CPV E-BOOK

IMPLEMENTING

CONTINUED
PROCESS
VERIFICATION
with Bio4C™ ProcessPad

CONTINUED
3.1 PROCESS 1.1
VERIFICATION
Process
(CPV) Process
Monitoring Concept
& Process
Improvement

PROCESS
DESIGN
PRODUCT
LIFECYCLE Identification
of the process
variables
1.2
2.2 PV (PPQ)

PROCESS
PERFORMANCE Control
QUALIFICATION Strategy

IQ, OQ, PQ
1.2

2.1

The life science business of Merck KGaA, Darmstadt, Germany

operates as MilliporeSigma in the U.S. and Canada.
Process Lifecycle Management
Table of Contents
Bio4C™ ProcessPad allows scientists and engineers to collect
process development data and leverage it throughout the
Bio4C™ ProcessPad provides a platform to capture, modify,
and track your critical performance parameters (CPPs) or
entire lifecycle for effective knowledge transfer during critical to quality attributes (CQAs). Over the lifecycle of the
Background
process characterization 3 through scale-up to large scale product, users can monitor the process’s evolution with full
GMP manufacturing. traceability of critical to quality parameters and control limits

Process Validation and Positioning of CPV

Core process functions create and consume process data in
applied to CPPs or CQAs. Using Bio4C™ ProcessPad, users
4
track, compare and build reports on process development
similar ways with similar underlying metadata and context.
and manufacturing datasets making tech transfer easy
Bio4C™ Parameter
ProcessPad helpsClassification and Statistical
manage and serve contextual Treatment of Data 4
and making historical process knowledge available during
process information throughout the lifecycle of Quality
Parameter Classification
by Design (QbD), Process Analytical Technology (PAT)
commercial manufacturing. Scientists and 4 engineers can
build discussion forums around development and
Statistical Treatment
and Continued Process Verification (CPV). of Data 5
manufacturing data reports to collect and archive subject
• Efficiently capture all development, scale-up, and
commercial
Determining Process Capability (Cpk) matter expert process knowledge for later search and retrieval
process execution data on a single platform (even if a subject matter expert leaves the organization –
and Process Performance (Ppk) 7
• Easily manage specification or control limits history the knowledge remains with you).
Estimating Process Capability for Normally
• Utilize historical data for investigations and
Distributed Data 8
process improvement
Estimating Process Capability for Non-normally
• Simplify the tech transfer process
Distributed Data 8
• Instant access to the latest version of process map/
genealogy
Parameter Monitoring
(CQAs, CPPs, sampling plans) 8
Preliminary Process Monitoring Phase 8
Statistical Process Control (SPC) Phase 8
Resetting or Revising Control Limits 8

Trending Rules 9
Examples of Rule Definitions 9
Rule Violation Investigation Approach 10

Documentation and Reporting 11

In-process Control and Monitoring
(IPCM) Document 11
Periodic Presentation of Process Trends 11
Quarterly Process Summary Reports 11
Site Trend Violation and CAPA Report (Quarterly) 11

Summary 12
Bio4C™ ProcessPad CPV Platform 13
Why is Bio4C™ ProcessPad a Comprehensive
CPV solution? 13
FDA CPV Requirements 13
Key Differentiators of Bio4C™ ProcessPad 14
Validated Data Source 14
Plug-and-Play Monitoring 14
Report Generation (CPV/APQR) 15
Salient Features of Bio4C™ ProcessPad 15

2
Background CONTINUED
3.1 PROCESS 1.1
VERIFICATION
Process
(CPV) Process
Monitoring Concept
& Process
Improvement
The Food and Drug Administration (FDA) guidance on
continued process verification and the EU GMP Annex PROCESS
15 requirements for ongoing process verification direct DESIGN
pharmaceutical and biopharmaceutical manufacturers to PRODUCT
ensure that their processes remain in a continual state LIFECYCLE Identification
of control (a validated state) during the lifecycle of the of the process 1.2
variables
product so that the strength, quality, and purity of the final
2.2 PV (PPQ)
drug product is maintained. Both regulatory agencies direct
manufacturers to develop sustained programs which collect
and analyze product and process data to evaluate the state PROCESS
PERFORMANCE Control
of control and to identify product or process problems or QUALIFICATION Strategy
opportunities to implement improvements.
IQ, OQ, PQ
1.2
A product’s lifecycle normally consists of 3 phases:
process design, process performance qualification, and the 2.1
last and lengthiest, continued process verification (CPV)
which carries on throughout the lifespan of the product
(Figure 1.) Figure 1. Product Lifecycle

