CVM GFI #159 (VICH GL36) Studies to Evaluate the Safety of Residues of Veterinary Drugs in Human Food: General Approach to Establish a Microbiological ADI

Administrative guidance, not enacted legislation. This captured publisher edition preserves the listed document and attachments; later developments are not independently consolidated.

#159
Studies to Evaluate the Safety of
Residues of Veterinary Drugs in
Human Food: General Approach to
Establish a Microbiological ADI
VICH GL36(R)
Guidance for Industry
This revision has updated the volume of the human colon.
Submit comments on this guidance at any time. Submit electronic comments to
https://www.regulations.gov. Submit written comments to the Dockets Management Staff
(HFA-305), Food and Drug Administration, 5630 Fishers Lane, Rm. 1061, Rockville, MD
20852. All comments should be identified with the docket number FDA-2003-D-0433.
For further information regarding this document, contact AskCVM@fda.hhs.gov.
Additional copies of this guidance document may be requested from the Policy and Regulations
Staff (HFV-6), Center for Veterinary Medicine, Food and Drug Administration, 7500 Standish
Place, Rockville MD 20855, and may be viewed on the Internet at https://www.fda.gov/animal-
veterinary, https://www.fda.gov/regulatory-information/search-fda-guidance-documents, or
https://www.regulations.gov.
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Veterinary Medicine (CVM)
January 2022

Contains Nonbinding Recommendations
2
VICH GL36(R2) (SAFETY – MICROBIOLOGICAL ADI)
February 2019
For Implementation at Step 7
S TUDIES TO E VALUATE THE S AFETY OF
R ESIDUES OF V ETERINARY D RUGS IN
H UMAN F OOD :
G ENERAL A PPROACH TO E STABLISH A
M ICROBIOLOGICAL ADI
Adopted at Step 7 of the VICH Process
by the VICH Steering Committee
in February 2019
for implementation in August 2020
This Guidance has been developed by the appropriate VICH Expert Working Group and is
subject to consultation by the parties, in accordance with the VICH Process. At Step 7 of the
Process the final draft has been recommended for adoption to the regulatory bodies of the
European Union, Japan, and the USA.

Contains Nonbinding Recommendations
3
Table of Contents
1. 	INTRODUCTION................................................................................................................... 5
1.1. 	Objectives of the guidance............................................................................................. 5
1.2. 	Background .................................................................................................................... 5
1.3. 	Scope of the guidance..................................................................................................... 7
2. 	GUIDANCE ............................................................................................................................. 7
2.1. 	Steps in determining the need for a microbiological ADI .......................................... 7
2.2. 	Recommendations for determining NOAECs and NOAELs for the 	endpoints
of concern ........................................................................................................................ 8
2.2.1. 	Disruption of the colonization barrier ............................................................. 8
2.2.1.1. 	Detection of colonization barrier disruption .................................... 8
2.2.1.2. 	Test systems and study design ........................................................... 9
2.2.1.2.1.
	In vitro tests ....................................................................... 9
2.2.1.2.2.
	In vivo tests ........................................................................ 9
2.2.2. 	Increase in the population(s) of resistant bacteria in the human colon
(as defined in section 1.2. Background) .......................................................... 10
2.2.2.1. 	Detection of changes in the population of resistant bacteria ........ 10
2.2.2.2. 	Test systems and study design ......................................................... 10
2.2.2.2.1.
	In vitro tests ..................................................................... 10
2.2.2.2.2.
	In vivo tests ...................................................................... 11
2.3. 	General recommendations .......................................................................................... 11
2.4. 	Derivation of a microbiological ADI .......................................................................... 12
2.4.1. 	Disruption of the colonization barrier ........................................................... 12
2.4.1.1. 	Derivation of an ADI from
in vitro data ......................................... 12
2.4.1.2. 	Derivation of an ADI from
in vivo data .......................................... 13
2.4.2. 	Increase in the population(s) of resistant bacteria ........................................ 13
2.4.2.1. 	Derivation of an ADI from
in vitro data ......................................... 13
2.4.2.2. 	Derivation of an ADI from
in vivo data .......................................... 13
3. 	GLOSSARY........................................................................................................................... 13
4. 	REFERENCES...................................................................................................................... 17
APPENDIX A .............................................................................................................................. 18
1. 	Experimental Conditions ............................................................................................... 18

Contains Nonbinding Recommendations
4
2. 	Inoculum .......................................................................................................................... 18
3. 	Study duration................................................................................................................. 18
APPENDIX B .............................................................................................................................. 19
APPENDIX C .............................................................................................................................. 21
1. 	Calculation of MICcalc ..................................................................................................... 21
2. 	Example Calculation....................................................................................................... 21
APPENDIX D .............................................................................................................................. 23
1. 	Introduction ..................................................................................................................... 23
2. 	Examples of Test Systems for the Assessment of the Fraction of Oral Dose
Available to Microorganisms ......................................................................................... 23
3. 	Methodological Aspects of Test Systems....................................................................... 25
4. 	Description of How Test Systems Could Be Used in Determining the Fraction
of Oral Dose Available to Microorganisms .................................................................. 26

