ASSESSMENT,

QUALIFICATION &
VALIDATION
Managing Risk and Ensuring
Success With a Contract BEG IN
Manufacturing Partner

Produced in coordination with:

 MORE THAN DUE DILIGENCE
Any long-term commercial relationship, especially between a drug innovator and its
CMO, requires a number of things to remain sound: ongoing dialogue, common expecta-
tions, milestones and metrics, responsiveness, secure data access and more. With time,
these relational and transactional exchanges build a collaborative bridge that pushes
risk out and business success into these highly technical and complex relationships.

The question remains, though, how do you get there from here? How does a prospec-
tive drug innovator successfully and strategically assess prospective partners to en-
sure a long-term and successful relationship? If you are reading this, chances are good
you are pondering the same thing. To help gain wisdom, we’ve gathered together a
sampling of the collective knowledge of both leading Pharma manufacturing experts
and the expert staff at WellSpring Pharma Services. What follows is an offering of
“best” advice on how to create a successful CMO partnership right from the start.

What Makes a Successful Due Diligence

When In-Licensing a Pharmaceutical Product? page 3

What Are Feasibility Studies and How Are

They Used in the Pharmaceutical Industry? page 5

What Considerations Should Be Made

When Planning A Feasibility Study? page 6

Tech Transfer’s New Framework page 7

Validation = Process Design page 16

What Are IQ, OQ, and PQ, and Why Are They Required? page 18

t PREVIOUS PAGE 2 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

What Makes a Successful Due

Diligence When In-Licensing a
Pharmaceutical Product?

WHEN A company is looking at acquiring or in-licensing your company hundreds of thousands of dollars in lost
a pharmaceutical product, there are a few technical factors revenue. The availability of a $10/kg material could cost
that should be carefully considered before the deal is com- you months of lost market sales.
pleted. Navigating through all of the technical documents We’ve had a couple of clients where they’ve acquired
can be very time consuming and confusing. The techni- or in-licensed a product and wanting to start the transfer
cal parameters that a company needs to evaluate can be work only to realize that in one case the API supplier
complex. You need to glean as much technical information discontinued manufacturing that API so it was no longer
as possible on the manufacturing and packaging process available. In another case, the lead time was close to eight
itself as well as from Quality Control, Quality Assurance, months in getting an API sourced. It could be a couple
Regulatory and Logistics of a pharmaceutical product. of years by the time you find and qualify that new API
You should go into the product due diligence with a supply, get it into your process, qualify the process and
detailed checklist to ensure you leave no stone unturned. get the qualification lots placed on stability. In extreme
The goal, of course, is that you come out of the due cases, you may not be able to launch the product at all.
diligence process fully aware of all the issues and
potential issues of acquiring a pharmaceutical product EXPERIENCED TECHNICAL HELP CAN HELP YOU
— new or existing. We’ve seen some clients acquire new AVOID POTENTIAL PITFALLS
products thinking everything was great, only to find out, If you’re a virtual company, you may not have the techni-
for instance, that their methods were not USP standard cal expertise to really sit down and understand what the
methodologies, even though their plan was to market the manufacturing and packaging processes are or what the
product as USP standard. product nuances are. How well were they documented in
development reports? Is there even a development report?
WHAT ARE THE ELEMENTS IN A TECHNICAL PACKAGE Because a lot of older products don’t have that documenta-
THAT A PHARMACEUTICAL COMPANY MIGHT LOOK tion. Is there alcohol in the product? Many manufacturers
AT WHEN TRYING TO IN-LICENSE A PRODUCT? do not handle alcohol or other solvents in their facility and/
Analytical methods are one of the critical factors to review or equipment due to safety or environmental issues — can
right off that bat. Analytical methods have to be current to yours? Due diligence of all of these factors needs a really
today’s regulatory standards, but it even goes back further critical, experienced eye looking at and understanding just
than that, based on the product’s formulation composition. exactly what it is you might be buying.
Is there supply of API? Is it reliable? Have you talked to the If there is specific equipment that’s required and you
API suppliers? Maybe the company has some API in house need to outsource the manufacturing of that product, how
but their supplier is dropping out. How do you check that easily is that equipment available at your potential contract
out? You need to call the suppliers of all the ingredients — manufacturing partner sites? Are some of the excipients
API and key functional excipients. It’s possible that they even available, like we mentioned before? How well were the
are no longer being made by these suppliers. If that is the process parameters developed? Does the product require
case, this could delay the launch of your product and cost a lot more development work in order to get it to the point

t PREVIOUS PAGE 3 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

where it’s approvable by a regulatory authority or is it an LOOK AT THE PRODUCT’S COMPLAINT HISTORY
easy transfer? What you may think is an easy transfer can TO IDENTIFY QUALITY CONCERNS
turn into a long and very costly exercise in order to have the Lastly, when considering product quality, make sure
product in a position where you can market it. you look into the complaint history for the product —
If your company is considering buying a it could give valuable insight into potential manufac-
pharmaceutical product, you should really consider turing issues and whether they have been adequately
engaging the help of your existing manufacturing partner addressed. Were there any FDA 483s issued related
in the process right up front. The sooner they’re involved, to the manufacturing or testing of the product and
and if that’s the site where the product will be transferred were those satisfactorily addressed? Were these issues
to, the fewer unexpected challenges you will face down adequately documented and reported to the regula-
the road. Hopefully those issues can be identified with tory agency? Are there any documented commitments
enough time so that solutions can be developed that will that the regulatory agency, like FDA, is expecting to
not delay your market introduction. be completed that your company may have to assume
The technical people at your contract manufacturing site once you acquire the product? Again, these issues
would probably have enough expertise to be able to look could delay the product getting to market and may be
at a file objectively and find potential pitfalls within that very costly to complete.
product. For example, if an analytical method is not specific Time to complete a due diligence is usually very tight.
enough or precise enough they may be able to identify You would be well served by developing a thorough check
that or uncover that for you in advance. If there’s a certain list that spans manufacturing, quality, regulatory and
processing parameter that hasn’t been well qualified or well logistics. By doing this upfront, it makes the transfer so
documented they’ll be able to tell you that also. They’ll be much easier — fewer surprises along the way means a
able to help you avoid potential problems down the road. smoother tech transfer.

t PREVIOUS PAGE 4 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

What Are Feasibility Studies

And How Are They Used?

