Peer-Review Research

Continued Process Verification

for Cleaning Validation–
Challenges and Pitfalls
Richard J. Forsyth and Sabine Imamoglu

V
Continued process verification (CPV) for a cleaning alidating cleaning in a pharmaceutical manufacturing
validation (CV) program begins once the validation facility is a regulatory requirement (1–4). In regula-
study is complete. Planning for the CPV needs to tory guidance documents, program basics, regulatory
expectations including prerequisites, and acceptance
be considered, however, as the cleaning validation
criteria are reviewed along with the strategy for selection of the
is planned. Otherwise, the necessary parameters product(s) and equipment to validate. Although cleaning vali-
for the CPV might not be captured in a way to allow dation (CV) execution is described in general terms, guidance
the smooth transition from the CV study to the CPV documents are limited to describing what to do, but not how to
program to maintain the validated state of cleaning. do it. The more recent the guidance update (2), the greater level
of specific expectations are included.
The guidance documents include general instructions on
how to proceed once cleaning validation is completed. The
validated state of cleaning is to be monitored using ongoing
testing of the cleaning process to demonstrate continued
control of cleaning. This linear approach to cleaning val-
idation resulted in the cleaning maintenance part of the
program being slowly neglected, resulting in programs that
fell out of compliance. The concept of lifecycle control of
manufacturing process validation (5) addressed the short-
comings of the linear approach to process validation using
Stage 3–Continued Process Verification (CPV). Using a sim-
ilar approach, an ongoing program to collect and analyze
cleaning parameter data can be applied to cleaning valida-
tion. Although the lifecycle approach better addresses the
post-validation cleaning program, it still leaves the details
up to the individual facility. And the concept of CPV is not
mentioned until after the initial validation is complete.
Waiting for the completion of cleaning validation to ad-
dress CPV will prove problematic as to how to document,
gather, and trend the appropriate cleaning process data,
critical process parameters (CPPs), hold times, and cam-
MUURAA- STOCK.ADOBE.COM

paign lengths to clearly demonstrate continued control of

the validated cleaning process. CPV is a factor to be con-
sidered as the cleaning validation strategy is defined. Ad-
dressing cleaning parameters for the long term needs to be
addressed as part of the cleaning strategy. Otherwise, there
Submitted: June 23, 2021 might not be a viable mechanism to capture and trend ongo-
Accepted: September 13, 2021 ing cleaning parameters. The CPV strategy should address
34 Pharmaceutical Technology  January 2022 P h a r mTe c h . c o m
what cleaning parameters to check only for compliance and rameters and their ongoing status. The worst-case product,
what parameters to check for compliance as well as to track cleaning limit, and equipment grouping will not be affected
and check trends. unless a change is made to the validated state. Changes
could include: a new product or a new piece of equipment
Background is introduced; a change is made in a product manufactur-
To understand the issues of addressing CPV after comple- ing process or equipment configuration; or a product or
tion of cleaning validation, start by considering the require- equipment is retired from service. Any change should go
ments for cleaning validation: through the change control system and an assessment made
• The hardest-to-clean product(s) as to whether an additional CV study is required. The as-
• The lowest cleaning limit sessment should include an update of the original product
• Equipment grouping or equipment assessment documentation. The assessment
• Product/equipment matrix and document update should fall on the CV representative
• CPPs to the Change Control Committee.
• Demonstration of acceptable and consistent cleaning Critical cleaning parameters. CCPs were established during
• Dirty equipment hold time (DHT) cleaning development and are those parameters that have a
• Clean equipment hold time (CHT) direct impact on the level on cleanliness from the cleaning
• Campaign length. process. A clean-in-place (CIP) cleaning cycle records all
There are alternatives for establishing the products and cleaning parameters and issues a report at the conclusion. A
equipment to execute cleaning validation, but a grouping manual cleaning procedure is a more problematic situation
strategy is accepted by the regulatory agencies (1–4) and is as far as identifying and recording CCPs. During devel-
the most practical approach. A hardest-to-clean product for opment, it can be shown that detergent concentration and
cleaning can be established using solubility of the formu- water temperature are not critical. Personnel can be trained
lation API, but this approach ignores excipients and prac- and qualified for consistent cleaning. Minimum cleaning
tical experience. A more rugged approach is to look at the times can be established, and specific cleaning tools can
cleanability of the formulations using a three-prong meth- be identified. During CV of a manual procedure, the CCPs
odology (6): formulation composition, personnel experience, can be recorded by an observer to corroborate the cleaning
and formulation cleanability. outcome. However, post-CV recording of CCPs falls on the
The cleaning limit should be established for every prod- personnel cleaning the equipment, which is problematic for
uct (6), and the lowest cleaning limit should be the target someone wearing wet, soapy gloves and could lead to CCPs
for the cleaning validation study. Equipment needs to be being captured immediately after cleaning.