Until recently biopharmaceutical manufacturers focused Bio4C™ ProcessPad is a data visualization, analytics, and
predominantly on the process design and process performance process monitoring platform that enables bioprocess lifecycle
qualification phases of the product lifecycle; given recent FDA management, reporting, investigations, and continued process
and EMA guidance to institute ongoing programs to collect verification. It allows biopharmaceutical scientist to collect and
and analyze product and process data, the industry is now manage data from paper-based records, spreadsheets, batch
giving equal attention to CPV. CPV has become a part of the record data, quality control data, external databases, data
observations in recent warning letters of regulators either due historians, and streaming machine data in a single software
to lack of complete understanding of the concept or lack of environment. Bio4C™ ProcessPad provides data visualizations
right tools to perform CPV in an efficient way. Below are some and analytical tools for straight forward statistical trending of
excerpts from the FDA warning letters. data control charts, correlation charts, box plots, etc. It also
offers easy reporting including campaign reports, process
In pharmaceutical and biopharmaceutical manufacturing, the
summary reports, Annual Product Reviews (APRs) and Annual
amount of process and analytical data per batch is very high.
Product Quality Review (APQR). Along with offline monitoring
If a structured procedure is not followed, manually collecting
of data, the Bio4C™ ProcessPad-RT collects, aggregates, and
the data and trending it statistically is time-consuming and
provides direct web-browser access to real-time streaming
error prone. Among many tasks, CPV requires collecting
data from processing equipment such as bioreactors,
process parameter data, trending it against statistical
chromatography systems, buffer systems, etc.
control limits, and calculating process capability (Cpk and
Ppk) at defined intervals or after every few batches. Since a
structured CPV program is now a regulatory expectation for
process manufacturing and quality operations, the industry
seeks an automated software solution built specifically for
ongoing process verification.
Process validation
and positioning of CPV
From the process design stage through commercial Successful process design and process qualification determine
production, process validation collects and evaluates data that the process can reproducibly produce a commercial
and establishes scientific evidence that a process is capable molecule. Continued process verification ensures that the
of consistently delivering quality product. process remains in a state of control and is steadily producing
a medicine of desired quality. Unlike process validation’s first
Process validation involves a series of activities taking place
two stages which have a distinct endpoint, CPV is sustained
over the lifecycle of the product and process. Its activities can
for the life of a drug molecule and requires a comprehensive
be divided into three stages:
strategy. Following are the key components required to
1. 
Process Design: The commercial manufacturing process establish a CPV program:
is defined during this stage based on knowledge gained
1. Parameter classification and statistical treatment of data
through development and scale-up activities.
2. Determining process capability (Cpk) and process
2. Process Qualification: During this stage, the process
performance (Ppk)
design is evaluated to determine if the process is capable
of reproducible commercial manufacturing. 3. Parameter monitoring
3. 
Continued Process Verification: Ongoing assurance is 4. Out-of-trend detection rules
gained during routine production that the process remains
5. Documentation and reporting
in a state of control.
In this eBook we will outline a strategy and all the associated
components required to establish a compliant and efficient
CPV program.

Parameter Classification and Statistical Treatment of Data

Parameter Classification subject matter expert prior manufacturing experience and
process development knowledge. A risk analysis can also be
Process Characterization, Process Validation or In-process
performed to from each parameter type.
Control process description documents define parameter
classifications and set initial action or specification limits. In the next step, limits are defined and applied to each
Table 1 summarizes the parameter definitions and their role parameter type.
in process monitoring. All the Critical Process Parameters Table 2 describes all the limits which need to be considered
(CPP) and Key Process Parameters (KPP) mentioned in the during process monitoring.
process description documents should be monitored in the
program. Monitored parameters (MP) are optional and ought
to be selected on a case-by-case basis based upon relevant

Table 1. Parameter Types

Parameter Type Abbreviation Description Routine Monitoring

Critical Process Parameter CPP A performance or input parameter that directly Must
impacts product identity, purity, quality, or
safety.