A FEASIBILITY study is probably the single most impor- should be made to the next set of experiments. Flexibility
tant consideration in the development of a drug. It’s where in experiment design is an invaluable asset; be prepared
we begin to elucidate the formula, the process, the equip- to set off on a novel path. Overly strict adherence to one’s
ment train — where we identify the upfront variables that plans can prove more wasteful in the long run than simply
must be reckoned with. When developing a new process, starting anew. Active Pharmaceutical Ingredients (API)
whether it concerns a new product or an existing product, may be in short supply and tend to be expensive; minimiz-
issues with quality or an inefficient process, it’s always ing their usage is of primary concern.
necessary and beneficial to develop a feasibility plan. Don’t be afraid to challenge raw materials, vendors,
A good feasibility plan should clearly state the grades of materials and orders of addition. It can be
objectives of the project and will entail a series of helpful to vary quantities to understand their impact on
experiments to determine what preexisting issues may both process and formulation: the physical properties of
exist. In the absence of potential issues, it should identify each raw material, the equipment itself and what both
the critical processing factors that coalesce to consistently lend to the final product.
manufacture and reproduce a quality product. Analytical methods need to be onboard, fully developed,
and stability indicating. Given proper expertise and a clear
DEVELOPING A FEASIBILITY PLAN definition of the properties and characteristics of the end
Feasibility plans can take a variety of forms, but should product, it should be possible to identify formulation and
always highlight: process within 10 to 15 sets of experiments.
• Exactly what experiments will be performed
• What nuances such experiments are designed to elucidate THE BENEFITS OF MANUFACTURING A SMALL-
• W hat materials will be used SCALE BATCH FOR YOUR PROTOTYPE
• The equipment that should be identified Once a viable prototype is in hand, it can’t hurt to make
All of these considerations dovetail into what we refer to a small-scale batch and put it on stability under acceler-
as QbD: quality by design. With each set of experiments, we ated conditions for three to six months. While clients may
try to challenge as many variables as possible. Well-studied tend towards impatience and gun-jumping, waiting six
and correctly designed, a given set of experiments can months will give you a clear reading on whether or not
challenge three, four, even five variables simultaneously. you’ve got a product. Assuming that all is well after the
At this stage, it is of primary importance to ensure that waiting period, scaling up can be aptly considered.
everyone involved in an experiment is clear on its ultimate Results from feasibility batches can be used not only for
objective. Making and recording visual observations at analytical development but also for packaging purposes
each juncture of the process is not just beneficial, but as well, providing product for other groups within the
absolutely essential. Written comments in the batch record organization to perform their tasks.
can be indicative of other issues extant in the process, often Small scale feasibility studies can provide valuable
suggesting a change of course in follow-up experiments. feedback in determining and optimizing critical process
parameters such as mixing speeds, mixing times, drying
ASSESSMENT OF THE EXPERIMENT DESIGN IS KEY times and parameters, tablet compression, weight,
As each experiment reaches its conclusion, it’s important thickness, drying and coating. Feasibility studies
to pull the team back together to review written comments, confirm that the parameters we transpose into our batch
visual observations, and the batch record itself, parsing out records are the right parameters needed to consistently
what worked, what didn’t work, and what modifications manufacture a given product.

t PREVIOUS PAGE 5 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

What Considerations Should

Be Made When Planning
A Feasibility Study?

WHEN DEVELOPING a drug product, feasibility • Be flexible in your experiment design: Assess your
studies are significant in fine-tuning your processes from visual and written observations; discuss what worked,
the formulation stage to the final production stages. These didn’t work, and what changes to the experiments you
studies give opportunity to confront potential process can make for the next round. Be open minded in even
variables for either new products or existing. Overall, challenging new grades or vendors of raw materials,
mapping out a feasibility plan can improve inefficien- orders of addition, etc.
cies in the process and equipment train. To start, clear • Analytical methods need to be onboard, fully developed,
objectives of the project must be clearly identified, and and stability indicating. Given proper expertise and a
from there, a series of experiments must be put into place clear definition of the properties and characteristics of
to test out what pre-existing issues may be present. In the the end product, it should be possible to identify formula-
absence of potential issues, it should identify the critical tion and process within 10 to 15 sets of experiments.
processing factors that coalesce to consistently manufac-
ture and reproduce a quality product. CONSIDER MANUFACTURING A SMALL
SCALE BATCH TO TEST YOUR PRODUCT
KEY WAYS TO APPROACH YOUR FEASIBILITY AND PROCESS PROTOTYPE
STUDY PLAN Once a possible prototype is in hand, it’s very beneficial
Feasibility plans will not always take on the same form to make a small-scale batch. Small scale feasibility studies
or shape for each project — there will always be unique can provide helpful feedback in optimizing key process
strategies and challenges for every project in drug devel- parameters, such as:
opment. However, these studies should be mapped out • Mixing speeds and times
being mindful of these major considerations, which tie • Drying times
nicely together to achieve QbD — Quality by Design: • Tablet compression
• Identify the key experiments that will need to be con- • Weight
ducted, their purpose, and what beneficial information • Thickness
their execution could reveal. • Coating
• Set your objective to test as many variables as possible
— a correctly designed set of experiments can confront Feasibility studies confirm that the parameters
as many as 3 to 5 variables concurrently. are cut right, and consistently deliver quality when
• K now the equipment at hand, and what materials need manufacturing a product.
to be used. Another great idea is to put a feasibility lot on stability
• Ensure that you have a clear-cut means of recording and under accelerated conditions for three to six months.
documenting every observation at every stage of the We know, we know — waiting six months can test
process being challenged. your patience, but in reality, that six months will reveal
• Be attentive: Review and dissect your observations to whether or not you’ve got a product, and it’s worth the
see if they reveal new issues in the process that may not wait. Assuming that all is well after the waiting period,
have been otherwise considered. scaling up your manufacturing process can be explored.

t PREVIOUS PAGE 6 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

Tech Transfer’s New Framework

HOW PROCESS VALIDATION GUIDANCE SIMPLIFIES TECH TRANSFER,
ESPECIALLY FOR LEGACY PRODUCTS.

By Bikash Chatterjee and Mark Mitchell, Pharmatech Associates

THE TECHNOLOGY transfer of a process, whether it organization, with references to the business unit and
is from R&D to commercial manufacturing or to another technology transfer team assembled for the project.
site or contract manufacturing organization (CMO) is a Through this real-life example, we will discuss not
critical step in the lifecycle of any drug product, involving only the approach taken to establish the design and
many steps. As major blockbuster drugs come off patent control space for the final process, but also the Process
and large pharmaceutical companies look to bolster their Performance Qualification (PPQ) study design and
pipeline through acquisition, the control and consistency acceptance criteria for Stage 2 and the approach taken to
of development data can vary dramatically. To make mat- satisfy Stage 3 of the new PV guidance.
ters more complicated, the new Process Validation (PV)
Guidance issued by FDA in January 2011 now defines THE NEW PV MODEL
three major stages of process validation that must be Under FDA’s 1987 guidance, Process Validation could be
satisfied to consider the process validated. characterized as “quality by sampling and testing,” while
With the present article, we will lay out a practical the new guidance would more appropriately describe
approach that addresses this complexity and propose to validation as “quality by design and control.” Let’s look
discuss and summarize the diverse factors required to closer at the three distinct stages that make up the new
describe the process, establish the control strategy and definition of process validation:
specify the acceptance criteria to successfully transfer a
legacy or newly acquired process to another process train • Stage 1: Process Design: The commercial manufac-
and satisfy the new guidance. turing process is based on knowledge gained through
To illustrate, we will take a closer look at the development and scale-up activities.
methodologies employed and the challenges encountered • Stage 2: Process Qualification: The process design is
as part of a recent technology transfer process validation evaluated to determine if the process is capable of repro-
exercise executed for a legacy product for a client ducible commercial manufacturing.