Contains Nonbinding Recommendations
5
Studies to Evaluate the Safety of Residues of Veterinary Drugs in
Human Food: General Approach to Establish a Microbiological
ADI
Guidance for Industry
This guidance represents the current thinking of the Food and Drug Administration (FDA or
Agency) on this topic. It does not establish any rights for any person and is not binding on FDA
or the public. You can use an alternative approach if it satisfies the requirements of the
applicable statutes and regulations. To discuss an alternative approach, contact the FDA staff
responsible for this guidance as listed on the title page.
1. 	INTRODUCTION
1.1. 	Objectives of the guidance
A variety of toxicological evaluations can be performed to establish the safety of veterinary drug
residues in human food. An issue that needs to be addressed for veterinary antimicrobial drugs is
the safety of their residues on the human intestinal flora. The objectives of this guidance are (1)
to outline the recommended steps in determining the need for establishing a microbiological
acceptable daily intake (ADI); (2) to recommend test systems and methods for determining no-
observable adverse effect concentrations (NOAECs) and no-observable adverse effect levels
(NOAELs) for the endpoints of health concern; and (3) to recommend a procedure to derive a
microbiological ADI. It is recognized that different tests may be useful. The experience gained
with the recommended tests may result in future modifications to this guidance and its
recommendations.
The contents of this document do not have the force and effect of law and are not meant to bind
the public in any way, unless specifically incorporated into a contract. This document is
intended only to provide clarity to the public regarding existing requirements under the law.
FDA’s guidance documents should be viewed only as recommendations, unless specific
regulatory or statutory requirements are cited. The use of the word should in Agency guidances
means that something is suggested or recommended, but not required.
1.2. 	Background
The intestinal flora plays an important role in maintaining and protecting the health of
individuals. This flora provides important functions to the host such as (1) metabolizing
endogenous and exogenous compounds and dietary components; (2) producing compounds that
are later absorbed; and (3) protecting against the invasion and colonization by pathogenic
microorganisms.

Contains Nonbinding Recommendations
6
Ingested antimicrobial drugs can potentially alter the ecology of the intestinal flora. They may
reach the colon due to incomplete absorption or may be absorbed, circulated and then excreted
via bile or secreted through the intestinal mucosa.
It is recommended that the microbiological endpoints of current public health concern that
should be considered when establishing a microbiological ADI are:
Disruption of the colonization barrier: The colonization barrier is a function of the normal
intestinal flora that limits colonization of the colon by exogenous microorganisms, as
well as overgrowth of indigenous, potentially pathogenic microorganisms. The capacity
of some antimicrobial drugs to disrupt this barrier is well established and known to have
human health consequences.
Increase of the population(s) of resistant bacteria: For the purposes of this guidance,
resistance is defined as the increase of the population(s) of bacteria in the intestinal tract
that is (are) insensitive to the test drug or other antimicrobial drugs. This effect may be
due either to the acquisition of resistance by organisms which were previously sensitive
or to a relative increase in the proportion of organisms that are already less sensitive to
the drug.
An extensive literature review did not reveal reports of human health effects (e.g.,
prolonged antimicrobial therapy, prolonged hospital stay, predisposition to infection,
treatment failure, etc.) that occur as a result of changes in the proportion of antimicrobial
resistant bacteria in the normal human intestinal flora. However, based on the
understanding of microbial ecology, such effects cannot be excluded.
Although the effect of antimicrobial residues in food on the human intestinal flora has been a
concern for many years, a harmonized approach to determine the threshold dose that might
adversely disturb the flora has not been established. International regulatory bodies have used a
formula-based approach for determining microbiological ADIs for antimicrobial drugs. These
formulae take into consideration relevant data including minimum inhibitory concentration
(MIC) data against human intestinal bacteria. Due to the complexity of the intestinal flora,
uncertainty factors have been traditionally included in the formula. However, the use of
uncertainty factors results in conservative estimates and it is recommended that more relevant
test systems be developed that allow a more realistic estimate of a microbiological ADI, possibly
without the use of these factors.
The present revised guidance is an attempt to address the complexity of the human intestinal
flora and reduce uncertainty when determining microbiological ADIs. The guidance
recommends a process for determining if a microbiological ADI is appropriate and discusses test
systems that take into account the complexity of the human intestinal flora. These test systems
could be used for addressing the effects of antimicrobial drug residues on human intestinal flora
for regulatory purposes.
Since further research is needed to confirm the reliability and validity of all test systems
discussed in this guidance (see Appendix A), this guidance does not recommend any one

Contains Nonbinding Recommendations
7
particular system for use in establishing an ADI. Instead, this guidance provides
recommendations for a harmonized approach to establish a microbiological ADI and offers
test options rather than specifying a testing regimen.
1.3. 	Scope of the guidance
This document provides guidance for assessing the human food safety of residues from
veterinary antimicrobial drugs with regard to effects on the human intestinal flora. However, it
does not limit the choice of studies that may be performed to establish the safety of residues in
human food with respect to adverse effects on human intestinal flora. This guidance does not
preclude the possibility of alternative approaches that may offer an equivalent assurance of
safety, including scientifically based reasons as to why microbiological testing may not be
appropriate.
2. 	GUIDANCE
If a drug intended for use in food-producing animals has antimicrobial activity, the safety of its
residues should be addressed with respect to the human intestinal flora. Derivation of a
microbiological ADI is only recommended if residues reach the human colon and remain
microbiologically active.
2.1. 	Steps in determining the need for a microbiological ADI
When determining the need for a microbiological ADI, the following sequence of steps is
recommended. The data may be obtained experimentally or from other appropriate sources such
as scientific literature.
Step 1. Are residues of the drug, and (or) its metabolites, microbiologically active against
representatives of the human intestinal flora?
• 	Recommended data:
- 	MIC data, obtained by standard test methods, from the following relevant genera of
intestinal bacteria (E. coli, and species of Bacteroides, Bifidobacterium, Clostridium,
Enterococcus, Eubacterium (Collinsella), Fusobacterium, Lactobacillus,
Peptostreptococcus/Peptococcus).
- 	It is recognized that the understanding of the relative importance of these
microorganisms is incomplete and that the taxonomic status of these organisms can
change. The selection of organisms should take into account current scientific
knowledge.
• 	If no information is available, it is recommended to assume that the compound and (or) its
metabolites are microbiologically active.
Step 2. Do residues enter the human colon?

