2 3 Cleaning Validation Nairobi May 2011

WHO Prequalification of Medicines ProgrammeWHO Prequalification of Medicines ProgrammeTraining in selected GMP topics for manufacturers and Training in selected GMP topics for manufacturers and

GMP InspectorsGMP InspectorsNairobi, 09 – 12 May 2011Nairobi, 09 – 12 May 2011

Deusdedit K. MubangiziTechnical Officer

World Health OrganisationEMP/QSM/PQM

[email protected]

2-3 CLEANING VALIDATION

mailto:[email protected]

Selected GMP topics for manufacturers and GMP inspectors Nairobi, May 9-12, 20112

Cleaning Validation

ObjectivesTo review:

• General requirements for cleaning validation

• Validation protocol requirements

• How to check limits

• Analytical requirements

• Sampling methods


Cleaning Validation: Points to consider

• When to use a matrix approach.

• Selection of the target molecule/"worst case" (including

detergents).

• Calculation of the acceptable carry over residues and

acceptable limits.

• Calculation of the product contact surface area.

• Determination of sampling locations.

• Determination and use of recovery factors.

• Suitable method of analysis for cleaning validation.


Cleaning Validation: General Principles• Written SOPs for cleaning processes – which should be validated• A company should have a cleaning policy and cleaning validation procedure

to cover:– contact surfaces;– cleaning after product changeover;– between batches in campaigns;– bracketing products for cleaning validation; and– periodic evaluation and revalidation of the number of batches

manufactured between cleaning validations.• The company has to prove consistency• What are the variables when a cleaning procedure is followed?• How many consecutive applications of the cleaning procedure

should be performed?– At least three consecutive applications of the cleaning

procedure should be performed and shown to be successful to prove that the method is validated.

• Must ensure training of personnel


Why cleaning validation is so important (1)

• The objectives of GMP include prevention of possible contamination and cross-contamination.

• Potential sources of contamination:

– contaminants (e.g. microbes, previous products (both API and excipient residues), residues of cleaning agents, airborne materials (e.g. dust and particulate matter), lubricants and ancillary material, such as disinfectants

– Also decomposition residues from product or detergents

• Adequate cleaning procedures play an important role in preventing contamination and cross-contamination.


Why cleaning validation is so important (2)• Validation of cleaning methods provides documented

evidence that an approved cleaning procedure will provide clean equipment, suitable for its intended use.– to prove that the equipment is consistently cleaned of

product, detergent and microbial residues to an acceptable level.

• It is considered important in multiproduct facilities - should be performed, e.g. for equipment, sanitization procedures and garment laundering.

• Cleaning validation may not be necessarily for non-critical cleaning, e.g. between batches of the same product (or different lots of the same intermediate in a bulk process), or of floors, walls, the outside of vessels, and following some intermediate steps.


Why cleaning validation is so important (3)What the Guidelines say

• “Particular attention should be accorded to the validation of … cleaning procedures” (WHO, 4.11)

• “Cleaning validation should be performed in order to confirm the effectiveness of a cleaning procedure” (PIC/S)

• “The data should support a conclusion that residues have been reduced to an ‘acceptable’ level” (FDA)


Cleaning process parameters• Factor that affect the quality of cleaning include:

– The cleaning agent– Time:

• length of leaning,• Dirty equipment hold-time (verification testing required if exceeded ± extended

cleaning)• Cleaned equipment hold-time.

– Action, e.g. Pressure of water spray with spray gun– Concentration– Temperature– Surface type and quality, e.g., stainless steel, glass, and a variety

of plastics.• As a general rule, cleaning of a given contamination type from glass-lined

vessels is easier than cleaning the same residue from stainless steel vessels because glass is a much smoother surface.


Factor that affect the quality of cleaning…(2)

– Contamination levels– Contamination conditions: "freshly deposited" ("wet"),

"dried," "baked," and "compacted."– Mixing– Water quality: water for final rinse must be at least of

the same quality as water added for manufacturing in the subsequently manufactured product.

