Transcript WHO Collaborating Centre for International Drug Monitoring

WHO Collaborating Centre for
International Drug Monitoring
the Uppsala Monitoring Centre
WHO Drug Monitoring Programme
Founding Members 1968
WHO Collaborating Centre
the Uppsala Monitoring Centre
•
•
•
•
established as a foundation 1978
based on agreement Sweden - WHO
international administrative board
WHO Headquarters responsible for policy
WHO International Drug Monitoring Programme
2004
Flow of information
Medical practice
National Centres
Manufacturers
WHO Headquarters
WHO Collaborating
Centre (UMC)
Wider scope of
pharmacovigilance
• Adverse reactions
– Type A
– Type B
• Lack of effect
•
•
•
•
– counterfeiting
– resistance
– interaction
Quality problem
Dependence and abuse
Poisoning
Medical error
PHARMACOVIGILANCE
WHO definition
The science and activities relating to
the detection, assessment,
understanding and prevention of
adverse effects or any other drugrelated problem
Expansion of WHO Drug Monitoring
Programme
• Database enriched – not only problems
seen in industrialized countries
–
–
–
–
Drugs against tropical diseases
Better monitoring of traditional medicines
Another kind of co-morbidity
Effects of malnutrition
Developments in ICH Countries
• Strong focus on regulatory
requirements/actions
– Serious, unlabelled reactions to new drugs
– Public health perspective in the background
– Risk management
• Industry stewardship has major role
• Standardized format for exchange of
case information (E2b)
– MedDRA
Cumulative Number of Reports per Year in Vigibase
April 2004
3 500 000
3 000 000
# Reports
2 500 000
2 000 000
1 500 000
1 000 000
500 000
0
1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003
Year
TOP 10 COUNTRIES
OTHERS 11%
NLD 2%
THA 2%
SWE 3%
ESP 3%
USA 46%
FRA 4%
AUS 5%
CAN 5%
DEU 6%
GBR 13%
Submitting ADR Reports
to WHO
• Vigibase online (Internet)
• CD produced by national computer system
• computer network (FTP/e-mail)
Submitting ADR Reports
to WHO
• ICH - E2b format
• WHO agreed format
Staff
– medical director
– pharmacists
– biomedical scientists
– IT specialists
– sales and marketing
– administrative
– total approx 40 persons
Functions 1
• Signal detection
– Identification of previously unknown drug
reactions
Signalling Procedure
What should be achieved?
• Signals should not be missed
• Signals should be found early
• ‘False’ signals should be kept to a
minimum
Why use a Bayesian neural
network?
• An automated procedure with a power to
consider all combinations
– drug - ADR
– drug - indication - age - ADR
• All combinations are considered in an
unbiased manner
• Strong associations are highlighted for
clinical assessment
Signal detection using a neural
network approach
• If the Posterior probability > Prior
probability
– The drug ADR combination is present more
often than expected
– This is represented by a high value of
Information Component (IC)
Information Component (IC)
• Definition
– IC=log2 (Posterior Probability/ Prior
Probability)
Captopril - Coughing
6
5
4
3
2
1
0
-1
-2
IC
79:1 81:1 83:1 85:1 87:1 89:1 91:1 93:1 95:1
Time(year)
Practolol, ATC C07AB - Peritonitis
8
6
4
2
0
-2
-4
Practolol
ATC C07AB
-6
71
74
77
80
83
86
year
89
92
95
98
Signal Detection & Follow-up
Combinations.db
(reported quarterly)
Quarterly analysis
BCPNN
Vigibas
e
National Centres
Combinations database - example
Signal Detection & Follow-up
Combinations.db
(reported quarterly)
Triage (filter)
Quarterly analysis
BCPNN
Vigibas
e
National Centres
The triage procedure
• IC-2std>0
• Substantial increase in IC from last
quarter
• New drugs
• Serious reactions (Critical terms)
• Reports from >2 countries
• Special interest reactions
• Not in literature
Signal Detection & Follow-up
Combinations.db
(reported quarterly)
Triage (filter)
Quarterly analysis
BCPNN
Review
panel
Vigibas
e
Pharma
Company
Yes
SIGNAL
National Centres
No
Follow-up
Panel of signal reviewers
• 38 clinical and ADR experts
• voluntary consultants recruited globally
– assess associations in specialist area for
clinical significance
– write assessment report for SIGNAL
Functions 2
• Signal strengthening
– Web-based search programme
– Search requests
Functions 3
• adverse reaction profiles
IBUPROFEN - ADR profile
System Organ Class
Vision
Skin
Repro
Neoplasms
Liver-bil
Foetal
Cardiovasc
Appl site
0
1000
2000
3000
4000 5000
No of reports
6000
7000
8000
Functions 4
• Comparing national experiences
International Differences
(Quantitative and Qualitative)
•
•
•
•
•
•
•
•
•
disease prevalence
genetic
social
cultural
healthcare systems
health professional practices
indication for, and use of medicines
pharmaceutical formulations
drug monitoring practices
Functions 5
• identification of risk factors
Potential Risk Factors
•
•
•
•
•
•
•
•
other drugs
sex / gender
age
genetic constitution
dosage
duration of treatment
route of administration
indication
Functions 6
• Combining ADR figures with other data
– drug utilization statistics