grouped for validation so that one piece of equipment is Trending of CCP data is even more challenging. The CCP
representative for the entire group. The product/equipment data must be extracted from the cleaning checklists, entered
matrix, which contains the equipment train for each prod- into a logbook or database, and trended over time.
uct, is used to decide the most efficient cleaning validation Acceptable and consistent cleaning data. The CPV plan should
strategy to cover all products and equipment groups. include periodic testing of equipment after cleaning to
The product and equipment selection must be made be- demonstrate continued control of the validated cleaning.
fore cleaning validation can commence. The cleaning data The frequency of the testing should be risk-based with the
to demonstrate acceptable and consistent cleaning as well rationale clearly defined in the CPV plan. Factors that might
as the critical cleaning parameters (CCPs) (e.g., time) are increase testing frequency include manual cleaning meth-
captured during cleaning validation execution. The agreed ods vs. CIP methods; low cleaning limits; and CV data that
upon DHT, CHT, and campaign length are targeted prior to passed but are close to the acceptable residue limit (ARL);
execution and then confirmed during execution. The CCPs, all of which increase the risk of subsequent cleaning fail-
DHT, CHT, and campaign length can be conveniently cap- ures. Capturing and trending CV and CPV data is probably
tured in the cleaning validation study protocol documenta- the clearest demonstration of continued control of equip-
tion along with the product and equipment being validated. ment cleaning in the facility. All data should be trended to
Once cleaning validation is successfully completed and demonstrate consistency of cleaning. The challenge here is
a final report documented, the controlled, validated state to update and maintain the CV database. Data processing
needs to be maintained. and trending could fall on the personnel assigned to per-
form sampling or testing. An alternative would be to mirror
Continued process verification the effort used to trend in-process manufacturing data.
Product and equipment. To ensure a smooth transition, a flex- Dirty hold time. DHT is established in the CV protocol
ible plan or protocol to address CPV should be in place as execution. The end of manufacturing activity is recorded
the cleaning validation study draws to completion. The CPV in the product batch record. After a designated DHT, the
plan should be drafted by the same team that implemented beginning of cleaning is recorded in the cleaning record or
the CV program and should address all the captured pa- checklist. The executed CV protocol captures both times
Pharmaceutical Technology  January 2022 35
Peer-Review Research
and determines the effective dirty hold time. DHT is cap- DHT is not critical (7). If the equipment is scraped, vacu-
tured for the three validation runs, and the longest of the umed, and wiped down with solvent (e.g., 70% isopropyl al-
three is designated the validated DHT based on the equiv- cohol) to minimize the amount of product residue left on the
alency of all CV data. A single DHT, which is applied to all equipment, then the DHT might be a non-issue because the
equipment for all products, is applied going into the CPV equipment would be rid of most residue and would already
stage of the cleaning lifecycle. be dry. This strategy would minimize the API released into
Post-validation, capturing and relating the two times with the wastewater stream during cleaning and limit exposure
the established DHT becomes more problematic. The two of personnel during cleaning.
times and DHT are in separate documents, and personnel Clean hold time. The issues for tracking and trending CHT
filling out the manufacturing batch record do not see the are comparable to the DHT issues. CHT is established in
cleaning record, while personnel cleaning the equipment a separate CV protocol execution. The end of cleaning ac-
do not have easy access to the manufacturing batch record. tivity is recorded in the cleaning record or checklist. After
The established DHT is reported in the CV final report, a designated CHT, the beginning of manufacturing is re-
and because the DHT is a cleaning parameter, the end of corded in the product batch record. The executed CV pro-
manufacturing time and established DHT need to be readily tocol captures both times and determines the effective CHT.