Stage 1: Process Design

• Define the Knowledge Space
• Identify Critical Process Parameters
• Determine Control Strategy
2011
FDA Process
Validation Stage 2: Process Qualification
• Equipment / Utility / Facility
Guidance Qualification
• Process Performance Qualification

Stage 3: Process Monitoring

• Monitoring of Critical Process Parameters as Part of
APR and Other Monitoring Programs

Figure 1. The FDA Process Validation Model

t PREVIOUS PAGE 7 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

• Stage 3: Continued Process

Verification: Ongoing assurance is
gained during routine production Max Set
Variability of actual data around set point
that the process remains in a state Point Run(s)

of control.
The PV roadmap uses a milestone- PAR
driven framework creating a phase- NOR
gate process for each stage of the new Target Set Point

process validation lifecycle as shown

Limit of Individual
in Figure 1. excursions

Min Set
FOCUS ON THE CONTROL OF Point Run(s)
PARAMETERS INSTEAD OF THE
TESTING OF ATTRIBUTES
Duration of Process
As the new PV guidance states:
• Quality, safety and efficacy are de- Figure 2. Relationship between PAR and NOR
signed or built into the product.
• Quality cannot be adequately
Establish
assured merely by in-process and fi- Product Design
PAR/NOR
nal product inspection and testing.
• Each step of a manufacturing
process is controlled to assure the Process/CPPs/Risk PPQ
Assessment Prerequisites
finished product meets all quality
attributes including specifications. Continuous
Improvement
Defining a knowledge space
Process Reproducibility
Process Understanding

Historical
PPQ
Performance
relating process parameters and

Process Monitoring
material attributes to quality Risk Assessment
Verification
attributes allows us to establish a Equipment Design
Risk Assessment
control strategy around the most Verification

critical process parameters. Stage

1, Process Design, encompasses Characterization
Change Control
and Stage 3
identification and control of critical Studies
Recommendation
process parameters to provide a
high level of assurance that the Figure 3. Technology Transfer/PV Framework
critical quality attributes for the
entire lot will meet the defined
limits. In-process and finished attributes. Process Monitoring also ICH Q8. The PAR is established with
product inspection and testing provides the continuing opportunity data; these data are usually gathered
on a relatively small sample of the to evaluate any emergent critical during Process Design. Commercial
lot become merely a confirmation process parameters, which may production lots produced outside a
of that control. Stage 2, Process occur as a process, or as materials, PAR for a critical process parameter
Qualification, is a demonstration equipment and facilities mature and represent unknown quality and would
of that control of critical process potentially drift over time. be technically unsuitable for release
parameters and their prediction The key to control of a critical despite acceptable in-process and final
of critical quality attributes, both process parameter is to characterize product inspection and testing.
within lot and lot-to-lot. Stage 3, the range for which operation within Many companies establish a
Process Monitoring, is the ongoing this range, keeping other parameters tighter range for production control
verification that critical process constant, will result in producing called a Normal Operating Range
parameters remain in control and product that meets certain critical (NOR), frequently seen on batch
continue to predict the outcome quality attributes, or the Proven records. In these cases, excursions
of the testing of critical quality Acceptable Range (PAR) as defined in of a critical process parameter

t PREVIOUS PAGE 8 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

outside the NOR require a quality 34.5

investigation to confirm that the PAR
has not been exceeded. The NOR UCL = 34.085
frequently represents the qualified 34.0

Temperature (ºC )
limits of the control system used for
the critical process parameter.
33.5
One possible relationship between X = 33.427

the PAR and NOR is shown in

Figure 2. The PAR limits are set by 33.0
the minimum and maximum set
LCL = 32.770
point runs for the critical process
parameter where the product meets 32.5
its quality attributes. The actual
5 10 15 20 25 30
data for the parameter will vary
Lot
around the chosen set point, shown
in the diagram by the shaded areas
Figure 4. Control Chart of Product Bed Temperature for the Granulation Process
around the set point. Here, the NOR
is shown as a narrower limit than LSL USL
the PAR. The NOR was determined Process Data
by the qualified control limits of the LSL....................................... 32
Target..................................... *
parameter when operating at its set USL....................................... 35
point; the NOR is used for the batch Sample Mean.............. 33.4273
Sample.............................. N 33
record limits of normal production
StDev (Overall)............. 0.28969
data. The extremes of individual
excursions around the set point Overall Capability
Pp..............................1.73
limits of the PAR may be used to PpL............................1.64
justify limited duration deviations, PPU ..........................1.81
which may occur in production. Ppk............................1.64
Cpm ..............................*

LEGACY PRODUCTS VS. NEW

MOLECULAR ENTITIES 32.0 32.4 32.8 33.2 33.6 34.0 34.4 34.8
Legacy products represent a unique Observed Performance Exp. Overall Performance
challenge for technology transfer and PPM <LSL....................0.00 PPM < LSL......................... 0.42
PPM>USL....................0.00 PPM > USL ........................ 0.03
PV because of the inconsistency in PPM Total....................0.00 PPM Total ........................ 0.45
terms of the development information
available. NMEs have the advantage
of gaining process understanding at Figure 5. Product Bed Temperature for the Granulation Process

small scale, with a focus on scale-up

and/or tech transfer. The ability to
identify critical process parameters at tion work as well as manufacturing THE TECHNOLOGY TRANSFER
small scale has economic advantages experience to continually improve FRAMEWORK
and also provides greater flexibility in their processes. Implementation of the Gone are the days of simply compar-
terms of experimental design. Using recommendations in this guidance ing product performance against its
the ICH Q8 definition, it is possible to for legacy products and processes release specification. The objective
move from the knowledge space to the would likely begin with the activities of technology transfer is to acquire
design space quickly and efficiently. described in Stage 3.1.” the necessary process and product
The new PV guidance recognizes this The big difference with legacy knowledge to establish a PAR and
and states: “Manufacturers of legacy products vs. NMEs as they relate to NOR for each unit operation that is
products can take advantage of the PV is that the baseline data gathering consistent with the predicate process
knowledge gained from the original activity begins in Stage 3 of the PV being transferred. Thus the new PV
process development and qualifica- lifecycle rather than Stage 1. guidance requires the demonstration

t PREVIOUS PAGE 9 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

of process reproducibility in the PPQ phase of Stage 2. Raw Material %w/w Function
Reproducibility effectively requires establishing acceptance
criteria that are consistent with the process stability dem- API 60 Active ingredient
onstrated in the predicate process. Reproducibility must be Microcrystalline
22 Excipient filler
defined for within lot and between lot variability as part of cellulose
the PPQ exercise. The technology transfer framework used Povidone K 29-32 5
Granulation
for this project is based upon Pharmatech Associates’ PV binder
model shown in Figure 3 and will be discussed as follows: Lactose 12 Excipient filler

