Download Visually Clean as a sole acceptance criterion for cleaning validations protocols

Downloaded from journal.pda.org on September 5, 2013
''Visually Clean'' as a Sole Acceptance Criterion for Cleaning
Validation Protocols
Destin A. LeBlanc
PDA J Pharm Sci and Tech 2002, 56 31-36
Downloaded from journal.pda.org on September 5, 2013
TECHNOLOGY/APPLICATION
“Visually Clean” as a Sole Acceptance Criterion for Cleaning
Validation Protocols
Destin A. LeBlanc
Cleaning Validation Technologies, San Antonio, Texas.
ABSTRACT: The role of visual examination as the sole acceptance criterion in cleaning validation protocols is
explored, including the proper definition of “visual limit” for a given residue. Such a visual limit is specific to
the surface the residue is on, and is further defined by viewing conditions, such as lighting, distance, and
angle of viewing. A visually clean standard may only be properly utilized if the visual limit is below any
scientifically calculated residue limit, such as that determined by a traditional dose-based limit determination. While such an approach, properly applied, has scientific justification and appears to be accepted by the
proposed Annex 15 to the European GMPs, the approach is still untested with regulatory authorities.
Keywords: cleaning validation, residue limits, visually clean
Introduction
Because of the extensive analytical effort required
t o d o c u m e n t a c c e p t a b l e l eve l s o f p o t e n t i a l l y
contamina ting residues in cleaning validation
protocols, some scientists are re-evaluating the
value of utilizing visual examination, supported by
data to demonstrate what that visual observation
means in light of actual residues that might be
present on cleaned surfaces (1,2,3). If visual
examination were the most stringent determination
of how clean surfaces would be, then it may be
justifiable to utilize a “visually clean” standard as
the only criterion for measuring the cleanliness of
surfaces. This approach was explored as long ago
as 1989 by Mendenhall, who stated that in most
cases “the visual cleanliness criteria was more rigid
and clearly adequate” as compared to a quantitative calculation of residue limits (4). A standard of
visually clean as the sole acceptance criterion was
also presented by the FDA in its cleaning validation guidance document (5); in that case it was
presented as an option for cleaning between lots of
* Author to whom correspondence should be addressed:
Cleaning Validation Technologies, 22215 Roan Forest, San Antonia, TX 78259, 210-481-7865. E-mail:
[email protected]
Note: This paper is based on presentations to the PDA Annual meeting, December 6, 2000, in Philadelphia, PA, and at the 2001 Japan
PDA Congress, February 21, 2001, in Kyoto, Japan.
Vol. 56, No. 1, January/February 2002
the same product (e.g., for dedicated equipment or
between batches of the same product in a campaign).
Three Types of Visual Examination
It is important to distinguish three different purposes
of a visual examination. First is visual examination
as part of an inspection before equipment use. This
is part of Good Manufacturing Practices (6). A second aspect of visual inspection is as a monitoring
technique following cleaning (7). A determination
of visually clean may be a precondition for the
release of the equipment for manufacture of the
subsequent product. The third aspect, and the focus
of this paper, is visual examination as part of a
va l i d a t i o n p r o t o c o l t o d e t e r m i n e w h e t h e r t h e
equipment surfaces are acceptably clean based on
scientifically justified acceptance criteria.
Residue Limits Criteria
The most common approach to setting residue
limits for actives in cleaning validation protocols
is expressed by Fourmen and Mullen (8), in which
they present the criteria Lilly utilized in setting
residue limits. These criteria were to calculate a
limit based on the minimum dose of the active in
the maximum dose of the next product. That calculated value was then compared to a “default” value
of 10 ppm of active in the next product, with the
l ow e r o f t h e t wo va l u e s u s e d f o r s u b s e q u e n t
calculations. This residue is then measured using
31
Downloaded from journal.pda.org on September 5, 2013
appropriate analytical and sampling techniques.
In addition to the analytical determination, the
equipment was required to be visually clean.
Figure 1. Comparison of two a pproaches to
residue limits.
In the PIC/S guidance document (9), and in the
essentially equivalent proposed Annex 15 to the
European GMPs (10), the same three elements are
considered, but with a slightly different presentation. The key for these European documents is that
the most stringent of three criteria – dose based limit
in the next product, 10 ppm in the next product,
and visually clean – be used as the acceptance
criterion for residues on surfaces.