Key Process Parameter KPP A performance or input parameter that directly Must
impacts CPPs or is used to measure the
consistency of the process step.

Monitored Parameter MP A performance or input parameter that may or Not all, case by case basis
may not impact KPPs and is used to measure
the consistency of the process step or routinely
trended for troubleshooting purposes.

4
Table 2. Limit Types and Applications
List Name Abbreviation Description Limits Source Applicable to
Parameter Type

Specification Limits USL, LSL These limits are defined based on process Process CPP
characterization limits. Any excursion from Characterization,
these limits will cause out of specification Process Development
and batch rejection.

Action Limits UAL, LAL These limits are process validation ranges. Process Validation CPP, KPP
Any excursion from these limits will cause
major process deviation or discrepancy.

Alert Limits or UCL, LCL These are monitoring ranges derived from Statistical: Process CPP, KPP, MP
Statistical Control historical runs for out of trend detection History >15
Limits and measurement of process consistency. commercial batches

Target CL The target (or centerline) is derived again Statistical: Process CPP, KPP, MP
from historical runs as a measure to keep History >15
the process consistent and proactively alert commercial batches
if process is deviating from set target.

Note: CL=centerline, UCL=upper control limit, LCL=lower control limit, SD = standard deviation

Specification limits and action limits are defined during process design and qualification and the first 15 to 30 batches can be
trended against these limits. Alert limits (or statistical control limits) should then be defined based on these historical/initial 15
to 30 batches.

Statistical Treatment of Data

Once the parameter criticality for the desired product quality is determined and parameters classified accordingly, the next step
is to perform statistical treatments on the data to determine statistical control limits for the desired performance parameter.

Determining the Distribution of Data for Each Process Parameter

Different types of data distributions and statistical treatment for control limits evaluation for each distribution type.

Table 3. Data Distribution and Statistical Treatments

Distribution Sample Graph Description and Examples Typical Limits Applied

Normal

CL Average
Most of the process parameters will follow this distribution
UCL Average + 3 SD
of a normal/Gaussian bell shaped curve.
LCL Average – 3 SD

Non-normal
(Beta or Gamma)

Some parameters will not follow a normal distribution

pattern and follow a skewed distribution. For example
CL Median
most of the data related to process impurities will be
skewed towards the lower bound (approaching a value UCL 99.865th Percentile
of 0). Some parameters like cell viabilities would be
LCL 0.135th Percentile
skewed toward the upper bound (approaching a value
of 100%).

5
Setting Up the Control Limits.
How data is distributed determines the procedure for setting up in-process control limits for the continued process
verification program (Figure 2).

DETERMINING CONTROL LIMITS

How the data is distributed determines the procedure for setting up
in-process control limits for the continued process verification program

QUALITY
ATTRIBUTE

NO Is the data YES

normally
distrubuted?

CALCULATE CALCULATE
LIMITS LIMITS
based on 99.865th and based on standard
0.135th percentile as deviations (1,2 or 3SD)
UCL and LCL and 50th for UCL and LCL. Calculate
percentile as center- Average for centerline
line (Exclude known (Exclude known
outliers) outliers)

SET CONTROL LIMITS

Figure 2. Determining Control Limits

Rules for Excluding or Including Data Points Setting Up Statistical Process Control (SPC)
The batches used for control limit estimation should Statistical process control (SPC) is an important element of
represent only the inherent process variability (common CPV and a process control chart (Figure 3.) plays the most
cause variations). To represent the true processes against important role in SPC and any process monitoring program.
which future process results can be compared, all batches A successful CPV system should not only create SPC charts
with special causes or assignable root-cause deviations from validated data, it should also store, display, and evaluate
(for example, arising due to process changes, equipment the control chart statistics based on historical limit changes.
failures, mechanical faults, operator errors etc.) should be Hence, an efficient CPV system should always have the
excluded from limit calculations. When a process change capability to capture not only historical data, but also the
is applied, the control limit calculations should restart for contextual knowledge (meta data) of historical changes in
monitoring parameters for the process steps affected with control limits.
change (and all impacted steps downstream of the affected
process step) starting from batche where the process change
is applicable.
Once the control limits for all the CPPs and KPPs (parameters
being monitored) are defined, it should be documented in
an In-process Control and Monitoring (IPCM) document.

