Cleaning validation in the pharmaceutical industry

INTRODUCTION 
Cleaning validation in the pharmaceutical industry has been a topic of ever-increasing interest and scrutiny in recent Food and Drug Administration (FDA) inspections. The validation of procedures used to clean the equipment employed during the various steps of a manufacturing process is a clear requirement of current Good Manufacturing Practice (cGMP). As such, FDA inspectors now expect to see a functioning cleaning validation program with appropriate documentation in place during their inspections.
The requirement that equipment be clean before being used is not a new concept. The equally important requirement that it also be sanitary is many times obfuscated by the word, clean.
In response to the often-asked question "what is clean," the FDA published a guidance document: the 2004 FDA "Guide to Inspections Validation of Cleaning Processes."
The FDA's guide to inspections, which "intended to cover equipment cleaning for chemical residues only," includes:
1. "FDA expects firms to have written procedures [Standard Operating Procedures (SOPs)] detailing the cleaning processes ..."
2. "FDA expects firms to have written general procedures on how cleaning processes will be validated."
3. These procedures will "address who is responsible for performing and approving the validation study, the acceptance criteria, and when revalidation will be required."
4. "FDA expects firms to conduct the validation studies in accordance with the protocols and to document the results of studies."
5. Besides assuring chemical cleanliness, "the microbiological aspects of equipment cleaning should be considered. This consists largely of preventive measures ..."
6. "Determine the specificity and sensitivity of the analytical method used to detect residuals or contaminants."
7. "The firm should challenge the analytical method in combination with the sampling method(s) used to show that contaminants can be recovered from the equipment surface and at what level ..."
8. "Direct sampling (e.g., with swabs) is 'most desirable,' although rinse sampling may be satisfactory." 

OBJECTIVES
The objectives of this article are to establish a broad basis for cleaning validation policy and programs, and to determine the requirements, procedures, acceptance limits, and working papers needed to support this vitally important activity.
Cleaning Validation Protocol
Cleaning validation protocols should be developed, approved, and executed in accordance with the SOPs covering these activities in place at the time. A typical cleaning validation protocol should consist of Objective, Sampling and Testing Methodologies, and Acceptance Criteria sections.
* Objective
This section defines the intention and scope of the cleaning validation exercise. Additionally, it will include information such as equipment names, identification numbers, the name(s) and type(s) of product being cleaned from the equipment, and the individual components of the product and equipment under investigation.
* Sampling and Testing Methodologies
This section should typically include a step-by-step explanation of sampling techniques and requirements, as well as the specific analytical procedures to be used in the analysis of those samples. It should specify which laboratories are to be involved in the testing and any precautions to be taken throughout the validation exercise.
A visual check should be incorporated into the cleaning assessment. The sampling technique chosen to evaluate the effectiveness of the cleaning procedure should be swabbing, the fluid rinse of samples, or a combination of both methods. The following sampling methods provide various levels of assurance concerning cleaning:
** Visual inspection
*** Active product contact parts of the equipment are individually examined (wherever possible) for cleanliness. This visual inspection allows the early localization and identification of any inadequacies in the cleaning procedure.
*** Qualitative -- dependent upon inspector and item sampled.
*** Subjective -- dependent upon inspector and item sampled.
** Rinse water sampling and analysis
*** According to 2004 FDA "Guide to Inspections Validation of Cleaning Processes:" "Two advantages of using rinse samples are that a larger surface area may be sampled, and inaccessible systems or ones that cannot be routinely disassembled can be sampled and evaluated."
*** Analysis can be quantitative, using pH, conductivity, particle count, microbial count, Total Organic Carbon (TOC) determination, spectrophotometry, bioassays, or limulus amebocyte lysate for pyrogens.
*** Recovery factor is uncertain; it involves dilution.
** Surface sampling and analysis
*** Removes adherent materials.
*** Analysis can be quantitative.
*** Precise definition of the area sampled is required.
** Surface sampling from coupons
*** Quantitative.
*** Depends on whether coupons are equivalent to the surface of interest.
*** Requires removing coupons from the system.
** Method Selection
Whenever possible, each piece of equipment should be dismantled into its individual components after cleaning and each part should be individually tested for cleanliness. In this manner, any inadequacies in the cleaning process will be more readily identified and localized.
