S I A

N E
DO
P IN
OU
R
R SG
GE
N A
CONTINUEDAPROCESS VERIFICATION
C M
Q 06 MAY 2021
QA
Objectives
S I A
N E
D O
I N and
• Have a basic understanding of the purpose
U P
benefits of CPV
R O
S G to the
• Explain why CPV is important
E R
A G the business, and
patient/consumers,
A N
C M
regulatory agencies
A Q
• Describe the basic steps for implementing CPV
Q
 What is Continued Process Verification
(CPV)?
•
S I A
Continued Process Verification (CPV) is part of the FDA guidelines on Process

NE
Validation with a product lifecycle concept. The guideline is also an integration

DO
of the International Conference on Harmonisation (ICH) guidance for industry

IN
Q8 (Pharmaceutical Development), Q9 (Quality Risk Management) and Q10

P
(Pharmaceutical Quality Systems)
U
R O
R SG
G E
NA
M A
Q C
QA
• The lifecycle concept links product and process development, qualification of
the commercial manufacturing process, and maintenance of the process in a
state of control during routine commercial production.
 What is Continued Process Verification
(CPV)?
•
S I A
Process Validation is defined as the collection and evaluation of data, from the

NE
process design stage through commercial production, which establishes

DO
scientific evidence that a process is capable of consistently delivering quality

IN
product. The FDA guidance describes process validation in three stages :

U P
R O
R SG
G E
NA
M A
Q C
QA
 What is Continued Process Verification
(CPV)?
•
S I A
Stage 3 CPV is the process by which we verify that our product control strategy

N E
remains capable and in its qualified state during routine commercial
manufacture
DO
P IN
U
• It is how we demonstrate that our processes can deliver products at highest
quality on a consistent basis
R O
S G
E R assessing our process capability
• It is the way in which we monitor both our process variation through routine

A G
data trending and periodically

A Ndefine the actions required to maintain a state of control and identify

M
• Helps
C process improvement opportunities. Helps us shift from “fire-fighting” or
A Q reactive to proactive mode.
Q • It is applicable to both new and existing products
 Regulatory Expectations on CPV

• Continued Process Verification (CPV) is now a requirement of many, if

not all, of the world’s regulatory agencies and WHO.
S I A
N E
D O
• FDA , Process Validation: General Principles and Practices, guidance for

I N
industry, January 2011 (Section D. Stage 3 – Continued Process
Verification):
P
Uto evaluate process stability
R O
– “Production data should be collected
and capability.”
S G
E R control techniques develop the data collection
– “We recommend that a statistician or person with adequate training

Gstatistical methods and procedures used in measuring and

in statistical process

N A
plan and

M A evaluating process stability and process capability.”

Q C – “If properly carried out, these efforts can identify variability in the

A
process and/or signal potential process improvements.”

Q
 Regulatory Expectations on CPV

• In 2018 and 2019, about

23% of the FDA
S I A
Inspection observations
NE
were related to gaps in
DO
IN
CPV whereby 9-10% was

P
attributed to lack of
“control procedures to
OU
monitor and validate
R
SG
performance” (source:

E R
https://www.fda.gov/ins

G
pections-compliance-
A
N
enforcement-and-
A
M
criminal-

Q C investigations/inspection

A
-references/inspection-

Q observations)
 Regulatory Expectations on CPV
• The goal is to provide continual assurance that the process remains in a
state of control (the validated state) during commercial manufacturing.

A
•
I
Regulatory agencies require the use of data to support the development
and manufacture of drug products
E S
•
N
Good manufacturing practices must be followed when collecting and
O
D
evaluating the data needed to assess product performance

IN
U P
R O
R SG
G E
NA
M A
Q C
QA
 Why is CPV important?
Through CPV, we are able to ensure the effective
execution of our robust Product Control Strategies and

S I A
• Build trust with our patients and consumers by
NE
D
consistently delivering products of the highest qualityO
P IN
•
OU
Detect, Understand, and control the sources of
variation in our processes
R
SG
• Improve Process and quality performance, to

E R
eliminate defects and waste

A G
•
A N
Build trust with our regulators by demonstrating

C M continual assurance that our processes are

Q
maintained in a state of control

QA
 S I A
N E
DO
P IN
OU
R
SG
R
GE
A
N CPV Implementation
MA
QC
QA
 When does CPV occur?
• CPV can begin at different phases in the product lifecycle depending on
whether it is a new product or an existing product (also known as legacy
product)
S I A
•
NE
New product will always begin as Stage 1 and follow the normal cycle into
Stage 3 CPV.
DO
•
IN
For existing or legacy products, most will begin at Stage 3 CPV unless there’s a
P
U
significant process change requiring revalidation, which will then put you back

O
into Stage 1 Process Design or Stage 2 Performance Qualification.
R
R SG New Product

G E
Legacy Products
Introduction

NA
M A
Q C
QA
 Who is involved in CPV?