Contains Nonbinding Recommendations
8
• 	Recommended data:
- 	Absorption, distribution, metabolism, excretion (ADME), bioavailability, or similar
data may provide information on the percentage of the ingested residue that enters the
colon.
- 	If no information is available in humans, it is recommended to use appropriate animal
data. If there is no available information, it is recommended to assume that 100% of
the ingested residue enters the colon.
Step 3. Do the residues entering the human colon remain microbiologically active?
• 	Recommended data:
- 	Data demonstrating loss of microbiological activity from in vitro inactivation studies
of the drug incubated with feces or data from in vivo studies evaluating the drug’s
microbiological activity in feces or colon content of animals.
If the answer to any of questions in steps 1, 2, or 3 is “no”, then the ADI should not be based
on microbiological endpoints and the remaining steps are not recommended.
Step 4. Assess whether there is any scientific justification to eliminate testing for either one
or both endpoints of concern. 	It is recommended that available information regarding
colonization barrier disruption and resistance emergence for the drug be considered. If a
decision cannot be made based on the available information, both endpoints should be
examined.
Step 5. Determine the NOAECs/NOAELs for the endpoint(s) of concern as established in
step 4. The most appropriate NOAEC/NOAEL should be used to determine the
microbiological ADI.
2.2. 	Recommendations for determining NOAECs and NOAELs for the endpoints of
concern
2.2.1. 	Disruption of the colonization barrier
2.2.1.1. 	Detection of colonization barrier disruption
Changes in bacterial populations are indirect indicators of potential disruption of the colonization
barrier. These changes can be monitored by various enumeration techniques in a variety of test
systems. A more direct indicator of barrier disruption is the colonization or overgrowth of an
intestinal ecosystem by a pathogen. In vivo test systems or complex in vitro test systems (e.g.,
fed-batch, continuous, or semi-continuous culture systems) have the potential to evaluate barrier
disruption as evidenced by colonization of a challenge organism added to the test system.
Challenge organisms (e.g., Salmonella, Clostridium) should be insensitive to the test drug.
Inoculation schemes with the challenge organisms should take into account the timing of the

9
Contains Nonbinding Recommendations
challenge relative to drug treatment, the number of organisms per challenge dose, and the
number of times that the test system is challenged.
2.2.1.2. Test systems and study design
2.2.1.2.1. In vitro tests
The use of MICs to assess the potential for a drug to disrupt the colonization barrier does not
take into account the complexity of the human intestinal flora. Therefore, the MIC50 of the most
relevant genus/genera for which the drug is active (see section 2.1. Steps in determining the need
for a microbiological ADI) results in a conservative estimate of a NOAEC for disruption of the
colonization barrier. The NOAEC estimate is conservative because, among other reasons, the
inoculum density is orders of magnitude lower than the bacterial population in the intestinal tract
(Ref. 1). Therefore, the NOAEC may be considered as an option to establish an ADI. The
isolates should be obtained from multiple healthy individuals and include a minimum of 10
isolates from each of the genera listed in section 2.1.
Each MIC test of a pure culture of a relevant isolate provides data for a single strain of a species.
Other in vitro test systems provide information for hundreds of bacterial species (>108 bacterial
cells/g) for each fecal inoculum. Each inoculum can be tested in replicate to determine treatment
effects. Based on all the above, in vitro systems using fecal batch cultures are inherently more
robust and relevant than the MIC test system.
Other test systems discussed below, which model the intestinal flora, may result in a more
appropriate NOAEC and possibly a higher ADI.
Fecal slurries provide a simple test system to derive a NOAEC for disruption of the colonization
barrier following short-term exposure to the drug and may be appropriate for dose-titration
studies. The slurries can be monitored for changes in bacterial populations and the production of
short chain fatty acids. These two response variables, when monitored together, can be used as
indirect indicators of barrier disruption. The NOAEC derived from this test system may prove to
be a conservative estimate of barrier disruption.
Semi-continuous, continuous and fed-batch cultures of fecal inocula may be appropriate to
evaluate disruption of the colonization barrier following prolonged exposure to the drug.
However, exploratory work using continuous and semi-continuous cultures has given various
NOAECs for barrier disruption because of differences in protocols. As a consequence, it is
recommended that study designs take into account the issues raised in Appendix A.
In the case of fecal slurries, semi-continuous and continuous cultures, and fed-batch cultures of
fecal inocula, there are unresolved issues such as the impact of fecal inocula (individual variation
and sex), dilution rate, duration of drug exposure, and reproducibility of the tests.
2.2.1.2.2. In vivo tests
In vivo test systems using human flora-associated (HFA) and conventional laboratory animals
may be suitable for the assessment of disruption of the colonization barrier. Compared to