6
 Process Control Limits

Process PHASE 1 PHASE 2 PHASE 3

Capability Cpk=1.39 Cpk=1.24 Cpk=1.35
UPPER LIMIT

2
2

Centerline 3

LOWER LIMIT
Fem1

Fem4

Fem7

Fem11

Fem14

Fem17

Fem20

Fem23

Fem27

Fem30
Trend Rules Violation Control Limits
Figure 3. Process Control Centerline Specification or Action Limits

Determining Process Capability (Cpk) and Process Performance (Ppk)

A manufacturing process is a unique combination of its characteristic or a process output parameter. Process
manufacturing environment comprised of machines, capability indices (Cp, Cpk,) provide a common metric
methods, and people engaged in the production process. to evaluate and predict the performance of processes
Process capability indices provide a quantitative way to and summarize process performance relative to a set of
determine the performance and future capability of the specifications (i.e. quality boundary).
process within the desired quality boundary for the given
Periodic capability studies should be conducted to predict the
manufacturing environment.
overall ability of a continuous distribution process to make
Process capability indices have been used in manufacturing products within the required specifications.
to provide quantitative measures on process potential and
performance. The output of a process can be a product

Estimating Process Capability for Estimating Process Capability for

Normally Distributed Data Non-Normally Distributed Data
Process capability (Ppk, Cpk) for a normally distributed Since the average and standard deviations will not represent
monitoring process parameter will be calculated using the non-normally distributed data correctly, process capability
the following: (Cpk) cannot be estimated for non-normal data. Instead
Process performance (Ppk) will be evaluated based on all of
data points in terms of percentile ranges. For a non-normally
distributed monitoring process parameter, Ppk will
be calculated using the following:

Where Where
USL = Upper Specification Limit (for CPP) or Upper Action
USL = Upper Specification Limit (for CPP) or Upper Action
Limit (for KPP)
Limit (for KPP)
LSL = Lower Specification Limit (for CPP) or Upper Action
LSL = Lower Specification Limit (for CPP) or Upper Action
Limit (for KPP)
Limit (for KPP)
X0.50 = Median of the population under analysis
Avg = Average or mean of the population under analysis
σ = Standard deviation of the population under analysis X0.99865 = 99.865th Percentile of the population
under analysis
σMR = Moving Range Standard Deviation
X0.00135 = 0.135th Percentile of the population under analysis

7
Parameter monitoring
During process monitoring, a performance parameter Preliminary Process Monitoring - PPM Phase
under the monitoring program will undergo different (Data gathering phase)
monitoring modes (or phases) depending upon the 1.10

number of data points and accumulated history of 1.05

the parameter being monitored. 1.00

0.95
Monitoring phases 0.90
0.85
Different monitoring phases (Figure 4) are explained
0.80
in the following diagram.
0.75

Preliminary Process 0.70

0.65
Monitoring Phase

2414

2514

2614

2714

2814

2914

3014

3114

3214

3314

3414

3514
The Preliminary Process Monitoring (PPM) phase is
primarily a data collection or accumulation phase
for obtaining enough historical information to make
reasonable assumptions on inherent variability
(common cause variation) of the parameter and
estimate a statistical control limit with reasonable Setting Alert Limits
1.20
confidence. The PPM phase should run for a minimum
1.15
of 15 data points. During this phase no statistical
1.10
control limits will be applied to any parameter types.
1.05

Statistical Process Control 1.00

(SPC) Phase 0.95

0.90
Once the basic history of a parameter (capturing 0.85
common cause variability) is accumulated, control 0.80
limits will be established using statistical methods
414