Mg Stearate 1 Lubricant
PRODUCT REQUIREMENTS SPECIFICATION (PRS)
To illustrate, here is a case in point: The business unit Purified water QS Solvent
of a pharmaceutical company acquired the rights to a Coating Solution
%w/w Function
controlled release anti-hypertensive tablet. The tablet Raw Material
had been manufactured for 15 years outside the U.S. and
Eudragit Coating Controlled release
was to be transferred to the acquiring company’s main Solution
12
polymer
manufacturing site. A PRS was given to the development
team defining the critical-to-quality attributes for the Triethyl Citrate 1 Plasticiser
final tablet, including: Talc 1.5 Glidant
• Greater than 50 percent Active Pharmaceutical Ingredi-
Water QS Solvent
ent (API)
• Round 200-mg tablet Table 1. Formulation Details
• Coated to mask taste
• 12-hour drug release with the following specifications: Process Step Original Process Transferred Process
• 4-hour dissolution 20-40 percent 100 Liter tank
250 Liter Tank with
• 8-hour dissolution 65-85 percent Compounding with integrated
Tri-blender
Impeller
TECHNOLOGY TRANSFER MODEL: Fluid Bed Same Mfg. 350 Same Mfg. 350 kg
PROCESS UNDERSTANDING Granulation kg product bed product bed
PRODUCT DESIGN Milling Fitzmill Comil
The technology transfer package included the formula-
tion, raw material, API and finished product specifica- Blending 30 cu ft. Blender 100 cu ft. Blender
tions and master batch records. No development report 24 station tablet 36 station tablet
was ever written for the product. The team looked at the Compression
press, manual press closed loop
control with pre- control, and pre-
Chemistry, Manufacturing and Control (CMC) sec-
compression compression
tion of the non-disclosure agreement to understand the
36” coating pan, 48” coating pan,
composition and functionality of each component of the
Coating 3 spray guns, 4 spray guns,
formulation. The formulation is shown below. peristaltic pump peristaltic pump
The final product design revealed two key considerations Table 2. Comparison of Process Equipment
for the downstream process characterization studies. First,
the product has a fairly large loaded dose. This translates to
a potentially lower risk of content uniformity issues which CRITICAL PROCESS PARAMETERS/RISK ASSESSMENT
could translate to a more forgiving PAR and NOR for the In the absence of a development report, the team turned
final blend step. Second, the primary controlled release to a tiered risk-assessment approach for insight into
component is limited to the coating step, which means if the process design and sources of variability. The risk
the upstream process steps can be shown not to impact the assessment was divided into two parts. The first evalu-
final drug release profile this will simplify the final process ation compared each process step against the defined
validation argument. The raw material specifications were Critical Quality Attributes (CQA) in order to identify
either compendial or cut-sheet specifications from the which process steps would require close characterization.
supplier. Limited API characterization studies had been Process steps with a High rating were then further evalu-
performed. A comparison of the original process train and ated. The second tier of the risk assessment evaluated the
the new process train is shown in Table 2. potential impact of the process parameters. Parameters

t PREVIOUS PAGE 10 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

Process Steps
CQA
Granulation Drying Milling Blending Compression Coating
Appearance Low Low Low Low Medium High
Assay Low Low Low Medium Low Low
Impurity Low Low Low Low Low Low
Blend Uniformity Low Low Medium High High Low
Drug Release Low Low Low Medium Medium High
Particle Size
Medium Low High Low Low Low
Distribution
Justifications for N/A N/A Milling screen size Blending can Compression The final appear-
High Rating and speed can affect blend can affect drug ance and drug
affect the PSD uniformity, uniformity in release rate are
and therefore the assay, and the tablet based affected by the
powder flow and drug release upon particle coating quality
tablet fill weight profile size variability and reproduc-
control and flow ibility

Table 3. Process Unit Operation Risk Assessment

were divided into scale independent an impact on the CQAs across the and NOR were not considered
and scale dependent variables. Those final PAR and NOR. This was a Critical Process Parameters (CPP)
parameters that were identified as significant definition, which could and would not become part of the
having a High potential impact on have a profound impact on the final Stage 3 monitoring program.
CQAs were targeted for further number of parameters tracked in
study. Scale-dependent parameters the Stage 3, Continuous monitoring HISTORICAL DATA ANALYSIS
required further experimental portion of the PV process. Since The absence of development data
characterization. Scale-independent the objective of every process establishing the PAR and NOR for
parameters focused on an analysis of development exercise is to identify the CPP can be ascertained to some
historical performance. An example a process design and control space extent by evaluating the historical
of the risk assessment at the process which does not have an impact on behavior of each parameter along
level is shown in Table 3. the final product CQAs, parameters with the corresponding behavior of
The team also defined a process that did not move the product the CQAs for the unit operation. Data
parameter as critical when it had CQAs based upon their final PAR should be extracted from multiple
batch records to determine whether
the process is stable within lot and be-
480 UCL = 481.13
tween lots. In some cases, only mean
data or composite data may be avail-
Particle Size d50 (microns)

470
able. To do this, the team went back
into the batch records of approximate-
460
ly 30 lots across a period of one year
450
X = 452.11 to extract the necessary data. This
exercise also gave some indication as
440 to whether the parameter was truly
a CPP, based upon whether it had an
430 impact on the corresponding CQA for
the unit operation. The data for each
LCL = 423.10
420 unit operation were plotted as control
1 4 7 10 13 16 19 22 25 28 31 charts and the process capability was
Lot determined. Excursions outside the 3
Figure 6. Particle Size for the Granulation Process sigma limit of the control charts were

t PREVIOUS PAGE 11 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

Fitted Line Plot

d50 = 559.1 – 3.200 Coating Temp the experimental design which re-
S............. 10.2818
duces the number of runs required.
470
R-sq.............0.8% The new guidance is moving the
Particle size d50 (microns)

