U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
Center for Biologics Evaluation and Research (CBER)
Center for Devices and Radiological Health (CDRH)
June 2006
Clinical/Medical
Guidance
for Industry
Chronic Cutaneous Ulcer and Burn Wounds — Developing Products
for Treatment
Additional copies are
available from:
Office of Training and
Communications
Division of Drug Information, HFD-240
Center for Drug Evaluation and Research
Food and Drug Administration
5600 Fishers Lane
Rockville, MD 20857
(Tel) 301-827-4573
http://www.fda.gov/cder/guidance/index.htm
or
Office of Communication,
Training, and
Manufacturers Assistance, HFM-40
Center for Biologics Evaluation and Research
Food and Drug Administration
1401 Rockville Pike, Rockville, MD 20852-1448
(Tel) 800-835-4709 or 301-827-1800
http://www.fda.gov/cber/guidelines.htm
or
Office of Communication,
Education, and Radiation Programs
Division of Small Manufacturers, International and Consumer
Assistance, HFZ-220
Center for Devices and Radiological Health
1350 Piccard Drive
Rockville, MD 20850-4307
DSMICA E-mail:
dsmica@cdrh.fda.gov
DSMICA Fax: 301-443-8818
(Tel) Manufacturers Assistance: 800-638-2041 or 301-443-6597
(Tel) International Staff: 301-827-3993
http://www.fda.gov/cdrh/ggpmain.html
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
Center for Biologics Evaluation and Research (CBER)
Center for Devices and Radiological Health (CDRH)
June 2006
Clinical/Medical
Guidance for Industry
Chronic Cutaneous Ulcer and Burn Wounds — Developing Products for
Treatment
This
guidance represents the Food and Drug Administration’s (FDA’s)
current thinking on this topic. It does not create or confer
any rights for or on any person and does not operate to bind FDA
or the public. You can use an alternative approach if the
approach satisfies the requirements of the applicable statutes
and regulations. If you want to discuss an alternative
approach, contact the FDA staff responsible for implementing
this guidance. If you cannot identify the appropriate FDA
staff, call the appropriate number listed on the title page of
this guidance.
The purpose of this guidance is to provide
recommendations to sponsors for the development of drugs,
biological products, and devices
to treat chronic cutaneous ulcer and burn wounds (i.e.,
wound-treatment products). The first part of this guidance
addresses specific preclinical considerations. The guidance then
addresses important considerations in clinical study design,
including endpoint selection and manufacturing.
This guidance specifically refers to venous
stasis ulcers, diabetic foot ulcers, pressure ulcers, and burn
wounds. For the purposes of this guidance, a chronic cutaneous
ulcer is defined as a wound that has failed to proceed through
an orderly and timely series of events to produce a durable
structural, functional, and cosmetic closure. A burn wound
is defined as a cutaneous wound induced by thermal, chemical, or
electrical injury.
In the Food and Drug Administration (FDA),
the Center for Drug Evaluation and Research (CDER), the Center for
Biologics Evaluation and Research (CBER), and the Center for
Devices and Radiological Health (CDRH) all regulate products to
treat cutaneous wounds. This guidance contains recommendations
applicable to the development of products regulated by any of the
three Centers. Center-specific issues and advice are noted where
appropriate.
FDA’s guidance documents, including this
guidance, do not establish legally enforceable responsibilities.
Instead, guidances describe the Agency’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 Agency guidances means that something
is suggested or recommended, but not required.
A product’s labeled indication for use
is based on substantial evidence and is reflective of the safety
and efficacy of the product as determined in clinical
investigations. In the case of wound-treatment products, the
stated indication should include the types of wounds for which a
product is intended (e.g., venous stasis ulcer, diabetic foot
ulcer, pressure ulcer, burn sites, and donor sites) as well as the
product’s benefits, risks, and limitations. Because wounds differ
in their pathophysiology, it is difficult to generalize results
obtained from a trial conducted in subjects with one wound type to
patients with another wound type. Therefore, separate clinical
trials should be considered for each type of wound indication
sought. However, if a scientific rationale and clinical data
support clinical activity of a product in more than one wound
type, it may be possible for studies performed in one wound type
to support another in establishing substantial evidence of
efficacy and safety.
This section consists of specific points to
consider for wound indications (including wound healing and wound
care) for drugs and biological products. It is not intended as a
general guidance for preclinical testing.
Animal wound models can be helpful in
establishing pharmacological responses, as well as assessing
potential toxicities of wound-treatment products. The choice of
an animal wound model should be based on the best science
available, as well as its applicability to the scientific
questions that one is attempting to answer. The animal species
selected should exhibit a biological responsiveness to the test
product (i.e., should be a relevant species). Although animal
models can be useful for establishing proof of concept for some
types of products, in general they can be inadequate predictors of
efficacy in clinical trials. Since currently there are no ideal
animal models for chronic wounds or extensive burns, multiple
animal models typically should be used to assess activity of
wound-treatment products. Fibroplasia and stroma formation can be
evaluated by subcutaneous injection of some products in various
animal models. Contraction and re-epithelialization can be
evaluated by topical application on full-thickness excisional
wounds or in a pig graft donor site model. Pigs can be useful
models because their cutaneous architecture is most similar to
that of human skin. Induction
of angiogenesis can be evaluated in chick chorioallantoic membrane
or rabbit cornea. Breaking strength can be tested in a rat linear
incision model.
