S P E C I A L F E A T U R E

C o m m e n t a r y

Misconception and Concerns about Bioidentical

Hormones Used for Custom-Compounded
Hormone Therapy

Bhagu R. Bhavnani and Frank Z. Stanczyk

Department of Obstetrics and Gynecology (B.R.B.), University of Toronto, Toronto, Ontario, Canada M5B

Downloaded from https://academic.oup.com/jcem/article-abstract/97/3/756/2536306 by guest on 03 May 2020

1W8; and Departments of Obstetrics and Gynecology, and Preventive Medicine (F.Z.S.), University of
Southern California Keck School of Medicine, Los Angeles, California 90033

he results of the Women’s Health Initiative hormone
T therapy (HT) trials have caused both apprehension
and confusion about the overall risks and benefits asso-
ciated with postmenopausal HT (1). Since the results were
published, patient and clinician interest in potential alter-
natives to conventional HT has grown immensely and
continues to grow. This appears to be particularly true for
products or regimens that claim to have fewer risks and
side effects than commercially available HT preparations.
One commonly used alternative HT involves custom-
compounded hormone preparations. The hormones in these
preparations include estrogen [17␤-estradiol (estradiol), es-
trone, and/or estriol], progesterone, testosterone, andro-
stenedione, and dehydroepiandrosterone. The products can
be prepared in individualized dosages and forms such as
creams, gels, lotions, sublingual tablets, troches (lozenges)
for buccal administration, and suppositories by compound-
ing pharmacies from a clinician’s prescription. They are
often touted by advocates as safer than commercially pre-
pared HT products. Proponents also claim that custom-
compounded HT is associated with fewer side effects and
may provide better symptom relief than conventional HT
because the hormones used in the preparations are
“bioidentical,” i.e. identical to those made in the body.
The majority of these claims are unsubstantiated, and
no rigorous randomized control trials have been carried
out to support any of the claims. Because the custom-
compounding of HT products is not regulated, a par-
ticular concern is that some women may be overdosed,
or treated with ineffective products, or subject to un-
identified risk.
The objectives of the present commentary are: 1) to
show that the so-called “bioidentical” hormones in cus-
tom-compounded HT preparations may not be identical
to those made in the body; and 2) to point out how the lack
of regulation of these hormone products can cause adverse
effects in postmenopausal women using them.
Classification of Steroids Used for
Postmenopausal HT
For simplicity and to document potential pharmacological
differences, various steroid hormones used for HT can be
divided into four groups: 1) natural (class A); 2) native to
the body and synthesized from natural precursors (class
B); 3) native to the body and synthesized from nonsteroi-
dal precursors (class C); and 4) synthetic and not native to
the body (class D).
Class A steroids
The steroids in class A are found in nature and are
formulated into drugs without undergoing any chemi-
cal modifications. For example, conjugated equine es-
trogens are estrogens in the form of sulfate esters, which
are simply extracted from pregnant mare’s urine and do
not undergo chemical modification. Approximately
50% of conjugated equine estrogens consist of estrone
sulfate, and the remaining approximate 50% consists of
equine estrogens. Equine estrogens are native to the
horse but not the human.
Class B steroids
The class B steroids are semisynthetic. They are steroids
that exist in nature and are biosynthesized by the human

ISSN Print 0021-972X ISSN Online 1945-7197 Abbreviation: HT, Hormone therapy.
Printed in U.S.A.
Copyright © 2012 by The Endocrine Society
doi: 10.1210/jc.2011-2492 Received September 6, 2011. Accepted December 6, 2011.
First Published Online December 28, 2011