R-sq(adj)......0.0% industry toward a quality-by-design

460 philosophy for process validation.
This translates to a more parametric
approach rather than an attribute-
450
based approach to process design.
The application of a risk-based model,
440 considering the process and product
design at the outset of the technology
transfer project, allows the application
430
of scientific understanding to filter
33.0 33.2 33.4 33.6 33.8 34.0 34.2 34.4 the potential list of parameters that
Product Bed Temperature (°C)
may affect the process and product
CQAs to a limited few. The analysis
of historical performance reduces the
Figure 7. Correlation Between Particle Size and Bed Temperature number of factors that may need to be
characterized at the next scale. It also
provides a foundation for establishing
investigated to determine if there were influencing the particle size data and a baseline PAR and NOR for scale
deviations associated with the events. disrupting the correlation. independent parameters when
An example of the control chart and This approach was repeated based moving to the next scale, factoring
capability histogram for fluid bed upon the parameters that had a in the larger scale equipment design
product bed temperature is shown medium or high rating in the risk table. and configuration. Finally, applying a
in Figures 4 and 5. Capability limits For these scale independent parameters DOE approach to the few remaining
are based on a previously established the existing PAR ranges were used for scale dependent parameters will
PAR for the product bed temperature. the next phase of scale-up studies. establish the corresponding PAR
In addition, the corresponding and NOR for the transferred process
CQA for the process — particle CHARACTERIZATION STUDIES before moving to the process Control
size — was evaluated to determine For those parameters that were Stage of the roadmap.
if there was any impact from the scale dependent additional charac-
excursion. Figure 6 shows the terization studies were required to DEVELOPING AN EFFECTIVE
control chart for the particle size, establish PAR and NOR that were SAMPLING PLAN AND ACCEP-
the CQA for this process. A linear consistent with the predicate pro- TANCE CRITERIA
regression between the process cess. For simply scalable processes In this part of the article, we will dis-
parameter and the critical quality like blending, single time-based cuss the considerations in developing
attribute is shown in Figure 7. This blend uniformity studies may be ad- an effective sampling plan and accep-
indicates no statistically significant equate to identify the PAR and NOR tance criteria for the Stage 2 Process
relationship between the product for the new scale. For more complex Performance Qualification (PPQ)
bed temperature and the particle unit operations, such as the coat- and how to transition to the Stage
size through the range of data ing operation, a Design of Experi- 3 Continuous Monitoring phase of
examined. It is likely that product ments (DOE) approach may be more the new PV guidance. With the new
bed temperature would not meet appropriate. The team developed guidance, as in the original 1987
our definition of “critical process a series of balanced orthogonal guidance, moving to PPQ requires
parameter” from this data. However, experiments to establish the PAR completion of the following:
since historical analysis is not a for these parameters. This raises an- • Facility and Utility qualification
controlled experiment where all other good point to consider when • Equipment qualification (IQ,OQ
other parameters are necessarily confirming CPPs. By conducting the and PQ or equivalent)
held constant, there may be other historical analysis first it is possible • Analytical Method Validation is
parameters or material attributes to reduce the number of variables in complete and Measurement System

t PREVIOUS PAGE 12 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

Analysis (MSA) has concluded that the resolution of the must develop a statistical prediction for the acceptance
method is appropriate criteria of quality attributes which is typically much more
• Cleaning Validation protocol; Cleaning method devel- rigorous than simply meeting the specification limit.
opment and validation
• Upstream processing validation such as Gamma irra- SAMPLING
diation of components, for the new batch size Since the new PV guidance focuses on quality by design
• Environmental Monitoring program for the new facility and control, there is greater interest in the identification
• Master Batch Record and control of critical parameters to ensure that critical
• Qualification of in-process testing equipment, SMA, quality attributes throughout the lot are predictable. We
validation of method and SOP in place. cannot test the entire lot for the quality attributes, but we
In a technology transfer exercise, these elements must can control the parameters, and they should predict those
be applied to the new equipment and include the larger quality attributes. Sampling and testing now become a
commercial batch size consideration. If all the elements verification of what we should already expect to occur.
are not complete prior to beginning the PPQ runs then a A sample from a lot does not tell us the value of a quality
strategy may be developed, with the participation of QA, attribute since that quality attribute could be variable
to allow concurrent processing of the PPQ lot and process throughout the lot. In statistical terms, this is known as the
prerequisites. For example, if cleaning validation has not population. However, statistics can help us infer a likely
been completed prior to the PPQ runs, and the PPQ lots range of a lot’s mean value for a quality attribute, expressed
are intended for commercial release, then a risk-based as a confidence interval. We could also calculate a similar
approach to the cleaning validation may be adopted with confidence interval for the standard deviation of the lot.
studies conducted concurrently with the manufacture The mean of the sample values is not as important as
of the lots with the caveat that the lots are not releasable the calculated confidence interval (usually chosen as 95
until the cleaning validation program is complete. percent confidence) for the lot’s mean. This is because it
If such an approach is adopted, then consideration is the limit of the confidence interval that must meet our
must be given to both the major clean procedure, acceptance criteria, since we want to be able to infer that
typically performed on equipment when changing the true mean — and the true standard deviation — meets
products, and the minor clean procedure, typically the acceptance criteria, not just individual tested samples.
performed during a product campaign. To determine the acceptance criteria for PPQ lots, we use
In our case study process, all prerequisites were the process knowledge from the process design to make an
complete with the exception of cleaning validation, which estimate of the process mean — in other words, where the
was conducted concurrently. The new process site used process centers — and the process standard deviation — or
a matrix approach to cleaning validation, bracketing how the process varies around the center — for each critical
its products based upon an assessment of the API/ quality attribute. This allows for a statistical comparison of
Formulation solubility, potency, LD50 and difficulty-to- the PPQ lots’ means to the expected process mean.
clean profiles. For the purposes of the PPQ runs, only The comparison between two means is done using the
the major clean procedure was used between lots since “t-Test,” to evaluate any difference in two independent
the minor clean procedure had not been qualified. To samples. The acceptance criteria is successful when the
establish a PPQ plan that is efficient in demonstrating t-Test concludes that the difference between the lot’s
process reproducibility, the considerations for sampling population mean and the predicted process mean is less
testing and establishing acceptance criteria must be than the largest predicted variation of the predicted
thoughtfully considered, especially for products with process mean, calculated from the process standard
limited development or performance data. deviation. In statistical terms, this describes the alternative
To cite the PV guidance, the objective of the hypothesis (H1) of the t-Test:
Process Performance Qualification is to “confirm the
process design and demonstrate that the commercial H1: μ1 – μ 2 < (Target Difference)
manufacturing process performs as expected.” The
PPQ must “establish scientific evidence that the process Where, μ1 and μ2 are the predicted process mean and the
is reproducible and will deliver quality products population mean of the PPQ lot and the Target Difference
consistently.” It is clear that producing three commercial is the predicted variation in the process mean. For the
lots in a row to meet its specification limits is no longer t-test, when the null hypothesis (H0) is not significant, the
sufficient to meet process qualification objectives. We alternative hypothesis (H1) is concluded to be true.

