The Safety and Regulation of Natural Products Used as Foods and Food Ingredients

Article (PDF Available)inToxicological Sciences 123(2):333-48 · August 2011with98 Reads
DOI: 10.1093/toxsci/kfr198 · Source: PubMed
The use of botanicals and dietary supplements derived from natural substances as an adjunct to an improved quality of life or for their purported medical benefits has become increasingly common in the United States. This review addresses the safety assessment and regulation of food products containing these substances by the U.S. Food and Drug Administration (FDA). The issue of safety is particularly critical given how little information is available on the toxicity of some of these products. The first section uses case studies for stevia and green tea extracts as examples of how FDA evaluates the safety of botanical and herbal products submitted for consideration as Generally Recognized as Safe under the Federal Food, Drug, and Cosmetics Act. The 1994 Dietary Supplement Health Education Act (DSHEA) created a regulatory framework for dietary supplements. The article also discusses the regulation of this class of dietary supplements under DSHEA and addresses the FDA experience in analyzing the safety of natural ingredients described in pre-market safety submissions. Lastly, we discuss an ongoing interagency collaboration to conduct safety testing of nominated dietary supplements.


TOXICOLOGICAL SCIENCES 123(2), 333–348 (2011)
Advance Access publication August 5, 2011
The Safety and Regulation of Natural Products Used as Foods and Food
Ali Abdel-Rahman,*
Njwen Anyangwe,*
Louis Carlacci,*
Steve Casper,*
Rebecca P. Danam,
Evaristus Enongene,*
Gladys Erives,
Daniel Fabricant,*
Ramadevi Gudi,*
Corey J. Hilmas,*
Fred Hines,*
Paul Howard,
Dan Levy,*
Ying Lin,*
Robert J. Moore,*
Erika Pfeiler,§
T. Scott Thurmond,
Saleh Turujman,*
and Nigel J. Walker{
*Division of Dietary Supplement Programs, Office of Nutrition, Labeling, and Dietary Supplements and Office of Food Additive Safety, Center for Food Safety
and Applied Nutrition, U.S. Food and Drug Administration, College Park, Maryland 20740; Office of Scientific Coordination, National Center for
Toxicological Research, U.S. Food and Drug Administration, Jefferson, Arkansas 72079; §Office of the Commissioner, Office of the Chief Scientist, U.S. Food
and Drug Administration, Laurel, Maryland 20708; and {National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle
Park, North Carolina 27709
These authors contributed equally to the preparation of this article.
To whom correspondence should be addressed at Office of Food Additive Safety, U.S. Food and Drug Administration, 5100 Paint Branch Parkway, HFS-265,
College Park, MD 21029. Fax: (301) 436-2972. E-mail:
Received May 13, 2011; accepted July 17, 2011
The use of botanicals and dietary supplements derived from
natural substances as an adjunct to an improved quality of life
or for their purported medical benefits has become increa singly
common in the United States. This review addresses the safety
assessment and regulation of foo d products containing these
substances by the U.S. Food and Drug Admin istr ation (FDA).
Theissueofsafetyisparticularly critical given how little
information is available on the toxicity of some of these
products. The first section uses case studies for stevia and green
tea extracts as examples of how FDA evaluates the safety of
botanical and herbal produ cts submitted for consideration as
Generally Recognized as Safe under the Federal Food , Drug,
and Cosmetics Act. The 1994 Dietary Supplement Health
Education Act (DSHEA) created a regulatory framework for
dietary supplements. The article also discusses the regulation of
this class of dietary s upplements under DSHEA and addresses
the FDA experience in analyzing the safety of natural
ingredients described in pre-market safety submissions. Las tly,
we discuss an ongoing interagency collaboration to conduct
safety testing of nominated dietary supplements.
Key Words: botanicals; dietary supplements; food; food
ingredients; safety; regulation.
The art of healing comes from nature, not from the physician.
Therefore the physician must start from nature, with an open mind.
Attributed to Phillip von Hohenheim (aka Paracelsus),1493–1541.
Poison is in everything, and nothing is without poison. The
dosage makes it either a poison or a remedy. Attributed to
Phillip von Hohenheim (aka Paracelsus), 1493–1541.
A large percentage of the U.S. population uses a botanical or
nutritional or dietary supplement on a daily basis as an adjunct to
an improved quality of life or for their alleged medical benefits.
The regulation of botanicals by the U.S. Food and Drug
Administration (FDA) is governed by the provisions of the
Federal Food Drug and Cosmetic Act (FD&C Act). How FDA
regulates the use of a substance is determined by its intended
conditions of use. Products intended to diagnose, mitigate, treat,
cure, or prevent disease are regulated by FDA as drugs. Drugs
derived from botanical sources are outside the scope of this article.
FDA’s Center for Food Safety and Applied Nutrition (CFSAN) is
responsible for regulating food ingredients and ensuring that those
ingredients derived from botanical and other sources are safe and
lawful. In 1994, the Dietary Supplement Health and Education
Act (DSHEA) created a regulatory framework for dietary
supplements that included provisions establishing current good
manufacturing procedures, mechanisms for pre-market safety
notifications for new ingredients, and a mechanism for establish-
ing claims used in product labeling.
This article discusses botanical or other naturally derived
product regulatory and/or safety assessment with emphasis on: (1)
the regulation and safety assessment of botanicals and herbals
submitted for consideration as food ingredients under the
Generally Recognized as Safe (GRAS) provisions of the FD&C
Act. This section uses case studies to illustrate the safety
assessment process for products submitted to FDA under this
program, (2) the regulation of these products as used under dietary
supplement provisions of the FD&C Act. This section also
presents a general framework for assessing the safety of these
Published by Oxford University Press on behalf of the Society of Toxicology 2011.
at FDA Library on September 27, 2011toxsci.oxfordjournals.orgDownloaded from
products when they are new dietary ingredients (NDIs), and (3) an
ongoing interagency collaborative effort to conduct safety
assessment studies of nominated naturalproductsusedasdietary
HERBAL SOURCES (Rebecca P. Danam and Gladys Erives)
Currently, there is an increased global interest in the use of
botanicals or botanical-derived products as food ingredients
because of the belief that they may be beneficial to health.
