Design and Analysis of Shedding Studies for Virus or Bacteria-Based Gene Therapy and Oncolytic Products: Guidance for Industry

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

Design and Analysis of Shedding
Studies for Virus or Bacteria-Based
Gene Therapy and Oncolytic Products
Guidance for Industry
Additional copies of this guidance are available from the Office of Communication, Outreach
and Development (OCOD), 10903 New Hampshire Ave., Bldg. 71, Rm. 3128, Silver Spring,
MD 20993-0002, or by calling 1-800-835-4709 or 240-402-8010, or email ocod@fda.hhs.gov, or
from the Internet at
http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guida
nces/default.htm.
For questions on the content of this guidance, contact OCOD at the phone numbers or email
address listed above.
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Biologics Evaluation and Research
August 2015

Contains Nonbinding Recommendations
i
Table of Contents
I. 	INTRODUCTION............................................................................................................. 1
II. 	SCOPE ............................................................................................................................... 2
III. 	BACKGROUND ............................................................................................................... 2
IV. 	WHY COLLECT SHEDDING DATA DURING PRODUCT DEVELOPMENT? ... 3
V. 	COLLECTION OF SHEDDING DATA IN PRECLINICAL STUDIES.................... 4
VI. 	DESIGN OF SHEDDING STUDIES: GUIDING PRINCIPLES................................ 5
A. 	Biological Characteristics ..................................................................................... 5
B. 	Route of Administration ....................................................................................... 6
VII. 	COLLECTION OF SHEDDING DATA IN CLINICAL STUDIES............................ 6
A. 	When to Collect Shedding Data in Clinical Studies?......................................... 6
B. 	Study Design .......................................................................................................... 7
VIII. 	ANALYTICAL ASSAYS TO MEASURE SHEDDING ............................................... 9
IX. 	ANALYSIS OF SHEDDING DATA ............................................................................. 12
A. 	The Nature of the Shed Material ....................................................................... 12
B. 	The Extent of Shedding ...................................................................................... 13
X. 	WHAT TO INCLUDE IN A CLINICAL SHEDDING STUDY REPORT ............... 13
XI. 	ASSESSING THE POTENTIAL FOR TRANSMISSION TO UNTREATED
INDIVIDUALS DUE TO SHEDDING ......................................................................... 15
A. 	What Information in the Shedding Data Can be Used to Assess Potential for
Transmission to Untreated Individuals? .......................................................... 16
B. 	Monitoring Untreated Individuals for Transmission ...................................... 16
XII. 	REFERENCES ................................................................................................................ 17

Contains Nonbinding Recommendations
1
Design and Analysis of Shedding Studies for Virus or Bacteria-
Based Gene Therapy and Oncolytic Products
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.
I. 	INTRODUCTION
The Center for Biologics Evaluation and Research (CBER)/Office of Cellular, Tissue, and Gene
Therapies (OCTGT) is issuing this guidance to provide you, sponsors of virus or bacteria-based
gene therapy products (VBGT products)1 and oncolytic viruses or bacteria (oncolytic products)2
with recommendations on how to conduct shedding studies during preclinical and clinical
development. For purposes of this guidance, the term “shedding” means release of VBGT or
oncolytic products from the patient through one or all of the following ways: excreta (feces);
secreta (urine, saliva, nasopharyngeal fluids etc.); or through the skin (pustules, sores, wounds).
Shedding is distinct from biodistribution because the latter describes how a product is spread
within the patient’s body from the site of administration while the former describes how it is
excreted or released from the patient’s body. Shedding raises the possibility of transmission of
VBGT or oncolytic products3 from treated to untreated individuals (e.g., close contacts and
health care professionals). This guidance represents FDA’s current thinking on how and when
shedding data should be collected for VBGT and oncolytic products during preclinical and
1 Gene therapy products are all products that mediate their effects by transcription and/or translation of transferred
genetic material and/or by integrating into the host genome and that are administered as nucleic acids, viruses, or
genetically engineered microorganisms. The products may be used to modify cells in vivo or transferred to cells ex
vivo before administration to the recipient. See section III. of FDA’s guidance entitled “Guidance for Industry:
Gene Therapy Clinical Trials - Observing Subjects for Delayed Adverse Events” dated November 2006.
http://www.fda.gov/biologicsbloodvaccines/guidancecomplianceregulatoryinformation/guidances/cellularandgeneth
erapy/ucm072957.htm.
2 Oncolytic products refer to replication competent viruses or dividing bacteria that are used as therapeutic agents to
mediate lysis of tumor cells. Some oncolytic products carry foreign genes (immune modifying genes, genes that
enhance oncolysis etc.), and mediate part of their anti-tumor effect by transcription and/or translation of these
foreign genes in the host. Hence, oncolytic products that carry foreign genes can also be classified as gene therapy
products.
3 Transmission could occur if the VBGT or oncolytic product is shed in the form of intact viruses or bacteria but not
when shed as viral or bacterial degradation products such as nucleic acid fragments.

