder) moeten doen?

Na 3 succesvolle fase I studies wordt een bioanalyse methode
enigszins aangepast: ander merk HPLC kolom en wordt bij de
MS/MS overgegaan van API-CI naar ESI (zachtere ionisatie).
De methode wordt prachtig gevalideerd en toegepast bij studie
nummer 4. De chromatogrammen van vrijwel alle subject
samples laten onverwacht in meer of mindere mate een
schouder zien op de analyte piek.
 Wat moet het BA lab doen?

 DARQA for giving me the opportunity to present at their
meeting.
 The regulators and inspectors for stimulating us to
continuously improve our work
 Philip Timmerman (JnJ) for some of the ‘History’ slides
 Silke Lüdtke (BI) for some of the ‘Anomalous’ slides
 You, for your attention

GCP meets GLP- The sponsor perspective

More Related Content

What's hot (18)

Viewers also liked (6)

Similar to GCP meets GLP- The sponsor perspective (20)

More from Peter van Amsterdam (10)

Recently uploaded (20)

GCP meets GLP- The sponsor perspective