It may not be practical or desirable to dismantle large or Clean-In-Place (CIP) equipment. Regardless, validation sampling and testing should commence as soon as possible after the cleaning process is complete to reduce the chance for contamination by outside sources. Equipment that has just been cleaned should be covered immediately by appropriate means to protect it from any contamination.
** Solvents
Aqueous or organic solvents used in the cleaning procedure, should be sufficient to remove residues, and at the same time, should be minimized to reduce the risk of reaction with or damage to the equipment, or the over-dilution of the residue and the resultant loss of analytical sensitivity.
Samples should be collected in clean or sterile containers. Sterile containers are suitable for this intended use. All validation samples must be properly labeled with complete information regarding the source of the sample, sampler's name, sampling date, reference number, product name, and the part of equipment from which the sample has been collected.
A sample of the rinse or swabbing solvent should always be included with the actual test samples to serve as a reagent blank for any chemical or microbiological determination when required.
All types of samples, physical, chemical, or microbiological, should be collected according to a written procedure and using techniques, reagents, equipment, and containers appropriate to the type of testing to be performed. Only trained personnel should perform the collection of these samples.
The environmental effectiveness of cleaning procedures should be assessed by surface sampling of non-product contact surfaces (e.g.: floors, walls, air ducts, exterior equipment surfaces, etc.). Samples should be collected and analyzed for potential contamination.
** Sampling Methods
The sampling method selection for cleaners, involves choosing between rinse water sampling, swabbing surfaces, coupon sampling, or placebo sampling. Rinse water sampling involves taking a sample of an equilibrated post-final rinse that has been re-circulated over all surfaces. Rinse samples should be correlated to a direct measuring technique such as swabbing.
Swabbing involves using a wipe or swab that is moistened with high purity water, such as Water-for-Injection (WFI) that is typically wiped over a defined area in a systematic multi-pass way always going from clean to dirty areas to avoid recontamination (e.g.: 10 cm side by side strokes vertically, 10 cm horizontally, and 10 cm each with the flip side of the swab in each diagonal direction). For TOC analysis, very clean swabs or wipes and sample vials should be used. (All of these are commercially available). The amount of residue is known to be uniformly distributed on the smooth surfaces of equipment parts. Also, the most difficult to clean or "worst-case" areas of the equipment should be identified and specifically targeted for sampling whenever possible.
** Residue Detection
Selecting a method to detect cleaner residues can involve specific methods for specific cleaner ingredients such as: High Performance Liquid Chromatography (HPLC), ion selective electrodes, flame photometry, derivative Ultra Violet (UV) spectroscopy, Thin Layer Chromatography, enzymatic detection, and titration. It can also involve non-specific methods that detect the presence of a blend of ingredients such as: TOC, pH, and conductivity. The FDA prefers specific methods, but will accept non-specific methods with adequate rationales for their use. For investigations of failures or action levels, a specific method is usually preferable.
** Analytical Evaluation
Analytical validation of the cleaning procedure should be performed after the approval of visual inspection (absence of stains or any materiel residue). The specificity, sensitivity, and percentage of recovery of the test method should be adequate to meet acceptance criteria.
For the swab method it may be necessary to determine:
*** The percentage recovery of the swab extraction procedure.
*** The effectiveness of the swab at recovering residues from equipment parts surface.
*** The interference of swab materials in the analysis.
For the rinse solution method it may be necessary to determine:
*** The percentage recovery of the rinse solution extraction procedure.
*** The effectiveness of the rinse solution at recovering residues from equipment parts surfaces.
*** The interference of the rinse solution in the cleaning procedure and analysis.
*** A correction for recovery efficiency in calculations for acceptable residue levels.
Percentage Recovery = 100 X Sample Concentration / Standard Concentration
The percentage recovery is important because it will be applied when evaluating the final residual concentration according to the relation:
Percentage of actual amount of residual = Calculated Amount X Percentage Recovery
It is very difficult to establish acceptable fixed limits for recovery percentage due to the individual difference in solubility of residues, the solvent used, and the nature of the manufacturing surfaces.