Contains Nonbinding Recommendations
2
clinical development and how shedding data can be used to assess the potential for transmission
to untreated individuals. This guidance finalizes the draft guidance of the same title dated July
2014.
FDA’s guidance documents, including this guidance, do not establish legally enforceable
responsibilities. Instead, guidances describe the FDA’s current thinking on a topic and should be
viewed only as recommendations, unless specific regulatory or statutory requirements are cited.
The use of the word should in FDA’s guidances means that something is suggested or
recommended, but not required.
II. 	SCOPE
The products covered by this guidance are VBGT and oncolytic products that OCTGT reviews.
The focus of this guidance is shedding studies, including both how and when shedding data
should be collected and how shedding data can be used to assess the potential for transmission to
untreated individuals.
This guidance does not cover plasmids, peptides, and genetically modified mammalian cells that
OCTGT also reviews because, unlike VBGT and oncolytic products, there is no potential for
plasmids, peptides, and genetically modified mammalian cells to be infectious or transmissible.
This guidance also does not address collection or submission of adverse event information,
including those adverse events that could be attributed to shedding. Please see the regulations at
Title 21 of the Code of Federal Regulations (CFR) Part 312, specifically 21 CFR 312.32 and 21
CFR Part 600, specifically 21 CFR 600.80, for information on the collection and submission to
FDA of adverse event information.
Finally, while assessment of shedding can be utilized to understand the potential risk to the
environment, the scope of this guidance does not include shedding as it may relate to potential
environmental concerns with respect to a specific VBGT or oncolytic product. For more
information on this topic, you may wish to consult FDA’s guidance document entitled
“Determining the Need for and Content of Environmental Assessments for Gene Therapies,
Vectored Vaccines, and Related Recombinant Viral or Microbial Products; Guidance for
Industry” dated March 2015.4
III. 	BACKGROUND
VBGT and oncolytic products are derived from infectious viruses or bacteria. In general, these
products are not as infectious or as virulent as the parent strain of virus or bacterium because of,
in part, the derivation methods and/or modifications made during product development that lead
to attenuation. Hence, it is likely that these products are shed to a lesser extent than during
4 Available at
http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guidances/CellularandG
eneTherapy/ucm401869.htm.

Contains Nonbinding Recommendations
3
natural infection by the parent strain. Nonetheless, the possibility that the shed VBGT or
oncolytic product may be infectious raises safety concerns related to the risk of transmission to
untreated individuals. To understand this risk, shedding studies that are conducted in the target
patient population(s) may be appropriate before licensure.
Typically, clinical shedding studies are not stand-alone studies but are integrated into the design
of a safety or efficacy trial. Because there are many product-specific factors and patient-specific
factors that can influence the design of a shedding study, sponsors should consult with OCTGT
in the early stages of product development for specific recommendations as to their product.
IV. 	WHY COLLECT SHEDDING DATA DURING PRODUCT DEVELOPMENT?
Shedding studies should be conducted for each VBGT or oncolytic product to provide
information about the likelihood of transmission to untreated individuals because historical data
alone may not be predictive of the shedding profile. Shedding data can be used to evaluate
measures to prevent transmission. Shedding data collected during product development should
provide a clear and comprehensive understanding of the shedding profile of VBGT or oncolytic
products in the target patient population(s). Note that it may be appropriate to describe these
data in the package insert for an approved Biologics License Application (BLA).
To inform the design of human shedding studies, shedding data may be collected in animals
following administration of the VBGT or oncolytic product. These data can help estimate the
likelihood and potential shedding profile in humans, particularly when there is concern about
transmission to untreated individuals. However, such data cannot substitute for human shedding
studies for several reasons. For example, a VBGT or oncolytic product may be derived from a
human-specific strain; therefore, animals may not adequately predict the shedding profile in
humans. Similarly, various animal species/models may not adequately address patient-specific
factors, such as differences in the immune status at the time of product administration, which
may contribute to the potential for shedding in humans (for more details refer to section VII.B. of
this guidance).
Product-specific variables may also affect shedding. For example, the biological characteristics
and route of administration (entry) of VBGT or oncolytic products can be different from that of
the parent strain of viruses and bacteria. Specifically, these products may be:
• 	Derived from laboratory-adapted wild-type, attenuated or engineered strains that may
not have been characterized in humans in prior studies.
• 	Replication competent or incompetent viruses; viruses that can infect a host cell and
amplify to produce progeny are replication competent and those that can infect a host
cell but cannot establish an infection, amplify, and produce progeny are replication
incompetent.