available to personnel performing the subsequent cleaning. The CHT is captured for the three validation runs, and the
There are several options how to capture, maintain, and longest of the three is designated the validated CHT based
trend DHT data. A logbook kept with the equipment is on the equivalency of all CV data. A single CHT, which is
the most straight-forward solution. The validated DHT is applied to all equipment after cleaning, is applied going into
known and is recorded in the logbook header. The date and the CPV stage of the cleaning lifecycle.
time of completion of manufacturing activity are entered The options to capture, maintain, and trend CHT data
each time the piece of equipment is used. At the beginning parallel those for DHT. For the logbook option, it might be
of cleaning, the date and time are recorded, the DHT is cal- able to capture both the CHT and DHT in the same logbook
culated and confirmed to be less than the established DHT. and trend quarterly. Downloading and trending the DHT
A second alternative can be capturing DHT electronically and CHT from a logbook for every piece of equipment is
in a production record system. This is preferable because labor intensive.
system checks can be included to prevent moving forward CHT criticality could also be minimized during valida-
without appropriate action if the DHT is exceeded. tion (7). If CHT is established for an extended period (e.g.,
A third alternative is to capture the DHT at the beginning > 45–60 days), then equipment held in the same controlled
of the equipment cleaning checklist or record. This would conditions can be used up to the validated CHT.
ensure that the DHT is captured but makes trending more For clean equipment held outside the manufacturing
difficult in that there is no centralized documentation of the area, a standard policy of recleaning any equipment being
DHT data in a logbook. brought into the manufacturing area is customary.
Attempting to trend the DHT data is more problematic, A second option to minimize concern for CHT post vali-
because collating DHT data from logbooks for all equipment dation is to routinely rinse or wipe the equipment with 70%
or individual cleaning records is resource intensive. The best isopropyl alcohol immediately before use. This action further
option is to periodically download the DHT information mitigates risk of bioburden proliferation during the CHT.
into a database for ongoing trending. Trending of CHT data also provides a level of control to
If an electronic manufacturing batch record is used and the CPV program. If CHT times increase over time, trend-
the equipment cleaning information is captured electron- ing could indicate a problem with scheduling of manufac-
ically, then DHT data can be captured. Trending of elec- turing batches or manpower resource issues, which could
tronically captured DHT would have to be programmed be addressed before a CHT excursion by extending the CHT
to download and trend, but once implemented, would not through a protocol execution.
require continued personnel involvement. Campaign length. The maximum campaign length is estab-
The value of the logbook option is limited because the re- lished in the CV protocol execution. The number of batches
cords are paper-based and all entries and data analysis are manufactured and the length of time to manufacture are
handwritten. However, the advantage is that the logbook stays captured both on the equipment logs and the protocol. The
with the equipment and compliance is more readily confirmed. campaigns are executed three times for validation, and the
Trending of DHT data provides another level of control to longest campaign length is designated as the maximum
the CPV program. If DHT times increase over time, trend- campaign length based on comparable cleaning data.
ing could indicate a problem with scheduling equipment Post validation, the batch campaign length is known
cleaning or manpower resource issues, which could be ad- based on scheduling and equipment use. However, the
dressed before a DHT excursion by extending the DHT. length of time to manufacture a campaign is not typically
Alternatively, depending on how the equipment is han- noted and would normally only be unduly extended due to
dled at the end of manufacturing, it could be argued that mechanical equipment issues. And while bulk hold times
36 Pharmaceutical Technology  January 2022 P h a r mTe c h . c o m
are considered for manufacturing process validation, the tency of cleaning and the level of risk for a potential cleaning
potential effect of increased campaign time on the subse- failure in the future. But once cleaning validation is com-
quent cleaning is not considered. One option is to have the plete, cleaning data are not generated after every equip-
equipment card designed to include a check of the maxi- ment cleaning. Based on the risk of a cleaning failure, the
mum campaign parameters, both number of batches and frequency of swab testing can decrease, and the lower fre-
number of days. The campaign data can be periodically quency of cleaning data points presents a periodic snapshot
downloaded into a database for ongoing trending. of cleaning rather than a continuous record of equipment
Non-compliance. The risk of not complying with validated cleanliness. In addition, relying solely on cleaning data for
parameters is low as long as the validated parameters are CPV is akin to solely relying on release testing for man-
readily available, limited in complexity (e.g., one DHT value ufacturing process control, which is not acceptable to the
of 10 days for all equipment), and the parameters are re- regulatory agencies.
