Guidance for Industry
Exocrine Pancreatic Insufficiency Drug Products –
Submitting NDAs
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 assist
manufacturers of exocrine pancreatic insufficiency drug products
in preparing and submitting new drug applications (NDAs). On
April 28, 2004 (69 FR 23410), the Food and Drug Administration
(FDA) announced that all orally administered pancreatic enzyme
products (PEPs) are new drugs that will be approved for
prescription use only, and explained the conditions for
continued marketing of these drug products. This guidance
pertains to products that contain the ingredients pancreatin and
pancrelipase; these ingredients, which are of animal origin,
contain the following enzymes: lipases, proteases, and amylases.
These enzymes break down fats (lipases), proteins (proteases),
and carbohydrates (amylases) into elementary units of small size
that can traverse the intestinal mucosa, incorporate into the
blood stream, and work as sources of energy and building blocks
of cells.
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.
Pancreatic enzyme preparations of porcine
or bovine origin have been available in the United States for
the treatment of exocrine pancreatic insufficiency (EPI) in
children and adults with cystic fibrosis (CF) and chronic
pancreatitis (CP) since before the enactment of the Federal
Food, Drug, and Cosmetic Act of 1938 (the Act). Under the Act,
beginning in 1938, new drugs were required to be the subject of
approved NDAs. With the exception of one PEP approved in 1996,
PEPs have been marketed without NDAs.
There are approximately 30,000 pediatric
and adult patients with cystic fibrosis in the United States.
Pediatric patients affected with cystic fibrosis and patients
with chronic pancreatitis who have significant reduction of
pancreatic function are unable to digest fats, proteins, and
carbohydrates. As a consequence, the absorption of these
nutrients is impaired, with the resultant malnutrition and a
host of secondary complications, including retarded growth and
development, impaired immune response, infections, and bleeding
tendencies, among others.
In the Federal Register of November
8, 1985 (50 FR 46594), the FDA published a notice of proposed
rulemaking to establish a monograph for over-the-counter (OTC)
EPI drug products. The Agency accepted the recommendations of
the Advisory Review Panel on OTC Miscellaneous Internal Drug
Products (the Panel) that EPI drug products be considered safe
(generally recognized as safe (GRAS)) and effective (generally
recognized as effective (GRAE)) and not misbranded.
Interested persons were invited to submit new data, written
comments, objections, or requests for an oral hearing on the
proposed rulemaking. Based on the information received, the FDA
reconsidered the approach described in the November 8, 1985,
proposed rulemaking and concluded that: (1) an OTC monograph
would not be sufficient to adequately regulate these drug
products; (2) preclearance of each product to standardize enzyme
bioactivity would be necessary; and (3) because continuous
physician monitoring of patients would be necessary as a
collateral measure to ensure the safe and effective use of these
products, such products should be available by prescription
only. In the Federal Register of July 15, 1991 (56 FR
32282), the FDA withdrew the November 8, 1985, proposed rule and
proposed a regulation to declare that OTC drug products used to
treat EPI are not GRAS and GRAE and are misbranded. The final
rule, which affected only OTC products, published on April 24,
1995 (60 FR
20162).
In the 1991 and 1995 proposed and final
rules, the FDA discussed its review of the scientific data that
provide the basis for the FDA’s decision to require approval of
PEPs through the new drug approval process under section 505 of
the Act.
At this time, the FDA expects to receive
only NDAs, including section 505(b)(2) applications, and not
abbreviated new drug applications (ANDAs) for these products.
For a pancrelipase or pancreatin product to be approved as an
ANDA under section 505(j), the proposed drug product must be
shown to contain the same active ingredients as an approved
reference listed drug (21 CFR 314.92(a)(1)). Because of the
complexity of pancreatic extract products, it is unlikely that
currently available physiochemical and biological analytical
tools would be able to demonstrate that the active ingredients
in pancreatic extract products from two different manufacturers
are the same. Therefore, the Agency has concluded that
pancreatic extract drug products currently are not likely to be
appropriate for ANDAs. Nonetheless, manufacturers with further
questions about the feasibility of submitting ANDAs for
pancreatic extract products are advised to contact the Office of
Generic Drugs (HFD-600, Center for Drug Evaluation and Research,
Food and Drug Administration, 7500 Standish Pl., Rockville, MD
20855).