In impaired-healing models, the window of
time for measuring treatment effects is extended.
Impaired-healing models include infection, necrotizing trauma,
irradiation, administration of corticosteroids or chemotherapeutic
drugs, or drug-induced or genetic diabetes mellitus in mice, rats,
hamsters, guinea pigs, and young pigs. Each model has one or more
of the characteristics that can be useful for evaluating a
product’s activity. For example, the rabbit ear dermal ulcer
model is useful for evaluating re-epithelialization because it
lacks the vigorous wound contraction seen in other rodent models
and, in addition, allows for the induction of ischemia in the
wound. There is extensive scientific literature that can be
consulted for a more comprehensive discussion of the advantages
and disadvantages of the models previously cited, in addition to
other available wound models.
Although there are no ideal animal models for
chronic wounds, in vivo biodistribution and pharmacokinetic (BD/PK)
studies generally provide helpful data for the design of
toxicology studies. Preferably, the pharmacokinetic (PK) profile
can be determined in the same animal species that will be used in
the toxicology assessment. For topical wound-treatment products,
application of the product to a wound site on the animal’s skin
may provide more relevant information than application to intact
animal skin. When technically feasible, the potential for
regional and systemic exposure to a product for a chronic wound
indication might be better approximated by subcutaneous
injection. Consideration should also be given to alterations of
the BD/PK profile and the potential for product accumulation with
repeated dosing. Information regarding the stability of the
product at the target site (target receptor levels for biological
products) contributes to a better understanding of the activity
and potential toxicity of the wound-treatment product.
The design of nonclinical toxicology studies
for wound-treatment products should reflect, as much as possible,
the intended clinical use of the product
with respect to route, dosing regimen, and duration of exposure. It
is important to assess any exaggerated pharmacological
responses and potential toxicities of wound-treatment products.
Administration of the wound-treatment product at multiples higher
than the anticipated therapeutic dose (potentially determined from
wound models) can provide an estimate of the therapeutic index
(toxic dose/effective dose) to aid in the selection of the initial
clinical starting dose. Vehicle and sham controls should
be employed where appropriate to evaluate any adverse or
beneficial effects of product formulation ingredients on wound
healing and adverse events. The vehicle control animals should
receive the same excipients and formulation as are in the intended
clinical product, without the active agent. The sham control
animals should be manipulated in the same manner as the vehicle
control animals, but should not receive the vehicle or the active
agent.
Cutaneous irritation and sensitization
testing should be generally indicated for all topically applied
wound-treatment products, since these adverse reactions can
seriously complicate human wounds. Products that will be
delivered in an aerosol formulation should be evaluated for
pulmonary toxicity, and possibly ocular toxicity (products known
to be cutaneous irritants are assumed to be ocular irritants, and
testing is generally waived).
The immunogenic potential of
biotechnology-derived wound-treatment products can be a
confounding factor in repeat-dose toxicology studies because
antibodies to the administered product may affect the PK profile,
the pharmacodynamic response, and the toxicity of the product.
Although the development of antibodies to antigenic products
generally has not been predictive of the clinical response, data
on antibody formation should be collected to provide a complete
preclinical safety assessment of the wound-treatment product.
Sponsors are encouraged to seek FDA input during development to
ensure collection of adequate immunogenicity information.
Carcinogenicity studies generally should be
conducted for drugs intended to treat chronic ulcers.
For biological products, the 2-year chronic bioassay and
carcinogenicity study currently used for drugs is generally
inappropriate because of species specificity and immunogenicity of
the product. However, data in rodent initiation-promotion
carcinogenesis models support the potential of various growth
factors to act as tumor promoters. Current unresolved issues
regarding the carcinogenic and tumorigenic potential of
wound-treatment products include the likelihood of tumor promotion
in the proposed patient populations and the additional
susceptibility of patients exposed to environmental or other
potential carcinogens (e.g., systemic chemotherapy) as well as the
possibility of inducing scar carcinomas in chronic wounds.
Sponsors are encouraged to address this issue by referencing the
existing scientific literature, and evaluating the potential of
the test product to stimulate the growth of normal and malignant
cells that express the receptor for the product. Sponsors are
encouraged to seek FDA input for product-specific questions that
can affect the carcinogenicity evaluation.
Reproductive and developmental toxicology
studies are recommended for wound-treatment products that might be
administered to women of childbearing potential.
Genotoxicity studies should be performed for
all nonbiological drug products. These studies are indicated for
a biotechnology-derived product only when supported by appropriate
scientific rationale.