Most stringent of dose calculation and 10 ppm
AND
Equipment visually clean
The difference between the two approaches is
presented in Figure 1. The Fourmen and Mullen
approach requires that surfaces be visually clean
and that the most stringent of the dose limit and 10
ppm be applied. The Eur opean a pproach only
requires that the most stringent of the three criteria
b e a p p l i e d . I n o t h e r wo r d s , i t a l l ow s f o r t h e
possibility that a determination of visually clean
a l o n e ( i f v i s u a l ex a m i n a t i o n i s t h e m o s t
stringent criterion) could be used in cleaning
validation protocols.
contamination of a residue based on calculations
using the dose-based limit and/or the 10 ppm
default limit. The rationale for this is that any
determination of a surface as visually clean should
be clear evidence that the surface level contamination is below VL. If VL is also at or below the
scientifically calculated acceptance criterion based
on a dose limit and based on a 10 ppm default limit,
then a visually clean surface should be below that
calculated acceptance limit.
Fourmen and Mullen:
Proposed Annex 15:
Most stringent of:
Dose calculation
10 ppm
Equipment visually clean
Determination of Visual Limit
“Visually Clean” as Sole Criterion
If the approach of visually clean as the sole acceptance criterion is to be considered, care must be
used in applying it properly. One key for the proper
sole use of a visually clean standard is to experimentally determine the visual limit (VL) for the
target residue. While various numbers, such as
4 µg/cm 2 and 1-4 µg/cm 2 , have been presented as
appropriate levels for visual detection (8,11), the
actual level may vary depending on the residue
itself, as well as the surface the residue is present
on and the viewing conditions. VL should be
defined as the lowest surface level of contamination (in units such as µg/cm 2) at which a specific
residue is consistently and uniformly seen on
surfaces by trained observers. In determining the
VL, it is actually higher VLs which represent the
worst case. This is somewhat counter-intuitive,
since a natural reaction is to think that residues that
may be visible at very low levels represent the worst
case. This can be explained by the second key for
proper use. This is the fact that, in order to utilize
a visual observation as the sole acceptance criterion, VL must be less than or equal to the surface
How is VL determined? The simplest procedure is
to perform laboratory studies in which model surfaces are spiked with the target residue at various
defined levels. These residue levels are typically
levels that bracket the range from about 0.5 µg/cm 2
to about 10 µg/cm 2 . The surfaces are independently
viewed by trained and qualified observers under
defined viewing conditions. Each spiked surface,
as well as appropriate controls, is rated according
to whether the surface is uniformly visually contaminated or not. A key in this evaluation is not to
approach the evaluation as a determination of
w h e t h e r t h e m o d e l s u r fa c e i n e a c h c a s e i s
completely clean or not. Rather the purpose is to
determine at what residue level the spiked surface
c o n s i s t e n t l y c a n b e c o n s i d e r e d v i s i b l e. W h a t
happens in real life experiments is this. A surface
area of about 25 cm 2 is covered with, for example,
a solution of the residue, and the solution is allowed
to dry. As solutions of low residue levels dry, there
may be migration of solids across the surface. In
such cases, after drying, only a portion of the originally spiked surface of 25 cm 2 (in some cases a very
small portion) will show signs of visible residue.
32
PDA Journal of Pharmaceutical Science and Technology
Downloaded from journal.pda.org on September 5, 2013
Figure 2. VL determination example.
Observer A
Observer B
Observer C
1
Y
Y
Y
Residue level in µg/cm2
2
3
4
5
Y
Y
N
N
Y
Y
Y
N
Y
Y
Y
N
6
N
N
N
7
N
N
N
8
N
N
N
Y = “clean” (clean or non-uniform residue)
N = “not clean” (uniformly visible residue)
This is not to be interpreted as an indication
that residues at that spiked level would be visible.
For example, if a surface is spiked at a level of
0.5 µg/cm 2 and is allowed to dry, and if only a small
portion of that surface shows visible residue, should
that be an indication that at a level of 0.5 µg/cm 2
the residue would be visible? While that might be
the case, it is at least possible that if uniformly
spread over the surface the residue would not be
visib l e. What could happen if that case wer e
allowed to set the VL? Setting a VL means that any
surface judged visually clean contains residues
below the VL. If one established 0.5 µg/cm 2 as the
VL based on that data, it is possible that one
might view a surface at a uniform residue level of
0.7 µg/cm 2 , judge the surface visually clean, and
then assume that the level is below the VL of
0.5 µg/cm 2 . In such a case one could be mistaken.