10
Contains Nonbinding Recommendations
conventional laboratory animals, the intestinal flora of HFA animals possesses greater similarity
to the human intestinal flora, both in terms of the range of bacterial populations and metabolic
activity. However, the intestinal flora derived from humans may not be stable in the HFA
animals. The relative importance of the stability of the implanted flora and the specific
composition of the flora is unknown. For technical reasons, the conventional laboratory animal
can be tested in higher numbers, which allows a more robust statistical analysis of the results.
It is recommended that study design take into account factors such as animal species, sex,
inoculum variability among donors, number of animals per group, diet, randomization of
treatment groups, minimization/elimination of coprophagy, housing of animals within an
isolator, cross contamination within the isolator and route of drug administration (e.g., gavage,
drinking water). It is recommended that germ-free animals be inoculated in sequence, first with
a Bacteroides fragilis strain, followed by the fecal inoculum.
2.2.2. Increase in the population(s) of resistant bacteria in the human colon (as defined
in section 1.2. Background)
The guidance below highlights the considerations that should be taken into account when
addressing this endpoint.
2.2.2.1. Detection of changes in the population of resistant bacteria
It is recommended that studies to evaluate the emergence of resistance take into account the
organisms of concern in the intestinal tract and the documented resistance mechanisms to the
drug class. Preliminary information regarding the prevalence of resistance in the human
intestinal flora, such as daily variation within individuals and the variation among individuals
can be useful in developing criteria for evaluating resistance emergence. MIC distributions of
sensitive and known resistant organisms of concern can provide a basis to determine what drug
concentration should be used in the selective agar media to enumerate resistant organisms in the
fecal samples. Since drug activity against an organism can vary with test conditions, the MIC of
the organism growing on selective medium should be compared to the MIC determined by
standard methods (e.g., Clinical and Laboratory Standards Institute [CLSI] (Ref. 2, 3)). Changes
in the proportions of resistant organisms during pre-treatment, treatment and post-treatment
periods can be evaluated by enumeration techniques on media with and without the antimicrobial
drug, applying phenotypic and molecular methodologies.
Changes in antimicrobial resistance can be influenced by factors other than drug exposure (e.g.,
animal stress) which should be taken into consideration in animal test systems.
2.2.2.2. Test systems and study design
2.2.2.2.1. In vitro tests
The duration of exposure required for resistance to develop in a population of bacteria can be
dependent on the drug, the nature of the resistance mechanisms, and how it evolves in nature
(e.g., by gene transfer between cells, by gene mutations). For these reasons, acute studies of pure

11
Contains Nonbinding Recommendations
cultures to assess the endpoint are not recommended. Therefore, MIC tests are not
recommended to be used to determine a NOAEC for increases in resistant populations.
Defined cultures may provide useful information to determine the potential for a resistant
population to emerge due to mutation in an isolate and/or gene transfer among isolates.
However, these test systems are not designed to evaluate changes in resistant populations and are
not recommended.
Test systems using short-term exposure of fecal slurries to a drug are not recommended for
resistance emergence testing because the duration of the test is inadequate to assess changes in
resistant populations.
Continuous and semi-continuous cultures and fed-batch cultures of fecal inocula provide a means
to evaluate long-term exposure of bacteria to the drug. Refer to Appendix A for issues that are
recommended to be addressed regarding study conduct and data evaluation.
2.2.2.2.2. In vivo tests
Changes in resistant populations can be assessed in HFA-rodents. General study design and
supporting protocol should follow the recommendations stated in section 2.2.1.2.2. In vivo tests.
The test system supports a complex flora and would be a source of genetic resistance
determinants. The system accommodates more replication than the continuous or semi-
continuous culture systems, but less than fed-batch cultures. The variability of the HFA-rodent
test has not been assessed; however it is useful for identifying sex differences. There are also
advantages to conducting resistance studies in conventional laboratory animals.
HFA-rodents and conventional animals provide means to evaluate the potential for resistance
emergence following long-term exposure of bacteria to the drug. Refer to Appendix A for issues
that are recommended to be addressed for study conduct and data evaluation.
2.3. General recommendations
• It is recommended that fecal samples or bacterial isolates from human donors be obtained
from healthy subjects with no known exposure to antimicrobial agents for at least 3 months.
• In the case of in vivo tests, it is recommended that the test species selected for testing should
allow for (1) maximum independent replication; (2) sufficient quantity of feces to be
collected for analyses; and (3) minimal coprophagy. Evaluation of the sex should be
considered unless data demonstrate that only one sex is appropriate.
• Statistical issues should be addressed when designing studies of antimicrobial residues (see
Appendix B).
• It is recommended that the pre-validation and validation process, such as that being
developed by OECD since 1996 (Ref. 4), be considered for subsequent validation of test
systems to assess the effects of antimicrobial drugs on human intestinal flora. The process
should be adapted and modified for this use depending on the test system being validated.

Contains Nonbinding Recommendations
12
• 	Study designs should take into account unresolved issues of the effects of storage and
incubation conditions on fecal inocula.
2.4. 	Derivation of a microbiological ADI
When more than one value can be determined for the microbiological ADI, in accordance with
the methods discussed below, the most appropriate value (relevant to humans) should be used.
2.4.1. 	Disruption of the colonization barrier
2.4.1.1. 	Derivation of an ADI from in vitro data
If the endpoint of concern is disruption of the colonization barrier, the ADI may be derived from
MIC data, fecal slurries, semi-continuous, continuous, and fed-batch culture test systems.
ADI derived from MIC data:
MIC calc : The MIC calc is derived from the lower 90% confidence limit for the mean MIC 50 of the
relevant genera for which the drug is active, as described in Appendix C.
ADI derived from other in vitro test systems:
NOAEC: It is recommended that the NOAEC derived from the lower 90% confidence limit for
the mean NOAEC from in vitro systems be used to account for the variability of the data.
Therefore, in this formula uncertainty factors are not generally needed to determine the
microbiological ADI.
Volume of colon content: The 500 mL value was estimated from the result of three-dimensional
abdominal magnetic resonance imaging measurement (Ref. 5).
Fraction of an oral dose available for microorganisms: It is recommended that the fraction of an
oral dose available for colonic microorganisms be based on in vivo measurements for the drug
administered orally. Alternatively, if sufficient data are available, the fraction of the dose
available for colonic microorganisms can be calculated as 1 minus the fraction (of an oral dose)
excreted in urine. Human data are encouraged, but in its absence, non-ruminant animal data are
recommended. In the absence of data to the contrary, it should be assumed that metabolites have
antimicrobial activity equal to the parent compound. The fraction may be lowered if the