– Rinsing– Environmental factors, e.g. Humidity and air quality


Cleaning validation protocol (1)

Should include :

• Objective of the validation

• Responsibility for performing and approving validation study

• Description of equipment to be used

• Interval between end of production and cleaning, and commencement of cleaning procedure

• Cleaning procedures to be used

• Any routine monitoring equipment used

• Number of cleaning cycles performed consecutively

• Sampling procedures used and rationale

• Sampling locations (clearly

defined)


Cleaning Validation Protocols (2)

• Approved by QC or QA and to cover, e.g.– disassembly of system;– pre-cleaning;– cleaning agent, concentration, solution volume,

water quality;– time and temperature;– flow rate, pressure and rinsing;– complexity and design of the equipment;– training of operators; and– size of the system.


Cleaning Validation Protocols (3)• The cleaning validation protocol should include:

– objectives, responsible people;

– description of the equipment including the make, model, serial number or other unique code;

– equipment used for routine monitoring (e.g. conductivity meters, pH meters and total organic carbon analysers);

– number of cleaning cycles; sampling procedures (e.g. direct sampling, rinse sampling, in process monitoring and sampling locations) and the rationale for their use


Cleaning Validation Protocols (4)

• The cleaning validation protocol should include (2):– data on recovery studies (efficiency of the

recovery of the sampling technique should be established);

– analytical methods; – acceptance criteria (with rationale for setting the

specific limits) including a margin for error and for sampling efficiency;

– cleaning agent to be used;– revalidation requirements.


Record of cleaning validation

• The relevant cleaning records:– should be signed by the operator, checked by

production and reviewed by quality assurance,– and source data (original results) should be kept.

• The results of the cleaning validation should be presented in cleaning validation reports stating the outcome and conclusion.


Cleaning validation: manual cleaning

Personnel

• Manual cleaning methods are difficult to validate

– Cannot validate people; can measure proficiency

• Personnel:

– Must have good training

– Must be effectively supervised


Cleaning validation: sampling methods (1)Swab samples • It is a direct sampling method• Most commonly used method• Use “swabs” (inert material) - type of

sampling material should not interfere with the test

• Disadvantages– inability to access some areas– assumes uniformity of contamination

surface• Factors to be considered include:

– Extraction efficiency– Area swabbed, number of swabs used,

whether they are wet or dry swabs,– Swab handling and swabbing

technique– Location from which the sample is

taken (including worst case locations, identified in the protocol)

– Equipment surfaces swabbed (e.g. glass or steel)

Rinse samples • It is an indirect method• Allows sampling of:

– a large surface – areas that are inaccessible

or that cannot be routinely disassembled

• Provides an "overall picture"• Non-specific analytical

methods used: pH, conductivity, TOC

• Useful for checking for residues of cleaning agents

• Insufficient evidence of cleaning

• Used in combination with other methods such as swabbing


Cleaning validation: sampling methods (2)

• Once collected, the swabs are extracted with a solvent that removes contaminants, and the solvent extract is then analyzed by an appropriate scientific method to determine the type and amount of contaminants in the extract, like:

– TOC (total organic carbon) analysis or,

– HPLC (high performance liquid chromatography) or,

– other methods.

• This amount is then extrapolated to calculate the total amount of contaminant present on the equipment surface that could be carried over into the next processed batch.

• In a like manner rinse water samples can be extracted with a solvent or concentrated and then analyzed.

• Should the analysis show that a particular contaminant level is above the established maximum acceptable level, the equipment is cleaned again and again tested, until all possible contaminants are below the maximum acceptable level.

• Once this is achieved the equipment is ready for use in processing the next batch of product or intermediate.


Cleaning validation: quality of the swab (1)• When using swabs for sampling, care must be taken to insure that

products extracted from the swabs themselves do not interfere with the sampling method.

• Sometimes materials extracted from the swabs can mask the presence of the contaminants collected from the equipment surface as they are detected in the same manner as the contaminant of interest.– run a “blank” on the swabs (extracting clean swabs with the solvent blend

and analyzing them by HPLC), and ensure that no peak is detected at the same wavelength as the target product/residue in the sample swabs.

– If the blank signal is not consistent, you cannot simply subtract the blank amount from the amount determined for the swabs used in the analysis.