This section consists of specific points to
consider for wound indication trial design. It is not intended as
general guidance on trial design.
A. Randomization and
Stratification
Randomization is particularly important for
reducing bias in wound indication trials because standard wound
care procedures and baseline wound characteristics generally have
a profound effect on outcome. If variation in standard wound care
procedures among clinical study sites is unavoidable,
stratification by study center is recommended to minimize any
imbalances among study arms. In some cases, it may be appropriate
to prospectively stratify randomization by other important
covariates, such as wound size or duration. Variables thought to
significantly affect outcome should be incorporated into the
planned efficacy analyses even if these variables are not used for
stratification in randomization (see section IV.I., Statistical
Considerations Specific for Wound-Treatment Product Trials).
A comparator arm is recommended for many
wound-treatment product trials involving drugs, biologics,
devices, and combination products (i.e., drug delivery studies or
cellular wound dressings). This is usually a vehicle control arm.
The vehicle control should contain the same formulation and
excipients as the study product, without the active agent. Such
trials should be performed with identical standard-of-care
procedures in both the control and investigational product arms.
If the effect of a product’s vehicle is not established, product
development should also include a study arm treated with standard
of care alone, usually in exploratory studies. In the selection
of comparator groups, careful consideration to optimizing the
dosage, frequency of administration, and method of use is
important.
Trial designs in which subjects serve as
their own control have been used to study topical products
intended for serious burns in an attempt to minimize the
heterogeneity characteristic of this patient population. However,
this approach compromises the evaluation of systemic toxicity,
necessitating additional controls or studies to collect adequate
safety data.
In general, blinding of subjects and
investigators to the assigned treatment reduces bias and should be
employed when feasible. Early trials of topical wound-treatment
products often include a treatment group that receives only
standard care (see section IV.B., Comparator Arms) to establish
whether the vehicle has an effect on healing. Often the standard
care only arm cannot be blinded. In other cases, especially for
trials of some medical devices, it is impractical or unethical to
implement a control treatment that mimics the test product for the
purposes of blinding. In these situations, blinded assessment by
a third-party evaluator should be considered.
For most ulcer types, we encourage the
selection of a single target lesion for efficacy determination
before subject randomization. In some cases, it may be
appropriate to treat other lesions with the investigational
product to obtain additional safety information, which is still
reported as per patient. An alternative approach, that may
be appropriate, is complete healing of all lesions reported as
per patient.
The tools to assess clinical trial endpoints
should be both prespecified and, for multicenter trials,
standardized across clinical sites. For example, if photographs
will be used for measurement and documentation of wound changes,
the lighting, distance, exposure, and camera type should be
specified and consistent at all clinical centers.
Regardless of the methodology, the following
variables should be addressed in all clinical trials for
wound-treatment products.
The type of chronic ulcer (venous stasis,
diabetic, pressure, arterial insufficiency) usually can be
determined by considering the subject’s history and performing a
physical examination. Objective tools to confirm the diagnosis
can include the following:
- Doppler sonography to qualify and quantify
vascular insufficiency: arterial or venous (deep, superficial,
or mixed)
- Transcutaneous oxygen tension (tcpO2)
measurements
- Ankle/brachial index
- Filament testing to quantify sensory
neuropathy
- Measurement of laboratory markers for
diabetes mellitus
- Histopathology of ulcer biopsies to
exclude neoplastic, immune-mediated, or primary infectious
disease
Quantitative measurements of wound size are
routinely used to assess initial wound size before and after
debridement, as well as progress toward closure. For ulcers that
are more superficial, such as venous stasis ulcers, the area of
the wound opening should be measured. For ulcers that extend
deeply into tissue, volume or surface area should be measured when
feasible. The extent of tissue undermining and sinus tracts is an
important part of the evaluation. In the case of diabetic ulcers,
qualitative assessment by determining the maximal depth is a
frequently used method. For other ulcers, such as pressure
ulcers, molds can be used to provide precise measurement of volume
and surface area. Alternatively, semiquantitative measurements
can be achieved using the maximal width/length/depth and shape
coefficient. There are also widely accepted criteria used to
classify the stages of ulcers (e.g., National Pressure Ulcer
Advisory Panel (NPUAP) for Pressure Ulcers: NPUAP Classification,
Wagner’s Classification for foot ulcers).
For acute burns, it is important to attempt a
determination of the burn depth as this parameter affects both the
choice of standard-of-care regimen and the expected time to
healing. The distinction between partial, full-thickness, and
indeterminate wounds is currently based on clinical judgment;
biopsy and Doppler measurement of blood flow are sometimes used as
well. Clinical parameters include appearance of the tissue,
sensation, and bleeding upon debridement. Wound depth
heterogeneity is often an impediment to quantitative measurement,
and burn depth extension in the first 24 to 48 hours post-injury
frequently necessitates reassessment. Because burn wound depth
often becomes better defined following the initial evaluation, the
initial clinical assessment of the body surface area affected by
deep partial full-thickness (2nd degree) and full-thickness (3rd
degree) wounds can be compared with the total body surface area
ultimately grafted.