Contains Nonbinding Recommendations
4
• 	Dividing and/or auxotrophic bacteria; auxotrophic bacteria are unable to synthesize
an organic molecule required for their growth and division but when this molecule is
available with the other nutrients they require, growth and division of the bacteria
may occur.
• 	Introduced into the human body through unnatural routes and hence, the infectivity,
replication, persistence and shedding from the human body may be different than that
of the parental strains.
• 	Engineered to carry transgenes, e.g., tropism-altering genes, immune modifying
gene(s) or genes that enhance oncolysis.
V. 	COLLECTION OF SHEDDING DATA IN PRECLINICAL STUDIES
The decision to assess shedding in preclinical studies5 is based on the biological characteristics,
derivation, and genetic make-up of the VBGT or oncolytic product. For example, preclinical
shedding data may be requested for an oncolytic or a replication competent VBGT product, if:
• 	Humans have not been previously exposed to the product, as in the case of a non-
human bacterial or viral strain.
• 	The product has been administered to humans, but has been modified to achieve a
different in vivo tropism than the parent strain.
• 	The product has been previously administered to humans; however, a change in the
route of administration is proposed.
• 	Humans have not been previously exposed to the product, and the route of
administration differs from the natural route of exposure/infection.
The use of the animal species/model(s) is an important factor that can affect the biological
relevance of the shedding profile generated in the animal. Considerations include the
permissiveness or susceptibility of the animal to infection from the VBGT or oncolytic product
under investigation, and any preexisting immunity that may affect infectivity or product
clearance.
Collection of shedding data is an endpoint that can be included in preclinical studies designed to
collect other data, such as safety and biodistribution. The decision to include an assessment of
the shedding profile of a VBGT or oncolytic product in an animal study will depend on various
5 For general information about preclinical assessment, including study design, of these products see FDA’s
guidance entitled “Guidance for Industry: Preclinical Assessment of Investigational Cellular and Gene Therapy
Products” dated November 2013,
http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guidances/CellularandG
eneTherapy/ucm376136.htm.

Contains Nonbinding Recommendations
5
product-specific factors, as described above and in sections I. and IV. of this guidance. We
recommend that sponsors initiate communication with the Pharmacology/Toxicology staff in
OCTGT early in their product development program to discuss the need for generating a
shedding profile for their VBGT or oncolytic product in animals, and the planned methodology
to collect this shedding data.
VI. 	DESIGN OF SHEDDING STUDIES: GUIDING PRINCIPLES
The key aspects in the design of shedding studies are: the choice of clinical samples that are
collected from subjects in a trial (e.g., feces, urine, nasal swabs); the frequency of sample
collection and duration of the monitoring period; and the assay methodology selected to test for
the presence of the shed VBGT or oncolytic product in the clinical sample (Ref. 1).
To guide the design of shedding studies, the following should be considered:
A. 	Biological Characteristics
• 	Replication competence: The ability of the VBGT or oncolytic product to
multiply and amplify in the human host greatly affects how it is disseminated in
the body and may increase the extent and duration of shedding.
• 	Immunogenicity: When the VBGT or oncolytic product is derived from viruses
or bacteria that elicit a strong immune response, the product may be more rapidly
cleared from circulation than a poorly immunogenic product, and may be shed for
a shorter duration. Similarly, when a product is administered multiple times, the
product may be shed for a shorter duration in the later dose cycles than after the
immune-priming first or early doses.
• 	Persistence and latency: The duration of a shedding study may be longer if the
VBGT or oncolytic product exhibits persistence or latency-reactivation in the
host, as in the case of an oncolytic herpes virus product that is capable of latency
(period of time during which a virus is present in the host without producing overt
clinical symptoms). Shedding of such products may be intermittent and
unpredictable.
• 	Tropism: Tropism of the product may affect what samples should be collected to
assess shedding. For example, VBGT or oncolytic products that are engineered to
carry tropism modifying gene(s) or mutation(s) may exhibit an altered shedding
profile than the parent virus because of retargeting of the product to different
tissues or organs.