VL Determination Example
Figure 2 illustrates an example of determining VL
for a residue. A model surface is spiked with the
target residue at levels of 1-8 µg/cm 2 . A panel of
three observers rate each panel as to whether the
surface is unifor m ly contaminated or not. As
expected, there are certain levels (5-8 µg/cm 2 )
where all observers agree and see uniform contamination. There are certain levels (1-3 µg/cm 2 ) where
all observers agree and do not see uniform contamination. And, there is one level (4 µg/cm 2 ) where
the observers disagree. In this case, the VL would
be established as 5 µg/cm 2 , meaning that in a
validation protocol, any surface judged visually
clean would be contaminated at residues below
5 µg/cm 2 . As mentioned before, there may be an
i n c l i n a t i o n t o p i c k t h e l ow e s t l eve l at wh i c h
observers see no residue, or the lowest level at
w h i c h a t l e a s t o n e o b s e r ve r s e e s u n i f o r m
c o n t a m i n a t i o n . S u c h a p p r o a ch e s w o u l d b e
misleading. In this example, if the VL was set at
4 µg/cm 2 (the lowest level at which any observer
saw uniform contamination), then it is possible that
a surface might be contaminated at a level of
4 µg/cm 2 and yet in a validation protocol, Observer
A would judge the surface as clean, and therefore
below a level of 4 µg/cm 2 . The establishment of VL
as the highest level at which all observers see
uniform contamination is the worst case. It avoids
the possibility that in any evaluation in a validation protocol there might be false negatives (that
is, cases in which the surface is viewed as clean and
yet below the dose-based acceptance criterion).*
Issues in VL Determination
In establishing a VL value for a given residue, care
must be exercised in techniques used to spike the
model surfaces. The first issue is the nature of the
model surface itself, including both the type (stainless steel, glass, plastic), as well as the finish or
texture of the surface. As much as possible, the
model surface used for spiking studies in a VL
determination should approximate the actual equipment surface being evaluated in the validation
protocol. For example, white residues that are not
visually apparent on polytetrafluoroethylene
surfaces may be readily apparent on stainless
steel surfaces.
A second issue is how the residue is applied to the
surface. The most common ways of applying the
residue are as a solution or as dispersion (although
other techniques could be used depending on the
particular residue). The solvent could be water or
an organic solvent. Ideally, one would apply the
*The author must confess that in his recent book on this subject, he also misrepresented the appropriate use of utilizing a visual limit (see
Validated Cleaning Technologies for Pharmaceutical Manufacturing, Interpharm Press, Englewood, CO 2000, pp.144-145).
Vol. 56, No. 1, January/February 2002
33
Downloaded from journal.pda.org on September 5, 2013
solution or dispersion uniformly across a fixed
surface area (such as 25 µg/cm 2 ) on the model
surface, and allow it to dry. One would prefer that
the drying result in a uniform deposition of the
residue across the spiked surface. However, in
actual VL determinations, it is usually the case that
there is some migration of solids as the water or
organic solvent dries. In certain cases, this will
result in the residue being non-uniformly deposited
on the surface, such that part of the spiked surface
is visually clean and part is not. In such a case, it
should be realized that those spiked levels might
represent actual levels at which the residue might
not be visible if it were present uniformly on the
surface. While it is in the best interest of a manufacturer to establish a VL as low as possible (in
order that the VL be below the dose-based calculation), this fact of non-uniform contamination in a VL
determination is not a fatal flaw. The reason for this is
that a higher VL is actually the worst case.
A final issue in preparation of spiked surfaces for
VL determination is storage and handling of spiked
surfaces. Ideally, the surfaces would be evaluated
immediately after preparation and drying. If not,
care must be used in storage and subsequent
handling of coupons. This is important to avoid the
spiked residue being removed from the surface, as well
as to avoid extraneous contamination during storage.
Issues in Viewing
The conditions of viewing the spiked coupons
should also be controlled. These include items such
as the level and angle of lighting, distance of
o b s e r ve r f r o m t h e s u r f a c e , a n d t h e a n g l e o f
viewing. Here again, the attempt should not be to
view the surfaces under the best viewing conditions
(bright light, short distance). Ideally the viewing
conditions should be the same or similar to those
expected in viewing of equipment surfaces during
the validation protocol. Since a higher VL is the
worst case, it may make sense to determine the VL
under less than ideal conditions (dimmer light,
longer viewing distance). An alternat ive is to
eva l u a t e t h e s p i ke d c o u p o n s u n d e r d i ff e r e n t
viewing conditions, and possibly have different
values for VL for different viewing conditions.
Figure 3. Cleaning agent example.