514

614

714

814

914

1014

1114

1214

1314

1414

1514
and procedures described in earlier sections.
All parameter data received after the PPM phase +3SD 1.15739 -3SD 0.825944 +250 1.102149
will be trended against these newly established -2SD 0.881185 +1SD 1.046905 -1SD 0.936426
control limits. Avg. 0.991667

Resetting or Revising
Control Limits
Control limits should be periodically re-evaluated,
1.4
revised, or reset as appropriate based on the PPM Phase SPC Phase
1.3
following:
1.2
Enough Batch History 1.1
Alert limits should be periodically re-evaluated after
1.0
every 25-30 batches after entering the SPC phase.
This will prompt tightening or widening of limits 0.9

based on further accumulated process knowledge. 0.8

The batches used for limit evaluation should be free 0.7

from any special cause variation. 0.6
414
514
614
714
814
914
014
114
214
314
414
514
814
914
014
114
214
314
414
514
614
714
814
014
115
215
315
415

Process Changes
If changes to processes are introduced after entering
Upper Alert 1.157 LowerAlert 0.826 Rule 1
the SPC monitoring phase, the alert limits should
Lower Action 0.7 Target 0.992
be reset, and all parameters affected downstream
of where the change was introduced should enter
the PPM phase to re-gather history for new process
conditions until it enters SPC phase again.

Figure 4. Process Monitoring Phases

8
Trending Rules
Some form of Nelson (or Western Electric) rules should be Examples of Rule Definitions
established and used for out-of-trend detection as a method to
The first four Nelson rules that can be used to trend
determine if some process parameters are out of control.
parameters and detect out-of-control batches are described in
Figure 5 below.
Rule 1: One point is more than standard
deviations from the mean

Rule 1. UCL 3σ One sample (two shown

in this case) is grossly
One point is more 2σ
out of control.
than 3 standard
1σ
deviations (UCL,
LCL) from the mean. X

LCL

Rule 2: Nine or more points in a row

are on the same side of the mean
UCL 3σ
Rule 2. Some prolonged
2σ bias exists.
Nine (or more)
points in a row are 1σ
the same side of the
mean. X

LCL

Rule 3: Six or more points in a row

are continually increasing or decreasing
3σ
Rule 3. UCL A trend exists.
2σ
Six (or more)
points in a row 1σ
are continually
increasing (or X
decreasing).

LCL

Rule 4: Fourteen or more points in a row

alternate in direction, increasing then decreasing
UCL 3σ
Rule 4. This much oscillation
2σ is beyond noise. This
Fourteen (or is directional, and
more) points in a 1σ the position of the
row alternate in mean and size of the
direction, increasing X
standard deviation
then decreasing. do not affect this rule.

LCL

Figure 5. Nelson Rules

9
Rule Violation Investigation Approach
A batch violating any of the four trending rules should be should be consulted appropriately with a formal feedback
properly investigated and closed with appropriate corrective form. These feedback forms should be documented within
and preventative actions (CAPA). Since these investigations the process monitoring group and summarized (and attached
do not impact the disposition of a batch, it will not follow as appendix) in the periodic process monitoring summary
the rigor of a formal quality management system; however, reports. An example of the investigation form that can be
the process monitoring group should be responsible used by SMEs is shown in Figure 6.
for tracking all such investigations related to trend rule
violations. Furthermore, subject matter experts (SMEs)