If a sponsor markets or wishes to market
more than one PEP, the number of NDAs it needs to submit will
depend on the composition of the products. If the products vary
by active ingredient (e.g., product 1: amylase and lipase;
product 2: amylase and protease), then separate NDAs should be
submitted. If the products vary only by potency ratios of the
same active ingredients (e.g., product 1: amylase, 15,000
amylase units, lipase, 1,200 lipase units, and protease, 30,000
protease units, and product 2: amylase, 15,000 amylase units,
lipase, 1,500 lipase units, and protease, 35,000 protease
units), then separate NDAs need not be submitted. Different
strengths or concentrations can be submitted in the same NDA.
To be approved, an NDA must meet the
requirements described in 21 CFR 314.50 regarding chemistry,
manufacturing, and controls information. We recommend that
applicants consult the FDA guideline Submitting Supporting
Documentation in Drug Applications for the Manufacture of Drug
Substances and the guidance for industry Submitting
Documentation for the Manufacture of and Controls for Drug
Products.
We also recommend that applicants consult relevant International
Conference on Harmonisation (ICH) guidance documents (e.g., Q1A,
Q2A, Q2B, Q3C, Q5A, Q5C, and Q6B). The following sections
describe additional information unique to PEPs that should be
provided in NDAs or Drug Master Files.
For the
starting material used in the manufacturing process, information
on animal species, tissue types, and countries of origin should
be provided. Animals used should have been raised with the
intent for use as human food. The source documentation should
include animal origin, identification, and movement since birth;
maintenance animal medical records; surveillance of herds; and
documentation of feeds. Feeds should not contain any
reprocessed animal products.
The
manufacturing process should be validated for its capability to
remove and/or inactivate viral agents as recommended in ICH Q5A.
A full viral risk assessment should be made and justified.
The
drug substance should be adequately characterized using
appropriate chemical, physical, and biological testing.
Batch-to-batch consistency with respect to chemical identity,
biological activity of different classes of enzymes including
specific activity, identity, and purity level should be
demonstrated. Purity can be evaluated by enzyme-specific
activity. Identity can be demonstrated by fingerprint analysis,
using (but not limited to) the following methods:
·
Chromatography
(e.g., ion-exchange or reversed phase high-pressure liquid
chromatography (HPLC))
·
Sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)
·
Isoelectric
focusing (IEF)
·
New analytical
technologies, when appropriate
Nondrug
substance-related impurities such as process-related impurities,
inorganic impurities, and residual solvents should be
controlled. Estimates of the amounts of inactive cell
components and product-related impurities including inactive
enzymes present in the drug substance should be made.
Specifications for the drug substance should include tests for
identity, biological activity of different classes of enzymes,
purity, and other relevant attributes. Appropriate acceptance
factors (e.g., limits and ranges) should be established and
justified for lipases, amylases, and proteases.
Specifications for the drug product should include tests for
identity, biological activity of different classes of enzymes,
degradants, dissolution, and other relevant attributes.
Pancreatic enzymes from natural sources are a mixture of
lipases, amylases, and proteases, which are present in varying
proportions. Appropriate acceptance criteria should be
established and justified for these enzymes. However, for
purposes of labeling, product potency should be expressed as
lipase activity. When a novel or
non-novel,
but noncompendial,
excipient is included in the formulation of the drug product,
manufacturing and control information on the excipient should be
provided. Refer to related sections in ICH Q6B.
Due to
the inherent lability that has been observed with PEPs,
stability data through 12 months at the recommended storage
temperature as well as 3 months of accelerated stability data
should be provided.
Additional stability data can be submitted as an amendment
during the review process, and an expiration date will be
determined based on the review of the stability data in the NDA.
Primary
stability data should be generated according to the guidance
developed in ICH Q1A and Q5C.