Also, such products are commonly referred to as ‘natural’
because their source is found in nature. An ingredient derived
from a botanical/herbal source might require extensive
purification for use in food or the botanical/herbal source
itself might be the product added to food.
Through the FD&C Act, FDA is responsible for overseeing the
safety of food ingredients (Under the FD&C Act, the legal status
of a substance is dependent on whether it is offered for use in
a conventional food or its use is as a dietary supplement or as
a dietary ingredient in a dietary supplement product. This section
of the article discusses only the use of a substance as an ingredient
in conventional foods). The FD&C Act defines the term ‘food
additive’ to refer to any substance the intended use of which
results in it becoming a component of food. The use of a food
additive requires pre-market review and approval by FDA that
results in a regulation prescribing safe conditions of use for the
food additive. An exemption to the food additive definition, and
as such, to the pre-market review and approval by FDA is for
those substances that are generally recognized, among qualified
experts, as having been adequately shown to be safe under their
intended conditions of use (GRAS). The safety standard,
‘reasonable certainty of no harm’ under the conditions of its
intended use, is the same for all substances added to food as
described under section 170.3(i) in Title 21 of the Code of Federal
Regulations (CFR) (21 CFR 170.3(i)). Therefore, a GRAS
substance is distinguished from a food additive based on the
common knowledge about the safety of the substance for its
intended use. GRAS status for a food ingredient can be achieved
either based on the history of common use in foods before the
year 1958 (21 CFR 170.30(c)(1) and 170.3(f)) or through
scientific procedures (21 CFR 170.30(b) and 170.3(h)).
The basis for GRAS determination through experience based
on common use in foods is rarely relied upon because it requires
documentation that, prior to 1958, there was a substantial history
of consumption for food use by a significant number of
consumers using the same substance under similar conditions of
intended use. For a substance to be GRAS through scientific
procedures, the scientific data about the safety must be widely
available and there must be consensus among qualified experts
that the scientific data establish the substance to be safe under
the conditions of its intended use. GRAS status does not require
FDA’s determination that the intended use of a substance is
GRAS. Persons making an independent GRAS determination
for a food ingredient can choose to inform the FDA of their
determination that the use of a substance is GRAS and seek the
agency’s review of the company’s basis for this determination
by submitting a GRAS notice to FDA under the GRAS
Notification Program (USFDA 1997).
Since the start of the GRAS Notification Program in 1997, more
than 350 GRAS notices have been submitted. Of these,
approximately 120 GRAS notices were for food ingredients that
either are botanicals or derived from botanicals (CFSAN, FDA,
2011). Food ingredients from botanicals or botanical extracts
present unique challenges for safety evaluation as they are complex
mixtures of numerous chemical substances that exhibit composi-
tional complexity and variation depending on several factors.
For the safety review of botanical substances, FDA considers
the taxonomic identity of the source, the chemical identity, purity,
and stability of the substance in question, its relationship to the
article of commerce, description of the intended use, and
consequent exposure. It is very important that the identity of the
starting plant material is authenticated and standardized to ensure
consistent quality of the desired final product. Safety studies
should be conducted with the end product/substance that is
substantially similar, if not identical to the article of commerce.
Variations with respect to harvesting, storing, and processing
methods can lead to compositional differences or contaminations
and give rise to conflicting data rendering some test results
irrelevant. Hence, validated methods are critical for the
evaluation of pharmacological, toxicological, and clinical studies
of botanical ingredients. Safety studies may need to be
conducted on a single component of the mixture if toxicity is
either known or suspected about a specific component. FDA
considers on a case-by-case basis, different types of toxicolog-
ical studies, including short- and long-term toxicity studies,
metabolism and pharmacokinetic studies, reproductive and
developmental toxicity studies, mutagenicity/carcinogenicity
studies, immunotoxicity, and neurotoxicity studies to assess
safety. Under the FD&C Act, only safety is considered when
evaluating a substance added to food. The purported benefits of
botanical/herbal ingredients are not considered in the safety
assessment. FDA may consider such benefits or health claims
in a separate notification process.
Herein, we discuss the different factors that affect the safety
assessment of substances derived from botanicals either as highly
purified single components or as ‘extracts containing several
components. As specific examples, we discuss two cases
involving botanicals. The two cases, Stevia rebaudiana Bertoni
(hereafter referred to as S. rebaudiana or stevia) and Camellia
sinensis (hereafter referred to as C. sinensis or green tea extract),
illustrate the various and complicated aspects involved in the
process of the safety evaluation.
Stevia rebaudiana leaves contain sweetening diterpene
glycosides, known as steviol glycosides (SGs), which constitute
at FDA Library on September 27, 2011toxsci.oxfordjournals.orgDownloaded from
4–20% of the dry leaf weight. The leaf extracts are complex
mixtures comprised of SGs (stevioside, rebaudioside A,
rebaudioside B, rebaudioside C, rebaudioside D, rebaudioside
E, rebaudioside F, dulcoside A, rubusoside, steviolbioside),
labdane-type diterpenes, triterpenoids, steroids, volatile oils, and
flavonoids (Kinghorn, 2002). FDA does not permit the use of
crude whole-leaf extracts of stevia or uncharacterized stevia-
derived compounds in conventional foods due to insufficient
information to support its safety for such use and due to reports
in the literature that raised safety concerns.