– This is a common with polyester swabs. Polyester fabric contains extractable “oligomers” (short chains of the ester molecule but much smaller than the polyester molecule). Oligomers are easily extracted from the polyester fibers by many of the solvents commonly used in such analyses.

– The oligomers are different for swabs from different manufacturers, reflecting different sources of the polyester material used in the swab construction, and different oligomers will have different light absorption wavelengths.


Cleaning validation: quality of the swab (2)• To solve this problem the customer had a choice of trying a different swab

from a different manufacturer or using a cleaner swab, which would aid in reducing the total amount of oligomer extractables in each sample.

– If a different manufacturer’s swab is used it is likely the source of the polyester material used in the swab construction would be different and different oligomers would be extracted. These different oligomers may have a light absorption wavelength that would be different from the wavelength with which the customer is concerned, so there would be no interference with the customer’s analysis procedure.

– It is possible to use a special cleaning procedure to temporarily reduce the amount of oligomer contamination in the competitor’s swab, even though the offending peaks would return slowly over time, requiring repeated treatment of batches of swabs used in the analysis.

• In summary such issues with the materials used in quantitative sampling procedures occur frequently and must be taken into account when designing a cleaning validation procedure. The choice of materials used in the analysis can sometimes effect the accuracy of the analytical procedure and complicate the performance of the procedure.


Rinse samples (indirect method)

l The manufacturer has to provide evidence that samples are accurately recovered

l What is considered acceptable in terms of recovery?

Recovery

< 50% is consideredquestionable

> 80% is considered good

> 50% is considered reasonable


Cleaning validation: analytical methods

Analytical method:• Should use validated analytical

methods• Must be sensitive assay

procedure:– HPLC, GC, HPTLC– TOC– pH – conductivity– UV– ELISA

Validation should include:

• Precision, linearity, selectivity

• Limit of Detection (LOD)

• Limit of Quantification (LOQ)

• Recovery, by spiking

• Consistency of recovery

•The detection limit for each analytical method should be sufficiently sensitive to detect the established acceptable level of the residue or contaminants.


Cleaning validation limits (1)

• The starting point for any determination of residue acceptance limits is the amount of residue from the cleaning process that could be present in the subsequently manufactured product without posing an unreasonable risk.– One would prefer that no residue is present. However, it is

impossible to measure "no residue." Even the criterion of being below the limit of detection (LOD) of the analytical procedure is not by itself a very good method for selecting residue limits. As methods are improved and have even lower LODs, a cleaning process which was previously viewed as acceptable can become unacceptable.


Cleaning validation limits (2)

• for contamination of the next product, the units typically are ppm or µg/g; requires information on area swabbed and the swab recovery factor.

• for surface contamination, the units are usually µg/cm2; requires information on batch size and equipment surface area.

• for the analyzed sample, the units are typically µg or µg/g; requires information on area swabbed and the swab recovery factor.– an acceptance limit of 3.2ppm in the subsequent product is not

necessarily the same as 3.2ppm in the analyzed sampled prepared by a swabbing procedure.


Cleaning validation: setting limits (3)Setting limits

• Uniform distribution of contaminants not guaranteed

• Decomposition products to be checked

• Setting limits; cleaning criteria:

– visually clean

– 10ppm in another product

– 0.1% of therapeutic dose• by Fourman and Mullen at Eli Lilly

• The most stringent of three options should be used

“Visually clean” • Always first criteria• Can be very sensitive but

needs verification• Use between same product

batches of same formulation

• Illuminate surface adequately• Spiking studies need to

determine level at which residues are visible


Cleaning validation: setting limits (4)

“10ppm”• Historical, found in:

– In some poisons regulations– Pharmacopoeial limit test

• Assumes residue to be harmful as heavy metal.

• Useful for materials for which no available toxicological data.

• Not for pharmacologically potent material.

not more than 0.1%

• Proportion of MINIMUM daily dose of current product carried over into MAXIMUM daily dose of subsequent product

• Need to identify worst case


Formulae and Examples:Limit in Subsequent Product (1)

• the information needed is:

– the minimum daily dose of the active being cleaned (Product A)

– the maximum daily dose of the subsequently manufactured drug product (Product B).