Primary stability studies should be performed with batches that
are formulated to be released at 100 percent of the
label-claimed potency for lipase.
Existing stability data not obtained under ICH conditions can be
submitted as supporting data.
Since
high doses of pancreatic enzymes have been associated with
safety problems (see 69 FR 23411), the finished product should
be formulated to 100 percent of the label-claimed lipase enzyme
activity. With suitable justification (e.g., manufacturing
losses), however, overages may be acceptable. Amylase and
protease activity in the formulation should remain within
justified limits.
An
appropriate in vitro release test method should be developed.
To be approved, an NDA must meet the
requirements described in 21 CFR 314.50 for nonclinical and
toxicology data.
A. Toxicology
No
toxicology studies are needed if excipients are classified as
GRAS for oral administration or are USP/NF compendial excipients
and are present at levels previously found acceptable.
If the excipients are not classified as GRAS or have not been
previously approved for the same route of administration,
amount, or therapeutic use, safety should be established through
toxicology studies. For new excipients without previous
clinical data, clinical trials of the drug product containing
the new excipients should also be performed. If the new
excipients are classified as GRAS but are present in quantities
in excess of the allowed levels, their safety should be
established at the higher levels through toxicological studies
of the excipients or the drug product containing the higher
levels of the excipients. To determine their safety, the
toxicology program for new excipients or for excipients with
higher levels than listed for GRAS should supply data from
long-term studies in both rodent and nonrodent mammalian species
plus standard reproductive toxicity and genotoxicity information
(Steinberg et al. 1996).
Information from published reports of toxicology studies should
also be included in the NDA.
Because of the extensive use of the
currently marketed PEP products, no new pharmacology studies for
such products are necessary. The FDA recommends that applicants
summarize the published literature about the pharmacology of
their particular PEP and submit this summary with the
bibliography as part of a 505(b)(2) application. In addition,
we encourage submission of all available nonclinical information
including any pharmacological data generated with the drug
substance and/or drug product.
To be approved, an NDA must meet the
requirements in 21 CFR 314.50 for human pharmacokinetic and
bioavailability information. The bioactivity and/or
bioavailability of the active ingredients should be determined
at the site of action (gastrointestinal tract). The lipase,
amylase, and protease activities should be determined from
aspirates from the stomach and duodenum. The data should be
obtained under fasting conditions as well as after a standard
meal stimulation.
The use of any inactive ingredient in the
formulation to prevent or minimize the hydrolysis of the enzymes
in the stomach should be supported with in vitro and/or in vivo
release data. An appropriate in vitro release test method
should be developed.
To be approved, a PEP NDA must meet the
requirements for clinical studies described in 21 CFR 314.50.
The Agency has determined that there is a considerable body of
evidence that replacement of pancreatic enzymes has clinical
benefit for patients with cystic fibrosis and chronic
pancreatitis (69 FR 23410). This section summarizes general
approaches to the design of clinical studies intended to provide
such evidence of effectiveness and safety in support of an NDA
for currently marketed PEPs of animal origin. The discussion
includes guidance on patient populations that should be studied,
endpoints (outcome measures) to evaluate efficacy and safety,
and suggestions for the design of clinical studies.
Currently marketed PEPs differ in their
composition, enzymatic activities, formulation, method of
manufacture, stringency of quality control during manufacturing,
stability, and bioavailability (i.e., bioactivity in the small
intestine). These differences have led to highly variable PEP
quality and therapeutic performance among manufacturers. For
any given manufacturer, such differences over time can lead to
batch-to-batch inconsistency and to unacceptable variability in
PEP quality and therapeutic performance. With improvements in
quality as outlined in the guidance, therapeutic performance may
be better predicted from in vitro studies or from in situ
measurements of PEP bioactivity in the small intestine.
For NDA approval of any particular PEP,
clinical studies should demonstrate a relationship between the
extent of clinical benefit and the amount of PEP administered
(e.g., empirical demonstration of dose-response relationships in
clinical trials).
NDAs filed under section 505(b)(2) of the
Act may include published articles along with a bibliography of
clinical trials in lieu of clinical data.