Initial toxicological studies in rats administered aqueous,
crude, or partially purified stevia leaf extracts reported adverse
reproductive (Planas and Kuc, 1968), renal, and cardiovascular
(Melis, 1995, 1996) effects. For example, Planas and Kuc (1968)
reported that female rats fed an aqueous leaf extract had reduced
fertility. This finding was also confirmed in studies with female
mice (Nun˜es and Pereira, 1988). In addition, Melis (1999)
reported that the crude leaf extract decreased fertility in the male
rat. Traditionally, Guarani Indians in Paraguay consumed the
decoction of leaf extracts as an oral contraceptive (Kinghorn,
2002). These reports and other studies on its toxicity raised
safety concerns about the use of stevia or stevia-derived
substances. Recently, studies have been conducted with purified
preparations of SGs, stevioside, rebaudioside A, or their
degradation product steviol to address these safety issues.
Rebaudioside A and stevioside, the principal sweetening SGs
of S. rebaudiana, were isolated and purified from the leaves, with
95% minimum purity (with respect to the total SGs content).
Using these high-purity SGs, extensive toxicological in vivo
studies were conducted in animal models as well as in humans. In
particular, two subchronic toxicity studies of purified rebaudioside
Ainrats(Curry and Roberts, 2008; Nikiforov and Eapen, 2008),
pharmacokinetic and metabolism studies in both rats (Roberts and
Renwick, 2008) and humans (Wheeler et al., 2008), and a two-
generation reproductive/developmental toxicity study conducted
with rebaudioside A in rats (Curry et al.,2008)demonstratedlack
of pharmacological activity and reproductive toxicity. Chronic
toxicity studies on purified stevioside or rebaudioside A showed
no adverse effects (Curry and Roberts, 2008; Toyoda et al.,1997).
In addition, clinical studies carried out in human volunteers to
evaluate the potential pharmacological effects of rebaudioside A
and stevioside on blood pressure and blood glucose levels
indicated no adverse effects (Barriocanal et al.,2008; Hsieh et al.,
2003; Maki et al., 2008) in contrast to the previous studies
conducted with crude leaf extracts or less pure SG preparations.
The results of these recent toxicological studies, together with the
clinical studies in humans, support the safety of high-purity
( 95%) rebaudioside A, stevioside, or SG mixtures. The Joint
FAO/WHO Expert Committee on Food Additives in 2008
reviewed the recent scientific data on SGs and established an
acceptable daily intake for SGs as 0–4 mg/kg body weight
(expressed as steviol equivalents).
Given these recent developments, FDA has received several
GRAS notices since 2008 for high-purity rebaudioside A,
stevioside, or SG mixtures containing primarily rebaudioside A
and stevioside. Based on the substantive toxicological data on
high-purity SGs that is now publicly available, FDA had no
questions about the notifier’s determinations that these
substances are GRAS for use as a sweetener in food. For the
reasons described above, S. rebaudiana whole-leaf extracts or
partially purified extracts containing low-purity SGs are not
considered safe for use as food ingredients and FDA has an
import alert in effect to exclude them from such use in U.S.
Green Tea Extract
Tea is the most commonly consumed beverage in the world,
second only to water. Green, black, and oolong tea are all
derived from the leaves of the C. sinensis plant. Tea has a long
history of consumption as a beverage (hot water infusion) since
ancient times. More recently, green tea extracts enriched with
certain components are being consumed. They are also
available as dietary supplements in the form of tablets and
Tea is a complex mixture comprised of several constituents.
It is an abundant source of polyphenols, especially the complex
group of compounds called flavonoids. Catechins (flavan-3-
ols) are the major flavonoid compounds and contribute up to
30% of the dry weight of the tea leaf. The four major catechins
in green tea are epigallocatechin-3-gallate (EGCG), epigallo-
catechin (EGC), epicatechin-3-gallate (ECG), and epicatechin
(EC). Their epimers include catechin (C), catechin-3-gallate
(CG), gallocatechin (GC), and gallocatechin-3-gallate (GCG).
Green tea also contains gallic acid, chlorogenic acid, alkaloids,
namely theophylline, theobromine, and caffeine, volatile
compounds, minerals, and a unique amino acid,
The most abundant catechin, EGCG, is considered the most
bioactive and the most studied.
The composition of green tea varies with growing
conditions, horticultural practices, harvesting, and processing
conditions. It is also influenced by environmental factors such
as climate, season, the variety and age of the plant, as well as
how the leaves are processed (Cabrera et al., 2006).
Recently, green tea catechins (GTC) have garnered consider-
able attention in the popular press for their claimed health
benefits such as ‘‘weight loss, reduction in body fat, maintenance
of normal blood glucose levels and antioxidative properties.’
Based on these purported beneficial effects, the marketing of
concentrated green tea extracts and/or its components, particu-
larly EGCG has been on the rise. Numerous studies in animal
models using green tea extracts that contain catechins at various
concentrations have attributed antioxidative, thermogenic, anti-
carcinogenic, and anti-inflammatory properties to these GTC
(Chacko et al., 2010). In spite of these reported favorable effects,
conflicting results have been reported from epidemiological
studies and cancer intervention trials (Boehm et al., 2009).