• L 1 = (0.001)(min. daily dose of active in Product A)

max. daily dose of Product B



• Assume that product A has an active at a level of 2,000µg/mL and is dosed at 5mL from 3 to 5 times daily. The minimum daily dose is:

• 2000µg/mL x 5mL/dose x 3 doses/day = 30,000µg/day• Also assume that product B is dosed at 5mL from 2 to 4 times a

day, then the maximum daily dose of Product B would be:• 5mL/dose x 4 doses/day = 20mL/day

– Note that the calculation for the subsequent product is independent of what the active is or at what level that active is present.

• The residue limit in the subsequent product:• L1 = 0.001 x 30,000µg/day = 1.5µg/mL

20mL/day



• This calculation of 1.5µg/mL (or approximately 1.5ppm assuming a specific gravity of 1.0) is independent of batch size and the surface area of the equipment.

• This means that this calculation can be done as soon as information on the composition and relevant dosing of the two products is available.

• The calculated value of 1.5ppm should be compared to the 10ppm "default" value and the lower value used for subsequent calculations.



• Safety factors other than 0.001 could be selected:– safety factors of 0.001 for oral dose product and 0.0001 for parenterals

have been suggested.– While more stringent safety factors may be easily justified, it would require

significant justification if a safety factor less stringent than 0.001 were used.

• Companies may also base limits on parameters other than therapeutic doses:– rather than using the minimum daily dose of the active, other measures

such as the no observable effect level (NOEL) or the minimum pharmacological effect level may be selected.

– Since these will result in residue limits more stringent than limits based on the minimum daily dose, there is no scientific reason not to select these criteria.

– It should be noted, however, that arbitrarily selecting more stringent criteria may result in unreasonable cleaning overkill and may also stretch the detection limits of available analytical methods.


Formulae and Examples:Limit per Surface Area (1)

• Once the residue limit in the subsequent product is determined (using the 10ppm and 0.1% criteria), the next step is to determine the residue limit in terms of the active ingredient contamination level per surface area of equipment.

• This limit (L2, in µg/cm2) depends on:– the limit in the subsequent product (the lower of L1 and

10ppm);– the batch size of the product B (in kg), and;– the shared equipment surface area (in cm2).– This is expressed mathematically as:

• L2 = (Ll)(batch size of subsequent product)(1,000)shared equipment surface area

• where 1,000 is a conversion factor to account for ppm (limit in subsequent batch) and to convert kg to µg (for the batch size of subsequent product)



• Continuing forward with the example used, the limit in the subsequent product is 1.5ppm. Assuming the batch size of the subsequent product is 200kg and the shared equipment surface area is 60,000cm2, then the L2 limit is:

• L2 = 1.5ppm x 200kg x 1,000 = 5.0µg/cm2

60,000cm2

• In determining the surface area, all shared product contact surfaces, including piping, baffles, and the like, should be considered.

• It should be noted that this calculation for L2 assumes that the residue will be evenly distributed over all surfaces. In fact, this is generally not the case.– However, this assumption is still the worst case. If one is doing swab

sampling on the most difficult-to-clean locations, then assuming an even distribution will make it more difficult to meet the residue limits for those locations.



• If more than one product (for example, products B, C, D, and E) could possibly be manufactured following product A, then the surface area limits (L2) for cleaning following product A should be calculated for each subsequent product.

– The residue limit for cleaning validation purposes should be set at the lowest of these L2 surface area limits. This gives the manufacturing department more flexibility to make products in any order.

– It should be recognized, however, that there may be circumstances in which residue limits may require restrictions on which products may follow product A on the manufacturing schedule.


Formulae and Examples:Limit in the Analyzed Sample (1)

• The established residue limits: (L1) and (L2), are not directly measured by the analytical procedure.

• The analytical procedure measures the active in solution as a result of either swabbing and desorbing that swab into a suitable solvent or by doing rinse sampling and measuring the active in the rinse solvent.

• Swab sampling:– a specified surface area of the equipment is sampled and the swab

is then desorbed into a fixed amount of solvent.