Two distinct populations have the largest
clinical need in practice for PEPs: (1) pediatric and adult
patients with cystic fibrosis; and (2) adult patients with
chronic pancreatitis. Both conditions can cause pancreatic
insufficiency and maldigestion, leading to malabsorption of
dietary nutrients and subsequent malnutrition. Different
dosages of PEPs may be recommended to treat these two
populations. At a minimum, because cystic fibrosis is
primarily a pediatric disease, the efficacy studies in the NDA
should include clinical studies in pediatric patients with
cystic fibrosis.
Although demonstrating a beneficial effect
on clinical outcomes is desirable in clinical trials (e.g.,
weight gain or nutritional status), efficacy can also be
demonstrated by showing a meaningful beneficial effect on
appropriate pharmacodynamic measures such as steatorrhea. Some
examples are provided here:
·
Demonstration that administration of the PEP to
patients with exocrine pancreatic insufficiency causes a
meaningful decrease in stool fat as evaluated in a 72-hour
quantitative stool collection
·
Demonstration that administration of the PEP to
patients with exocrine pancreatic insufficiency causes
significantly more responders than in a comparison group (e.g.,
stool fat originally higher than 14 g/day decreased to less than
7 g/day)
·
Demonstration that administration of the PEP to
patients with exocrine pancreatic insufficiency causes
significantly fewer patients to withdraw from blinded therapy
because of steatorrhea than in a comparison group
·
Other quantitative endpoints can be considered
Safety variables that should be assessed in
clinical trials with PEPs include symptoms and signs of
malabsorption, such as manifestations of steatorrhea; complaints
of bloating; flatus; abdominal pain; loose and frequent stools;
overt diarrhea; blood in the stool; and uric acid elevations.
With regard to safety, we note that the
etiology of fibrosing colonopathy has not been completely
elucidated. In an effort to minimize development of fibrosing
colonopathy that has been assumed to be related to high doses of
PEPS, the FDA, in conjunction with the Cystic Fibrosis
Foundation (CFF), recommends a starting dose of 500 to 1,000
lipase units /Kg/meal with titration to less than 2,500
units/Kg/meal or less than 4,000 lipase units/g fat/day (FitzSimmons
et al., 1997; Borowitz et al., 2002). Doses in excess of 2,500
USP lipase units/Kg/meal should be used with caution and only if
their benefit is documented by 3-day fecal fat. Doses in excess
of 6,000 USP lipase units/Kg/meal have been associated with
fibrosing colonopathy. This dosing recommendation, applicable
to any formulation, was made on the basis of concern over
dose-related colonic strictures in cystic fibrosis and the
likelihood that maximal efficacy is achieved at the recommended
ceiling dose of 2,500 USP lipase units/Kg/meal.
The clinical
studies confirming efficacy of the specific PEP can be: (1)
parallel; (2) randomized withdrawal; or (3) crossover designs.
The designs of these studies for PEP products are discussed
below. Other designs, such as those in which patients are
challenged with increases in dietary fat, can also be
considered.
The clinical studies confirming efficacy of
the specific PEP should include appropriate controls, such as
dose-comparison controls, or active treatment controls. Placebo
may be appropriate with a rescue protocol to protect patients.
As noted in the following sections, if a placebo is not used
(such as in a comparison of two doses of a PEP, or in a
comparison of one PEP with another (e.g., an active control)),
differences between treatments should be demonstrated to help
interpret results. If desired, the efficacy and dose response
of the PEP can be demonstrated in the same study.
Duration of the entire trial could be days
to 2 to 3 weeks, depending on the design chosen. Blinding and
randomization are recommended to reduce bias. Diets may need to
be standardized. The total number of patients in the study can
be between 10 and 25, depending on study design. Either two
studies or one adequate and well-controlled clinical
investigation and confirmatory evidence may be appropriate.
Parallel studies can be used to demonstrate
efficacy of a PEP, such as when the effects of the PEP are
compared to other doses of a PEP and/or to another active
product (such as another PEP) or placebo.