Contains Nonbinding Recommendations
6
• 	Stability of product attenuation: It is common for VBGT or oncolytic products to
be attenuated either for replication in normal (non-tumor) cells, or for loss of
virulence or latency in the human host. However, for some products that have a
higher potential for recombination or reversion in the patient, the shedding pattern
and/or what is shed may change.
B. 	Route of Administration
In addition to the tropism of the VBGT or oncolytic product, the route of product
administration should be considered in the selection of sample types to collect in a
shedding study. For example, to assess shedding in patients administered an oncolytic
virus by the intradermal route, we recommend the collection of skin swabs at the site of
injection, in addition to the other samples routinely assessed for shedding (e.g., urine,
feces, and saliva). Similarly, we recommend the collection of nasopharyngeal washes
when an oncolytic virus is administered by inhalation or via the intranasal route.
VII. 	COLLECTION OF SHEDDING DATA IN CLINICAL STUDIES
Shedding data collected in clinical studies provides a shedding profile of a product in the target
patient population and is used to estimate the potential of transmission to untreated individuals.
Depending on the shedding profile, it may be appropriate to include the information on shedding
in the Investigator Brochure and in the Informed Consent for Investigational New Drug (IND)
studies. Depending on the shedding profile, it also may be appropriate to include shedding data
in the package insert for licensed products. This information will inform patients and physicians
if shedding could occur with the use of a VBGT or oncolytic product, the potential for
transmission of the product to untreated individuals, and of the measures to take to prevent such
transmission.
A. 	When to Collect Shedding Data in Clinical Studies?
• 	For VBGT and oncolytic products classified as replication competent, we
recommend that sponsors begin collecting shedding data in Phase 1 trials.
Considering that replication competent products are associated with a higher
potential for release as infectious viruses or bacteria, sponsors may need to
continue collecting shedding data during Phase 2 and Phase 3, after a dose and
regimen have been determined, to better characterize shedding.
• 	For VBGT products that are classified either as replication incompetent or
replication deficient, we recommend that sponsors collect shedding data later in
product development (e.g., during Phase 2 studies), after a dose and regimen have
been determined. Compared to shedding of replication competent products,
shedding of replication incompetent or replication deficient products is expected
to be low, for a limited duration, and associated with a lower potential for release
as infectious viruses or bacteria.

Contains Nonbinding Recommendations
7
• 	After shedding data are collected in early phase studies, it is not uncommon that
dose, route, regimen (frequency of product administration, concomitant therapy,
preconditioning regimens, etc.), or indication are modified. These changes can
alter how product is shed. In such cases, shedding data collected in early clinical
trials may not be adequate or relevant to predict the shedding profile of the
product in its current state. Additional data should be collected in subsequent
clinical trials where the route, dose regimen, and indication are the same as in the
pivotal trial.
B. 	Study Design
The plan to collect shedding data in clinical studies can be based on prior clinical
experience with the same or similar product, but when there is no such experience, as in
the case of first-in-human VBGT or oncolytic products, the shedding profile generated in
animals can be informative. We recommend that sponsors prospectively design and
incorporate the sampling plan in the clinical study to collect shedding data.
There are four critical choices in the design of a sampling plan:
• 	Frequency of sample collection;
• 	Duration of sample collection;
• 	Type(s) of samples collected; and
• 	Storage conditions for types of samples collected.
However, there are many aspects that can influence these choices, as described earlier in
section VI. of this guidance and further elaborated below.
• 	Frequency of sample collection: Shedding is most likely to occur in the period
immediately following product administration, irrespective of replication
competence of the VBGT or oncolytic product. A second peak of shedding may
be noted in the days/weeks after administration of a replication competent product
as a result of its multiplication/amplification in vivo. Accordingly, sampling
should start immediately after product administration, with frequent sampling
during the initial weeks following treatment to capture the shedding pattern
accurately (e.g., sampling on day 1, 3, 7, 10 and then weekly). Analysis of
samples should continue until three consecutive data points are obtained at or
below the limit of detection (LOD) of the shedding assay. If the level of shedding
does not reach the LOD of the assay but there is a continual decreasing trend,
collection should continue until the results demonstrate that a plateau has been
reached in at least three consecutive data points.