Contains Nonbinding Recommendations
13
applicant provides quantitative in vitro or in vivo data to show that the drug is inactivated during
transit through the intestine.
2.4.1.2. 	Derivation of an ADI from in vivo data
The microbiological ADI is the NOAEL divided by the uncertainty factor.
Uncertainty factors for in vivo studies should be assigned as appropriate, taking into
consideration the class of compound, the protocol, numbers of donors, and sensitivity of the
measured outcome variables.
2.4.2. 	Increase in the population(s) of resistant bacteria
2.4.2.1. Derivation of an ADI from in vitro data
If the endpoint of concern is an increase in the population(s) of resistant bacteria, NOAECs
derived from semi-continuous, continuous, and fed-batch culture test systems may be used to
establish a microbiological ADI.
NOAEC: It is recommended that the NOAEC derived from the lower 90% confidence limit for
the mean NOAEC from in vitro systems be used to account for the variability of the data.
Therefore, in this formula uncertainty factors are not generally needed to determine the
microbiological ADI. However, where there are concerns arising from inadequacies in the
quality or quantity of in vitro data used in determining the NOAEC, the incorporation of an
uncertainty factor may be warranted.
2.4.2.2. 	Derivation of an ADI from in vivo data
The microbiological ADI is the NOAEL divided by the uncertainty factor.
Uncertainty factors for in vivo studies should be assigned as appropriate, taking into
consideration the class of compound, the protocol, numbers of donors, and sensitivity of the
measured outcome variables.
3. 	GLOSSARY
The glossary includes terminology referred to in the Appendices as well as in the text.
Acceptable Daily Intake
(ADI)
An estimate of the amount of a substance, expressed on a
body weight basis, that can be ingested daily over a lifetime
without appreciable risk to human health.

Contains Nonbinding Recommendations
14
Antimicrobial Activity
Antimicrobial Agent
Balanced Design
Batch Culture
Blocking Factor
Challenge organism
Colonization
Colonization Barrier
Complete Design
Continuous Culture
Conventional Laboratory
Animal
The effect of an antimicrobial agent on a bacterial
population.
A drug substance that is either biologically derived or
chemically produced with antimicrobial activity as its major
effect.
A statistical design is balanced if each combination of
values or levels of all factors in the design (treatment
factors, factors of interest such as sex, or blocking factors)
have the same number of experimental units or replicates. A
partially balanced design is not balanced, but combinations
of treatments and other factors occur in a regular way such
that the analysis remains relatively simple.
A culture where neither substrate nor waste products are
removed until completion of incubation, normally incubated
for short periods, generally up to 24 hours.
An experimental factor whose values or levels define groups
of experimental units that are similar or that can be expected
to respond in a similar manner. Systematic variation among
blocks can be removed from the estimate of error in the
statistical analysis, resulting in greater precision. An
example is a cage containing several animals, which are the
experimental units, or an isolator containing several cages.
An organism added experimentally to a test system to
evaluate colonization barrier disruption.
The establishment of microorganisms in the intestinal tract.
A function of the normal intestinal flora that limits
colonization of the colon by exogenous microorganisms, as
well as overgrowth of indigenous, potentially pathogenic
microorganisms.
A statistical design is complete if all combinations of factors
or groups in the design have at least one observation. An
incomplete design is one in which no observations are made
for some combinations of factors.
A culture maintaining continuous growth of microorganisms
by the simultaneous supply of nutrient and removal of spent
medium, maintaining a constant microbial load within a
fixed incubation volume.
A laboratory animal with its natural indigenous intestinal
flora.

Contains Nonbinding Recommendations
15
Coprophagy
Defined Culture
Dilution (Flow) Rate
Donor (Fecal) Inocula
Drug Residue
Experimental Unit
Factorial Design
Fecal Slurry
Fed-Batch Culture
Human Flora-Associated
Interaction Effect
Intestinal Flora
The ingestion of feces.
A microbial culture in which all microbial species are
known.
The rate of supply and removal of medium from a
continuous culture system. Dilution rate controls the
microbial growth rate within a continuous culture system.
Fecal flora obtained from human volunteers and used to
inoculate the test system. Fecal flora is considered to be
equivalent to the intestinal flora.
The drug, including all derivatives, metabolites and
degradation products that persists in or on food.
The standard subject to which a treatment is applied and a
measurement is made. Examples include a whole animal or
a specific organ or tissue, a cage containing several animals,
a cell culture.
An experimental design that involves combinations of a
number of factors, including a treatment factor, each having
two or more values or levels. Other factors may include
stratification (e.g., sex) or blocking factors (e.g., cage).
Typically, the outcome variable is measured on a number of
experimental units at each combination of levels of the
various factors. The statistical analysis of the data involves
a multifactorial analysis of variance.
Human feces or fecal solids minimally diluted in anaerobic
buffer.
A batch culture fed continuously or semi-continuously with
nutrient medium. Portions of the fed batch culture can be
withdrawn at pre-determined intervals. A constant culture
volume is not maintained.
A germ-free host animal implanted with human fecal (HFA)
Animal flora.
Treatment effects that are modified by the presence of other
factors. For example, the effect of a treatment may be
greater or less in males than females, or may change over
time.
The normal microbial flora of the colon.