The following three factors contribute to the difficulty of establishing fixed limits for the recovery percentage:
1. The residues behavior toward the solvent used.
2. The solvent used.
3. The nature of the manufacturing surfaces.
Some products such as proteins, for example, have a very low solubility, so the percentage recovery may be as low as 10-20%. For soluble residue, a higher percentage recovery should be expected. In general we can expect an ideal percentage recovery that falls between 60% and 90%. It is very important to continuously develop the sampling and swabbing methods and reproducibility to improve percentage recovery values.
** Microbiological Cleaning Considerations
All equipment that comes in contact directly with raw material--intermediate as well as final product--must be considered for inclusion, because of its potential to act as a possible source for microbiological contamination. In addition, the facilities must be considered for the level of microbiological contamination appropriate to the area classification.
Microbiological samples should be collected prior to and throughout the cleaning procedure to assist in selection and confirmation of the efficacy of disinfectants and detergents. Microbiological cleanliness is assessed as<200 cfu / 100 [cm.sup.2] for non-sterile production.
It is important to determine the type of organism present. It is necessary to demonstrate the absence of pollution indicator organisms such as, Escherichia coli, Salmonella spp, and Pseudomonas aeroginosa, from all locations monitored. It is necessary, as well, to ensure that high levels of other microbial flora do not mask these organisms.
Within sterile production, attention must be paid to the number of organisms present rather than their type. The level of microbiological contamination of the rinse water should be [less than or equal to] 10 cfu / 100 ml. Sampling is repeated three times during the validation.
** Worst-Case Determination
Worst-case determination of cleaning validation is a crucial step in defining contamination limits and in cleaning procedure efficacy. A worst-case determination study should be based on: active product solubility; active product toxicity; smallest batch size that can be manufactured using the equipment concerned; the maximum daily dose of this product; the number of dosages that can be made from next batch (contaminated); the product in its largest available tables mass, or in case of ampoules or vials, the largest available filling volume, and in both cases, the highest daily dose; the total area with which the product comes into contact; the area of one tablet or the volume of one individual fill; and the total amount of residual contaminant (see Figure 1).
After completing the worst-case determination table, we can easily identify the product representing this case (A, B, or C). The table should list all products to be manufactured in the same equipment whatever the chemical and bioactivity types of actives.
* Acceptance Criteria
In determining the final acceptance criteria for a cleaning validation exercise, the calculation of the acceptable level of contaminant in the next product maximum therapeutic patient dose is of primary importance. Acceptance criteria are established by considering the contaminant type, the facility, and the risk to the operator, product, and patient.
More stringent acceptance criteria are required in the case of highly biologically active materials compared to some excipients.
Facilities that produce product based on a single chemical entity (dedicated facility) shall not be subjected to as stringent a standard as multi-purpose facilities. Dedicated areas offer a low-risk potential, whereas a multi-product area tends toward a higher risk of contamination.
The use of automated cleaning process will tend toward more reproducible results when compared with manual systems.
When a number of materials are potential contaminants, consider which items are to be removed by the cleaning process (e.g.: chemical intermediates, active ingredient, detergent, excipient, colour, flavour, degradation product, micro-organism, endotoxin, particulates, lubricants, residual solvents, moisture, etc.).
The equipment should demonstrate the absence of obvious liquid, liquid and solid residues, and be free from any noticeable "off" odor.
A worst-case approach should be adopted and the cleaning procedure should be validated for the least soluble and most difficult to clean active or finished product as well as any residual cleaning agent. When more than one piece of equipment or stage is involved in the processing, the cumulative effect of each should be taken into consideration.
Microbiological acceptance criteria for cleaning procedures should be established based on product type. There may be a requirement for all or certain specific microorganisms to be absent dependent upon product type.
Acceptance criteria should be justified in a rational, written, approved, document prior to the commencement of the validation exercise.