• To determine the residue limit (L3 in µg/g or ppm) in the analytical sample (the solvent the swab is desorbed into), one must know:– the surface area residue limit (L2), the surface area swabbed (in

cm2), and the amount of solvent the swab is desorbed into (in g).


Formulae and Examples:Limit in the Analyzed Sample (1)

• The limit in the analyzed sample is calculated as follows:• L3 = (L2)(swabbed surfaced area)

amount of desorption solvent

• Continuing with the example used so far, if L2 is 5.0µg/cm2 and assuming the surface area swabbed is 25cm2 is and the amount of solvent used for desorption is 20g, then the limit L3 in µg/g is:

• L3 = 5.0µg/cm2 x 25cm2 = 6.3µg/g (or 6.3ppm)

20g


Recovery Factors• The acceptance limit in the analyzed sample should be adjusted

by a swab recovery factor.• There are two ways to do this:

– One is to include the swab recovery factor in the actual analytical calculation. For example, if the swab recovery factor was 0.80 (80 percent), and one measured 1.3ppm in the analytical procedure, then that value is adjusted by dividing the analytical result by the recovery factor to arrive at a determination of 1.3ppm/0.80 = 1.6ppm.

– The other alternative is to include the recovery factor in the numerator of the equation for L3. In this case, the recovery factor of 0.80 should be included in the numerator.

• While the numbers used will be different, the net effect of comparing the analytical result to the calculated limit will be logically the same.– One should standardize how this is performed in order to avoid

situations in which the recovery factor is used in both the calculation of the L3 limit and the determination of the analytical result on the desorbed solvent.


Visual Cleanness (1)• Visual cleanness is significant. If a surface is visually dirty, then

either the cleaning procedure is not acceptable or a once acceptable procedure is now out of control.

• The dividing line between visually clean and visually dirty is usually regarded as being in the range of 4µg/cm2*.*Fourman, G. L., and M. V. Mullen. 1993. Determining cleaning validation acceptance limits for pharmaceutical manufacturing operations. Pharmaceutical Technology17 (4):5 460.

– If the L2 surface contamination acceptance limit is calculated and found to be significantly above 4µg/cm2, then provided that critical surfaces are readily visible, it may be possible to default to visually clean as the only acceptance criteria.

– For potent drugs where the L2 acceptance criterion would typically be well below 1µg/cm2, a determination of visual cleanliness would have no significance as to the adequacy of cleaning.

• In this case, visually dirty would still be an indication of cleaning failure, but visually clean could not clearly indicate whether the residue was at an acceptable level.


Visual Cleanness (2)

• For cases where determination of visually clean is critical, determine the highest level that is not visible for that specific residue.– This can be done by spiking model surfaces (for example,

stainless steel coupons) with different levels of the residue (for example, 0.5,1.0, 2.0, 4.0, and 8.0µg/cm2) and having a trained panel of observers look at the coupons in a "blinded" manner to determine whether or not the coupons are visually clean.

– This should be done under viewing conditions (lighting, angle, distance) that simulate the viewing of actual equipment.

– The highest residue level at which all panel members consider the coupons visually clean establishes an acceptance level for that particular residue.


Microbiological Contamination• Setting acceptable limits for microbiological contamination is a more

difficult issue. No clear guidelines for process equipment exist.• One cannot expect the equipment to be free of all microorganisms,

especially if any final rinse involves non-sterile water, unless a final sanitizing or sterilization step is used.

• As a minimum, the criteria used for critical clean-room surfaces should be used*.

*Pharmacopeial Forum. 1997. Microbial evaluation of clean rooms and other controlled environments, in-process revision. Pharmacopoeia1 Forum 23 (1):3494-3520.

Grade Air sample

(CFU/m3)

Settle plates

(90mm diameter)

(CFU/4hours)

Contact plates

(55mm diameter)

(CFU/plate)

Glove print

(5 fingers)

(CFU/glove)

A < 3 < 3 < 3 < 3

B 10 5 5 5

C 100 50 25 -

D 200 100 50 -

•The presence of enteric organisms such as Escherichia coli or Enterococcus would ordinarily be unacceptable.