A randomized withdrawal study should have
two phases: a run-in phase and a randomized withdrawal phase.
In the run-in phase, patients should be administered the PEP
under study and the dose should be adjusted (e.g., titrated) to
achieve and stabilize at the desired clinical outcome (e.g.,
control of stool fat excretion). An open-label design is
appropriate for this phase. In the next phase (the withdrawal
phase), patients who have apparently responded to the PEP should
then be randomized in a double-blind fashion to either continued
treatment with the PEP or, as is typical, to placebo. At the
end of the withdrawal phase the effects of the two treatments
should be compared. For example, the primary efficacy endpoint
could be a quantitative measure of stool fat over 72 hours
(e.g., the mean change in stool fat or the number of
nonresponders who have recurrent steatorrhea). In some cases at
the outset of the randomized withdrawal period, it may be
desirable to discontinue treatment gradually to avoid sudden
onset of symptoms of pancreatic insufficiency.
Patients should be monitored even during
the withdrawal phase to allow discontinuation from randomized
study treatment if clinically appropriate (e.g., for clinically
worrisome diarrhea). Patients who discontinue study treatment
can then be given appropriate medical therapies. If
prespecified in the protocol, a count of these treatment
failures (nonresponders) can be incorporated into the primary
efficacy analysis. In such cases, the protocol should define
specific discontinuation criteria for patients who fail
treatment.
A randomized withdrawal design also can be
adapted to incorporate a dose-response evaluation of a PEP. At
the outset of the withdrawal phase, for example, patients can be
randomized to placebo and to two or more dosage levels of a
PEP. The response of patients at the different dosage levels
(including placebo) can then be compared. Although inclusion of
a placebo arm is often the most usual and straightforward way of
demonstrating efficacy, this arm can sometimes be excluded.
In a crossover study, each patient in the
study is treated with all or most of the treatments under
investigation, usually in a randomized sequence.
A crossover study allows for a paired
statistical analysis of the data (i.e., each patient serves as
his or her own control), thereby decreasing the effects of
interpatient variability, which otherwise might obscure true
drug effects. In general, fewer patients are needed to perform
a crossover study than a parallel study. However, because each
patient is administered several treatments, each patient’s study
involvement is longer than in a parallel study. Moreover,
sponsors are strongly cautioned that if baseline conditions are
not reestablished between treatment periods, or if treatment in
one period carries over into the subsequent period or periods,
the results likely will not be interpretable using a paired
statistical analysis. Although data from the first period could
still be analyzed as in a parallel study (unpaired statistical
analysis), the main advantage of using a crossover design would
have been lost.
In a randomized, two-period,
placebo-controlled, crossover study of a PEP, for example,
patients should first be stabilized on existing therapy to
establish baseline conditions. Patients should then be
randomized to receive one of two treatment sequences:
placebo-PEP versus PEP-placebo. If quantitative determination
of stool fat is used as the primary endpoint, each period should
last at least 72 hours to allow for adequate collection of stool
specimens. Reestablishment of baseline conditions should be
documented between periods.
A significant portion of the target
population for PEPs includes pediatric patients with cystic
fibrosis, a congenital genetic disease in which there is chronic
exocrine pancreatic insufficiency dating from birth. These
patients represent the majority of pediatric patients with
exocrine pancreatic insufficiency. At the time of publication
of this guidance, the only PEP approved for use in pediatric
cystic fibrosis patients is an immediate-release formulation,
and that product is not currently marketed.
To comply with the Pediatric Research
Equity Act of 2003 (PREA) (21 U.S.C. 355c), the application must
contain data that are adequate to assess the safety and
effectiveness of the PEP for the claimed indications in each of
the appropriate pediatric subgroups (newborns, infants,
children, and adolescents). The data should be adequate to
support dosing and administration in each pediatric
subpopulation for which the drug has been assessed to be safe
and effective. Studies may not be needed in each pediatric age
group, if data from one age group can be extrapolated to
another. Whether or not pediatric studies in more than one age
group are necessary depends on expected therapeutic benefit and
use in each age group, and on whether safety and effectiveness
data from one age group can be extrapolated to other age
groups. As with the use of adult data, the extrapolation can be
supplemented with data to define dosing and safety for the
relevant age groups. Because solid dosage forms of PEPs cannot
be swallowed by young pediatric patients (i.e., generally six
years of age or younger), under PREA, sponsors must attempt to
develop age-appropriate formulations for this patient
population.