Contains Nonbinding Recommendations
8
• 	Duration of sample collection (monitoring period for shedding):
o 	In general, when the VBGT or oncolytic product is a replication competent
virus or bacteria, the monitoring period for shedding is longer (for sample
collection) as compared to when the product is replication incompetent or
replication deficient. This is because you will want to capture the second peak
of shedding associated with multiplication or amplification in vivo.
o 	The immune status of the patient population should be considered. Patients
who are immune compromised may have an extended or even different
shedding profile than their immune competent counterparts. For replication
competent products, the immune competence of the patient population is a
relevant factor because many of these products are used in cancer patients
who have had immunosuppressive chemotherapy. When treated with
replication competent VBGT or oncolytic products, immunosuppressed
patients may become persistently infected and may shed the product for
extended periods of time (Ref. 2). Therefore, the monitoring period for
shedding may be longer for immunosuppressed patients treated with a
replication competent VBGT or oncolytic product.
o 	When a VBGT or oncolytic product is administered in multiple cycles, or
when there is pre-existing immunity, the duration of shedding may be
shortened because of product-specific immune responses. Data from single
dose administration of product may be used to guide the timing of sample
collection following multiple administrations.
o 	If an oncolytic product is based on a virus that has the potential for latency
reactivation, we recommend the collection of additional samples for shedding
analysis when clinical signs warrant, i.e., when patients show signs of
infection due to reactivation.
• 	Type(s) of samples collected: The types of clinical samples (e.g., urine, fecal
swabs, saliva, etc.) collected to assess shedding depend on a variety of factors
including the route of administration of the product, the tropism of the virus or
bacteria, the natural route of transmission and shedding of the parent virus or
bacterium from which the product is derived (as described in section VI. of this
guidance), and biodistribution or shedding data from preclinical studies. For
example, if an oncolytic herpes virus product is administered intradermally for
treatment of skin cancers, there is the potential of transmission of the oncolytic
herpes virus product through infected scabs/skin secretions because that is a
natural route of transmission for herpes viruses. In this case, skin swabs or
dressing from injection sites should be analyzed for shedding. Likewise, if an
oncolytic adenovirus product is administered intranasally, there is the possibility
of transmission of the adenovirus product through respiratory secretions,
therefore, nasopharyngeal swabs or washes should be collected in the shedding
study. For tropism-modified products, knowledge about the route of transmission

Contains Nonbinding Recommendations
9
and shedding of the parent virus or bacterium may not be relevant or sufficient to
guide in sample collection.
• 	Blood is not typically analyzed for shedding but should be collected as part of
pharmacokinetic analysis to understand the extent of product dissemination from
the site of administration and the rapidity of product clearance. This information
can be particularly useful to assess the extent of product shedding when limited
vascular spread is expected for products administered locally (e.g., intratumoral,
intramuscular, intracranial, subretinal routes).
• 	Storage conditions for type(s) of samples collected: The appropriate storage
conditions for different type(s) of samples need to be established in order to
minimize degradation of product-specific nucleic acids or loss of product-specific
infectivity or bacterial viability. Multiple aliquots of samples may be needed for
different tests, such as for a product-specific infectivity assay after a sample is
scored positive for product-specific nucleic acids.
VIII. 	ANALYTICAL ASSAYS TO MEASURE SHEDDING
An analytical assay that measures shedding is designed to detect product in the clinical sample,
either by detection of nucleic acids or for the presence of infectious viral particles or dividing
bacteria. Based on the design and output (nature of the assay readout), shedding assays can vary
greatly in their performance and suitability. Hence, the choice of a shedding assay can greatly
affect the quality of the data collected, and is important in the generation of meaningful shedding
data, i.e., data that provides a complete shedding profile for a product and can be used to
estimate the potential for transmission to untreated individuals.
We recommend that sponsors consider the following in the selection of the analytical assay to
measure shedding:
• 	At least one of the assays used to measure shedding should be quantitative. We
recommend that sponsors report the extent of shedding of VBGT or oncolytic
products in terms of the number of genome copies or infectious units to provide a
quantitative assessment of shedding. Often an assay with a quantitative readout, like
quantitative polymerase chain reaction (qPCR), is used because of the ease of
performing/standardizing the assay, high throughput format, rapid turnaround time,
and assay sensitivity.
• 	Because detection of nucleic acids by qPCR may not indicate the presence of
infectious virus, for replication competent products, detection of nucleic acids should
be followed up with infectivity or growth-based assays. Replication competent
products are capable of growth or multiplication in humans and if shed, can be
infectious. Since only infectious viruses or bacteria are potentially transmissible, we
recommend that sponsors follow a step-wise approach for the analysis of shedding of
replication competent products. Specifically, a steady rise in the PCR signal for

Contains Nonbinding Recommendations
10
product-specific nucleic acids in clinical samples collected over successive time-
points is suggestive of bacterial growth or viral multiplication in vivo. In such cases,
clinical samples should be further analyzed for infectivity in cell culture (for viruses)
or growth (for bacteria).
• 	If shedding of conditionally replicating VBGT or oncolytic products is noted by
qPCR assay at a level above the LOD, we recommend that sponsors further
characterize the shed material to confirm infectivity or growth because such products
could be shed in their infectious form even if replication is confined, mostly, to
tumors or to a particular tissue-type. The assessment of infectivity for conditionally
replication competent products, such as conditionally replicating adenovirus or
auxotrophic bacteria, should take into account product-specific in vitro cell culture or
growth conditions; for example growth in differential media for auxotrophic bacteria
followed by a selective method for product identification.
• 	There are many different approaches to assess infectivity or growth with assays that
have a quantitative read-out (Ref. 3). For example:
o 	For detection of infectious viruses: Assays that measure infectivity in terms
of Tissue Culture Infectious Dose 50 (TCID50 ), plaque-forming units (PFU),
focus-forming units (FFU).
o 	For detection of dividing bacteria: Assays that measure bacterial growth in
colony forming units (CFU).
Sponsors may justify limiting the shedding analysis of a replication competent or
conditionally replicating product to qPCR assay, if:
o 	A correlation between qPCR and the infectivity assay or growth-based assay
is established, and the signal noted in the qPCR assay is at or below the LOD
of the infectivity or growth-based assay; or
o 	The cell culture step in an infectivity assay is demonstrated to be unsuitable
for the analysis of clinical samples with complex composition such as excreta
due to adverse effects on cell viability.
• 	Shedding analysis of replication incompetent or deficient products by qPCR may be
adequate. Most VBGT products are replication incompetent or replication deficient;
for example, adeno-associated virus (AAV) vectors, E1-deleted adenovirus (Ad)
vectors, and some herpes virus vectors (HSV). Replication incompetent or
replication deficient products are incapable of multiplying in humans, and therefore,
are shed to a lower extent and in a form that is incapable of establishing an infection.
Hence, qPCR may be adequate as the primary assay to assess shedding of replication
incompetent or replication deficient products.