Grouping Strategies

• The two methods of grouping for cleaning validation are:– "product" grouping– "equipment" grouping.

• In either case, products or equipment are grouped together, and then a representative case (usually the worst case) is selected for the three PQ cleaning runs.


Product Grouping (1)• Conditions for product grouping:

– They must be similar products.• "similar" products may include products:

– with the same excipients and different levels of actives.– of similar types - all liquids, creams, tablets, and so on.

– The products are all made on the same equipment.– The products are cleaned with the same cleaning process:

• cleaned with the same cleaning SOP, including the method of cleaning, the cleaning agent, concentration, time, and temperature of cleaning.

• It is not acceptable to group together two products cleaned with the same cleaning agent but at different washing parameters.

• The next step is to select a representative product.– The goal here should be to select the product that is the worst

case or most difficult to clean.


Product Grouping (2)• There are 4 types of justifications used to select the worst case

among a group of similar products:– (1) lab simulation cleaning studies – ease of cleaning,– (2) solubility characteristics of the drug substance,– (3) solids or activity level of the drug substance, and– (4) operational experience in previous manufacturing.

• If none of these criteria produce a product that clearly should be the representative product for grouping purposes, select the product with the lowest analytical acceptance limit.– The acceptance limit for the representative product should not necessarily

be the acceptance limit calculated for that representative product but rather the lowest limit among all the products in that category.

– To determine the lowest calculated residue limit among the product group, it is necessary to calculate the residue limit per surface area for each product in the group.


Illustration of Residue Limit Determination for Representative Product in Grouping Strategy

Product Active Species Calculated Residue Limits

Group representative A 3.5 µg/cm2 of A

Product 2 B 1.5 µg/cm2 of B

Product 3 C 2.5 µg/cm2 of C

• The representative product in this case (the most difficult to clean) does not have the lowest calculated acceptance limit, but product 2 has.

• Therefore perform the three PQ cleaning runs with the representative product but measure the active A not to a limit of 3.5 µg/cm2 but to the lowest limit in the group (1.5 µg/cm2 of A).

– The logic here is that if one can clean the group representative to a limit of 1.5 pg/cm2 of A, then one should also be able to clean product 2, which is easier to clean, to a limit of 1.5 pg/cm2 of B.


Equipment Grouping (1)• If equipment is to be grouped, it must be

equipment cleaned with the same cleaning SOP.

• Group pieces of equipment with a similar design having either the same or different sizes:– such as storage vessels e.g. stainless

steel storage vessels of the same design that are 300L, 500L, and 1,000L

– It is generally unacceptable to group different types of equipment together.

• E.g. a ribbon blender and a V-blender.

Ribbon blender

V-blender


Equipment Grouping (2)• Once an equipment grouping is established, the next issue is to select

the representative piece of equipment in that group for the three PQ cleaning runs.– Choose the size difficult to clean (from other evidence) for cleaning

validation runs.– Otherwise, conduct three PQ runs on the equipment group, making sure

that at least one PQ run involves the largest size and at least one PQ run involves the smallest size.

• A special case of equipment grouping involves smaller items that are cleaned manually or in an automated parts washer:– the key to grouping smaller parts is that they are cleaned with the same

cleaning SOP– If some parts are cleaned by a manual sink scrub and others are cleaned

with high pressure spray cleaning, then all of the items cannot be grouped together.


Analytical Methods for Cleaning Validation (1)

• The selection of an analytical method for measuring residues is closely tied to:– the chemical nature of target residues, and– the analytical limits established for those residues.

• Chemical nature includes whether the target residue:– is organic or inorganic,– is soluble in water or other solvents,– its degree of polarity, and– its stability in the cleaning environment.


Analytical Methods for Cleaning Validation (2)

• An appropriate analytical method produces a result that has a logical, scientific link with the target residue.– If the target residue is an organic, non-ionized drug active (XYZ),

with acceptance criterion of 2ppm, then using conductivity as an analytical tool would be inappropriate because there is no scientific relationship between the presence of the target residue in the analytical sample and the measurement of conductivity in the test sample.