Key
A = 0.20 µg/cm2
B = 0.40 µg/cm2
34
C = 0.81 µg/cm2
D = 1.6 µg/cm2
E = 3.2 µg/cm2
F = 6.5 µg/cm2
G = 12.9 µg/cm2
H = water control
I = unprocessed control
PDA Journal of Pharmaceutical Science and Technology
Downloaded from journal.pda.org on September 5, 2013
Training of observers in the VL determination is
also important, since the observers will be evaluating surfaces that are uniformly clean, surfaces that
are partially clean, and surfaces that are uniformly
contaminated. Additionally, appropriate controls,
such as surfaces spiked with solvent or water only
and unspiked surfaces, should also be utilized.
Cleaning Agent Example
Figure 3 illustrates the types of surfaces obtained
when a VL determination is made. In this case, the
target residue was an alkaline cleaning agent. It was
diluted in water and applied to 25 cm 2 of a stainless
steel surface at solids levels of between 0.20 and
1 2 . 9 µ g / c m 2 ( C o u p o n s A t h r o u g h G ) . C o n t r ol
applications include the application of water alone
(Coupon H) and an unprocessed clean coupon
(Coupon I). The spiked coupons were then oven
dried. Even at the lowest application level, there
was some residue apparent on the surface; however,
that residue was not uniformly spread across the
surface. In evaluating the surfaces, Coupons A, B,
and C were contaminated, but the residue was not
uniform. For Coupons E, F and G, the residue was
clearly visible across the entire spiked surface. For
coupon D, there was a difference of opinion on
whether the residue was uniform across the spiked
surface. Based on the criterion previously discussed,
VL for this product (based on solids) would be
established at 3.2 µg/cm 2, meaning that any surface
j u d g e d v i s u a l ly c l e a n wo u l d d e fi n i t e l y b e a t
residue levels below 3.2 µg/cm 2 .
Application of VL to Validation Protocols
It is important to consider protocol evaluation
issues related to the sole use of the visually clean
criterion. The critical surfaces evaluated (that is,
those most difficult to clean) must be accessible for
visual examination. Generally, if a surface can be
swabbed, it can also be examined as to whether
it is visually clean. However, there may be other
surfaces, such as inside pipes, which do not lend
themselves to visual examination. Other evaluation
issues include lighting and distance of viewing,
which should be the same or better than what was
used in the VL determination experiments.
It should be remembered that determination of
“visually clean” is a non-specific procedure. If an
active, for example, has a VL of 2.7 µg/cm 2, and if
Vol. 56, No. 1, January/February 2002
the surface in the validation protocol is judged
visually dirty, then this does not mean that the
active is present at a level of 2.7 µg/cm 2 or higher.
The residue could be from a source other than the
active. Such results only mean that, if the residue
were all active, it would be present at a level of
2.7 µg/cm 2 or higher. This is not a serious drawback to use of the visually clean standard as the
sole residue criterion, since under conventional
acceptance criteria (dose calculation plus visually
clean), the protocol would also fail.
As a practical matter, one would expect that, for
many potent drugs, the use of visually clean as the
sole acceptance criterion would probably not be
applicable. The reason for this is that, in such cases,
the VL most likely would be above any doserelated calculation of surface contamination. On the
other hand, it is more likely to be applicable to most
non-potent drug products because the VL will be
significantly below the dose-based calculation. In
either case, it is important not to make assumptions
about applicability, but rather to actually calculate
the dose-based limit, perform the VL determination experiments, and compare the values. It should
also be noted that a visually clean criterion will
not address issues related to microbial or endotoxin
contamination.
Regulatory Issues
As mentioned earlier, proposed Annex 15 to the
European GMPs would seem to permit such an
approach of visually clean as the sole acceptance
criterion, provided that it is the most stringent
residue acceptance criterion. On the other hand, the
FDA personnel have written that “…relying only
on visual examination would not be scientifically
sound” (12). Since there is no further elucidation
of this topic, there may be two interpretations of
this statement. One is that visually clean as the sole
acceptance criterion is unacceptable. A second
interpretation is that it is only acceptable if supported by something else. An argument can be made
that, if it is supported by a scientific determination
of what the VL is (based on experimental data) and
by a calculation of the dose-based limit for comparison, then a scientific justification for its sole
use as a residue acceptance criterion can be made.
It should be noted that such an approach is untested
with the FDA. It is critical that if manufacturers
35
Downloaded from journal.pda.org on September 5, 2013
consider this approach, they carefully consider the
implications and carefully implement the process.
Poor or improper implementation can only lead to a
similar situation that occurred at the beginning of
cleaning validation, where the misuse of rinse sampling (e.g., using only the U.S.P. Purified Water specifications as the acceptance criteria) gave rinse sampling an undeserved negative reputation. On the other
hand, proper justification and implementation may
result in simplifying cleaning validation protocols
without sacrificing quality or scientific credibility.