When the target wound is an autograft donor
site, the protocol should clearly delineate the method for
harvest, and the size, thickness, and anatomic location of the
donor site.
Infection should be assessed clinically by
symptoms and signs that include purulent drainage, erythema,
warmth, exudation, odor, pain, fever, and leukocytosis as well as
wound size and time to wound healing. Fever, pain, and
leukocytosis may be absent, however, especially in subjects with
diabetic foot ulcers. Quantitative and qualitative culture of a
tissue biopsy can be used at baseline to help determine if the
wound is infected or merely colonized, and to guide appropriate
antimicrobial therapy. This method is generally preferred to
quantitative and qualitative culture of swab specimens.
The patient population to be included in
clinical trials should be appropriate for the type of wounds to be
studied. In general, the patient population chosen should be one
that optimizes the study’s ability to detect a treatment effect,
but should also be a population that reflects the population for
which the product will be indicated and used.
Three of the major categories of chronic
cutaneous ulcers are diabetic ulcers, venous stasis ulcers, and
pressure ulcers. As previously stated, separate trials generally
are appropriate for each type of chronic ulcer because these
ulcers have different etiologies and potentially different
responses to therapy. If demonstration of efficacy is restricted
to ulcers with a limited range of size or duration, and the
ability to extrapolate to larger or more recalcitrant ulcers is
unclear, then the labeled indication can be limited to ulcers with
similar parameters to those studied.
Healing of chronic ulcers can occur as
subject compliance with standard treatment improves during the
initial involvement in a clinical trial. Variability in aspects
of care that affect outcome can interfere with a study’s ability
to demonstrate the treatment effect of the investigational
product. Approaches for minimizing this variability can include
the use of: a) a study design with an initial phase (e.g., 1 to 2
weeks) when subjects receive only standard of care, and b) an
entry criterion that excludes subjects whose study ulcer size
decreases by a stated amount (e.g., 30 percent to 50 percent)
during this initial standard-of-care phase. Thus, the
randomized study population would exclude subjects demonstrating
substantial healing resulting solely from improved compliance with
standard care.
The population for burn trials is usually
characterized by the depth, surface area, and location of the burn
injury, as well as subject characteristics. Important
characteristics of a burn wound include its cause (thermal,
chemical, electrical), anatomic location, depth (full or partial
thickness), duration, and extent (percent total body surface
area). Subject characteristics that affect burn wound healing
include age, nutritional status, underlying medical conditions,
concomitant injury (e.g., head trauma, inhalation injury, bone
fractures), and scores that represent an overall severity of
illness or injury (e.g., the American Society of Anesthesiologists
(ASA) Classification, the Trauma-Injury Severity Score (TRISS), or
the Acute Physiology and
Chronic Health Evaluation (APACHE) III score).
Patients with serious burns commonly receive multiple concomitant
treatments, making it sometimes difficult to detect a treatment
effect. For this reason, stratification by injury severity and
other potentially confounding factors that are clinically
significant should be considered to minimize imbalances among
treatment groups. In early clinical development, to protect
subject safety, generally it is not advisable to include
investigational treatment of burns in anatomic locations such as
the face and hands as well as sites deemed high risk for
developing compartment syndrome; similarly, the body surface area
treated with the investigational product should be limited.
In clinical trials of subjects with
full-thickness burn wounds, donor sites for autografts are
sometimes selected as the target wound for study. However, the
demonstration of safety and efficacy of a product for a donor site
wound does not support the safety and efficacy of the product for
burn wounds, because burn wounds differ in clinically significant
ways from surgical wounds.
Standard care refers to generally
accepted wound care procedures, other than the investigational
product, that will be used in the clinical trial. Good standard
care procedures in a wound-treatment product trial are a
prerequisite for assessing safety and efficacy of a product.
Since varying standard care procedures can confound the outcome of
a clinical trial, it is generally advisable that all participating
centers agree to use the same procedures and these procedures are
described within the clinical protocol. If it is not practical to
apply uniform standard care procedures across study centers,
randomization stratified by study center should be considered. It
is also important that the sample size within study centers and
wound care records be adequate to assess the effect of wound care
variation.
A number of standard procedures for ulcer and
burn care are widely accepted. Several professional groups have
initiated development of care guidelines for ulcers and burns.
The Agency does not require adherence to any specific guidelines,
the basic principle being that standard care regimens in
wound-treatment product trials should optimize conditions for
healing and be prospectively defined in the protocol. The
rationale for the standard care chosen should be included in the
protocol, and the study plan should be of sufficient detail for
consistent and uniform application across study centers. Case
report forms (CRFs) should be designed such that, at each visit,
investigators describe the type of ulcer or burn care actually
delivered (e.g., extent of debridement, use of concomitant
medications). For outpatients, the CRF should also capture
compliance with standard care measures, including wound dressing,
off-loading, and appropriate supportive factors, such as dietary
intake.