Beverley, DW, J Kelleher, A MacDonald, JM Littlewood, T Robinson,
and MP Walters, 1987, Comparison of Four Pancreatic Extracts in
Cystic Fibrosis, Archives of Disease in Childhood,
62:564-568.
Briars, GL, DM Griffiths, IE Moore, PH Williams, K Johnson, and CJ
Rolles, 1994, Letter to the Editor, Lancet, 343:600.
Campbell, CA, J Forrest, and C Musgrove, 1994, Letter to the
Editor, Lancet, 343:109.
Cystic Fibrosis Foundation, 1993, Results of a Survey of 114
Cystic Fibrosis Care Centers in United States, Patient Registry
1992 Annual Data Report, Bethesda, MD, October 1993, in OTC Vol.
17BFR, Docket No. 79N-0379, Division of Dockets Management.
Dutta, SK, VS Hubbard, and M Appler, 1988, Critical Examination of
Therapeutic Efficacy of a pH-Sensitive Enteric-Coated Pancreatic
Enzyme Preparation in Treatment of Exocrine Pancreatic
Insufficiency Secondary to Cystic Fibrosis, Digestive Diseases
and Sciences, 33:1237-1244.
Fatmi, AA and JA Johnson, 1988, An In Vitro Comparative Evaluation
of Pancreatic Enzyme Preparations, Drug Development and
Industrial Pharmacy, 14:1429-1438.
FitzSimmons, SC, GA Burkhart, D Borowitz et al., 1997, High-Dose
Pancreatic Enzyme Supplements and Fibrosing Colonopathy in
Children with Cystic Fibrosis, New England Journal of Medicine,
336:1283-1289.
Graham, DY, 1979, An Enteric-Coated Pancreatic Enzyme Preparation
that Works, Digestive Diseases and Sciences, 24:906-909.
Graham, DY, 1977, Enzyme Replacement Therapy of Exocrine
Pancreatic Insufficiency in Man: Relation Between In Vitro Enzyme
Activities and In Vivo Potency in Commercial Pancreatic Extracts,
New England Journal of Medicine, 296:1314-1317.
Hendeles, L, A Dorf, A Stecenko, and M Weinberger, 1990, Treatment
Failure After Substitution of Generic Pancrelipase Capsules:
Correlation with In Vitro Lipase Activity, Journal of the
American Medical Association, 263:2459-2461.
Knabe, N, M Zak, A Hansen, J Moesgaard, N Kvist et al., 1994,
Letter to the Editor, Lancet, 343:1230.
Littlewood, JM, J Kelleher, MP Walters, and AW Johnson, 1988, In
Vivo and In Vitro Studies of Microsphere Pancreatic Supplements,
Journal of Pediatric Gastroenterology and Nutrition, 7
(Supplement 1):S22-S29.
Mischler, EH, S Parrell, PM Farrell, and GB Odell, 1982,
Comparison of Effectiveness of Pancreatic Enzyme Preparations in
Cystic Fibrosis, American Journal of Diseases of Children,
136:1060-1063.
Oades, PJ, A Bush, PS Ong, and RJ Brereton, 1994, Letter to the
Editor, Lancet, 343:109.
Regan, PT, JR Malagelada, EP DiMagno, SL Glanzman, and VL Go,
1977, Comparative Effects of Antacids, Cimetidine and Enteric
Coating on the Therapeutic Response to Oral Enzymes in Severe
Pancreatic Insufficiency, New England Journal of Medicine,
297:854-858.
Smyth, RL, D van Velzen, A Smyth, DA Lloyd, and DP Heaf, 1994,
Strictures of Ascending Colon in Cystic Fibrosis and High-Strength
Pancreatic Enzymes, Lancet, 343:85-86.