Acknowledgement
The support of STERIS Corporation for the work
in this paper is gratefully acknowledged.
References
1. Mullen, M. V., “Cleaning Limits: Determining
W h a t i s G o o d S c i e n c e ,” . P r e s e n t e d a t 6 t h
Annual Cleaning Validation for Pharmaceutical
and Biotechnology Manufacturers, Institute for
International Research, Princeton, New Jersey
(October 27-29, 1999).
2. Vey s o g l u , T. , “ I m p l e m e n t i n g a P r a c t i c a l
A p p r o a c h t o S e t t i n g C l e a n i n g Va l i d a t i o n
Residual Acceptance Limits,” Presented at 7 th
A n n u a l C l e a n i n g Va l i d a t i o n f o r P h a r m a ceutical and Biotechnology Manufacturers,
Institute for International Research, Philadelphia, Pennsylvania (September 18-20, 2000).
3.
Lauer, H. C., “Cleaning and Cleaning
Validation in the Field of Dry Pharmaceutical Production,” Presented at ISPE European Seminar,
Zurich, Switzerland (September 18, 2000).
4. Mendenhall, D. W., “Cleaning Validation,”
Drug Development and Industrial Pharmacy,
15 (13), 2105-2114 (1989).
5. FDA, Guide to Inspections of Validation of
Cleaning Processes (1993).
8. Fourmen, G. L. and M. V. Mullen, “Determining Cleaning Validation Acceptance Limit
for Pharmaceutical Manufacturing Operations,”
P h a r m a c e u t i c a l Te c h n o l o g y, 1 7
(4),
54-60 (1993).
9. Pharmaceutical Inspection Cooperation Scheme,
“Recommendations on Cleaning Validation,”
Document PR 1/99-2. Geneva, Switzerland
(April 1, 2000).
10. Draft 4 of Annex 15 to 1997 EU Guide to Good
Manufacturing Practice – Eudralex Volume 4.
“Validation master plan/design qualification/
installation and operational qualification/
non-sterile process validation/cleaning validation,” European Commission, Working Party on
Control of Medicines and Inspections
(September 17, 1999).
11. Jenkins, K. M. and A. J. Vanderwielen, “Cleani n g Va l i d a t i o n : A n O ve r a l l P e r s p e c t ive ,”
Pharmaceutical Technolog,y, 18 (4), 60-73
(1994).
12. FDA, Center for Drug Evaluation and Research,
Human Drug CGMP Notes, 6:2 (1998).
6. 21 CFR Part 211.67(b)(6).
7. P DA Tec h n i c a l R ep o r t N o . 2 9 , “ Po i n t s t o
Consider of Cleaning Validation,” Bethesda,
Maryland (1998).
36
PDA Journal of Pharmaceutical Science and Technology
Downloaded from journal.pda.org on September 5, 2013
An Authorized User of the electronic PDA Journal of Pharmaceutical Science and
Technology (the PDA Journal) is a PDA Member in good standing. Authorized Users are
permitted to do the following:
·Search and view the content of the PDA Journal
·Download a single article for the individual use of an Authorized User
·Assemble and distribute links that point to the PDA Journal
·Print individual articles from the PDA Journal for the individual use of an Authorized User
·Make a reasonable number of photocopies of a printed article for the individual use of an
Authorized User or for the use by or distribution to other Authorized Users
Authorized Users are not permitted to do the following:
·Except as mentioned above, allow anyone other than an Authorized User to use or access the
PDA Journal
· Display or otherwise make any information from the PDA Journal available to anyone other than an
Authorized User
·Post articles from the PDA Journal on Web sites, either available on the Internet or an Intranet, or in
any form of online publications
·Transmit electronically, via e-mail or any other file transfer protocols, any portion of the PDA
Journal
·Create a searchable archive of any portion of the PDA Journal
·Use robots or intelligent agents to access, search and/or systematically download any portion of the
PDA Journal
·Sell, re-sell, rent, lease, license, sublicense, assign or otherwise transfer the use of the PDA
Journal or its content
·Use or copy the PDA Journal for document delivery, fee-for-service use, or bulk reproduction or
distribution of materials in any form, or any substantially similar commercial purpose
·Alter, modify, repackage or adapt any portion of the PDA Journal
·Make any edits or derivative works with respect to any portion of the PDA Journal including any text
or graphics
·Delete or remove in any form or format, including on a printed article or photocopy, any copyright
information or notice contained in the PDA Journal