Contains Nonbinding Recommendations
16
Minimum Inhibitory
Concentration (MIC)
MIC 50
Microbiological ADI
No-Observable Adverse
Effect Concentration
(NOAEC)
No-Observable Adverse
Effect Level (NOAEL)
Outcome Variable
Semi-continuous
Culture
Short Chain Fatty Acid
Solid Phase
Systematic Variation
Test System
Uncertainty Factor
The lowest concentration of an antimicrobial compound that
inhibits growth of the test organism as determined by
standardized test procedures.
The concentration of an antimicrobial compound at which
50% of the tested isolates within a relevant genus are
inhibited.
An ADI established on the basis of microbiological data.
The highest concentration that was not observed to cause
any adverse effect in a particular study.
The highest administered dose that was not observed to
cause any adverse effect in a particular study.
A specific parameter measured in an experiment. Specific
outcome variables should be defined as part of the protocol,
and are the measurements actually made in the study.
A culture where substrate and/or waste products are added
and/or removed in a semi-continuous manner maintaining a
fixed incubation volume.
The volatile fatty acids that are produced by the intestinal
flora. The principal acids are acetic, propionic and butyric.
The particulate matter in an in vitro test system.
Factors that affect outcome variables. Such variation is
systematic in the sense that it represents an effect that is
reliably present. Systematic variation is distinguished from
random variation, which is not predictable. Systematic
variation may be caused by factors that are of interest, such
as sex, or by factors such as the particular isolator, which
are not.
A method used to determine the effects of antimicrobial
residues on the human intestinal flora.
A correction factor that takes into account the characteristics
of the test data as described in sections 2.4.1.2. Derivation
of an ADI from in vivo data, 2.4.2.1. Derivation of an ADI
from in vitro data, and 2.4.2.2. Derivation of an ADI from in
vivo data.

Contains Nonbinding Recommendations
17
4. 	REFERENCES
1. 	Cerniglia, C.E., and Kotarski, S. 1999. Evaluation of Veterinary Drug Residues in Food
for Their Potential to Affect Human Intestinal Microflora. Regulatory Toxicology and
Pharmacology. 29, 238-261.
2. 	National Committee for Clinical Laboratory Standards (NCCLS). 2004. Methods for
Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow Anaerobically;
Approved Standard – Sixth Edition. NCCLS document M11-A6. NCCLS, 940 West
Valley Road, Suite 1400, Wayne, PA, USA. (Note: In 2005, the organization’s name
changed to Clinical and Laboratory Standards Institute.)
3. 	National Committee for Clinical Laboratory Standards (NCCLS). 2003. Methods for
Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically;
Approved Standard – Sixth Edition. NCCLS document M7-A6. NCCLS, 940 West
Valley Road, Suite 1400, Wayne, PA, USA. (Note: In 2005, the organization’s name
changed to Clinical and Laboratory Standards Institute.)
4. 	OECD. 2001. Series of Testing and Assessment No. 34, Environment, Health and Safety
Publications. Draft Guidance Document on the Development, Validation and Regulatory
Acceptance of New and Updated Internationally Acceptable Test Methods and Hazard
Assessment. Organization for Economic Cooperation & Development, Paris.
5. 	Pritchard SE, Marciani L, Garsed KC, Hoad CL, Thongborisute W, Roberts E, Gowland
PA, Spiller RC. Fasting and postprandial volumes of the undisturbed colon: normal
values and changes in diarrhea-predominant irritable bowel syndrome measured using
serial MRI. Neurogastroenterol Motil. 2014 Jan;26(1):124-30. doi: 10.1111/nmo.12243.
Epub 2013 Oct 17. PMID: 24131490; PMCID: PMC3995006.

Contains Nonbinding Recommendations
18
APPENDIX A
Issues that Should be Investigated in Developing Test Systems and Data Interpretation
1. Experimental Conditions
Data generated for continuous flow, semi-continuous flow and fed-batch studies will be affected
by the growth conditions (e.g., growth medium, pH, dilution rate). Different bacterial species
may have different growth rates under the experimental conditions used for the test system. If
the dilution rate of the culture exceeds the growth rate of a bacterial species, then this species
ultimately will be eliminated from the test culture. The test system should be designed to
maximize the retention of the different bacteria and maintain the complexity of the initial
inoculum.
Test antimicrobial agents can affect growth rates of various bacterial groups. This may lead to
loss of components of the mixed culture by a reduction of growth rate below that of the dilution
rate used in the test system, which might cause some components of the flora to be washed out of
the culture. This may be minimized by developing test conditions with lower dilution rates.
Antimicrobial susceptibility is influenced by the physical condition of the exposed organisms,
which will be influenced by the growth conditions used in the test system. Based on the above,
further work is needed to determine the impact of different growth conditions on the NOAECs
derived for colonization barrier disruption and the increase in the population of resistant bacteria.
A number of factors should be considered in protocols for in vivo test systems. For example,
cross-contamination is a major issue when performing animal studies within a germ-free isolator.
The protocol should be designed to minimize cross-contamination.
2. Inoculum
The composition of the intestinal flora may vary among individuals with respect to bacterial
groups and resistant organisms. The bacterial populations are relatively stable within a single
individual, but this is not necessarily the case for resistant bacterial groups.
Multiple donors should be used to account for differences in flora between individuals. Pooled
inocula do not account for differences in flora between individuals. Therefore, test systems that
use fecal inocula obtained from individual donors are preferred to determine the effect of
antimicrobial residues on the intestinal flora. In addition, the composition of the donor inocula
should be taken into account when interpreting study results.
3. Study duration
The optimum incubation time to monitor for changes in bacterial populations in fecal batch
cultures needs to be determined. Likewise, in the case of complex long-term in vitro or in vivo
test systems, it is important to determine the period during which the integrity and complexity of
the intestinal flora remains stable and representative of the intestinal flora.