Figure 6. Feedback Form

10
Documentation and Reporting
Routine monitoring status should be periodically presented to Periodic Presentation of Process Trends
key process stakeholders within the organization and reports
All monitored parameters should be routinely trended on a
published to document the process knowledge as explained in
weekly basis for each commercial process with respect to
the following sections.
each processing location (manufacturing site or manufacturing
suite). When commercial operations are ongoing, key findings
In-process Control and Monitoring
(alerts, etc.) from routine monitoring should be presented to
(IPCM) Document key stakeholders at a regular frequency (preferably every 2
An IPCM document should be prepared for each commercial to 3 weeks; not exceeding 4 to 6 weeks). Key stakeholders
process for each manufacturing location. The document should should include people from manufacturing, process
list all performance process parameters (CPP, KPP and MP) development or manufacturing sciences, quality control
that will be routinely monitored with appropriate justification. and quality assurance.
This document should also list process limits (specification,
action and control limits) for all monitored parameters along Quarterly Process Summary Reports
with suggestions on statistical treatment of each parameter
Process Summary Reports should be published every quarter
based on distribution of data. This document should be
for each commercial process per manufacturing location. At
periodically revisited for any additions or removals of a
a minimum, the summary report should contain the following
monitored parameter, changes in process limits, or statistical
sections:
treatment. All changes to the monitoring plan should be added
as addendums or annexures. It is suggested to have a limit • List of batches produced
change log for each monitored parameter in this report. This • High Impact Process Deviations
document should be version controlled. At a minimum, the
• Root cause (if identified)
IPCM document should contain the following sections:
• CAPA (if identified)
• Process flow chart •  Impact on monitoring (include/exclude from future
• 
Table providing a list of process monitoring parameters monitoring statistics)
(CPP, KPP, and MP). The table will include the following: • Trend Rule Violations
• 
Justification for including the parameter in the • Nelson 1 violations
monitoring program and reason for monitoring • Nelson 2,3 and 4 violations
• 
If any KPP or CPP is excluded from the monitoring • Assessment of Process Capability
program, a justification as to why should be recorded
• CAPA
• 
Report ID of the reference report where justification of • Trend CAPAs identified
why a parameter is a CPP or KPP is given
• Status of CAPA for last quarter report
• 
Table providing Specification Limits for CPP and Action
• 
Duly filled Nelson assessment forms for each violated
Limits for CPP and KPP which should also include:
parameter (Annexure)
• 
Observed distribution of model data set for
• Process Trend Charts (Annexure)
limits calculation
• 
Report ID of reference report which where the source of Site Trend Violation and CAPA
specification and action limits is given Report (Quarterly)
• 
Table providing Control Limits (UCL, LCL, and CL) of all Overall operations trend violations and CAPA reports should
monitored parameters (CPP, KPP and MP) be prepared at periodic intervals (quarterly recommended)
• 
Change log of all control limits when the limits are revised for all products and processes for all manufacturing locations.
referencing the report ID of investigation that lead to The reports should highlight the differences or similarities of
change of limits violations across products, processes and locations.

• 
Distribution Q-Q plots of all monitored (CPP, KPP, MP)
parameter (Annexure)

11
Summary
CPV Flow Chart
As described in earlier sections, the entire CPV process flow is 11 steps (Figure 7).
CPV PROCESS

Identify
CPP, KPP Establish a
and MP from Identify Get validation technique for
performance and parameters ranges and statistical control
1 operating parame-
ters from process
2 that need to
be trended
3 specification
limits (if any) 4 limits and
frequency at
characterization/ periodically for these which to update
validation parameters the control
documents limits

Evaluate the
root cause of each Publish an
rule violation and Monitor in-process control Establish
parameters and monitoring
its impact on
product and 8 against set 7 document for 6 trending rules
(Nelson rules) 5
process. Initiate limits each commerical
appropriate CAPA process
if needed

Publish
quarterly process Update
summary reports control limits
(can feed APRs). Update in the system
9 Will have recom-
mendations for
10 control limits 11 upon every limit
update via a
process improve- change control
ments and limit
updates

Figure 7. The entire CPV process flow

CPV steps automated with Bio4C™ ProcessPad

Identify
CPP, KPP Establish a
and MP from Get validation technique for
Identify
performance and ranges and statistical control
parameters that
specification
1 operating parame-
ters from process
2 need to be
trended
3 limits (if any) 4 limits and
frequency at
characterization/ for these which to update
periodically
validation docu- parameters the control
ments. limits

Evaluate the
root cause of each Publish an
rule violation and Monitor in-process control Establish
parameters and monitoring
its impact on
product and 8 against set 7 document for 6 trending rules
(Nelson rules) 5
process. Initiate limits each commerical
appropriate CAPA process
if needed