Moreover, hepatic and intestinal toxicities associated with the
at FDA Library on September 27, 2011toxsci.oxfordjournals.orgDownloaded from
consumption of high doses of GTC preparations were
reported in animal studies (Galati et al., 2006; Isbrucker
et al.,2005). EGCG appears to act as an antioxidant as well as
a pro-oxidant agent. The cytotoxic effects of EGCG and tea
extracts to cancer cells have been attributed to their pro-
oxidant effects (Weisburg et al., 2004). EGCG-mediated
mitochondrial toxicity and reactive oxygen species formation
were implicated as the possible mechanism for the cytotox-
icity to isolated rat hepatocytes and hepatotoxicity in mice
(Galati et al., 2006). Further, recent reviews of published case
reports suggests a possible causal relationship between
concentrated green tea extracts, particularly EGCG, and liver
damage. A causative role has been suggested because the
symptoms resolved upon cessation of consumption and
reinjury were observed following rechallenge with the same
tea-derived preparations. It is intriguing that the liver damage
occurred mostly in women, suggesting gender differences in
susceptibility (Mazzanti et al., 2009; Sarma et al., 2008). It is
also reported that consumption of high doses of green tea
could alter thyroid function adversely (Chandra and De,
Bioavailability is a factor positively associated with the severity
of the toxicity of these catechins. For example, fasting increases
the bioavailability of EGCG (Chow et al., 2005; Isbrucker et al.,
2005) leading to increased severity of adverse effects when
consumed on an empty stomach. Also, the risk of adverse effects
is likely increased by factors that increase the bioavailability of
these flavanols such as the genetic polymorphisms in metaboliz-
ing enzymes (Miller et al., 2010) and herb-drug interactions. In
light of the accumulating scientific evidence, it is becoming
increasingly apparent that these catechins may have deleterious
effects at pharmacological concentrations and in certain sensitive
populations. Further research is necessary to better understand the
nature and mechanism of action of the bioactive components, the
differences in their bioavailability in humans and the safe range of
The two botanicals discussed above exemplify the complexity
of the toxicological analysis of botanicals and the contrasting
nature of toxicity scenarios presented by food ingredients
derived from botanicals. In the case of S. rebaudiana, the whole-
leaf extract was shown to exhibit toxic effects, whereas its
purified constituents (SGs) were found to be safe for use as
a sweetener. The reverse scenario is observed with C. sinensis.
Green tea consumption has a long history of safe use whereas
the isolated, purified, and concentrated catechin components of
green tea appear to have adverse effects and currently are not
considered safe for food ingredient use.
The safety assessment of botanical substances is complicated
by various factors. Compositional diversity is a key factor
because botanicals are complex mixtures, for which the
identity of all the individual components is not known and the
proportion of individual components varies with the source.
There are various other factors, among which are the lack of
standardization of the botanical (in terms of both materials and
analytical methods), lack of identity of the active ingredients,
and the use of different formulations of the botanical in the
article of commerce when compared with the test substance.
The paucity of data on the toxicology of whole extracts or the
individual components of botanicals makes it very challenging
to determine the safety of botanical substances for use in
conventional food. As such, the review of botanical substances
for safe use in conventional food must be approached with
some skepticism, an open mind and utilization of the full
arsenal of scientific tools available to assess the safety of such
substances. There is no set formula for dealing with the safety
evaluation of such materials or combination of materials. Each
new submission must be dealt with on a case-by-case basis.
(Corey J. Hilmas, Ali Abdel-Rahman, Njwen Anyangwe, Louis
Carlacci, Steve Casper, Evaristus Enongene, Ramadevi Gudi,
Fred Hines, Dan Levy, Ying Lin, Robert J. Moore, Erika A. Pfeiler,
and Saleh Turujman, Daniel Fabricant)
The botanical and animal food sources we consume on a daily
basis as part of the human diet contain a complex array of
naturally occurring compounds with a wide range of chemical
structures and physiological effects. The study and understand-
ing of diet, its structural complexity, and its ability to both
enhance and impair health presents a great scientific challenge to
toxicologists, chemists, botanists, and microbiologists. The
dietary supplement industry is an ever-increasing global market.
In this section, the U.S. regulatory requirements for dietary
supplements will be reviewed, followed by a discussion of the
general framework for assessing the safety of botanical and
herbal products as NDIs. Finally, examples from FDA response
letters and landmark cases will highlight identity and safety
issues regarding dietary supplements.
Overview of Regulation of Dietary Supplements in the
United States
Dietary Supplement Use in the United States
Use of dietary supplements has increased over the past 20
years in the United States. Over one-half of the U.S. population
report regularly consuming dietary supplements during 2003–
2006 (Gahche et al., 2011). Data from the National Health and
Nutrition Examination Survey (NHANES) has been used to
monitor use of dietary supplements since the 1970s. NHANES
III data suggested that supplement use was highest among three
population groups: children between 1 and 5 years of age,
middle aged, and the elderly (Ervin et al., 1999). The most
recent analysis of NHANES (2003–2006) suggests that dietary
supplement use is widespread among U.S. adults aged 20
and over (Gahche et al., 2011). The percentage of the U.S.
at FDA Library on September 27, 2011toxsci.oxfordjournals.orgDownloaded from
population who used at least one dietary supplement increased
from 42% in 1988–1994 to 53% in 2003–2006 (Gahche et al.,
2011). In particular, botanical products, more commonly used
in older than younger age groups, are reportedly used by
approximately 20% of all adults (Bailey et al., 2011).
Between 1994 and 2000, dietary supplement sales increased
80% to near $16 billion annually (Blendon et al., 2001).
Dietary supplement sales exceeded $25 billion in 2008, and
a recent report indicates that American consumers spent an
estimated $27 billion on dietary supplements in 2009 (Nutrition
Business Journal, 2010). In 2009, 17 million U.S. adults were
regular or heavy users of herbs or botanicals (Nutrition
Business Journal, 2010). The upward trend in use of dietary
supplements has been predicted to continue growing due to the
aging baby boom generation, increased interest in self-
sufficiency, and the rise in popularity of alternative botanical
and herbal products over their pharmaceutical counterparts. In
2009, FDA estimated 55,600 dietary supplement products on
the market (see Docket ID: FDA-2007-D-0209 [Document ID:
FDA-2007-D-0209-0023]). In FY 2010, the CFSAN devoted
34 full-time equivalents and $4.511 million for salaries and
operating costs to dietary supplement work.
Regulatory Requirements for Dietary Supplements in the
United States
The term ‘‘dietary supplement’’ is defined in section §201(ff)
of the Federal Food, Drug, and Cosmetic Act (the FD&C Act).