– population statistics
UMC - a communication centre
• WHO Pharmaceuticals Newsletter
12
UMC - a communication centre
• WHO Pharmaceuticals Newsletter
• Uppsala Reports
Uppsala Reports
UMC - a communication centre
• WHO Pharmaceuticals Newsletter
• Uppsala Reports
• Internet home page
http://www.who-umc.org
• Vigimed e-mail discussion group
Pharmacovigilance Training
• Training course
– Uppsala, Canberra
– 2 weeks
– 25 participants
– 9th course November 2004
• Regional and local activities
UMC involvement in local activities
2000 -2004
• 2000
– India, China, Kuwait, Romania, Uruguay
• 2001
– Oman, Russia, Ghana, Fiji, Singapore, Vietnam
• 2002
– Cuba, Chile, Morocco, Cyprus
• 2003
– Serbia, Zambia, Saudi-Arabia, Ghana, India
• 2004
– Hong Kong, China, Brazil,
Pharmacovigilance and
Public Health Programmes
• New medicines for tropical diseases
– Artimisinine derivatives against malaria
– HIV/AIDS
– Vaccination Programmes
– etc
•
•
•
•
Global Fund
’3 x 5’ initiative by WHO
Challenges for data collection
Need for pharmacovigilance training
Technical support
• literature coverage
Technical support
• literature coverage
• guidelines
Software for National Centres
Vigibase on-line
E2B
Search
Analysis
Dr
RC
NC
Vigibase
UMC Functions
• Harmonisation
Definitions established
within the WHO Programme
–
–
–
–
–
–
Adverse reaction
Adverse event
Side effect
Signal
Serious reaction
Causality:
»
»
»
»
»
»
certain
probable/likely
possible
unlikely
conditional/unclassifiable
unassessable
Worldwide network of
knowledge and competence
• Annual meeting of representatives of
National Centres
• Working relations with relevant
organizations
– CIOMS, ISoP, ISPE, DIA, IPCS, HAI, IFPMA, etc
Projects
• Methods for signal identification and analysis
– data-mining approach to signal analysis
• Improved monitoring of traditional medicines
– collaboration with Royal Botanical Gardens, Kew, UK,
Dept Systematic Botany, Uppsala
• Good communications practice in
pharmacovigilance; collaboration with:
– University of Verona
– EQUUS
– CIOMS
Common name
Problems in Herbal Pharmacovigilance
Chinese, Asian Ginseng
American Ginseng
Tienchi Ginseng
Siberian Ginseng
Russian Ginseng
Brazilian Ginseng
Wild red Am. Ginseng
Alaskan Ginseng
Wild Ginseng
Ayurvedic Ginseng
Ginseng of the Andes’
Botanical name
Chemical relation
Panax ginseng Meyer
Panax quinquefolius L.
Panax pseudoginseng Wall.
Eletherococcus
senticosus Maxim.
Acanthopanax
senticosus Harms.
Rumex hymenosepalus Torr.
Pfaffia paniculata (Mart.) Kunze.
Echinopanax
horridum (Sm.) Decne.)
Aralia nudicaulis L.
Withania somnifera (L.) Dunal
Lepidium meyenii Walpers
Standard
Similar
Similar
Different
Different
Different
Different
Different
Different
Different
Different
WHO herbal ADR database
Valid scientific botanical names
• No internationally standardized and accepted
classification of all botanical names of medicinal
herbs exist.
• Therefore the UMC has created a list of
preferred botanical names and their synonyms.
• The preferred names are...
...the complete Latin binomial name including
the author.
...assigned in collaboration with Royal Botanical
Gardens of Kew, London.
...specified with part and extract (in English)
if given.
WHO herbal ADR database
BOTANICAL
PREFERRED
NAME
Valid scientific
plant name_ID
Part_ID
Extract_ID
SYNONYMS
Latin name or
common name
HCN
WHO herbal ADR database
Herbal Code Number (HCN)
•
•
•
eg.
Similar to CASnumber used for chemical substances
- one unique number for each herbal substance
- same number of digits
But the HCN
- always begins with a ’9’
- developed by the UMC
Possible to link all herbal substances of the same plant by linking the
first five digits.
Valeriana officinalis
Valid scientific name ID
9005100000
90051
Valeriana officinalis root
Part ID (root)
9005100500
005
Valeriana officinalis root dry extract
Extract ID (dry extract)
9005100502
02
WHO herbal ADR database
ADR REPORT
ID
SUBSTANCE
ID
INGREDIENT
ID
MED.PROD.
Drug name
Substance name
Filenumber
CAS/HCN
Manufacturer
Literature source
ID
Source version
Source
Source version
BOTANICAL
PREFERRED
NAME
Valid scientific
plant name_ID
Part_ID
Extract_ID
SYNONYMS
Latin name or
common name
Country
ATC/HATC
ATC code
HCN
Official/Herbal
Data available to non-members
• By request to WHO Collaborating Centre
• To degree health professionals
• Caveat document
Requirements for joining the
WHO Programme
• programme for collection of spontaneous
ADR reports established
• a National Centre designated by Ministry
of Health
• technical competence to fulfil WHO
reporting requirements
Process for joining WHO Programme
1.
Ministry of Health (or
equivalent) designates
National Centre
2. Ministry of Health sends
formal application to WHOHQ, Geneva
Ministry of Health
2
5
5. WHO-HQ advises Ministry of
Health of admittance to the
Programme
National
Centre
3
3. National Centre sends
sample reports to the UMC
4. UMC notifies WHO-HQ
that reports are compatible
1
the UMC
WHO-HQ
Geneva
4