t PREVIOUS PAGE 13 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

There are several methods of predicting the process Power Curve for 2-Sample t Test
mean and its variation from process design data: One-sided test: Alt. Hypothesis Less Than
1.0

1) Use a predictive model: When DOEs are used during

0.8
process design and a strong relationship (correlation
and mechanism) is shown between critical process
0.6
parameters and critical quality attributes, a math-

Power
ematical model can be used to predict how variation
in the process parameter affects the quality attribute. 0.4

It is assumed that the PAR of the process parameter is

such that the quality attribute will be within specifica- 0.2

tion. Variation in the model itself must be considered

since the model equation usually predicts the quality 0.0
-4 -3 -2 -1 0
attribute on average rather than for individual PPQ Difference
lots, which will vary from the average. Alternatively,
scale-up models can also be useful. Sample Size
3
Assumptions
Alpha 0.025
4 StDev 1
5 Alternative <
6
2) Analyze Historical Performance: When performing a
technical transfer from one commercial site to another,
the historical process mean and its variation can be Figure 8: Sample Size by Power Curve for T-test
calculated to predict performance at the new site. Operating Characteristic (OC) Curve,
AQL = 0.1, Lot = 100,000
ANSI Z1.4-2008 Gen. Inspection Level II (black) vs. III (red)
3) A
 nalyze Development Performance: Development lots The power calculation uses the conceptions of alpha risk
produced during Process Design are used to determine (Type1 .0I error, the risk of failing a criteria which actually
the PAR for critical parameters. Consequently, these passes) and beta risk (Type II error, the risk of passing a
Probability of Acceptance

extreme set point runs will produce critical quality attri- criteria
0 .8 which actually fails). Power is 1– beta is targeted at
butes at their highest deviation from the process mean. either 0.8 (20 percent beta risk) or 0.9 (10 percent beta risk);
Variation in the raw materials lot (and any critical the actual risk of the sampling plan is determined after the
0 .6
material attributes) must be considered in the predicted number of samples is known. Calculating the sample size
process variation. A limited number of development lots using a power calculation will require the significance level
0 .4
may not have experienced the full variation due to the (alpha risk), the estimated maximum standard deviation
limited number of raw material lots used. (between samples), and a target difference.
As mentioned before, the t-Test is a statistical comparison Figure
0 .2 8 is an example power curve showing the number
LQ 1 0 %
of means. To compare standard deviations between lots, the of samples for different target power (0.8 and 0.9) with a
statistical test is the F-test (for normally distributed data) or standard
0 .0
deviation of 1. The sample size is determined by
Levene’s test (no assumption of normal distribution). The the first curve above the target power for a given target
0 .0 0 .2 0 .4 0 .6 0 .8 1 .0
acceptance criteria for the standard deviation of a quality difference. Our choiceLot of Percent
target Defective
difference is determined
attribute (variation between samples in a lot) must consider by the t-Test acceptance criteria: the largest variation
how the attribute varies from lot to lot in addition to the predicted in the process n mean.
c
500 1 n sample size
variation within each lot to ensure all portions of the lot have 800 2 c acceptance number

a high likelihood of meeting specification. LOT ACCEPTANCE SAMPLING PLANS

Certain sampling plans commonly used during PPQ When sampling for attributes that are discrete (pass/fail)
are predefined in various guidance and standards. rather than continuous (a numeric value), the sampling
One example is blend uniformity in which both the plan is determined by an operating characteristic curve
minimum sampling requirements and the acceptance instead of a power curve. Frequently used for visual
criteria are defined. Another is Bergum’s Method for defects, these plans are either calculated or selected from
Content Uniformity. For user-defined plans (e.g., t-Test) the ANSI Z1.4-2008 standard for sampling by attributes.
the minimum number of samples must be calculated to In our case, the manufacturer’s quality assurance group
ensure that a valid statistical conclusion may be drawn. chose the Acceptance Quality Level (AQL) for the at-
For the t-Test, F-test, or Levene test the number of samples tribute, because it represented the maximum process
is calculated using a power calculation for the specific test. average of defects for that attribute over time.

t PREVIOUS PAGE 14 NEXT PAGE u

 0.0
-4 -3 -2 -1 0
Difference
ASSESSMENT, QUALIFICATION & VALIDATION
Sample Size Assumptions
3 Alpha 0.025
4 StDev 1
5 Alternative <
6

The desire for PPQ lots is to increase the number of Operating Characteristic (OC) Curve,
samples (i.e. discrimination of the sample plan). However, AQL = 0.1, Lot = 100,000
shifting the AQL is not recommended since the AQL ANSI Z1.4-2008 Gen. Inspection Level II (black) vs. III (red)

is not representative for individual lots in isolation. To 1 .0

create a more discriminating sampling plan for PPQ, the
Limiting Quality (LQ, also called Lot Tolerance Percent

Probability of Acceptance
0 .8
Defective, LTPD) is the preferred method for creating a
more discriminating plan for PPQ.
Figure 9 compares a standard lot plan under Z1.4 0 .6

(General Inspection Level II) to a more discriminating

PPQ lot plan (General Inspection Level III). The number 0 .4
of samples increases from 500 to 800 and the LQ at 10
percent acceptance changes from approximately 0.77
0 .2
percent defective to 0.65 percent defective. LQ 1 0 %
These types of sampling plans are only suitable for
individual lot acceptance; they do not determine the 0 .0

actual percent defective for a lot. These plans only assure 0 .0 0 .2 0 .4 0 .6 0 .8 1 .0

that lots above the LQ have a low (10 percent or less) Lot Percent Defective

probability of being accepted under this plan.