Under the statutory definition, dietary supplements are defined
as a food and must be one of several enumerated ‘dietary
ingredients’ such as a vitamin, mineral, or herb or other
botanical. In addition, the product must be intended for ingestion
and ‘not represented for use as a sole item of a meal or of the
For regulatory purposes in the United States, dietary supple-
ments are regulated as foods under the FD&C Act, Title 21
Sections 301–399 of the United States Code (U.S.C.). Similar
products may be defined and regulated in different ways in other
countries. For example, in Canada, ‘‘natural health products’ are
a regulatory category separate from foods. Prior to 1994, the
FD&C Act did not specifically define the term "dietary
supplements" or the scope of products that could be dietary
supplements. Through DSHEA, Congress amended the FD&C
Act to define the term ‘dietary supplements and establish
a regulatory framework for safety and claims for dietary
supplement products. DSHEA was intended to strike a balance
between increased consumer access to dietary supplements and
consumer protection against the toxicological health risks
associated with consumption. See the following sections of the
FD&C Act for the regulatory authority governing dietary
supplements: §301 [21 U.S.C. §331] (Prohibited Acts), §402
[21 U.S.C. §342] (Adulterated Food), §411 [21 U.S.C. §350]
(Vitamins and Minerals), and §413 [21 U.S.C. §350b] (New
Dietary Ingredients).
The following example illustrates how the definition of
a term like ‘ingestion,’ when used in a statute, is ultimately
defined in a legal context. Ener-B Nasal Gel was ‘a Vitamin B-
12 supplement in gel form designed to be applied to the inside
of the nose and absorbed into the blood stream through the
nasal mucosal membranes’ (72 F3d 285, 1995). In United
States versus Ten Cartons, the U.S. district court ‘determined
that Ener-B is not a ‘dietary supplement’ within the meaning of
21 U.S.C. Sec. 321(ff)’ because ‘ingestion’ means to take
into the stomach and gastrointestinal tract by means of enteral
administration through the mouth. ‘The district court con-
cluded that Ener-B’s method of intake precludes its classifica-
tion as a dietary supplement’ (see 888 F Supp at 392–95).
Additional Federal Legislation on Dietary Supplements
Passage of DSHEA was followed by the Dietary Supplement
and Nonprescription Drug Consumer Protection Act (Public Law
109-462) signed into law on 22 December 2006. This law
amends the FD&C Act with respect to reporting of serious
adverse events related to dietary supplements and nonprescrip-
tion drugs. The law has four major provisions, (1) requires the
collection of all adverse event reports by manufacturers,
distributors, and retailers of dietary supplements; (2) requires
the reporting of serious adverse event reports to the FDA; (3)
requires firms to maintain records of reports of all adverse events
and requires that FDA be allowed to inspect those records; and
(4) requires that dietary supplement labels bear information to
facilitate the reporting of serious adverse events associated with
the use of dietary supplements by consumers. The requirements
of this law became effective on 22 December 2007.
Although DSHEA amended the FD&C Act by adding
dietary supplement-specific requirements, the statutory defini-
tion contained express language categorizing dietary supple-
ments under the broader heading of foods in general.
Accordingly, the legal requirements that apply to the
manufacturing and marketing of dietary supplements are not
limited solely to those found in DSHEA. For example, dietary
supplement firms must register their physical facilities with the
FDA under the Public Health Security and Bioterrorism
Preparedness and Response Act of 2002 (Pub. L. 107-188)
and provide prior notice when importing dietary supplements
or dietary supplement ingredients. Dietary supplements must
also conform with labeling requirements imposed by DSHEA
as well as other broader labeling amendments such as the
Nutrition Labeling and Education Act of 1990 (Publ. L. 101-
535) and Food Allergen Labeling and Consumer Protection
Act of 2004 (Pub. L. 108-282).
Under the authority of the FD&C Act and the Public Health
Service Act (42 U.S.C § 201 et seq.), the FDA has imposed
a number of regulatory requirements that address quality and
safety of dietary supplements. Regulations prescribing Current
Good Manufacturing Practices require that all manufacturers
and distributors of dietary supplements have in place
at FDA Library on September 27, 2011toxsci.oxfordjournals.orgDownloaded from
procedures to ensure quality, potency, and identity of dietary
supplements (21 CFR 111). Manufacturers may also have to
comply with other manufacturing regulations if their products
fit within the scope of products subject to manufacturing
regulations for seafood HACCP (21 CFR 123), juice HACCP
(21 CFR 120), acidified foods (21 CFR 114), or low acid foods
(21 CFR 113).
Dietary Supplement Safet y
The safety of dietary supplements is addressed in §402
(Adulterated Food), which has provisions that apply to all
foods and provisions specific to dietary supplements. The
FD&C Act has adulteration standards specific to supplements
including ‘significant and unreasonable risk of illness or
injury’ although the FD&C Act makes clear that the
government bears the burden of proof in showing that
a supplement is adulterated using this standard. Supplements
are also adulterated if they are prepared, packed, or held under
conditions that do not meet current good manufacturing
practice regulations.
In addition, the DSHEA amended the FD&C Act to require
a pre-market safety notification for some ingredients. Products
containing NDIs are adulterated in the absence of a required
notification or if there is inadequate information to provide
reasonable assurance that such ingredient does not present
a significant or unreasonable risk of illness or injury. The
statute specifies that the notification, when required, should
include the ‘history of use or other evidence of safety that the
dietary ingredient, when used under the conditions of use
suggested or recommended in the labeling of the dietary
supplement will reasonably be expected to be safe ... (see
§413 of the FD&C Act).
General Framework for Assess ing the Safety of Natural
Products as New Dietary Ingredients
FDA has received over 410 NDI notifications containing
some form of botanical ingredient as of February 2011.
Botanical ingredients represent 61% of the entire NDI
notification portfolio, which contains over 700 NDIs since
1994. Figure 1 illustrates the various ingredient types found in
NDI notifications received by FDA to date and their
representation as a percent among the entire NDI notification
portfolio. This figure takes into account that NDIs often include
more than one category due to the presence of multiple
ingredients. The statute created a notification system whereby
FDA is required to respond to the notification, acknowledging
receipt, and after 90 days, posting the notification on the public
docket. FDA does review the notifications and frequently
comments on the adequacy of the basis for the safety of the
product. Figure 2 represents the types of letters, issued by FDA,
for NDIs derived from botanical products. Approximately 18%
of letters are acknowledgements without objection or comment.