756 jcem.endojournals.org J Clin Endocrinol Metab, March 2012, 97(3):756 –759

J Clin Endocrinol Metab, March 2012, 97(3):756 –759
body, but for these to be formulated as therapeutic agents,
they need to be chemically synthesized from a natural
starting material, most commonly from a plant source
such as the Mexican yam and soybean. These plants con-
tain sterols such as diosgenin and stigmasterol, which are
used as precursors for the synthesis of a variety of steroids.
Essentially, steroid hormones such as estradiol, estrone,
estriol, progesterone, dehydroepiandrosterone, andro-
stenedione, testosterone, cortisol, aldosterone, synthetic
conjugated estrogens, etc., have been chemically synthe-
sized from these sources for decades. Contrary to many
claims, it is important to note that ingestion of the Mex-
ican yam or soybean does not result in the formation of any
of the above steroids because the human body lacks the
enzymes needed to convert diosgenin or stigmasterol to
those steroids. The chemical synthesis of a steroid hor-
mone such as estradiol from diosgenin requires at least
15 reactions. Another misconception pertains to the oc-
currence of mammalian steroid hormones in plants. It is
only recently that some steroid hormones such as proges-
terone and androstenedione have been positively identi-
fied in plants, using rigorous assay methodology. How-
ever, their concentrations are very low (2). There are no
hormonal preparations on the market that are derived by
extraction of a steroid hormone from a plant.
It is important to recognize that on a biomolecular level
steroids synthesized from plant precursors such as dios-
genin by the semisynthetic process are not actually iden-
tical to the corresponding endogenous steroids in the hu-
man body. There are two naturally occurring isotopes of
carbon, 12C and 13C, in the carbons that comprise the
chemical structure of steroids in plants and humans. Stud-
ies show that semisynthetic steroids consist of a different
13
C/12C ratio, when compared with the corresponding hu-
man endogenous compounds (3). This observation is
based on the fact that endogenous steroid hormones re-
flect an average of the 12C and 13C isotopes of carbon from
vegetal and animal food eaten by humans, whereas plant
sterols such as those found in soy exhibit a fixed 13C/12C
ratio. However, from a physiological standpoint, there is
no evidence showing that there is any difference between
a semisynthetic steroid and its corresponding endogenous
form in the human.
Class C steroids
The steroids in the class C group are “synthetic” com-
pounds that exist in nature but are synthesized from simple
nonsteroidal starting materials by a process generally
called “total synthesis.” This is one of the oldest means by
which steroid hormones were first synthesized. Examples
include estrone, estradiol, equilenin, progesterone, etc.
Without vigorous analytical analysis, it is extremely dif-
jcem.endojournals.org 757
ficult to differentiate between class B and class C steroids.
However, there can be significant differences in both the
chemical and biological properties of the steroids made by
these two distinct methods.
In total synthesis, there is one fundamental requirement
and, that is, the steroid synthesized has to have the same
precise stereochemistry as that of the naturally occurring
hormone. It is only then that the synthesized hormone will
have the same biological activity. It is obligatory that the
very specific three-dimensional structure be the same for a
Downloaded from https://academic.oup.com/jcem/article-abstract/97/3/756/2536306 by guest on 03 May 2020
compound to behave as an agonist because the activity of
the compound is expressed by its interaction with its spe-
cific endogenous receptor. Unfortunately, total synthesis
can give rise to a number of isomers; for example, in the
total synthesis of estrone, there are eight possible race-
mates (16 isomers), as shown in Fig. 1. Only one of these
is the natural hormone; others have different physical
properties, and some of them are totally inactive (4, 5).
Because estrogen preparations compounded by pharma-
cists do not undergo any vigorous testing or approval by
regulatory bodies, it is very possible that some mixtures,
depending on the route of synthesis, may not be as potent
or can be totally inactive therapeutically, as discussed
later. In other words, the so-called bioidentical hormones
may not be identical to the natural hormones produced in
vivo by the human body.
Class D steroids
The group of class D steroids includes steroids not
found in humans, animals, or plants, and includes drugs
such as medroxyprogesterone acetate, ethinyl estradiol,
norethindrone, norgestrel, etc. These manmade/designed