Contains Nonbinding Recommendations
11
• 	Use of a qualitative assay to assess shedding may be justified. When a qualitative
shedding assay is demonstrated to be more sensitive than a quantitative assay in the
detection of a specific product, a sponsor may justify the use of the former. However,
if there is detectable shedding with a qualitative assay, a follow-up quantitative
analysis should be performed to understand the extent of shedding in positively
scored samples.
• 	Shedding assay(s) should be demonstrated to be specific, sensitive, reproducible and
accurate. We recommend testing of clinical samples in a shedding assay in replicates
to determine reproducibility. The specificity of the assay should be well understood
to avoid false-positive or false-negative results, particularly since retesting is not
always feasible with clinical samples that are limited in quantity, such as nasal or skin
swabs. The sensitivity of the assay should be determined in terms LOD and the limit
of quantitation (LOQ), if using a quantitative assay. While the Agency does not
expect shedding assays to be validated, the assays should be qualified to meet
minimal performance capabilities and be suitable for the intended purpose.
The effect of sample type and composition on assay performance should be well understood.
Clinical samples such as feces and urine are rich in complex organic matter that can adversely
affect the performance of an assay and lead to an underestimation of shedding. Also, samples
such as feces, saliva, and nasal swab are rich not only in host proteins and nucleic acids but also
in the body’s natural flora and in circulating strains of viruses and bacteria from the environment.
Thus, the assay conditions should be optimized to selectively analyze for the product under
investigation. For that, the specificity of the reagents used in the assay should be assessed and
the quality of the reagents should be controlled. Certified and contaminant-free reagents should
be used in the analysis of clinical samples in a shedding assay.
Interference from clinical sample matrix can lead to a false-negative result or an underestimation
of the amount of shedding. For example, clinical samples like urine, saliva, and feces are rich in
proteases, nucleases, ions and salts that can affect the amplification process in a PCR;
specifically, nucleases in saliva/feces can degrade template DNA, bile salts in feces or urea in
urine can affect the activity of thermostable DNA polymerases in the PCR mixture. When PCR
inhibition is suspected due to interference from components in a clinical sample, the clinical
sample can be diluted to a limited extent to reduce the interfering component. Since dilution of
the clinical sample also leads to template dilution, the sensitivity of PCR assay should be
assessed in each assay run. For that, each diluted sample should be tested in parallel with one
that is spiked with a reference standard or positive internal control prior to dilution. If
interference cannot be decreased by limited sample dilution, alternative or additional extraction
procedures should be considered to remove the interfering component(s) in the clinical sample.
An underestimate of the level of shedding may also result due to degradation of viral or bacterial
nucleic acids in enzyme-rich clinical samples such as feces and saliva during storage,
handling/shipping and nucleic acid extraction. To account for such effects, we recommend that
mock/donor sample types be spiked, soon after collection, with the reference standard or internal