The majority of FDA response letters provide comments on
various aspects of identity and safety. The category of ‘Other
represents letters where the filing date has been reset due to
receipt of a substantial amendment from the notifier because of
extra time required to evaluate the submission, as authorized by
the regulation that implements the statute.
Establishing NDI identity is the first step toward evaluating
safety of a dietary supplement product containing an NDI. If
the identity of the ingredient is described only in a general way,
it may become unclear how the NDI qualitatively and
quantitatively relates to the substances described in safety
evidence. Therefore, the relevance of the safety evidence as
a basis for a reasonable expectation of safety of the NDI under
the intended conditions of use becomes unclear. As a new
ingredient, the NDI is typically different from previously
consumed material, and relating the composition of the various
versions of an ingredient to one another is often the subject of
discussion in the notifications. The approaches to evaluating
botanical and live microbial product NDIs will be addressed in
the following section.
General Safety Considerations for Botanical Products
The statue refers to ‘history of use or other evidence of safety’
as the basis for the safety of dietary supplements described in NDI
FIG. 1. Category types of NDIs received by FDA.
at FDA Library on September 27, 2011toxsci.oxfordjournals.orgDownloaded from
notifications. Notifications typically include published literature
and unpublished toxicology or clinical studies, in addition to
descriptions of historically used materials that are the same as or in
some ways similar to the new ingredient. One general method for
ranking botanical ingredients according to possible safety concerns
is to examine the history of use. Botanicals that are traditionally
consumed as food rely on that history of food use as evidence of
safety. A key consideration is the assumption that the conditions of
use of the botanical are the same or very similar to the form,
quantity, and frequency of consumption that is documented in the
history of use. Foods are often consumed either intermittently over
the course of time or seasonally. This pattern is different from
many traditional medical uses such as when a botanical is used for
a short period once or only intermittently to treat an acute medical
condition. Safety problems that might be obvious once a large
number of individuals begin regular daily use might not be evident
when the botanical is used only intermittently.
Safety Considerations for Live Microbial Dietary Ingredients
The concept of natural products as dietary supplements
includes live bacteria and yeast. Live microbial dietary ingredients
(LMDIs), commonly promoted for probiotic properties, have
gained popularity among consumers. Although fermented foods
which retain live microorganisms have been consumed by
humans for centuries, factors such as the use of newly isolated
strains, changes in methods of industrial cultivation, or dosage
at levels higher than those typically found in foods might alter
the safety profile in the context of novel uses of the organisms.
A recent evidence-based review of clinical literature on
probiotics found significant gaps in knowledge about the
safety of probiotics particularly with respect to chronic use and
evaluation of frequency of infections (AHRQ, 2011). To date,
FDA has received notifications for new LMDIs that include
species of Lactobacillus, Bifidobacterium, Bacillus,
Enterococcus, Saccharomyces, Streptococcus, and Clostrid-
ium. Because of their uniqueness among natural products,
safety reviews examine parameters based on the potential for
particular risks of consuming LMDIs. A number of published
assessments focus on the safety of these ingredients
(Bernardeau et al.,2006; FAO/WHO, 2002; Meile et al.,
2008; Sanders et al., 2010; Wassenaar and Klein, 2008;
Wright, 2005).
Participants at a workshop on nomenclature for organisms
used as probiotic food ingredients concluded that scientifically
valid names for the genus and species should be used on the
label and that the organisms should be identified below the
species level to the strain (Sanders and Levy, 2011). Other
factors that might be relevant to the safety of the ingredient
include: (1) the number of viable microorganisms per serving,
(2) properties of the organism that are dependent on the
fermentation medium and growth conditions, (3) any phylo-
genetically related microbes that are pathogens that produce
mammalian toxins, (4) the ability of the LMDI to persist in the
gastrointestinal tract, and (5) the resistance of the LMDI to
clinically important antibiotics with special attention given to
any transferable genetic elements encoding antibiotic resistance
genes (Sanders et al., 2010).
History of Use
Although evaluation through toxicology or clinical studies
are the most common ways notifiers have used to establish
a reasonable expectation of safety (Section 413(a)(2) of the
FD&C Act), a documented history of safe food use without any
additional animal toxicity or human clinical data have also
been used. Factors described in notifications relying on this
strategy include (1) the dose (amount per serving), (2) duration
of use, (3) frequency of intake, and (4) a comparison of the
historically consumed material to the NDI. Exposure assess-
ments for food frequently rely on estimations of a distribution
of intake within the population (e.g., the amount consumed by
the mean or the highest 90% of the population). Both the
European Union (EU) and Canada expect at least 25 years of
historical use data before the historical information can be used
as a basis for a safety determination. The EU has spelled out
their history of use criteria in the Council of the EU Regulation
(EC) No. 509/2006 March on agricultural product and
foodstuffs as traditional specialities guaranteed. FDA has not
published guidance with respect to history of use for NDIs.
FIG. 2. FDA response letter types for NDIs involving botanical product ingredients.
at FDA Library on September 27, 2011toxsci.oxfordjournals.orgDownloaded from
Evaluation of Other Evidence of Safety
FDA has not yet published guidance concerning the safety
evaluation of NDIs, although a guidance document is under
development. However, the agency receives approximately 50
NDI notifications a year, and the comments in FDA response
letters, which raise concerns about the adequacy of the safety
information in the notifications, illustrate some of the issues
raised by the safety of natural products used in dietary
Evaluation of NDI Notifications and Dietary Supplements
The Division of Dietary Supplement Programs NDI Review
Team in the Office of Nutrition, Labeling and Dietary Supple-
ments at the Center for Food Safety and Applied Nutrition
reviews NDI notifications. An informal characterization of the
types of FDA response letters to those notifications is shown in
Figure 3. The following sections describe issues raised in letters
sent to notifiers.