n c
The PV Guidance no longer defines the number of lots 500 1
800 2
n sample size
c acceptance number
required for PPQ; it is left to individual manufacturers
to justify how many lots are sufficient. There is no safe
harbor for producing three PPQ lots since justification Figure 9: Limiting Quality Comparison between Z1.4 Sample Plans
must be made for any number of lots. In order to make
any reasonable argument of reproducibility, it would be
expected that the minimum number of lots be no less sources of variability. The number of PPQ lots can then
than two to three. be determined by matrix design of the sources with
It is usually not necessary to operate process the highest risk to variation of quality attributes. Those
parameters at the extremes of the NOR since this should sources of variability, which cannot be included in the
have been previously established. As such, the setpoints PPQ, should be considered for monitoring during Stage
of process parameters are not changed between PPQ lots 3 - Continuous Process Verification.
and do not impact the number of PPQ lots required. In After completing the PPQ analysis, the team revisited
determining the number of lots consideration should the risk matrix to reflect the commercial operation. This
be given to understanding the source and impact of data was included in the Stage 2 final report.
variation on quality attributes.Suggested sources of
variation to consider are: STAGE 3 – DATA MONITORING
The last stage of the new PV lifecycle is process monitor-
• Number of raw material lots, especially when a critical ing. While monitoring has been part of the normal drug
material attribute is identified; quality management system (QMS), the new PV guidance
• Number of commercial scale lots previously produced advocates moving beyond the normal CQAs reported in
during Process Design; a product’s Annual Product Review (APR) and extend-
• Number of equipment trains intended for use; ing them to include the CPPs that have been identified as
• Process complexity and number of intermediate steps; critical to process stability. For the product in question,
• History of performance of commercial scale equipment a protocol was drafted to gather data over the next 20
on similar products; lots to establish alert and action limits relating to process
• Number of drug strengths; variability. This data was intended to be reported in the
• Variation of lot size within commercial equipment; product scorecard and included in the APR.
• In-process hold times between process steps;
• Number of intermediate lots and mixing for down- References:
stream processes. Editor’s Note: For the complete list of references associated with this
It is recommended to perform a risk analysis of these article visit: www.pharmamanufacturing.com

t PREVIOUS PAGE 15 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

Validation = Process Design

PROCESS DESIGN IS THE KEY TO EFFECTIVE PROCESS VALIDATION
By Parveen Bhandola, PASQ CQE, Nostrum Laboratories, Inc.

VALIDATING ANY pharmaceutical manufacturing

process is a legally enforceable requirement to ensure
that drug products are made with the highest possible
assurance that all quality attributes are met. Validation
establishes scientific evidence by collecting and evaluat-
ing data, to provide assurance that a process is capable
of consistently delivering quality products within the
commercial manufacturing conditions.
Beyond being a regulatory requirement, process
validation makes good business sense. Last year,
the FDA issued new process validation guidance [1],
consistent with its previous guidance [2]. However,
the newer guidance offers a much more elaborate and
detailed description, defining validation as a continuous
process consisting of distinct stages, each aligned with Process Qualification activities begin with the
the product lifecycle stage, as outlined in FDA/ICH’s selection of the design for the process equipment,
guidances Q8(R2) Pharmaceutical Development [3], Q9 utilities and facility and verification of the selected
Quality Risk Management[4], and Q10 Pharmaceutical design for its suitability for the intended process within
Quality Systems [5]. This article summarizes key the complete range of operating conditions. The Process
points within the revised guidance, emphasizing the Performance Qualification (PPQ) is performed under
importance of process design (for a more detailed cGMP conditions, and it involves demonstrating
version of this article, visit www.pharmamanufacturing. controls over the commercial manufacturing process.
com). It breaks validation down into three stages:
Stage 1: Process Design Stage 3 (Continued Process Verification) involves the moni-
Stage 2: Process Qualification toring of critical process parameters and quality attributes
Stage 3: C
 ontinued Process Verification during routine manufacturing to provide ongoing assur-
ance that the manufacturing process continues to remain
Stage 1 (Process Design) involves generating process knowl- under the state of control.
edge and understanding through well-designed and docu- Although all stages of Process Validation are
mented studies for the purpose of developing a strategy for important, Stage 1 (Process Design) is the foundation of
process control for each stage of manufacturing. Process the overall Process Validation.
Design activities should guide the development of master
production and control records. An effective Process De- BUILDING A FOUNDATION FOR PROCESS DESIGN
sign is essential for developing a process that is capable of During Stage 1 of Process Validation, process develop-
reliably and reproducibly producing the product with the ment and scale-up activities are performed to establish
intended attributes in routine manufacturing. scientific knowledge and data to define the commercial
process with the overall objective of accomplishing the
Stage 2 (Process Qualification) consists of the following Process Design. The process knowledge and information
two elements: gained at this stage help in developing the control strat-
a) Design and qualification of the process equipment, egy for routine production.
utilities and facility During the Process Design stage, the manufacturer
b) Process Performance Qualification is expected to understand the sources of variability, and

t PREVIOUS PAGE 16 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

detect the presence, degree and

potential impact of variations on the Similar to the necessity of a strong
manufacturing process. Strategy for
controlling the variations should foundation for a building, a strong
be developed on the basis of the Process Design is necessary for
assessment of the impact of variations
on the product quality attributes. supporting Process Validation.
Process variations may be caused
PARVEEN BHANDOLA
by materials, equipment, production
environment and manufacturing
procedures changes. Since a cations for certain critical parameters can prove beneficial during the
manufacturing process must remain may impact the process performance. Process Design stage. It is important
in a state of control over the process Selection of equipment design to involve operators in Process
lifecycle even as the materials, suitable for a manufacturing process is Validation activities, starting at the
equipment, production environment extremely important, and the impact Process Design stage and continuing
and manufacturing procedures of the variability originating from the throughout the process lifecycle. Most
change, the role of Process Design equipment must be established at the manufacturing managers are well
becomes crucial as it involves Process Design stage. For example, a aware that operators are the eyes and
understanding, detection, impact manufacturing process may allow the ears of any process, and generally
assessment and control of the process use of two alternative pieces having the first to observe any abnormal
variations that may ultimately impact similar design and working principles, process trends. Hence, it is critical to
the product quality attributes. but the changeover from one piece have an open dialogue with operators
The revised guidance emphasizes of equipment to the other may cause regarding their observations about the
the significance of process variability, process variability, and the impact process and equipment performance
and Process Design mainly involves of such variability on the process on a regular basis, throughout the
detection, impact assessment and performance should be considered at product lifecycle. Most success
controlling variations. The following the Process Design stage. Secondly, stories about significant process
section highlights some examples of the performance of some equipment improvements would not have been
potential sources of process variability. parameters may vary over a period of possible if operators had not shared
time, causing variation in the process first hand information and become
MATERIALS VARIATIONS performance. Variation in the force fully involved in overall process
For example, variability in physical of shaking the filter bags of a Fluid design and improvement.
characteristics of raw materials can Bed Drier, for instance, may cause a FDA’s revised validation guidance is
significantly impact the dry-blending variation in the extent of dislodging holistic, but hinges on proper process
process. Hence, it is important to con- fines from the filter bag. Such variation design. Since flexibility and change
sider the extent of controls exercised is not easy to detect, but it may cause are the precursors for improvement,
through the raw materials specifica- significant variation in the Fluid Bed it should help facilitate continuous
tions to provide assurance that the Drying process, thus causing variation improvement in pharmaceutical
process will continue to perform at in the quality attributes of the product. manufacturing.
the expected level. If the specifications Though operators may be properly
for some critical raw materials permit trained, variations are still common. References
wide ranges for parameters like For one thing, there may be a 1. “ Process Validation: General Principles &
particle size distribution, bulk density, difference in the performance of Practices,” FDA’s Guidance for Ind. (Jan. ’11)
tap density and moisture content, it experienced operators and newly- 2. “ Guideline for General Principles of Process
can cause significant variation in the trained operators. Also, operators Valid.,” FDA’s Guidance for Ind. (May 1987)
process performance even when all working during night shifts may not 3. “Q 8(R2) Pharmaceutical Development,”
raw materials meet the specifications. be as alert as those working during FDA/ICH Guidance for Industry (Nov 2009)
A sound Process Design should ques- the day, and this difference may also 4. “ Q 9 Quality Risk Management,” FDA/
tion the basis for developing the raw cause process variations. ICH Guidance for Industry (June 2006)
materials specifications and establish A strong interdisciplinary approach 5. “ Q 10 Pharmaceutical Quality Systems,”
how the variation within the specifi- and systematic brainstorming sessions FDA/ICH Guidance for Industry (May 2007)