The value of study site consistency in
standard care regimens within a trial cannot be over-emphasized
because of the profound effects these procedures have on clinical
outcome for burns and chronic wounds. Consistency in standard
care regimens is important for minimizing variability and allowing
assessment of treatment effect. It may be reasonable to evaluate
a single standard care regimen in early trials to minimize this
variability. If comparison of an investigational product to more
than one commonly used standard care option is desired, the
overall development plan should include specific assessment of the
effect of these standard care options on the experimental
treatment. These common options should be identified and
addressed prospectively in clinical trial design including being
clearly described in the clinical protocol and compliance captured
via the CRFs; criteria for data poolability should be defined
prospectively. Every attempt should be made to minimize
deviations from the procedures described in the protocol and
subject compliance recorded in CRFs. If more than one standard
care regimen is used in the same clinical trial, then randomized
treatment allocation within strata defined by these options in
standard care is important.
Parameters for consideration in choosing
standard care procedures for chronic cutaneous ulcer trials
include the following:
- Removal of necrotic or infected tissue
- Off-loading
- Compression therapy for venous stasis
ulcers
- Establishment of adequate blood
circulation
- Maintenance of a moist wound environment
- Management of wound infection
- Wound cleansing
- Nutritional support, including blood
glucose control for subjects with diabetic ulcers
- Bowel and bladder care for subjects with
pressure ulcers at risk for contamination
The presence
of necrotic tissue, sinus tracts, exudation or transudation, and
infection of soft and hard tissues can interfere with ulcer
healing. Appropriate debridement procedures for the indicated
ulcer should be specifically defined in the protocol. To avoid
bias and confounding of potential treatment effects, ulcer
debridement should generally precede evaluation of ulcer extent
and infection. Enzymatic debriding products, like other
concomitant topical products, can confound results in topical
wound-treatment product trials and generally should be avoided.
The need for additional sharp surgical
debridement, performed after study treatment has started, may
indicate product-induced wound deterioration. Any surgical
debridement should be documented on CRFs and included in analysis
of product safety and efficacy. In general, the study protocol
should specify that investigators discontinue study treatment if
repeat debridement is needed to address wound deterioration. For
example, ongoing removal of callus as part of standard diabetic
ulcer care would not require discontinuation of study treatment,
but deep debridement for new onset osteomyelitis would.
b.
Off-loading and compression
Relief of pressure is critical in determining
the outcome for certain chronic ulcers. Off-loading is often
difficult to standardize because equipment (e.g., type of bed) may
not be available at all sites, and compliance with study
procedures is labor intensive (e.g., turning). If these critical
aspects of effective therapeutic intervention cannot be
standardized across all sites, it is important to specify the
actual care delivered in CRFs and to consider concomitant care in
the efficacy analyses. For diabetic foot ulcers, compliance with
off-loading is a strong factor associated with wound closure and
should be monitored throughout the clinical trial. Off-loading
approaches (e.g., some form of casting and elevation) should be
weighed against the need to apply study treatments and monitor
outcome. Similar considerations are important in choosing
compression methods for venous stasis ulcers.
Maintenance of a moist wound environment
generally is accepted as a component of standard care for all
chronic cutaneous ulcers. Methods of maintaining a moist wound
environment should be identified in the clinical protocol and
optimized for the specific patient population, the wound type to
be studied, and the type of product being tested. If there is a
sound rationale for the expected benefit of a test product that
cannot be used with established standard care dressings,
alterations in standard care might be safely implemented. This
can be done by including in the protocol adequate criteria for
discontinuing subjects from study treatment or through the
appropriate implementation of an independent data safety
monitoring board to oversee trial safety.
Effective control and management of infection
is critical for treatment success of all wound-treatment products,
regardless of the claim. For this reason, wound-treatment
products that are not intended for use as an anti-infective
generally should be studied in subjects in whom the target ulcer
is uninfected. To this end, it is often appropriate to include in
the trial design a run-in period with standard therapies to
control infection.
In specific cases, such as the case of
diabetic foot ulcers, wound-treatment products that are not
anti-infective can be studied in wounds where infection is limited
and responding to standard systemic antimicrobial therapy. In
trials involving infected ulcers it is especially important that
the protocol clearly delineate adequate criteria for discontinuing
subjects from study treatment because of wound deterioration
during the treatment period (see section IV.H.3., Discontinuation
of Treatment).
If a subject’s ulcer becomes infected during
a study for a topical wound-treatment product, and the
investigator prescribes topical antimicrobial treatment, the
subject generally should be discontinued from study treatment.
(As for all discontinued subjects, safety assessment should
continue throughout the trial and these subjects should be
included in efficacy analysis). However, if sponsors anticipate
that post-approval clinical use would likely entail concomitant
topical products, the trial should be designed to address this
variable.