Worst-case criteria are to be calculated for a product group manufactured in common equipment and then are to be applied to a selected reference product. Adding a new product formulation to the existing product group may act on the existing acceptance criteria. Also, if the product is less soluble than the existing reference product, then the new product can become the reference product and full cleaning validation must be performed. The new product should become a "stand alone" case when the cleaning method required is not suitable for other products.
A safety factor of not more than 1 / 1000 (0.1%) of the active under investigation (contaminant) found in a single unit of the lowest dosage form of the next product should remain in the equipment after the cleaning procedure. A list summarizing the batch size of products manufactured through the same equipment should be prepared in order to determine the smallest batch size. This is an important step to calculate the carryover limit.
The calculation of acceptance criteria should be based on the following parameters:
** Residual limit of active (contaminant) expressed in mg / [cm.sup.2]: R
** 1/ 1000 of concentration of active (contaminant) per dose units: L
** Maximum allowable number of doses per day of next product (contaminated): D
** Smallest batch size in mg: B
** Concentration of active in unit dose of next product (contaminated) or the number of total dose units manufactured: C
** Total surface area of equipment parts in contact with the product (contaminant) expressed in [cm.sup.2]: T
** Surface swabbed in [cm.sup.2]: S
Thus, the residual limit in the cleaned equipment of active (contaminant) mg / [cm.sup.2] is calculated as follows:
R mg / [cm.sup.2] = L / D X B / C X S / T
For example:
Concentration of active (contaminant) per dose unit = 30 mg
Maximum allowable number of doses per day of next product (contaminated) = 4 Units
Batch size (smallest) in mg = 50 X [10.sup.6]
Concentration of active in unit dose of next product (contaminated) or the number of total dose units manufactured = 200 mg
Total surface area of equipment parts in contact with the product (contaminant) expressed in [cm.sup.2] = 45000 [cm.sup.2]
Surface swabbed in [ccm.sup.2] = 100 [cm.sup.2]
Residual limit of active (contaminant) in mg / [cm.sup.2] = 30 X 1/1000 / 4 X 50 X 106 / 200 X 100 / 45000
Residual limit of active (contaminant) in the equipment after cleaning = 4.17 mg / [cm.sup.2]
For the automated systems and where rinse is used and rinse volumes are known, the following equation could be used taking into consideration the total volume of final rinse in ml V:
R mg / ml = L / D X B / C X 1 / V
The active (contaminant) carryover from product A to the next product B (contaminated) per unit dose is calculated as follows:
For example, the concentration of active (contaminant) A per unit dose is 2 mg per day. Based on a safety factor of not more than 1 / 1000 (0.1%) of the active under investigation (contaminant) found in a single unit of the lowest dosage form of the next product, we can say:
The limit of concentration of active A found in total (standard) daily dose units of next (contaminated) product B is:
1 / 1000 X 2 mg = 0.002 mg or 2 [micro]g
Supposing that the total amount of contaminant product A found in the total product contacting parts of the equipment after cleaning is 9 mg (9000 [micro]g). The next smallest batch size worst-case (contaminated) B is 100 Kg and B unit dose weight (tablet weight for example) is 90 mg. The total allowable daily dose of B is three doses. The concentration of active contaminant A carried over to unit dose of product B is calculated:
Contaminant [micro]g / mg = 9000 [micro]g / 100 X [10.sup.6]
Contaminant [micro]g / mg = 9 X [10.sup.-5]
Contaminant [micro]g / Unit dose = 9 X [10.sup.-5] X 90
Contaminant [micro]g / Unit dose B = 0.0081
Contaminant [micro]g / Total Daily dose B = 0.0081 X 3
Contaminant [micro]g / Total Daily dose B = 0.0243
The active (contaminant) A carried over to total daily units dose of product B (contaminated) is inferior to the established limit of 2 [micro]g.
Another more conservative acceptance criteria limit could be adopted. It considers that no more than 5 ppm from any active product can be left on any part of equipment for potential carryover to next product.
This conservative limit could not be applied for all types of products of pharmaceutical forms. An acceptance limit of 5 ppm could be applied for products having a bioactivity or strength less than 10 mg per unit dose "highly bioactive" or for products having a high level of toxicity. The main inconvenience of the application of this limit is related to the analytical method Limit of Detection (LoD or LD) and to equipment sensitivity in detecting this value.