• Continued process verification is a cross-functional responsibility, requiring

S I A
individuals from different business groups. Some of the typical business groups
involved in CPV are
N E
DO
– Technical
P IN
U
– Engineering
– Quality
R O
SG
– Production
–
E R
Statistics
CPV
AG
– Informatics

A N – Analytical

C M
A Q
Q
 How is Stage 3 CPV Implemented?
• Below is the basis high-level steps for implementing CPV

S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 How is Stage 3 CPV Implemented?
• Stage 3 CPV is consists of two phases:

I A
– Phase 1 is the “continued monitoring and sampling of process parameters
S
E
and quality attributes at the level established during the process

N
qualification stage until sufficient data are available to generate
O
significant variability estimates”

IN D
– Phase 2 is the “routine sampling and monitoring which is adjusted to a

U P
statistically appropriate and representative level”

R O
G
S that the Stage “3A” and “3B” terminology are not
• For ease of discussion, we will refer CPV Phase 1 as Stage 3A and Phase 2 as

specifically usedE
R
Stage 3B. It’s worth noting

A G in the FDA Guidance, but the two phases are described.

A N
C M
A Q
Q
 How is Stage 3 CPV Implemented?

S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 How is Stage 3 CPV Implemented?

S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 How is Stage 3 CPV Implemented?

S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 How is Stage 3 CPV Implemented?

S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 How is Stage 3 CPV Implemented?

S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 How to transition from Stage 2 PQ to
Stage 3 CPV?
Stage 2 - PQ
• Upon completion of Stage 2 – PQ
S I A
E
there are two options that can be
followed when moving to Stage 3
O N
(CPV)
IN D
•
U
The first option is to move into Stage P
3A where higher
R O
SG
Stage 3a
sampling/monitoring of a Routine commercial

R
manufacture with
process/product is performed prior Requires

E
heightened sampling justification

G
to moving to routine manufacture

NA
and sampling routines coupled with

A
periodic data trending and review.
M
Q• C
The second option moves directly
A
Stage 3b

Q
from Stage 2 to Stage 3B without the Routine commercial
manufacture and
need for heightened sampling or sampling
monitoring. CAPAs
Planned change

• But when are these stages

applicable?
 How to transition from Stage 2 PQ to
Stage 3 CPV?

S I A
N E
Stage 3A CPVND
O
P I
O U
• Identify the CQAs and
CPPs (if needed) to
GR
• Define and document the • Implement the CPV
Governance
S
Stage 3A CPV monitoring

R
monitor) plan • Input into Product

E Quality Reviews

AG
• Collet data and calculate
process capability and

A N alert limits

M
• Document the product

C
Stage 2 PQ and process Stage 3B CPV

A Q understanding gained

Q
from Stage 3A and justify
the transition to Stage 3B
 Stage 3A CPV Monitoring Plan
Continued Process
Verification
Monitoring Plan
S I A
NE
DO
❑ Number of batches to be
I N
trended
U P
requirementR
O
❑ Sampling and monitoring

S G
E R Process Parameters to
❑ Critical Quality Attributes and

A G be trended
Critical

MAN ❑ Data trending frequency

Q C ❑ Statistical analysis tools and

QA control charts
❑ Action tracking and escalation
route
 Stage 3A CPV Monitoring Plan
• Number of batches to be trended- typically the first 30 batches after Stage 2 are

I A
subjected to Stage 3A as this is considered a “statistically significant number/population”

S
E
which will provide an accurate representation of the voice of the process (VoP)

• Sampling and monitoring requirement- this includes theO

N
points/location, sampling interval, and how manyN
I D number of sampling

P
samples will be collected

O U
G R
• Critical Quality Attributes and Critical Process Parameters- the critical quality

S and compliance such as API assay, preservative

attributes or process parameters that can be trended are those which are

E R
impacting efficacy, safety,

A
assay, etc
G
N
MA trending frequency- this will depend on the level of risks identified, it can
• Data

QC
be weekly, monthly, etc

QA • Action tracking and escalation- actions identified must be logged and closely
monitored. Escalation route must also be clearly defined. This should also
include procedure on how to manage any atypical or out of trends which will
sometimes trigger investigation or root-cause analysis to identify process
improvement initiatives
 Statistical Process Control Charts,
Process Capability, and Alert Limits
• Control charts are used to evaluate process stability. The Alert Limits are