Systemic antimicrobial therapy for target
wound infection may become necessary during the treatment period
of the study. Whether or not study treatment should be
discontinued in this situation should be discussed prospectively
and the plan included in the protocol. For example, treatment
discontinuation might be indicated in early trials, where little
is known about product safety and where infection may signal test
product-induced deterioration of the wound, but not in later
trials in subjects for whom systemic antimicrobial therapy would
be considered a component of standard care.
Nutrition is important to healing and immune
competence, and should be tracked to allow assessment of this
aspect of a subject’s ability to heal. Caloric intake and
metabolic status should be captured in the CRFs if the product is
known to have metabolic effects (e.g., anabolic steroids). For
products not known to have metabolic effects, these data may be
useful if the inclusion criteria encompass subjects significantly
above or below ideal body weight (e.g., cachectic subjects with
pressure ulcers). In studies of subjects with diabetic ulcers, it
is important to record blood sugar levels because nutritional
support with adequate blood sugar control is considered standard
care for wound healing in these subjects.
Standard care for serious burns includes
careful attention to the following parameters:
- Hemodynamic resuscitation
- Management of co-morbidities
- Timely burn debridement and excision
- Wound closure
- Management of wound infection
- Pain control
- Nutritional support
- Measures to inhibit excessive scar
formation
- Rehabilitation, including passive range of
motion when burns overlie joints
Because large burn centers tend to have
well-established, distinct standard care regimens, randomization
within multicenter burn trials may require stratification by
center. Since standard care procedures have profound effects on
clinical outcome, every effort should be made to reach agreement
among site investigators and to capture actual care delivered in
the CRFs.
In general, clinical outcomes associated with
the use of a wound-treatment product can be broadly grouped into
two categories: improved wound healing and improved wound care.
Each outcome category includes a variety of potential endpoints
for clinical trials. Suggestions for possible outcome measures
and clinical trial endpoints offered in this guidance are based on
current knowledge of the natural history and management of burns
and ulcers.
Complete wound closure of a chronic,
nonhealing wound is one of the most objective and clinically
meaningful wound healing endpoints. Complete wound closure
is defined as skin re-epithelialization without drainage or
dressing requirements confirmed at two consecutive study visits 2
weeks apart. Generally, trials to support an indication of
complete wound closure measure incidence of complete wound closure
in the treatment group and the control groups by a specified time
(landmark analysis). In the simplest case, a treatment effect
would be established if a clinically and statistically significant
greater proportion of subjects in the treatment group achieved
complete wound closure compared to the control arm. Timing of the
endpoint measurements should be based on the natural history of
the disease process, the anticipated time course for the treatment
effect, and the expected response
to standard care in the control arm.
Trial subjects
generally should remain in the study for follow-up evaluation at
least 3 months following complete wound closure. The purpose for
this follow-up period is to help distinguish actual wound healing
from transient wound coverage, determine if the product affects
the strength of wound closure relative to standard care, and
monitor for adverse effects on surrounding tissue (e.g., skin,
bones, supporting structures).
Measurement of partial wound healing in early
phase clinical trials, if prospectively defined, may indicate
relevant biological activity and help guide subsequent trial
design. Partial healing would not, however, suffice as a primary
endpoint in phase 3 studies because the clinical benefit of
incremental wound size changes has not been established. However,
partial healing that facilitates surgical wound closure can
be a measurable trial endpoint of clinical benefit.
An indication of accelerated wound closure
should reflect clinically meaningful reduction in the time to
healing using a time-to-event analysis (the event being
complete closure). Assessments should be performed at intervals
sufficiently frequent to detect a meaningful difference in
time-to-closure between treatment groups. If claims are sought
for both increased incidence of wound closure and
accelerated healing, the study should be designed to detect both
effects.
Because partial thickness donor sites
generally heal in 2 to 3 weeks with standard care regimens, when
considering clinical endpoints for a product that accelerates
closure of donor sites, the anticipated clinical benefit should be
taken into account. For example, a product that accelerated donor
site healing by only 1 or 2 days might provide clinical benefit if
it could be safely used in extensively burned patients requiring
repeated re-harvesting of donor sites. In studies where donor
site wounds are chosen as the primary target for efficacy
determinations, it is important to also assess the engraftment of
tissue obtained by re-harvesting as a safety evaluation, because
graft take should not be worsened by a product that accelerates
healing of donor sites.
If the claim sought is facilitation of
surgical wound closure by partial healing, studies should be
designed to measure the incidence of complete wound closure
following the needed surgical procedure (such as flap closure of
an extensive chronic pressure ulcer). The durability and quality
of the surgical wound closure should be assessed over time to
ensure that the product does not have a deleterious effect.
Studies
evaluating the cosmetic aspects of quality of healing (cosmesis)
can be designed to demonstrate a clinically significant effect on
outcomes such as scarring, the contour and feel of the healed
skin, or normalization of skin markings or pigmentation. In
choosing clinically relevant endpoints for an improved cosmesis
claim, it is important to consider the type and location of wound
(e.g., facial versus plantar surface of the foot) and whether a
reliable assessment tool exists, or can be developed.