Detergent and cleaning agents should be treated by using the safety factor of 1 / 1000 of L[D.sup.50] value or less than 10 ppm, whichever is the lowest. Another assessment of detergent residuals could be adopted such that residues should not exceed the detection limit of the method of analysis for the relevant active detergent substance.
The effective removal of residues having pharmacological or toxicological activity is the primary concern in any cleaning procedure and validation of that procedure. In addition to chemical assay and microbiological testing, other tests, such as pH, TOC, and conductivity may be desirable.
Whenever there is a change in manufacturing process, product formulation, manufacturing equipment, or cleaning procedure, revalidation of the cleaning procedure must be considered.
Cleaning Standard Operating Procedure
The standard cleaning operation procedure should include the following essential points:
* Objective
* Scope
Describing the range of application for the SOP, equipment, and products.
* Responsibility
Identification of who is responsible for performing the cleaning operations.
* Procedure
Description of cleaning method to be used including cleaning agent; concentration of the detergents surfactants, and sanitizing agents used during the cleaning procedure; temperature of the wash and rinse water or other solvent(s); flow rate and/or pressure at which the wash and rinse solvents are delivered; volume or amount of water or other solvents used to wash and rinse the equipment; diagrams describing the location of difficult to clean areas and "trap points:" inspection and/or testing regime to assess cleanliness and dryness; and status labeling of equipment and facility to ensure cleanliness status to all personnel.
* Other Items
Additional concerns that should be considered in the cleaning SOP include the following:
Where preparation of a cleaning solution is performed locally, it must be against a procedure that includes manufacturer instructions, batch numbering, and expiration dating.
Training records of operators should be shown for each cleaning procedure.
When a validated, automated cleaning procedure is in place, which produces a validated printout of critical processing stages, a triple check of critical stages should be performed.
The validation review or re-validation status should be re-assessed based upon any changes to the operating situation, equipment replacement, cleaning procedure changes, regulatory requirements, or adverse market comments that are related to the cleaning validation. Re-validation should take place once a year at a minimum.
The use of non-specific test methods could be permissible for re-validation exercises provided that the limits set can be related to a specific result in the initial validation or can be justified by some other means. Methods such as drain water conductivity and TOC analysis may be employed.
* Analytical Method
The analytical method used to determine the residual amount of active should be validated. A proper performance qualification protocol and report should be appropriately established and approved prior to starting the cleaning procedure. This demonstrates that the laboratory equipment and techniques are capable of evaluating with precision, according to written and validated analytical methods, the small amount of residual contaminant (s) (active or others).
A simulation exercise using the active product and the same material surface of production equipment would be suitable for the validation of the analytical method. Serial dilutions of active standard preparations could be used along with a placebo preparation containing the additives and excipients. These preparations would be suitable for simulating contamination during this exercise.
Prior to the collection of samples from the cleaned equipment, it is important to prove the effectiveness of the swabbing method and the swab materials to be used.
Cleaning Analytical Method Validation
The following sections should be included:
* Objective
* Scope
Describe the active product (s) that could be evaluated by the method.
* Acceptance Criteria
Describe the method followed to determine the acceptance criteria. The major and critical acceptance criteria to be mentioned are as follows: active product recovery percentage and active residual (contaminant) [micro]g per [cm.sup.2], or [micro]g of active residual (contaminant) per maximum daily dose units of next product.
* Method
Description of analytical methods used: standard preparation, sample preparation, analytical equipment used, analytical parameters, equipment parameter, sample volume, materials used, and the determination of the following values (which are specific to the analytical method and are relative for each active product):
** Precision
** Accuracy
** Limit of Detection
** Limit of Quantitation (LoQ)
** Linearity (where appropriate, linearity of detector response for standard solution over a range of concentrations)
** Recovery percentage
** Absence of interference between swab materials and active product
** Absence of interference between solvent and active product
* Calculations
Residual Limit of active (contaminant) A in mg / [cm.sup.2] of cleaned equipment or the concentration of active (contaminant) A carried over to unit dose of product B (contaminated) is calculated.