corded contemporaneously. If there is a non-compliance Therefore, other cleaning parameters that continue to be
with any parameter, it should be immediately recognized captured after every post-validation cleaning become more
and addressed before a non-compliant situation develops. indicative of continued cleaning verification. They include
For example, if the DHT is exceeded, testing after clean- CCPs (times, temperature, detergent concentration as neces-
ing can be arranged after cleaning to verify a successful sary), DHT, CHT, and campaign length. Along with a visual
level of cleaning before the cleaned equipment is used again. inspection performed by qualified personnel, trending of
Additionally, if this type of occurrence could be addressed these parameters individually might not provide assurance
proactively and a protocol documented and approved prior of CPV, but taken together, provide a picture of a cleaning
to equipment cleaning, the cleaning data might be used to system that continues to be in a state of control.
lengthen the DHT for the cleaning process. Any delay in
recording data increases the risk of non-compliance. Conclusion
Non-compliance investigation. If a non-compliance for a Tracking and trending CPV parameters are necessary to
cleaning parameter is only recognized after the equipment demonstrate that a cleaning validation system is maintained
is reused, an investigation is necessary. If a critical cleaning in a state of control. Options are available, but the easiest
parameter (time, temperature, detergent concentration), DHT, path forward is often the more labor-intensive approach.
or campaign length is non-compliant, then the equipment Consideration for long-term use of CCPs, cleaning swab
might not have been sufficiently cleaned, and carry-over into test data, DHT, CHT, and campaign length should be taken
the next batch is possible. A risk assessment should determine during the CV planning phase rather than waiting until CV
the level of risk of the non-compliance. The higher the clean- is complete, to ensure a robust ongoing CV program.
ing limit of the cleaned API, the lower the risk of unaccept-
able carry-over. Conversely, an API with a low cleaning limit References
increases the risk of an unacceptable carry-over. 1. FDA, Guide to Inspection of Validation of Cleaning Processes
If not already established, a visible residue limit (VRL) of (Division of Field Investigations, Office of Regional Operations,
Office of Regulatory Affairs, Washington, D.C., July 1993).
the cleaned API should be determined (8, 9). If the VRL is 2. EC, EudraLex Volume 4, EU Guidelines for Good Manufactur-
lower than the cleaned API cleaning limit, it provides a clear ing Practice for Medicinal Products for Human and Veterinary
indication that the visually clean equipment was sufficiently Use, Annex 15: Qualification and Validation, Section 10: Cleaning
cleaned and further batch investigation might be avoided. Validation (2015) Section 39, Page 243.
The cleaning records and equipment log need to be checked 3. Health Canada, Canada Health Products and Food Branch
to verify the extent of the non-compliance and what subse- Inspectorate Guidance Document, Cleaning Validation Guidelines
GUIDE-0028 (2008).
quent batches are potentially impacted. As necessary, the 4. PIC/S, PIC/S Validation-Master Plan, IQ, OQ, Non-sterile Process
subsequent batches should be tested for the presence of the Validation, Cleaning Validation (PI 006-3) (September 2007).
carry-over API, which might require some analytical method 5. FDA, Guidance for Industry Process Validation: General Principles
development and validation. Personnel should be interviewed and Practices, (Division of Field Investigations, Office of Regional
to determine the root cause and corrections implemented. Operations, Office of Regulatory Affairs, Washington, D.C.,
January 2011).
6. R. J. Forsyth, Pharm. Technol. 45 (8) (2021).
Compliance verification vs. trending 7. R. J. Forsyth, Pharm. Technol. 32 (4) 64–73 (2008).
Verification of cleaning validation parameters on an ongo- 8. R. J. Forsyth, Pharm. Technol 33 (3) 102–111 (2009).
ing basis is critical for maintaining the validated state of 9. R. J. Forsyth, Pharm. Technol, 40 (4) 50–57 (2016). PT
the cleaning program. If a parameter is exceeded, then ac-
tion must be taken to correct the non-compliant condition. Richard Forsyth* is a Principal Consultant with Forsyth
The value of trending some cleaning validation parameters Pharmaceutical Consulting. Dr. Sabine Imamoglu is Product
might not be as obvious. Certainly, trending cleaning data Supply, Pharmaceuticals, Pharmaceutical Affairs, Bayer AG.
in the form of swab sample or comparable data results adds *To whom all correspondence should be addressed.
value in that it provides an ongoing picture of the consis-
Pharmaceutical Technology  January 2022 37