19
Contains Nonbinding Recommendations
APPENDIX B
Recommended Statistical Issues to be Considered When Designing Studies of
Antimicrobial Residues
Two broad endpoints of current public health concern were identified, disruption of the
colonization barrier and increases in population(s) of resistant bacteria. The experimental design
must depend on which of these is to be addressed and should take account of the particular
outcome variables. It is recommended that a design paradigm for these test systems involves
choice of the test system, application of treatments and follow-up of the system over time. The
choice of test system depends on the characteristics of the human intestinal tract that is
represented by the test system. Since the MIC tests are simple in design, many of the issues
discussed below do not apply to this method.
The experimental unit is a central component of the study design. For an in vivo test system, for
example, the unit may be an individual animal or an entire cage. If cages are grouped within
isolators, some or all of the treatments to different cages within each isolator can be applied. In
this case the isolator becomes a blocking factor, since cages within the same isolator would be
expected to respond in a similar fashion. The use of blocking factors is an important tool for
reducing systematic variation. A related question is whether there are other systematic factors
such as sex that should be included, that is, whether a factorial design should be used. If there
are multiple factors, then the design involves choices of what combinations of these should be
included. It is important that this be done in such a way that the resulting design is balanced. In
a complete, balanced design, all combinations are represented, and occur the same number of
times. It is also possible to have incomplete designs, as well as various kinds of partial balance.
The analysis of variance may be recommended for such designs; these designs can be useful
when, for example, experimental resources are limited. An example of an incomplete design is
the standard two period cross-over design.
It should be decided how the treatments are applied to the experimental units. In some cases a
two-stage treatment, involving a drug treatment and a bacterial challenge, may be recommended.
There should be at least three antimicrobial treatment groups in addition to appropriate control
groups. The choice of antimicrobial treatment levels depends on the desired range of doses, but
should cover both effect and no-effect levels. The duration and the method of drug
administration depends on the test system. An important aspect of some studies is the evolution
of effects over time, and repeated measurement of outcome variables may be recommended.
Common issues are the timing and spacing of the measurements and bias caused by missing data.
Control of random variation due to biological variability and to measurement error depends on
the number of experimental units and number of samples. This number can be determined from
previous knowledge of the test system and outcome variables, either from past experience or
through a sample size computation, which should be employed where possible. Sufficient
replication should be included to allow precise measurement of treatment effects and appropriate
interaction effects, e.g., treatment effects that change over time. In some studies, it may be
important to examine such interaction effects as part of the statistical analysis. Another type of
replication is the pooling of fecal samples from animals in a single cage or the pooling of fecal

Contains Nonbinding Recommendations
20
samples from different donors. In both cases, we have the benefits of averaging, but not the
ability to estimate variability among replicates. Pooling may obscure individual effects (of
treatment and/or inoculum), and thus its use should be considered in terms of study objectives.

Contains Nonbinding Recommendations
21
APPENDIX C
1. 	Calculation of MICcalc
The MIC CALC is derived from the lower 90% confidence limit for the mean MIC 50 of the most
relevant genera for which the drug is active. The lower 90% confidence limit is calculated using
log transformed data. Thus, the mean and standard deviation are calculated using the log
transformed MIC 50 values. This also implies that the lower 90% confidence limit needs to be
back-transformed to obtain the correct value. The formula for the confidence limit is:
where: 	Mean MIC 50 is the mean of the log transformed MIC 50 values,
Std Dev is the standard deviation of the log transformed MIC 50 values,
n is the number of MIC 50 values used in the calculations,
t0.10,df is the 90th percentile from a central t-distribution with df degrees of freedom, and
df = n-1.
Examine the MIC 50 of relevant genera (see section 2.1. Steps in determining the need for a
microbiological ADI). The MIC calc is based on a summary value of those genera which are not
inherently resistant to the compound. Thus, the MIC calc is based on MIC 50 of those genera for
which the compound is active. Ensure that all MIC 50 values are not characterized as “</=”, so
they may be used in the calculation of the MIC calc .
2. 	Example Calculation
Any base log transformation of MIC 50 values can be used. However, if 2-fold dilutions of drug
are used in the MIC testing procedure, a base 2 log transformation conveniently will provide
integer values for the calculation. In the following example, the MIC 50 values were transformed
as follows:

Contains Nonbinding Recommendations
22

Contains Nonbinding Recommendations
23
APPENDIX D
Supplement to Section 2. Guidance Regarding the Determination of the Fraction of Oral
Dose Available to Microorganisms
1. 	Introduction
VICH GL 36 has been implemented since 2005. Having gained experience in working with the
guidance, regulators from all VICH regions agreed that additional guidance and clarity were
needed regarding in vivo and in vitro testing methods to determine the fraction of oral dose
available to microorganisms.
This Appendix is based on review of new data, scientific literature, and information from
disclosed sponsor submissions.
This Appendix contains three sections: a table of examples of test systems for the assessment of
the fraction of oral dose available to microorganisms, general considerations regarding
methodological aspects of the implementation of these test systems, and a description of how the
test systems could be used in determining the fraction of oral dose available to microorganisms.
2. 	Examples of Test Systems for the Assessment of the Fraction of Oral Dose Available to
Microorganisms
Various in vitro and in vivo test systems could be used separately and in combination to
determine the fraction of oral dose available to microorganisms. The table below provides
examples of such test systems, the type of data generated and considerations relevant to their use.