steroidal compounds can be synthesized from the same
plant starting materials such as diosgenin or stigmasterol,
or alternatively they can be manufactured from nonsteroi-
dal starting material by total synthesis, as described above
in Class C steroids.
Regulation of Custom-Compounded
Hormone Products
The U.S. Food and Drug Administration (FDA) regards
the use of the terminology “bioidentical hormone” as a
marketing ploy, implying a benefit for a drug for which
there is no medical or scientific basis (6). Although drug
compounding is generally subject to FDA oversight, the
agency relies on states to regulate the practice as part of
their overall regulation of the practice of pharmacy (7).
However, the ability of states to oversee the quality and
safety of compounded drugs is influenced and limited by
758 Bhavnani and Stanczyk Misconception and Concerns of Bioidentical Hormones J Clin Endocrinol Metab, March 2012, 97(3):756 –759
FIG. 1. Eight isomeric racemates of estrone.
the availability of resources for standard inspections and
enforcement.
A major concern about the custom-compounded prod-
ucts is that the practice of drug compounding is not subject
to the same degree of regulations and oversight by the FDA
and by regulatory authorities in other countries, which is
required for the commercial production and marketing of
prescription medicines. Unlike commercially available
drugs, compounded medicines are not tested routinely by
any regulatory agency for quality, purity and potency. In
addition, there are no product labeling requirements for
the custom formulations. This differs from commercially
available drugs, which are required to be sold with a pack-
age insert that details the product’s indications for use,
contraindications, pharmacokinetics, and adverse events,
using language approved by regulatory authorities.
Concerns about Accuracy of Hormone
Doses in Custom-Compounded Hormonal
Preparations
Because there is no strict regulation and oversight of com-
pounded medicines, postmenopausal women using cus-
tom-prepared hormonal preparations may be subject to
underdosing or overdosing of the products. Steroid doses
of products prepared by custom-compounding pharma-
cies are not regulated, and most often little is known about
the pharmacokinetics of the products. The doses can be
based on a physician’s prescription, report(s) from the lit-
Downloaded from https://academic.oup.com/jcem/article-abstract/97/3/756/2536306 by guest on 03 May 2020
erature, or information that is not evidence-based. In most
instances, serum levels achieved by a particular custom-
prepared hormonal preparation are not determined by the
physician. In addition, custom-compounded hormonal
preparations can be purchased and used by women with
no oversight by physicians as to the medical history of the
patient, potential presence of risk factors, or the serum
levels of active hormones attained with these products.
After becoming aware of product quality control prob-
lems associated with custom-compounded hormone prep-
arations, the FDA’s Division of Prescription Drug Com-
pliance and Surveillance examined 29 compounded drugs
obtained from 12 compounding pharmacies (8). The
agency reported that 10 of the products (34%) failed one
or more of their standard quality tests, including nine
products (one of which was progesterone) that failed be-
cause they contained less of the active ingredient than in-
dicated on the label.
Potential underdosing with progesterone has been re-
ported in three women who developed endometrial cancer
after using custom-compounded HT to relieve meno-
pausal symptoms (9). The HT preparations included tro-
ches containing varying doses of estrone, estradiol, estriol,
progesterone, and dehydroepiandrosterone, and were
used for 2– 4 yr. The final diagnosis of these subjects was
stage 1A or 1B, grade 2, endometrial cancer. The authors
of the report raised the possibility that the estrogenic com-
ponent of the troche was significantly absorbed but the
J Clin Endocrinol Metab, March 2012, 97(3):756 –759
progesterone dose was inadequate, thereby causing endo-
metrial hyperplasia.
Although hormones in custom-compounded HT have
risks similar to those in regulatory-approved products, e.g.
increased risk of a blood clot, heart attack, stroke, breast
cancer, gall bladder disease, etc., there is even a higher risk
if the hormone dose in a product is greater than expected.
In one study, healthy postmenopausal women were
treated for 2 wk with a custom-compounded testosterone
preparation, either a gel containing 1 mg of testosterone or
a lozenge containing 1 mg of testosterone propionate (10).
In the women receiving the lozenge, mean maximum tes-
tosterone levels occurred 1 h after treatment; the levels