t PREVIOUS PAGE 17 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

What Are IQ, OQ and PQ, and

Why Are They Required?

PARDON THE alphabet soup, but these acronyms identify PERFORMANCE QUALIFICATION (PQ) PUTS YOUR
common terms that you’ll see in many FDA and Health EQUIPMENT TO THE FINAL TEST
Canada guidance documents, and typically refer to equip- In the PQ — performance qualification — phase, we like
ment. IQ stands for Installation Qualification. OQ is Opera- to challenge the equipment, much like in the OQ phase,
tional Qualification and PQ is Performance Qualification. but now under load. While it’s great that it runs at 50
Before you even get to IQ, OQ, PQ, if you’re acquiring a RPM or 150 RPM when it’s empty, what happens when
new piece of equipment, you’ll need design specifications there’s 300 kilos of material in it? Can it still achieve
that define exactly what’s in that piece of equipment. those speed ranges? That’s the essence and focus of the
Everything from the type of power source it will utilize to PQ phase. Once you’ve completed these three phases,
the exact materials used in its construction. Once you have the equipment is available for use in whatever process
your final/approved design specs, you order the equipment, you intended for it.
it comes in and now you’re developing your IQ and OQ.
Quite often, the basis for the IQ and OQ will be the QUALIFICATION PROCESSES WILL GIVE YOU,
equipment manual itself. To save time and, prior to delivery, AND YOUR CLIENT, PEACE OF MIND
we’ll ask the equipment manufacturer for the manual and Why does the pharmaceutical manufacturing industry need
we’ll use the manual as the basis of our protocols. this qualification process? Validation executed as a global
exercise is a method of establishing documented evidence
INSTALLATION QUALIFICATION (IQ) that shows that we have a high degree of assurance that our
For installation qualification, we’ll first look at the equip- manufacturing process will consistently yield a product of
ment material. For example, if we specified 316 stainless, predetermined quality. If a manufacturer fails to do that, the
we’ll test to verify it is in fact 316 stainless. Sometimes results can be disastrous. It can cost the client hundreds of
stainless steel is passivated and you can test to verify there thousands of dollars, increase the risk of product recall and
are no further residues from the passivation process. potentially contribute to a loss of market share. We try to
You might have specified a 5 force power motor in your prevent that, and we do so by controlling change and ensur-
equipment, for example. You want to check to ensure it’s ing that we account for change in our day-to-day operations.
a 5 force power motor. You’ll also confirm that the power Occasionally, we get clients who ask, “Can you make
output and the power requirements are consistent with this product for us?” We may not have the specific
your specifications, and the room that the equipment is equipment needed, but the client wants our involvement
installed in can accommodate that power source. Once and would like us to bring the equipment in. The IQ, OQ,
you have completed your review of the installation and PQ process is very important here. We’ve helped clients
everything is in order, you can trust that the equipment is spec out the equipment they need and then quickly and
going to operate the way in which it was designed. efficiently bring that equipment in, install it, set it up, and
get it ready to go. The way we run the IQ, OQ, PQ process
OPERATIONAL QUALIFICATION (OQ) really expedites the transfer of a product into our facility.
The next phase is OQ, operational qualification. At this In one case, we obtained approval from a client in
stage, if you’ve specified that your equipment is going to February, specified the equipment, received it in March,
run in a range of 50-150 RPM and will draw a specific set it up and began manufacturing clinical trial batches
amount of power, you want to verify that the equipment is and brought them to the clinical site by September 1st. Due
achieving those operational requirements. So, review those to the rigorous process we go through, we identified the
parameters and challenge them. Again, make sure your equipment, knew what we had to do to get it up and running
equipment actually runs the way it’s supposed to run. optimally and did it quickly. That really helped our client.

t PREVIOUS PAGE 18 NEXT PAGE u

 ASSESSMENT, QUALIFICATION & VALIDATION

WellSpring provides peace of mind to pharmaceutical compa-

nies for their contract manufacturing and packaging needs.
When you work with WellSpring, you’ll have a whole team of
professionals backing your project. Our team will oversee the
entire operation; while a project manager will help you plan all
stages. You’ll also have a dedicated technical team to manage
all aspects of the project.

WellSpring Pharma Services’ capabilities include the following:

Quality
• Analytical Method Development, Validation and Transfers
• Quality Assurance
• Microbial Testing, Stability Storage and Testing
• Compendial and Non-Compendial Analytical Testing

Manufacturing
• Tablets, Capsules, Powders
• Topical Creams, Lotions, Ointments and Gels
• Non-Sterile Liquids

Packaging
• Child Resistant Packaging
• Bottles (glass/plastic), Jars, Tubes, Blisters
• Enema Bag Format
• Secondary Packaging

Storage
• Narcotics Vault
• Temperature Mapped Warehouse
• Validated Refrigerators and Freezers
• Facility Generator Back-Up
• Stability Chambers 5°C, 25°C, 30°C and 40°C

For more information, please contact:

WellSpring Pharma Services

400 Iroquois Shore Rd. • Oakville, Ontario, Canada L6H 1M5
Telephone: 844-879-7427 • Fax: 844-879-7427
Web: www.wellspringcmo.com • Email: info@wellspringcmo.com

t PREVIOUS PAGE 19