Contains Nonbinding Recommendations
24

Contains Nonbinding Recommendations
25
3. 	Methodological Aspects of Test Systems
This section provides general considerations regarding the experimental conditions used in
designing and conducting studies to determine the fraction of oral dose available to
microorganisms.
a) 	Dose and concentration of drug:
• 	Dose and drug concentration range to be used in the test systems and the experimental
objective should be justified.
• 	Dose and drug concentrations for testing should include levels that are expected with
residue ingestion, as well as higher levels.
b) 	Fecal parameters:
• 	Source and number of fecal samples:
o 	Donors should be healthy with no known exposure to antimicrobial agents for at
least 3 months before fecal collection (see section 2.3. General recommendations
of the guidance).
o 	Variability among donors (e.g., age, sex, diet) is inherent, and the implications of
donor variability for experimental design should be taken into account. The
number of fecal donors should be based on the experimental objective, and a
minimum of six donors are recommended (Figure 1).
o 	It is recommended that fresh samples (first motion of the day) should be
processed within the day of collection. Anaerobic storage for up to 72 hours at
refrigerator temperatures is acceptable.
• 	Physical characterization of fecal samples (e.g., fecal viscosity, water content, pH,
and solid content) is recommended. This information may be useful in interpreting
variability in subsequent study results.
• 	Fecal concentrations:
o 	At least one fecal concentration should be considered. A 25% fecal preparation (1
part fecal sample + 3 parts diluent) is recommended as representative of colon
contents.
• 	Diluent used to prepare fecal slurries:
o 	The chemical components used in diluting fecal material should be standardized
to minimize variability.
o 	An anaerobic buffer that is based on minimal salts should be used.
• 	Fecal incubation:
o 	Consider an initial experiment using a minimum of two donor samples to
determine an appropriate protocol. This should include a relevant range of
residue concentrations, incubation time and sampling at multiple time points, so
as to enable kinetic calculations.

Contains Nonbinding Recommendations
26
o 	The data for a minimum of six donors should be used for the final determination
of the fraction of oral dose available to microorganisms.
• 	Use of non-sterile or sterile fecal samples:
o 	Consider the impact of sterilization of feces on drug binding to fecal suspensions
in initial studies using a chemical assay.
o 	Non-sterilized feces should be used where possible when conducting in vitro
drug-binding/inactivation studies. Small differences between binding to non-
sterilized and sterilized fecal suspensions may allow further studies to be based on
sterilized feces only.
c) 	Methods to quantitatively determine the fraction of the microbiologically active drug
available to microorganisms:
• 	While either microbiological or chemical assays may be used in these experiments,
justification of the specific type of assay should be provided. If chemical assays are
used, they should be bridged to the microbiological activity.
• 	The strain of the indicator bacterial species will depend on the spectrum of activity of
the drug.
• 	The sensitivity and reproducibility of the assays should be considered.
• 	Study controls should be considered according to the test system used.
d) 	Reversibility of observed drug binding:
• 	A time course approach is recommended which will reveal possible reversibility of
drug binding.
• 	Further work to define the mechanism of binding is not essential for the purpose of
establishing the fraction of oral dose available to microorganisms.
4. 	Description of How Test Systems Could Be Used in Determining the Fraction of Oral
Dose Available to Microorganisms
In vivo and in vitro approaches, using different test systems considered applicable to determine
the fraction of oral dose available to microorganisms, were identified and reviewed. Conceptual
approaches of their application in deriving this fraction are outlined below and illustrated in
Figure 1.
APPROACH 1: In vivo test systems. Animals dosed with the drug, followed by one of the
following options:
• 	Option A: chemical extraction and analysis of the intestinal content and (or) feces to
determine the total drug concentration, is used to establish the fraction of oral dose
available to microorganisms.

Contains Nonbinding Recommendations
27
• 	Option B: both chemical and microbiological activity assays of the intestinal content and
(or) feces of dosed animals is used to establish the fraction of oral dose available to
microorganisms.
APPROACH 2. In vitro test systems. This approach comprises two steps (Phase A and B)
using in vitro fecal slurry test systems (see Figure 1). Phase A is an initial experiment, with fecal
samples from two donors, used to identify the incubation times and relevant range of added drug
concentrations sampling at multiple time points. This phase includes both chemical and
microbiological assays. Phase B is conducted based on results from Phase A with samples from
four additional donors and uses microbiological assays. The data for all six donors are used for
the final determination of the fraction of oral dose available to microorganisms.
APPROACH 3: Approach 1[Option A] + Approach 2. This approach combines both in vivo
studies and in vitro studies.

Contains Nonbinding Recommendations
28
Figure 1. Schematic Representation of Test Systems to Determine the Fraction of Oral Dose
Available to Microorganisms