Contains Nonbinding Recommendations
12
positive control and the percent recovery of the reference standard or internal positive control
should be determined on a one-time basis. Sample collection, storage, shipping, extraction and
analysis should be performed with the same methodology as that planned for the clinical (test)
samples.
IX. 	ANALYSIS OF SHEDDING DATA
In order to assess the potential of transmission to untreated individuals due to shedding, the
analysis of shedding data for VBGT or oncolytic products should address the following:
A. 	The Nature of the Shed Material
When clinical samples are scored positive for product in a shedding assay, the subsequent
analysis of these samples should provide answers to the following questions:
1. 	Do the clinical samples contain product-specific nucleic acids (full-length
genomes) suggestive of the presence of infectious viruses or bacteria or do the
clinical samples contain mostly degraded product-specific nucleic acid (genome)
fragments found in the absence of infectious viruses or bacteria?
An example of a clinical sample containing product-specific nucleic acid (full-
length/complete genomes) suggestive of the presence of infectious viruses is one in
which product-specific nucleic acids are amplified (by PCR) after treatment with
nucleases. Under such conditions, the only genomes that are amplifiable are the full-
length/complete genomes protected within intact nuclease-resistant viral particles that
may be infectious. When nuclease treatment is not feasible, amplification of full
length/complete viral genome by long PCR may suggest the presence of infectious
viral particles. If only small product-specific nucleic acid fragments are amplifiable,
then the clinical sample is not likely to contain infectious viruses.
2. 	Can viral or bacterial growth attributable to the shed product be detected in the
clinical sample?
Ideally, the shedding assay should be able to discern infectious from non-infectious
viruses, or dividing from non-dividing bacteria. We refer you to section VIII. of this
guidance for details on shedding assays and for our recommendations. If qPCR is the
only assay you have relied on in your shedding analysis of replication competent
products, or the shedding assay screens for small genome fragments of the product,
then we will assume that the shed material in the positively scored samples is
infectious.

Contains Nonbinding Recommendations
13
B. 	The Extent of Shedding
In the analysis of shedding data, the extent of shedding noted for each sample type as a
factor of time, dose (amount of product administered) and regimen (number of doses)
should be reported for all the patients monitored in the study. Raw data in the shedding
report should be accompanied by a corresponding analysis that is comprehensive and
describes the following:
1. 	The number of patients that are shedding as a percentage of the total patients in
the study for each sample type, dose and regimen studied.
2. 	The duration of shedding, including the first and last day of shedding, and the
peak period(s) of shedding in each sample type. The period when shedding stops
in most patients in the study also should be clearly identified.
3. 	The clinical sample(s) where shedding was consistently noted (type and time
point) and samples that were consistently negative for all the patients in the study.
4. 	The quantity of product shed in a clinical sample. The amount shed should be
reported taking into account the final volume/mass of the clinical sample, (e.g., 10
PFU of virus per mL of urine, or 10 CFU of bacteria per mL of urine, or 10
genome copies per microgram of stool). When assessing the quantity of
shedding, you should factor in the stability of the product in the clinical sample,
and whether there could be an underestimate of the level of shedding because of
loss during sample storage, handling and shipping (for details, please refer to
section VIII. of this guidance).
Note that your analysis of shedding data should be accompanied by a summary of the
shedding profile of the product in patients treated for a specific indication. While it is
common practice in clinical development of VBGT and oncolytic products to study a
product for different indications in multiple trials, the shedding pattern may be distinct in
each study population. We recommend against pooling shedding data from multiple
trials in which the same product is studied for different indications because results from a
shedding study in a given indication may not be generalizable to other indications.
Finally, we recommend that the shedding data be submitted in a format as described in
the next section.
X. 	WHAT TO INCLUDE IN A CLINICAL SHEDDING STUDY REPORT
In order to address the potential for transmission to untreated individuals due to shedding of
VBGT or oncolytic products from the patient, a full shedding report should be provided in the
BLA.6 The following should be provided in the report:
6 Interim data and study reports can be submitted in your IND annual report, for FDA review, if guidance is desired.

Contains Nonbinding Recommendations
14
A. 	A comprehensive shedding profile of the product in the target patient population(s)
which includes the following:
1. 	Background information on the product: Derivation history; biological characteristics
of parent viral or bacterial strain; route of transmission of parent strain; replication
competence; attenuation; and tropism of the product.
2. 	A summary of the biodistribution profile in animal models and the findings from
preclinical shedding studies, if conducted.
3. 	The rationale for both the clinical shedding study design (i.e., choice of clinical
samples, frequency and procedures of collection and storage), and the analytical
method selected to assess shedding.
4. 	Your data collection/sampling plan and your procedures for storage, shipping, and
handling of the product.
5. 	An assay description that includes the following:
(a) Test sample preparation or nucleic acid extraction procedures including the
dilution factor and amount of nucleic acid extracted per sample.
(b) If you use qPCR assays, provide for each assay the sample volume, amount of
nucleic acid per reaction, cycle numbers, primers, and size of the amplified
DNA.
(c) If you use infectivity/growth-based assays, provide the permissive cell
line/growth media, the conditions for adsorption and infection or growth, and
the nature of the read-out (TCID50, FFU, PFU or CFU; or Cycle threshold
(Ct) for assays with qPCR read-out).
(d) Assay qualification, controls and sensitivity: Description of the qualification
studies, standards, spikes, controls, number of replicates, assay variability and
sensitivity (LOD and LOQ, if applicable).
6. 	Analysis of shedding data:
(a) Tabulation and/or graphical representation of the shedding data.
(b) Analysis of the data and summary of the findings from the study.
7. 	Your estimate of the potential for transmission to untreated individuals of the product.
In this guidance, see section VIII. for a discussion of analytical assay types and
conditions, section VII.B. for a discussion of clinical sample collection, section IX. for