CPV Steps Automated by

Publish Bio4C™ ProcessPad
quarterly process Update
summary reports Validated data
(can feed APRs). Update
control limits
in the system source 12
9 Will have recom-
mendations for
10 control limits 11 upon every limit
update via a
process improve- change control
ments and limit
updates

Figure 8. CPV steps automated with Bio4C™ ProcessPad

12
Bio4C™ ProcessPad CPV Platform
Bio4C™ ProcessPad is a data visualization, analytics, and Why is Bio4C™ ProcessPad a
process monitoring platform that enables bioprocess lifecycle Comprehensive CPV Solution?
management, reporting, investigations, and continued process
verification. It stores the batch processing and analytical Continued Process Verification is the lengthiest and most data
testing data in a validated state so that the data can be heavy phase of process validation. It requires continuous
easily available for process troubleshooting and process monitoring of processes against defined limits (specification,
benchmarking operations. It also aids organizations in action and statistical control limits) throughout the commercial
life of a product. Bio4C™ ProcessPad is a comprehensive CPV
CPV Documentation
their compliance efforts by meeting all the applicable
solution which automates many of the business processes in
requirements of 21 CFR Part 11.
a CPV program (Figure 9).

2
1
IPCM Document
CPV Governing (In-process Control and Monitoring)
SOP A comprehensive monitoring document
for each product to be monitored
Overall procedures,
program requirements
and flow charts
SME
Identification

SME SME
Input Input IPCM
Docuent Preparation
• Approach for statistical
Process Control (SPC)
• Handling normal vs.
non-normal data etc.
Parameter • Frequency Monitoring
Identification

Parameters are
classified as CPP,
KPP or MP

3 One IPCM per product. Revised

every 30-50 batches with
revised control limits
CPV Reports
Trend report of each product
for monitored parameters
published periodically.
Each report will have trends of
performance parameters and a
brief investigation of each trend
observation with appropriate
CAPA if needed

One per product.

Typically published
every quarter or
on-demand

Figure 9. Documentation required to roll-out a CPV program and its organization.

13
 FDA Expectations for CPV
An Ongoing Program for Collecting and Analyzing Product and Process Data that
Relate to Product Quality
• Procedures for data collection and trending
• Data collected should verify the quality attributes
• Intra-batch and inter-batch variation
• Data should be collected to evaluate process stability and capability
• Data should be statistically trended
• I t is recommended that a statistician or person with adequate statistical training develop the data collection plans and
methods for analysis

Must Have a System for Detecting Unplanned Departures from the Process
• Evaluate the performance of the process
• Identify problems
• Determine if corrective action is necessary
• Anticipate and prevent problems to ensure control

Key Differentiators of Bio4C™ ProcessPad

Validated Data Source Plug-and-Play Monitoring
Bio4C™ ProcessPad is a validated data source. All the data Bio4C™ ProcessPad was designed from the ground up for
entered in Bio4C™ ProcessPad goes through two layers process monitoring. The following features make Bio4C™
of verification to ensure that data entered is accurate ProcessPad an ideal tool to monitor your processes:
and reliable. Bio4C™ ProcessPad has stringent access
1. Create and schedule creation of control charts
level control. For example, a user uploading data does
not have permissions to approve the data; similarly, an 2. 
Version control statistical limits revisions and log
approver does not have permissions to upload data. Bio4C™ historical contextual process knowledge
ProcessPad also has audit trail which captures every action 3. 
Storage of historical limits and displaying in a single chart
performed by all users. limits revisions and process performance within each limit
revision
4. Automatic calculation of statistical trend rule violation
5. Automatic capability analysis (Cpk and Ppk calculation)
6. 
Prediction of data distribution (normal or non-normal
data distribution)
7. 
Perform linear correlations with parameters across
process steps
Bio4C™ ProcessPad offers out-of-the-box data visualizations
and analytics than can accelerate investigations and root.