Wound healing resulting in hypertrophic scar
formation can be a source of long-term morbidity by inhibiting
function (e.g., inhibition of range of motion across joints). As
such, wound-treatment products that reduce hypertrophic scarring
may also lead to improved function. Studies designed to evaluate
the outcome of improved function through clinically relevant
endpoints can be useful in supporting a labeling claim for
function.
We recognize that products intended for wound
management may provide important patient benefit without improving
the incidence or timing of wound closure relative to standard
care. However, it is important to demonstrate that such products
do not significantly impede healing. Thus, wound closure should
be evaluated as a safety outcome for all products with a wound
care claim.
Infection at the wound site impairs healing.
Primary efficacy outcomes for topical antimicrobial
wound-treatment products can thus be healing, prevention
of, or cure of infection. Such antimicrobial products
generally should have an established and appropriate spectrum of
antimicrobial activity.
Debridement of necrotic tissue generally is
considered part of standard care for most ulcers and burns.
Although there is debate about the optimal design of trials to
assess the efficacy of debriding products, examples of clinically
relevant endpoints include improved wound healing (increased
incidence or acceleration of complete closure), reduced pain
associated with the debridement process, or decreased blood loss
during or immediately following debridement. When wound closure
is not the chosen primary efficacy endpoint, the study should
evaluate whether the debriding product impairs healing relative to
standard of care.
Wound pain amelioration endpoints should be
accompanied by assessment instruments that are prospectively
defined and appropriate to measure the type of pain for which an
indication will be sought.
These studies should include, as a safety endpoint, an assessment
of product effects on the healing process itself.
Temporary dressings, including interactive
temporary dressings, are intended to provide supportive care until
definitive closure can be accomplished. Temporary dressings are
expected to function as a barrier, much like human skin. Trial
endpoints other than healing should reflect one or more clinically
relevant barrier functions, such as retardation of fluid loss or
reduced infection rates. If healing is not the primary efficacy
endpoint, it should be evaluated as a safety endpoint.
Wounds can negatively affect many aspects of patients’ lives. Clinically
significant improvement in certain aspects of daily living
(e.g., improvements in functional abilities) might support a
labeling claim if appropriate trial endpoints measure a direct
clinical benefit or if the endpoints are assessed with a
clinically relevant validated instrument.
H. Safety Considerations
For biological products and some drugs,
immunogenicity generally is addressed by measuring antibody titers
before and after the treatment. Further immunological
characterization with measures of the effect of antibody formation
upon bioactivity, safety, and efficacy may be appropriate, since
the development of an immune response can render the product
inactive (neutralizing antibodies), and induce acute or chronic
immune reactions (e.g., anaphylaxis, contact sensitization,
autoimmune disease).
Because the patient populations in burn and
chronic ulcer trials often have a high background incidence of
serious adverse events, it is recommended that a safety data
monitoring board be used for blinded trials when the known or
suspected risk is significant, and the study population is
critically ill (e.g., seriously burned patients). Early in
clinical development, sponsors may need to incorporate specific
trial stopping rules based upon the occurrence of certain
important adverse events, such as death or respiratory failure.
Since the
active ingredients or the vehicle of a topical wound-treatment
product may have a deleterious effect on healing, subjects
participating in early stage clinical trials generally should be
discontinued from study treatment if signs or symptoms suggest
wound deterioration. Wound deterioration can manifest as erythema,
pain, discharge, infection, tissue necrosis, requirement for
repeat debridement or other surgical intervention (i.e.,
amputation), and increase in ulcer size. Undesirable alterations
of soft tissues, ligaments, periosteum, or joint capsules
underlying deep wounds also should be evaluated. Once reasonable
assurance has been achieved that the product does not harm the
wound, it may be appropriate to continue study treatment in later
trials, depending on the type of wound and patient population.
Subjects who are discontinued from study treatment should remain
in the study for safety assessment and efficacy analysis.
For topical
drug, biological, and combination products, early clinical
evaluations should include quantitation of absorption through the
wound. Systemic bioavailability of topically applied products is
generally assessed using standard pharmacokinetic measurements
with serial serum sampling. Systemic uptake is influenced by
wound factors such as size and vascularity, as well as product
characteristics such as molecular weight, chemical composition,
and the presence of excipients. Systemic availability of a
topical product may be profoundly influenced by wound bed
preparation. Thus, such studies should be performed when possible
on a well-debrided wound as long as the debridement process is
consistent with the methods of applying the investigational
product and standard methods of wound care.
In the case of growth factors, relatively little (less
than 1 percent) absorption typically occur from chronic ulcer
sites, but these amounts might be clinically significant because
some growth factors are active in vitro at nanogram
concentrations. For this reason, it is important to perform
sensitive assays against serum background of the endogenous growth
factor.