Contains Nonbinding Recommendations
15
analysis of shedding data, and section XI. for details on assessing the potential for
transmission to untreated individuals due to shedding.
B. 	As needed, data and analysis of clinical monitoring for transmission to untreated
individuals in the target patient population (refer to section XI.B. of this guidance for
additional information).
C. 	Other relevant information on the ability of the VBGT or oncolytic product, or its
parental/related strain of viruses or bacteria to potentially infect humans and cause
disease. When there is the potential to cause disease in humans, the following should
be discussed:
1. 	The spectrum of disease symptoms caused by the parental strain, including atypical
presentation of the disease, or occurrence of asymptomatic shedding;
2. 	The attenuation of the product compared to the parent strain of the virus or bacterium
circulating in the community;
3. 	The natural or acquired immunity of the general population that could potentially
protect against infection from the shed product;
4. 	The therapeutic options to treat the infection/disease in case of transmission to
untreated individuals of the shed product; and
5. 	Preventive/containment measures that can limit spread of the shed product beyond the
treated individual to minimize exposure of third parties, particularly, immune-
compromised adults, neonates and seniors. Note that the data you collect on onset
and duration of shedding can inform appropriate preventive/containment measures.
For example, if peak shedding occurs soon after treatment when the patient is
monitored in a health-care setting, then the possibility of transmission is mainly
confined to health care professionals (HCP) and individuals that come into close
contact with the patient. If shedding is prolonged or if there is a second peak of
shedding in the days following discharge of a patient from a health care setting, there
is the possibility of transmission to contacts beyond the health care and home setting.
XI. 	ASSESSING THE POTENTIAL FOR TRANSMISSION TO UNTREATED
INDIVIDUALS DUE TO SHEDDING
Our current understanding is that in most cases, the potential for transmission to untreated
individuals is extremely low when VBGT or oncolytic products are shed because of the
derivation methods and/or modifications that are designed to attenuate the product when
compared to the parent strain of virus or bacterium. Nevertheless, you should discuss the
potential for transmission based on the analysis of the shedding data collected in the clinical
studies and taking into consideration the factors described below.

Contains Nonbinding Recommendations
16
A. 	What Information in the Shedding Data Can be Used to Assess Potential for
Transmission to Untreated Individuals?
• 	Whether the VBGT or oncolytic product was shed.
• 	Whether the shed product was determined to be infectious.
• 	Whether the amount of infectivity in the clinical samples was comparable to that
needed to initiate infection in a third party. For example, adenovirus infective
dose is reportedly >150 PFU when given intra-nasally (Ref. 4) or lower when
aerosolized (Ref. 5). The minimum infectious human dose may vary when
viruses or bacteria are administered or acquired through different routes, or
among different strains, but for many disease-causing viruses and bacteria, the
minimum infectious dose in humans may be undefined.
• 	Whether the clinical sample containing the shed product represents the natural
route of transmission. For example, a respiratory virus that is shed in feces may
not be as infectious and transmissible when compared to that shed in
nasopharyngeal secretions.
B. 	Monitoring Untreated Individuals for Transmission
Because transmission to untreated individuals is an extremely low probability event,
monitoring such individuals for transmission is usually not required during the clinical
development of a product. However, if there is a potential for transmission, additional
data will be needed to assess that possibility; in which case, we recommend that sponsors
consult with OCTGT in connection with developing a monitoring plan.

Contains Nonbinding Recommendations
17
XII. 	REFERENCES
1. 	ICH Considerations: General principles to address virus and vector shedding, June 2009.
Available at
http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Multidisciplina
ry/M6/Concept_Paper/Considerations_on_Viral_Vector_Shedding.pdf.
2. 	Simon, A. et al., Viral infections in paediatric patients receiving conventional cancer
chemotherapy. Arch Dis Child. 93:880-889 (2008).
3. 	Schenk-Braat, E. et al., An inventory of shedding data from clinical gene therapy trials. J
Gene Med. 9: 910-921 (2007).
4. 	Health Canada, 2001. Material data safety sheet – infectious substances. Available at
http://www.phac-aspc.gc.ca/lab-bio/res/psds-ftss/msds3e-eng.php.
5. 	Couch, R.B. et al., The minimal infectious dose of adenovirus type 4; the case for natural
transmission by viral aerosol. Trans Am Clin Climatol Assoc. 80: 205-211 (1969).