14
Report Generation (CPV/APQR) Bio4C™ ProcessPad has contextual connection between
process steps (unit operations) giving users easy access to
CPV reporting is a tedious task. It involves collection of data
parameter correlations between parameters from across
from all the batches executed in that time frame and trending
process steps with advanced data visualization technologies.
them with statistical control limits, calculation of process
Investigators can easily generate complete lot genealogies
capabilities (Cpk and Ppk) and sighting the statistical trend
on demand. Other commercially available machine analytics
rule violation.
software applications are primarily focused on upstream
Standardized campaign (CPV) reports can be prepared in process data with monitoring via MVDA techniques and
Bio4C™ ProcessPad. Report templates are prepared by ignoring important offline/at-line data analysis and trending.
selecting the desired parameters from all the process stages There are no capabilities in these software where users
into a trending template. This template can then be scheduled can find downstream process data monitoring or CTQ/CPP
for automatic periodic report generation. management and monitoring. Bio4C™ ProcessPad’s golden
tunnel dashboards for machine data provides a simple
and extremely easy way to monitor live processes in
Salient Features of Bio4C™ ProcessPad historical context.
in Comparison to Traditional Software Bio4C™ ProcessPad is the only truly browser-based software
There is numerous software available on the market which available in the market for CPV. Being 100% browser based
claims to do CPV reporting but none of them perform the and built using the latest web technologies (which are
combined task. For example, Microsoft® Excel can be used inherently collaborative), Bio4C™ ProcessPad is well suited for
to make control charts with defined control limits, but it is collaboration, scale, and performance. Offered as a site license
extremely difficult to define (and validate) trend rules and for unlimited users makes it economical and easily adopted
calculate process capability within the charts. Similarly, other by various stakeholders in the organizations. Applications
desktop spreadsheet-based advanced statistical software can be accessed via a web browser within the company IT
packages in the market although good spreadsheet software network. Other software in the market are licensed per user
with the ability to trend charts preparation and perform and requires organizations to install software on each local PC
process capability calculations, they lack any ability to making it expensive and difficult to manage/scale.
validate archived data, which breaks the very first rule Other competing software applications have long learning
of CPV i.e. data sources must be validated. curves and are difficult to operate and master. Bio4C™
Bio4C™ ProcessPad has an embedded database with a ProcessPad user interface is extremely simple with minimal
workflow to capture and store batch record or experiment training requirements. Bio4C™ ProcessPad can be deployed
execution data in a 21 CFR Part 11 compliant manner. All within weeks at a site as compared to months in the case of
data entered into Bio4C™ ProcessPad is part of an audit trail. other software.
This is lacking in traditional software which does not have any The whole architecture of Bio4C™ ProcessPad has been
process of capturing and storing in-process parameters. As designed and purpose-built for pharmaceutical and
Bio4C™ ProcessPad stores the data in a database, this data biopharmaceutical organizations making it easily adoptable
can be made available to the end user on demand. As far within these organizations at all user levels. On the other
as competing software are concerned, they are standalone hand, other software is focused on advanced analytics which
desktop applications without any proper data capture or only expert data scientist in the organization can understand
data storage mechanism. These traditional software and use to the full extent. Bio4C™ ProcessPad is a tool for all
applications need to be fed data from an externally levels of expertise rather than few statisticians.
aggregated spreadsheet whenever performing analysis.
Bio4C™ ProcessPad has an evolved system to perform
continued process verification with automated reports on CTQs
and CPPs. It supplies out-of-the- box trend analysis with trend
violation detection using Nelson or Western Electric rules.
Furthermore, it has an advanced system of recording control
limits and their revision history to give insights on evolving
process capabilities which is required for any MSAT or process
support R&D team.

15
For additional information
please visit EMDMillipore.com/Bio4CProcessPad
To place an order or MilliporeSigma
receive technical assistance 400 Summit Drive
Burlington, MA 01803
please visit EMDMillipore.com/contactPS
or email ProcessPadSupport@EMDGroup.com

Preparation, Separation,
Filtration & Monitoring Solutions

© 2020 Merck KGaA, Darmstadt, Germany and/or its affiliates. All Rights Reserved. MilliporeSigma, the vibrant M, Millipore, BioContinuum and Bio4C Lit. No. MS_WP6539EN Ver. 1.0
are trademarks of Merck KGaA, Darmstadt, Germany or its affiliates. All other trademarks are the property of their respective owners. /2020
Detailed information on trademarks is available via publicly accessible resources.