For products that are absorbed from the wound
bed, the systemic dose depends on several factors including the
concentration of the active ingredient, the total body surface
area treated, the volume applied, frequency of application, and
duration of contact with the wound. Safety and pharmacokinetic
studies for topical wound-treatment products usually should be
conducted in subjects with the indication sought, since absorption
through intact skin of a healthy volunteer would not predict
absorption in a wound.
When preclinical studies or previous clinical
experience suggest that a topical product might induce clinically
significant dermatitis, irritation, or sensitization, we recommend
that testing be done on intact skin before conducting clinical
trials in chronic ulcers, wounds, or burns. The testing is
recommended because superimposed dermatitis can be deleterious to
wounds. The inclusion of routine testing of the final formulation
depends on the product, and sponsors are encouraged to discuss
dermal toxicity testing with the appropriate center before
initiating the trials.
This section
addresses issues that present special considerations for
wound-treatment product trials.
Analysis should be prespecified in the
protocol. For incidence of closure endpoints, categorical
techniques are recommended (e.g., Chi-square, tests of
homogeneity, or logistic regression). For time-to-closure
endpoints, outcome survival analyses should be performed.
For wound-treatment product trials, the
center or investigator is frequently included as a factor in the
analysis, because of variations in standard of care. Some
analytical tests (e.g., Mantel-Haenszel statistic and the Cox
Proportional Hazards Model) allow for covariate adjustment.
In the
clinical trial design, prospective stratification in randomization
should attempt to balance the trial arms for the one or two most
important variables that are highly likely to affect the trial’s
primary endpoint. Covariate analyses can be employed to adjust
for variables that affect the outcome. If covariate analyses are
used, the covariates and the analyses should be prespecified to
avoid concerns about interpretability of significance tests.
When analyzing covariates, experience
suggests that it is generally not useful to transform continuous
variables into dichotomous variables (e.g., baseline ulcer size
greater than or equal to 5 cm2 and duration of the
ulcer more than 1 year). The covariate should be used as a
continuous variable. Exploratory analyses can examine subgroups
defined by various cut points, but when a particular cut point is
deemed to be important in guiding the use of the product (e.g.,
ulcers greater than 10 cm2 do not respond), this cut
point should be prospectively identified and studied in a clinical
trial.
A wound represents a breach in the body’s
natural barrier to microbial invasion; therefore, the final
formulation of topical products used for the treatment of chronic
ulcers, wounds, or burns should be sterile to avoid introducing
exogenous microorganisms. Guidance on validation of the
manufacture of sterile products can be found in the guidance for
industry Submission Documentation for Sterilization
Process Validation in Applications for Human and Veterinary Drug
Products. Methods for performing sterility tests on drug
products can be found in USP <71> Sterility Tests.
To avoid contamination of a sterile product,
it is preferable for wound-treatment products to be packaged in
single-use containers. However, if packaged in multi-use
containers, wound-treatment products should either include a
preservative system or possess innate antimicrobial activity.
Antimicrobial preservatives should not be used as a substitute for
good manufacturing practices. The antimicrobial activity of the
product, with (or without) a preservative system, should be
demonstrated by performing a microbial challenge test such as USP
<51> Antimicrobial Effectiveness Test. The minimum
acceptable limit for the content of preservatives (or other
antimicrobial material, as in the case of an innately
antimicrobial API) in a product should be demonstrated as
microbiologically effective by performing a microbial challenge
test of the formulation with an amount of preservative less than
or equal to the minimum amount specified as acceptable. For the
purpose of application approval, stability data on pilot-scale
batches should include results from microbial challenge studies
performed on the product at appropriate intervals. Typically,
microbial challenge studies should be conducted initially,
annually, and at expiration. Chemical assays of preservative
content also should be performed at all test points. Upon
demonstration of the antimicrobial effectiveness of the minimum
specified preservative concentration, chemical assays of the
preservative may be sufficient to demonstrate the maintenance of
adequate antimicrobial activity for annual batches placed into
stability testing. For biological products, testing should be
done to ensure that the preservative does not compromise
biological activity.
Some products cannot withstand sterilization
processes because they degrade when heated or irradiated, and they
are not filterable. If a wound-treatment product cannot be
manufactured to be sterile, it should have a low bioburden. Bioburden
testing should be performed according to a validated test
procedure such as in USP <61> Microbial Limit Tests and
<62> Microbiological Examination of Nonsterile Products: Tests
for Specified Microorganisms (when implemented) at
appropriate, defined time points during stability studies.
Additionally, bioburden testing should include identification of
recovered microorganisms to exclude potentially deleterious
organisms.
Refer to the following standards for
validation of sterilization of medical devices:
- ISO 11137:1995 Sterilization of health
care products — Requirements for validation and routine control
— Radiation sterilization
- ISO 11135:1994 Medical devices —
Validation and routine control of ethylene oxide sterilization
- ISO 11134:1994 Sterilization of health
care products — Requirements for validation and routine control
— Industrial moist heat sterilization (available in English
only)