S I A
determined from the data and express the inherent routine variability of the

N E
process. The alert limit provide a means to identify atypical or out of trend

O
performance from a statistical quality control (SQC) perspective
D
I Nfor the data trending as part
P
• There are several control charts that can be used
U
of Stage 3A CPV.
R O
S G
– Trend plots
E Raverage
Ganalysis
– 5 point moving
A
N
– Cu-sum
A
C M – Histogram

A Q
Q
 Statistical Process Control Charts,
Process Capability, and Alert Limits
• Trend plot
S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 Statistical Process Control Charts,
Process Capability, and Alert Limits
• 5- point moving average
S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 Statistical Process Control Charts,
Process Capability, and Alert Limits
• Cu-Sum charts
S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 Statistical Process Control Charts,
Process Capability, and Alert Limits
• Histogram
S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 Statistical Process Control Charts,
Process Capability, and Alert Limits
•
I
Process capability measures how well the “voice of your process (VOP)”
S A
E
matches up with the “voice of the customer (VOC)”. It compares the variability
of a process to specifications
O N
IN D
U P
R O
R SG
E
N AG
MA
QC
QA
 Statistical Process Control Charts,
Process Capability, and Alert Limits
•
S I
Alert limits can be defined in different ways, depending on the risks and A
product understanding.
N E
o Use of historical data- most often than not, N D O
P I legacy products have alert

O Umaximum value from previous data

limits which are based on historical process performance, traditional

R
approach is to use the minimum and

G
S alert limits based on ± 3 standard deviation
E R
o Use of Standard Deviation-

A G
are commonly used

A N
C M o Use of stability data- there are quality attributes which tend to increase

Q
or decrease throughout the product shelf-life. Stability analysis are

QA conducted for those quality attributes in order to predict the appropriate

alert limits which will minimize risk of out-of-specification during stability
 Statistical Process Control Charts,
Process Capability, and Alert Limits

S I A
N E
How do we identify atypical or “out-of-trends (OOT) ? D
O
I N
U P
O
Rprovide a potential early warning to process
• Atypical data or out-of-trends are breakages in your statistical process control

S G
charts. Identifying this breakages

E R
shift or drift that may indicate a change in performance over time.

A G
A N
C M
A Q
Q
 Statistical Process Control Charts,
Process Capability, and Alert Limits
•
S I A
Below are the control chart rules which can help in identifying atypical data or
OOT.
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
• Commonly used rules are Rule 1, 4, and 5.
 Statistical Process Control Charts,
Process Capability, and Alert Limits
• Control Chart divided into Zones
S I A
N E
DO
P IN
OU
R
R SG
E
(+3 SD)

N AG (+2 SD)

MA
C
(+1 SD)

A Q
Q
(-1 SD)

(-2 SD)

(-3 SD)
 Statistical Process Control Charts,
Process Capability, and Alert Limits
• Rules 1-4
S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 Statistical Process Control Charts,
Process Capability, and Alert Limits
• Rules 5 and 6
S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 Statistical Process Control Charts,
Process Capability, and Alert Limits
• Rules 7 and 8
S I A
N E
DO
P IN
OU
R
R SG
E
N AG
MA
QC
QA
 What is CPV Governance?

• CPV governance
S I A
– formally assess product and process performance
N E
DO
– It confirms, prioritizes, and manages process risk areas

P
– It drives process performance improvement IN
OU
– Should include all applicable functions (e.g. Production, Engineering,

R
SG
Technical, Informatics, Quality)

•
E R
Regular meetings help drive the governance process, and include a review of

AG
data trends, which drive actions and decision making
N
A
• MTypical agenda includes review of data trends, identification and updating of
QC actions, sharing of success stories, and escalations or support needed.
QA
 Link Between CPV and PQR

•
S I A
Product Quality Review (PQR) is the over-all review of your product including

N E
assessment of different quality indicators such as deviations, change control,
complains, out-of-specifications, etc
DO
P IN
•
U
Data from the CPV is an input to the PQR.
O
R
R SG
E
N AG
MA
QC
QA
KEY LEARNINGS
S I A
N E
• CPV provides an assurance that products are DO
manufactured with highest quality on aP IN
OU consistent
and continuous manner
G R
R S remains capable
• It verifies that control strategies
G E
A
and its qualified state
N
M A
• It helps identify process improvement opportunities
Q C the approach from reactive to proactive
and shift
•QA
Applicable to new and existing products
 S I A
N E
DO
P IN
OU
R
SG
R KASIH
TERIMA
AG
E
A N
C M
A Q
Q