– If XYZ is degraded during the cleaning process, then an HPLC method specific to XYZ may not be appropriate for analyzing the target residue. The results most likely would be below the detection limit of the method.


Analytical Methods: Detection Limits• The LOD/LOQ of the analytical method should be at or

(preferably) below the acceptance criterion in the analyzed sample.– To build in a safety margin, most manufacturers prefer an

analytical method with an LOD of at least 25% of the target residue limit in the analyzed sample.

– Some companies have established overly restrictive requirements for their analytical methods because they have established requirements for the methods based on limits in the subsequent product (L1) rather than in the analyzed sample (L3).

– In the example used: L1 = 1.5ppm and L3 = 6.3ppm.• With a recovery factor of 0.8, L3 of 6.3ppm is equivalent to

6.3 x 0.8 = 5.0ppm measured by the method.• The LOD/LOQ should be ≤5.0ppm (or 1.3ppm for safe

margin) and not 1.5 x 0.8 = 1.2ppm (or 0.4ppm for safety margin).


Analytical Methods: LOD and LOQ

• LOD is the assay value at which it is still possible to say that the material is present, but it may be not possible to quantify with a specific value.– for chromatographic techniques, LOD is estimated at three times

the standard deviation of a baseline response.

• LOQ is the lowest assay value for which a reasonable confidence exists that the value is precise. For chromatographic procedures, the LOQ can be estimated– as 10 times the standard deviation of the baseline noise.– The LOQ can also be determined experimentally; as a practical

matter, it can be considered the lower limit of the validated range of the assay.


Analytical Methods: Specificity

• Specificity– an analytical procedure that measures only that species and

excludes all potentially interfering species.

• Nonspecific Methods– Nonspecific methods are usually methods that measure a gross

property that results from contributions from a variety of chemical species, e.g. conductivity and total organic carbon (TOC).

– an organic target residue may be measured by TOC of the analytical sample and assuming that all the carbon present is due to the target organic residue. If the amount of the target residue calculated by this method is below the acceptance criterion, then it is scientifically sound to say that the residue is less than the acceptance criterion.


Choice of the cleaning agent (1)• should be documented and approved by the quality unit

and should be scientifically justified on the basis of, e.g.– the solubility of the materials to be removed;– the design and construction of the equipment and surface

materials to be cleaned;– the safety of the cleaning agent;– the ease of removal and detection;– the product attributes;– the minimum temperature and volume of cleaning agent and rinse

solution;– the manufacturer’s recommendations;

• Detergents that have persistent residues such as cationic detergents which adhere very strongly to glass and are difficult to remove, should be avoided where possible.


Choice of the cleaning agent (2)

• The composition of the detergent should be known to the manufacturer and its removal during rinsing, demonstrated.

• Acceptable limits for detergent residues after cleaning should be defined. The possibility of detergent breakdown should also be considered when validating cleaning procedures.

• Detergents should be released by quality control and, where possible, should meet local food standards or regulations.


Cleaning validation

Questions for the GMP Inspector to ask

• How is equipment cleaned?

• Are different cleaning processes required?

• How many times is a cleaning process repeated before

acceptable results are obtained?

• What is most appropriate solvent or detergent?

• At what point does system become clean?

• What does visually clean mean?


Conclusion• The manufacturer needs a cleaning validation strategy• Assess each situation on its merits• Scientific rationale must be developed

– equipment selection– contamination distribution– significance of the contaminant

• “Visually clean” may be all that is required• Dedicated equipment should be used for:

– products which are difficult to clean, – equipment which is difficult to clean, – products with a high safety risk where it is not possible to achieve

the required cleaning acceptance limits using a validated cleaning procedure.

Cleaning validation


Workshop: Fitzmill Impact Mill

• Which are the critical areas for sampling?

Equipment Sampling Location Product Contact Material Sample Type Rationale


Workshop: Top mixing tank with discharge valve

• Which are the critical areas for sampling?

Equipment Sampling Location Product Contact Material Sample Type Rationale


Workshop: Sieve

• What structural issues should be considered in validating the cleaning procedure for a sieve?

• How do you determine the product contact surface area of a sieve?

Documents

2 3 Cleaning Validation Nairobi May 2011