NDIs Failing to Meet Administrative Requirements
The administrative requirements for an NDI submission in 21
CFR §190.6 are critical to evaluate an NDI submission.
Notifications which do not comply with the basic information
required in 21 CFR §190.6 generally cannot be reviewed for
safety. The FDA letter concerning an ingredient made from
Toona sinensis, dated 1 September 2005 in [Docket Report #339
(Docket ID: 95s-0316-rpt0294)], illustrates a response related to
compliance with §190.6. ‘Federal regulations found at 21 CFR
190.6 specify the requirements for a pre-market notification for
a new dietary ingredient. Your notification concerning Toona
sinensis does not comply with the requirements of 21 CFR
190.6 and is incomplete. The following items were not included
with your submission: (1) an original and two copies of the
notification, (2) a description of the dietary supplement or
dietary supplements that contains your new dietary ingredient,
(3) the level of the dietary ingredient in the dietary supplement,
and (4) the conditions of use recommended or suggested in the
labeling of the dietary supplement. Your notification describes
the physical characteristics and edible uses for the tree, fruit, and
leaves, but no information for the product that you intend to
market. Your notification provided three reference articles; the
remainder consisted of citations and abstracts of the references
that you relied on as evidence of safety. Any references to
published information offered in support of the notification shall
be accompanied by reprints or photostatic copies of such
references. In addition, your notification did not include
documented history of use of your new dietary ingredient as
an article present in the food supply.’
NDIs Excluded from the Definition of ‘‘Dietary Supplement’’
Sec. 201(ff) [21 U.S.C. 321(ff)] defines a ‘dietary supple-
ment.’ If the NDI does not fall within the meaning of 201(ff),
it is not a dietary supplement. One notification involved
‘obestilin’ (3#-hydroxystilbene), which was determined not to
be a dietary ingredient. In this case, the FDA letter, dated 12
March 2010 in [Docket Report #625], responded with the
following: ‘‘Based on the information in your notification, your
synthetically derived 3#-hydroxystilbene is not a dietary
ingredient within the meaning of Section 321(ff)(1) as
explained below. Although you state in your notification that
3#-hydroxystilbene has been reported to be a constituent of
plants such as Pterocarpus marsupium and Sphaerophysa
salsula, your synthetically derived 3#-hydroxystilbene is not
prepared by extracting and purifying 3#-hydroxystilbene from
a plant source. Therefore, your synthetic 3#-hydroxystilbene is
not a dietary ingredient under 21 U.S.C. 321(ff)(1)(F) because
it is not a constituent or an extract of a botanical, nor does your
notification present information establishing that synthetic 3#-
hydroxystilbene qualifies as any other type of dietary in-
gredient listed in Section 321(ff)(1). For these reasons,
synthetic 3#-hydroxystilbene cannot be marketed as a dietary
ingredient in a dietary supplement.’ This response highlights
that FDA has stated that synthetic copies of botanical
constituents are not equivalent to the natural counterpart in
the regulatory determination of the status of the ingredient
because they do not fit under 201(ff)(1)(F). The Federal
Register announcement of the 21 CFR 119 Ephedra Final Rule
explains ... synthetic sources of ephedrine cannot be dietary
ingredients because they are not constituents or extracts of
a botanical, nor do they qualify as any other type of dietary
ingredient. For these reasons, products containing synthetic
ephedrine cannot be legally marketed as dietary supplements’
(USFDA, 2004).
FIG. 3. FDA response letter to NDI notifications.
at FDA Library on September 27, 2011toxsci.oxfordjournals.orgDownloaded from
Another reason for a product to be excluded from the
definition of a dietary supplement is that it was authorized for
investigation as a new drug. The FDA response letter for
Cotinine, dated 20 April 2004 in [Docket Report #277],
demonstrates this point. ‘Your notification identified the
substance ‘Cotinine’ [derived from] (Duboisa hopwoodii
(F. Muell.)) as the substance that you intend to market as
a new dietary ingredient. .. . Cotininemaybeexcludedfromthe
definition of ‘dietary supplement’ under 21 U.S.C. 321(ff)(3)(B).
‘Cotinine’ is an article authorized for investigation as a drug for
which substantial clinical investigations have been instituted in
the United States, and the investigations have been made public,
and which was not before such authorization marketed as
a dietary supplement or as a food. Because the information in
your submission does not specify the form for your new dietary
ingredient, we are unable to determine if your ingredient has been
the subject of previous investigational new drug applications.
Finally, an NDI can be excluded from the meaning of 201(ff)
based upon its route of administration or intended effect. The
notification describing a Hoodia gordonii extract’ in ‘mint’
form was excluded from the definition in the FDA response
letter dated 15 January 2004 [Docket Report #262]. Hoodia
gordonii in the Asclepiadaceae family contains steroidal
glycosides with cardiac activity. ‘An article that is delivered
orally, but that exerts its effect prior to being swallowed (for
example, a gum, lozenge, or mint that stimulates salivation) or
that is a delivery system for a substance that is absorbed
buccolingually is not ‘intended for ingestion’. A gum, lozenge
or mint preparation which is not intended for ingestion can
function as a delivery system for a substance that is absorbed
buccolingually. As stated above, the definition of dietary
supplement in 21 U.S.C. 321(ff) states that a dietary supple-
ment is a product ‘intended for ingestion.’ The term ‘ingestion’
has been addressed by the court in United States v. Ten
Cartons, Ener-B Nasal Gel, 888 F. Supp. 381, 393–94
(E.D.N.Y.), aff’d, 72 F.3d 285 (2d Cir. 1995), which states:
The ordinary and plain meaning of the term ‘ingestion’ means
to take into the stomach and gastrointestinal tract by means of
enteral administration. See Stedman’s Medical Dictionary (4th
Lawyer’s Ed. 1976) (defining ingestion as the ‘introduction of
food and drink into the stomach.’); Webster’s Third New
International Dictionary (1976) (defining ingestion as ‘the taking
of material (as food) into the digestive system’.)...