* Conclusion
Cleaning Procedure Tools (Matrices)
Tables, schemas, and matrices are the main tools used during the preparation of cleaning validation protocol and procedure especially for multi-product areas and equipment. This will help cleaning procedure developers to determine the worst-case and to calculate by the most efficient method the limit of contamination.
* Equipment Parts Schema
Equipment parts schema are useful to illustrate the different critical parts of the equipment that are in direct contact with the product, parts area, location of difficult to clean areas or "trap points," and swab locations. It contains equipment name, identification number (tag number), equipment location, name of each item, item surface, swab factor, and products to be manufactured in the machine.
* Products Type Matrix
A table grouping the different active products according to their biological activity, physical characteristics, and toxicity should be established. This type of grouping gives an overview about the products' manufacturing area and equipment to be used. Products having the same characteristics may require a dedicated facility or equipment. A product type matrix helps to choose appropriate cleaning procedures for each group of products.
Based on this information, similar products should be gathered into one group.
Multi-Product Equipment Contamination Acceptance Criteria Matrix
Where equipment is used for multi-products manufacturing, it is useful to establish a matrix listing the contamination acceptance limits of products compared to the worst-case. This matrix gives a quick overview about the maximum contaminant carryover per product (contaminant followed by contaminated).
A, B, C, D, and E are the different products manufactured using the same equipment. Contaminant limit for each product expressed in [micro]g / [cm.sup.2] is determined during the active product carryover acceptance criteria determination exercise. Values are reported properly in the matrix. For example, after manufacturing product D (contaminant), the equipment is cleaned according to the cleaning procedure then sampled (by swabs). Samples after analysis give the result of 18 [micro]g / [cm.sup.2] as total residue of active product D (contaminant) in all product contact parts of the equipment. If the next product to be manufactured is B, the allowable limit of contamination figured in the matrix for this case (D followed by B) is 25 [micro]g / [cm.sup.2]. Consequently, the equipment will be declared "clean" and labeled accordingly.
Cleaning validation matrices should be reviewed and revalidated following any change of cleaning elements e.g., modification of cleaning procedure, use of new equipment, and equipment modification could result in surface change of product contact parts, changes in regulatory requirements, introduction of new product, etc.
 
CONCLUSION
It is practically impossible to prove that production equipment is "clean" at the level of 100%. However, it is possible to prove that the traces of active product remaining, spread through the equipment parts, are within an acceptable limit and that we are capable of detecting and quantifying these trace levels.
Cleaning validation provides a means of proving that the contamination levels have been reduced below contamination acceptance limits.
The cleaning validation program should involve a rational monitoring program to maintain a validated state. Cleaning validation activity should cover active residue identification, active residue detection method selection, sampling method selection, the establishment of residue acceptance criteria, methods validation, recovery studies, and the identification of equipment parts in direct contact with the product.
The good preparation and proper implementation of cleaning validation tools (matrices and tables) is a determinant factor in the success of a cleaning validation program.

ABOUT THE AUTHOR
Mowafak Nassani Ph.D. has a Doctoral degree from U.S.T.L. University of Montpellier, France, in Analytical, Industrial Chemistry. He is the General Director of Pharmaceutical Validation Services at PVS Canada. He worked as a senior consultant in QA/QC, GMP, compliance, auditing and validation for I.C.C.E, Brussels, Belgium. He has occupied the post of QA/QC Manager in leading multinational pharmaceutical companies.
Dr. Nassani can be reach by phone at: +1 (514) 991-2494 or by e-mail at: mnassani@pharmavs.com
REFERENCES
1. Guidance for Industry, "Non-clinical Studies for the Safety Evaluation of Pharmaceutical Excipients," 5/18/2005
2. FDA, "Guide to Inspection of Validation of Cleaning Processes," July 2004.
3. International Conference on Harmonization (ICH), "Guidance for Industry: Q3A Impurities in New Drug Substances," 2/11/2003
4. Validation of Analytical Procedures: Methodology, FDA Guidance, December 1997.
5. FDA, "Guide to Inspection of Pharmaceutical Quality Control Laboratories," July 1993.