rose from a mean baseline level of 16 to 692 ng/dl on d 1
and 836 ng/dl on d 14, and then declined. As for the
women receiving the testosterone gel, testosterone levels
showed a prolonged rise from a mean baseline value of 20
ng/dl to mean maximal levels of 97 and 100 ng/dl, reached
on d 1 and 14, respectively. The testosterone levels ob-
served in these women after treatment are substantially
above the upper normal range for women, i.e. 60 –70 ng/
dl. Such high testosterone levels present for prolonged pe-
riods may lead to deleterious effects.
Overdosing can also occur with progesterone creams
that are obtained from compounding pharmacies and used
by many postmenopausal women because the bioavail-
ability of progesterone after its topical application is
poorly understood (11). After the administration of dif-
ferent progesterone doses, a dose response of serum pro-
gesterone levels is not observed. The levels do not exceed
3.5 ng/ml, even when relatively high doses of the cream are
applied. However, the low progesterone levels do not re-
flect what is occurring in tissues because levels of proges-
terone in saliva can increase from baseline levels at least
two orders of magnitude after its topical application. Fur-
thermore, the levels in capillary blood are even much
higher. In addition, an antiproliferative effect on the en-
dometrium has been reported at serum progesterone levels
that were below 3.5 ng/ml. Therefore, it appears that tissue
levels of progesterone may be very high, although this is
not reflected in its circulating levels. Because progesterone
is metabolized to many different metabolites, some of
which are active, e.g. neurosteroids such as allopregnano-
lone, excessive intake of progesterone may cause adverse
effects such as a hormonal imbalance and mood swings.
In conclusion, health care providers need to diligently
consider the scientific evidence to determine the safety and
efficacy of all hormonal preparations used for HT. Use of
jcem.endojournals.org 759
the misleading term “bioidentical hormone” is inappro-
priate, and its use should be discouraged.
Acknowledgments
Address all correspondence and requests for reprints to: Frank Z.
Stanczyk, Ph.D., Professor of Research, Departments of Obstet-
rics and Gynecology, and Preventive Medicine, University of
Southern California, Keck School of Medicine, Livingston Re-
search Building, 1321 North Mission Road, Room 201, Los
Downloaded from https://academic.oup.com/jcem/article-abstract/97/3/756/2536306 by guest on 03 May 2020
Angeles, California 90033. E-mail: fstanczyk@socal.rr.com. Or,
Bhagu R. Bhavnani, Ph.D., Professor of Obstetrics, Gynecology,
University of Toronto, Li Ka Shing Knowledge Institute, St. Mi-
chael’s Hospital, 30 Bond Street, Toronto, Ontario, Canada
M5B 1W8. E-mail: bhagubhavnani@gmail.com.
This work was supported by a Medical Research Council of
Canada (CIHR) Grant MT-11329 (to B.R.B.).
Disclosure Summary: The authors have no conflicts of interest.
References
1. Clark JH 2006 A critique of Women’s Health Initiative studies.
Nuclear Receptor Signaling 4:1–10
2. Simersky R, Novak O, Morris DA, Pouzar V, Strnad M 2009 Iden-
tification and quantification of several mammalian steroid hor-
mones in plants by UPLC-MS/MS. J Plant Growth Regul 28:125–
136
3. Becchi M, Aguilera R, Farizon Y, Flament MM, Casabianca H,
James P 1994 Gas chromatography/combustion/isotope-ratio mass
spectrometry analysis of urinary steroids to detect misuse of testos-
terone in sport. Rapid Commun Mass Spectrom 8:304 –308
4. Shoppee CW 1958 Chemistry of the steroids. London: Butterworths
Scientific Publications; 105–127
5. Akhrem AA, Titov YA 1970 Total steroid synthesis. New York:
Plenum Press; 247
6. FDA Consumer Health Information. Bio-identicals: sorting myths
from facts. http://www.fda.gov/ForConsumers/ConsumerUpdates/
ucm049311.htm
7. 23 October 2003 Prescription drugs: state and federal oversight of
drug compounding by pharmacies. GAO-04-195T. Washington,
DC: U.S. Government Accountability Office
8. January 2003 FDA Center for Drug Evaluation and Research Re-
port: limited FDA survey of compounding drug products. Available
at http://www.fda.gov/Drugs/GuidanceComplianceRegulatory
Information/PharmacyCompounding/ucm155725.htm
9. Eden JA, Hacker NF, Fortune M 2007 Three cases of endometrial
cancer associated with “bioidentical” hormone replacement ther-
apy. Med J Aust 187:244 –245
10. Slater CC, Souter I, Zhang C, Guan C, Stanczyk FZ, Mishell Jr DR
2001 Pharmacokinetics of testosterone after percutaneous gel or
buccal administration. Fertil Steril 76:32–37
11. Stanczyk FZ, Paulson RJ, Roy SR 2005 Percutaneous administra-
tion of progesterone: blood levels and endometrial protection.
Menopause 12:232–237