Transcript Storage Decisions 2003

Hosted by
From Regulation To
Desperation
A Life Sciences case study in
compliance
Richard Scannell, GlassHouse
Hosted by
Background
 IT affecting regulations are increasing in number
and visibility
 Most organizations are ill-prepared to respond
and react
 Vendors have finally found their next Y2K
 Audits shift from faith-based to evidence-based
 Failures are well-publicized; successes are
unknown
Hosted by
Introduction
A Pharmaceutical Case Study
 Understanding the business lifecycle
 Mapping to the data lifecycle
 The Audit Process
 Classifying Datasets
 Mapping to technical requirements
Hosted by
Who are your regulators? Why?
Pharmaceutical World
 FDA – Keep Bad Products Off Market
•
Regulating Companies processes in bringing products to
market
•
Companies have to provide documented evidence that
their products meet quality standards
 Companies – Get Products To Market
•
•
FDA Approval Times Increases
Declining Profitability
The New Drug Development Process
PRE-CLINICAL
RESEARCH
Hosted by
CLINICAL STUDIES
NDA REVIEW
Phase 1
E
E
SYNTHESIS
AND PURIFICATION
Phase 2
Phase 3
Accelerated Development Review
ANIMAL
TESTING
E
Short-Term
Treatment IND
Parallel Track
Long-Term
INSTITUTIONAL
REVIEW BOARDS
INDUSTRY TIME
FDA TIME
IND SUBMITTED
SPONSOR/FDA MEETINGS ENCOURAGED
ADVISORY COMMITTEES
Source: www.fda.gov
NDA SUBMITTED
EARLY ACCESS:
SUBPART E
E
REVIEW
DECISION
SPONSOR ANSWER
ANY QUESTIONS
FROM REVIEW
Hosted by
Pre-Clinical Research
Internal
Data
Data Requirements
Clinical Trials
(Phase 2)
New Drug
Application (NDA)
Contract Service Organizations
Internal
Data
•
•
•
•
Biotechnology
Partners
Investigational New
Drug Application (IND)
Electronic Submission
Animal studies
Clinical tests
Drug components
Manufacturing process
Hosted by
Align Value with Cost
Occupancy:
 Appropriately congregate data
by value and rationalize
storage requirements at each
phase
Architecture:
 Hardware/software
requirements
 Processes and standards
Governance:
 Validation and qualification
 Movement of data up or
down the value chain
Hosted by
More Typical Approach
 Data placement based
on device management
rather than data
management
 No alignment between
business value of data
and infrastructure
 Unknown or high TCO
 Very little focus on
Process Management
Hosted by
Audience Response
1. I understand the business lifecycle and corresponding
data lifecycle of my business.
2. I know how we manage data but I don’t have much
appreciation how this ties to our business
performance.
3. I’m the IT guy; the business never talks to me.
4. I keep the lights on; what more do you want?
Hosted by
Regulations & Audits
 21 CFR Part 11 – Electronic Records and Electronic
Signatures
 50% of all FDA Inspections include some aspect of
computer system quality and compliance
 Computer Systems Compliance one of most common
reasons for a 483 (Violation)
 Expectation is that companies implement global
procedures – major IT implications
Hosted by
Validation – The Missouri Factor
Validation - documented proof that a product or process will
repeatedly produce a product to the stated specifications, i.e.,
Prove a Process Works.
"Establishing documented evidence which provides a high
degree of assurance that a specific process will consistently
produce a product meeting its pre-determined specifications
and quality attributes".
FDA Guidelines on General Principles of Process
Validation, May 1987.
Specifications; Qualifications; Verifications; Calibrations; Certifications
Hosted by
Noted Recent Issues in FDA
Inspections
 Insufficient Change Control
 Lack of or Inconsistent documentation
•
•
System Overviews
GMP applications used over a network
 Failure to control and archive GMP information
communicated via e-mail
 System security and user authorization
 Insufficient GMP training of IT Staff
Hosted by
Data Classification Maps to
Technology
 Define Classifications
• Location in the Lifecycle
• Requirements and Attributes
 Regulatory issues
 Policies
• Storage
• Backup/Restore/DR
• Archive
• Security
Hosted by
Clinical
Pre Clinical
(3.5 – 6 yrs)
(7m-1yr)
(m-2yrs)
(~ 3 yrs)
Patent Application
Risk
Phase 1
Clinical Trials
IND
Pre-Clin
Submission Production
Phase 2
Clinical Trials
Phase 1
IND
Pre-Clin
(Approx 2.5 yrs)
Phase 3
Clinical Trials
Phase 2 and 1
IND
Pre-Clin
NDA Submission
And Approval
Clin Data
IND
Production
Preclin data
All Data
Cost
Retirement
Pre-Clinical
Data
Pre-Clinical
Data
IND
Value
NDA
Architecture
Architecture
Data Classification
Hosted by
Data Classification Maps to Technology
Define:
1.Logical
Data Classes
2.Occupancy
3.Data
Management
Policies
Physical
Mapping &
Tier Design
System
Policies
Technical
implementatio
n on
standardized
technology
Regulated
Business Critical
Other
High
Med
Low
High
Med
Low
High
Med
Low
R-H
R-M R-L
B-H
B-M
B-L
O-H
O-M
O-L
TIER1
TIER2
TIER3
R-H B-H
R-M R-L B-M O-H
B-L O-M
Tape
Based
Backup on
RAID set
Multiple SNAP
copies
Tape Based
Backup on RAID
set
TIER4
O-L
Remote replication
Multiple SNAP
copies
Tape Based
Backup on RAID
set
Hosted by
Implementing Data Quality
 Data Classification Model
•
Drives procedures for managing data
 Documentation
•
Policies and Procedures for Backup/Restore/Archive; Disaster
Recovery; Change Control; Security; Configuration Management
•
Validation Documentation
 Training
 Validation/Process Training
•
Maintenance of Training Records
 Risk Audit
•
Requirements are on-going
Hosted by
What Does All This Mean To You?
 Analysts can help you understand requirements.
 Vendors are talking about compliance, but there is
little vertical capability in MOST suppliers.
 E-mail management and archiving has the most focus
(Sarbanes Oxley).
 Process management will become as important as
technology management.
 Establishing policies and evidence trails will be key to
audit success.
Hosted by
Creating the Business Case (Pharma)
The Carrot

“The organization Pharmaceutical Research
and Manufacturers of America estimates
that only five in 5,000 compounds that
enter preclinical testing make it to human
testing, and only one of those five may be
safe and effective enough to reach
pharmacy shelves.”

Gartner estimates that 30% of research in
Life Sciences is duplicated due to poor
information sharing

“Just keeping a drug pipeline full involves
investing $2 billion to $5 billion a year in
research, and that doesn’t include
production and distribution costs.”
Source: “Mastering the Value Chain: An
Interview with Mark Levin of Millennium
Pharmaceuticals”, Harvard Business Review
The Stick
 Violations in an FDA audit will
slow down time to market – for
a blockbuster drug that’s worth
up to $2M per day. Based on
patent timeframes and generic
manufacturing pressures, that
revenue is likely unrecoverable
 Failure on a given audit will
impact not only that submission
but other submissions in
process also.
 At its most extreme, businesses
shut down, executives go to jail
and patients die.
Hosted by
Thank You
[email protected]