The interpretation of the term ‘ingestion’ to mean enteral
administration into the stomach and gastrointestinal tract is also
supported by the language of the statutory sections immedi-
ately preceding and following Section 350(c)(1)(B)(ii).’ See
the §411 [21 U.S.C. §350] (Vitamins and Minerals) of the
FD&C Act for the statutory language.
Identity Issues with Old and New Dietary Ingredients
Chomper. In 1997, FDA demonstrated that Chomper, an
early dietary ingredient with laxative properties and multiple
botanical ingredients, and the plant material identified as the
ingredient plantain was adulterated and misbranded (Slifman
et al., 1998). In its 6 October 1997 warning letter to
Neutraceutical, FDA ‘received a complaint regarding injuries
sustained by a young woman who experienced an abnormal
heart rate with complete heart block, a potentially life-
threatening condition. The consumer’s symptoms were consis-
tent with an overdose of digitalis-like cardiac glycosides. The
young woman experienced this condition after ingesting
a regimen of dietary supplements. FDA’s investigation de-
termined that the problem was due to the ingredient plantain
found in your dietary supplement ‘Chomper’. FDA collected
multiple samples of plantain.’ ‘FDA analyses of these samples
showed that the plant material identified as ‘plantain’ contained
lanatosides (cardiac glycosides). The presence of lanatosides
support that the plant material contains Digitalis glycosides.
Digitalis lanata has been reported to contain these lanatosides.
Plantain has not been reported to contain any cardiac glycosides.
FDA also conducted an analysis of a sample of plantain to
determine whether the material identified as plantain actually
contained plantain. The analysis found that the characteristic
trichomes for plantain were low in concentration in the sample
when compared with reference specimens. These analyses
indicate that the plantain was contaminated with Digitalis.’
FDA determined the involved parties responsible for the
adulteration and misbranding product under Sections
402(a)(1), 402(f)(1)(A) and 403(a)(1) of the FD&C Act.
Milkweed seed oil. Milkweed seed oil presented another
ingredient that raised concerns about cardiotoxic glycosides,
this time in the context of an NDI notification. The FDA
response letter for Milkweed Seed Oil, dated 17 September
2009 (Report 597—Natural Fibers Corporation (Milkweed
Seed Oil), describes safety concerns raised by this notification.
‘Your notification concerns the new dietary ingredient
‘Milkweed Seed Oil (MSO)’ derived from the pod seeds of
milkweed plants that you call Asclepias Syriaca’[sic] and
Asclepias Specosia’[sic] that you intend to market as a dietary
supplement. FDA was unable to establish the identity of your
new dietary ingredient ‘Milkweed Seed Oil (MSO).’ For
example, plants in the genus Asclepias are widely regarded as
toxic to humans and grazing animals due not only to the
cardiotoxic glycosides (e.g. cardenolides) mentioned in your
notification but also due to toxic resinoids (e.g. galitoxin) and
alkaloids (e.g. phenanthroindolizidines) found in the seeds and
other parts of the plants. In addition, your notification states
that you will use petroleum ether or other, unspecified organic
solvents to produce your ingredient. Your notification does not
provide adequate information about the processing or
manufacturing of your ingredient, specifications for your
ingredient or specifications for the materials that you will use
to make your ingredient that address these poisonous and
deleterious substances which may be present in a dietary
supplement made by petroleum ether extraction of milkweed
seed pods. ... Therefore it is unclear how your ‘Milkweed
at FDA Library on September 27, 2011toxsci.oxfordjournals.orgDownloaded from
Seed Oil (MSO)’ is qualitatively or quantitatively related to the
materials used to establish the composition of your product or
how the composition of those materials can be used to provide
a basis for the safety of a product containing ‘Milkweed Seed
Oil (MSO).’ Because your notification asserts the safety of
‘Milkweed Seed Oil (MSO)’ based on the safety of each and all
of its constituents, FDA cannot determine the safety of
‘Milkweed Seed Oil (MSO)’ until the identity of the ingredient
you intend to manufacture can be established. For the reasons
discussed above, the information in your submission and the
other scientific literature the agency has reviewed FDA
disagrees with your conclusion that ‘Milkweed Seed Oil’,
when used under the conditions recommended or suggested in
the labeling of your product, will reasonably be expected to be
safe. Therefore, your product may be adulterated under
21 U.S.C. 342(f)(1)(B) as a new dietary ingredient for which
there is inadequate information to provide reasonable assurance
that such ingredient does not present a significant or unreason-
able risk of illness or injury. Introduction of such a product into
interstate commerce is prohibited under 21 U.S.C. 331(a) and
Lessons learned. Consequently, the two cases presented
above indicate that composition, adequate species identification,
and setting proper limits in a specification table for known toxic
constituents in a botanical are critical factors in assessing the
identity and safety of dietary supplements containing dietary
ingredients derived from herbs or botanicals. Botanical identi-
fication is critical and should be a major factor for dietary
supplement manufacturers to consider. Simply inspecting the
leaves of an herbal to verify herbal/botanical identification may
not be adequate to describe the identity of the plant. Indeed,
distinguishing between species at such a superficial level can be
problematic even for expert taxonomists. Many genera contain
species that are nearly indistinguishable at a macroscopic level.
Knowledge of morphological characteristics is crucial to proper
identification and may require microscopic or chemical analysis
to properly determine identity.
Aristolochic acid. Aristolochic acids are a family of
nitrophenanthrene compounds found in plants of the family
Aristolochiaceae, particularly those in two genera, Aristolochia
and Assarum. They came to worldwide attention following an
outbreak of about 100 cases of renal failure, many with urinary
tract carcinoma, in Belgium in the 1990s (Debelle et al., 2008).
In 1992,