0021-972X/07/$15.

00/0 The Journal of Clinical Endocrinology & Metabolism 92(10):3844 –3853

Printed in U.S.A. Copyright © 2007 by The Endocrine Society
doi: 10.1210/jc.2007-0620

Androgen Receptor Gene CAG Repeat Length and

Body Mass Index Modulate the Safety of Long-Term
Intramuscular Testosterone Undecanoate Therapy in
Hypogonadal Men
Michael Zitzmann and Eberhard Nieschlag
Institute of Reproductive Medicine, University Clinics, Muenster D-48149, Germany

Context: A reliable form of androgen substitution therapy regarding
kinetics, tolerance, and restoration of androgenicity is paramount in
hypogonadal men. Intramuscular injection of the long-acting ester
testosterone undecanoate (TU) offers a new modality.
Objective: The objective of the study was to assess the safety of TU
regarding metabolic and pharmacogenetic confounders.
Design: This was a longitudinal one-arm open observation trial. A
minimum of five individual assessments was a prerequisite. Putative
modulators of safety parameters entering regression models were
nadir and/or delta total testosterone concentrations, body mass index,
androgen receptor (AR) gene CAG repeat length, and age.
Setting: The study was conducted at an andrological outpatient
clinic.
Patients: Patients included 66 hypogonadal men (mean age 38 ⫾ 9.9
yr).
Main Outcome Measures: A total of 515 data time points each
related to prostate, erythropoiesis, lipoproteins, and circulation dur-
ing 118 treatment-years with 1000 mg TU at 10- to 14-wk intervals.
Results: Testosterone substitution resulted in significant decre-
ments of serum levels of low-density lipoprotein-cholesterol, resting
diastolic and systolic blood pressure, and heart rate. Erythropoiesis
was stimulated and concentrations of high-density lipoprotein-
cholesterol increased. Parameters remained stable after four injec-
tions. No adverse effects regarding the prostate were observed. Sig-
nificantly increased hematocrit greater than 50% was predicted by
enhanced androgen action (shorter AR CAG repeats per higher tes-
tosterone levels). However, insufficient androgen action (longer AR
CAG repeats per lower testosterone levels) caused pathological safety
parameters (high blood pressure, adverse lipid profiles). In addition,
a body mass index 30 kg/m2 or greater represents a clinically relevant
factor for the occurrence of all pathological safety parameters. Risk
calculations for obese patients and nonlinear pharmacogenetic mod-
els to tailor androgen substitution are presented.
Conclusions: Testosterone substitution with im TU is generally well
tolerated. Modifications of androgen action are due to both AR CAG
repeats and testosterone levels. Adverse observations are mostly seen
in obese patients. (J Clin Endocrinol Metab 92: 3844 –3853, 2007)

T HE CLINICAL PICTURE of male hypogonadism man-

ifests itself with a diversity of symptoms that are due
to a deficiency of testosterone (T) or its action; replacement
has to be accompanied by standardized procedures of sur-
veillance to avoid adverse effects (1, 4).
Recently metaanalyses revealed that erythropoiesis, lipid
is the treatment of choice (1). Today the clinician is able to constellations, and the prostate may be especially susceptible
choose from a broad spectrum of preparations that exhibit to undesired changes during T substitution (5, 6). However,
distinct pharmacokinetic properties and offer different these studies comprise data involving older injectable T
routes of application. In addition to short-acting transdermal preparations, producing unfavorable supraphysiologic se-
or buccal modalities, a long-acting intramuscular prepara- rum levels. These are no longer seen with the new injectable
tion of testosterone undecanoate (TU) has recently become TU, which may result in fewer or different side effects. We
available (reviewed in Refs. 2 and 3). have long-term experience with this preparation in non-
These preparations effectively establish, restore, and human primates as well as substitution therapy of hypogo-
maintain T-dependent functions in men. Diagnostic proce- nadal men (7–11), which has also been evaluated by other
groups (12, 13).
dures identifying the underlying cause of hypogonadism are
In addition, it has been previously shown that metabolic
a prerequisite for the start of therapy. T substitution therapy
aspects such as body mass index (BMI) or age in general may
influence the effect of T on safety parameters (14). The phar-
First Published Online July 17, 2007 macogenetic influence of the CAG repeat polymorphism of
Abbreviations: ANCOVA, Analysis of covariance; AR, androgen re- the androgen receptor (AR) gene on these parameters might
ceptor; BMI, body mass index; CV, coefficient of variation; HDL, high-
density lipoprotein; LDL, low-density lipoprotein; PSA, prostate-spe-
also play a long-term role concerning safety aspects. It has
cific antigen; T, testosterone; TU, T undecanoate. been previously demonstrated that the effects of other T
JCEM is published monthly by The Endocrine Society (http://www. preparations, such as short-acting im injected esters, trans-
endo-society.org), the foremost professional society serving the en- dermal and oral preparations are modulated by this poly-
docrine community. morphism (15–18). These pharmacogenetic reports did not

3844

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 May 2015. at 00:15 For personal use only. No other uses without permission. . All rights reserved.
Zitzmann and Nieschlag • Safety of Testosterone Undecanoate
specifically evaluate the effects of BMI on the outcome of T
substitution.
Therefore, we investigated the side effects of T substitution
by im applied TU in 66 hypogonadal men during 118 treat-
ment-years, and it is the first time that long-term aspects of
safety in combination with both metabolic features and phar-
macogenetics are considered.
Patients and Methods
Figure 1 illustrates the process of patient selection: all patients cur-
rently receiving im TU were identified from our patient database. These
men presented for examination and possible treatment of hypogon-
adism-related symptoms. Such symptoms were fatigue, loss of libido,
depressiveness, change in body composition/weight, decreased phys-
ical performance, decrease in aggressive behavior, disability to cope,
decreased work performance, and lack of androgenization. For inclu-
sion, at least one of these symptoms had to be accompanied by low total
T levels (⬍12 nmol/liter).
Exclusion criterion for T therapy were prostate-specific antigen (PSA)
levels above 4 ng/ml. All patients gave written informed consent for the
use of their data for scientific evaluation as approved by the Ethics
Committee of the Medical Faculty, University of Muenster, Muenster,
Germany, and the state medical board. By its design, this study cannot
be considered a safety study in its own right, and there has been no
systematic recording of all adverse events in its broadest interpretation,
without preconceived opinion as to their possible relevance, as is done
in clinical trials with systematic assessment of safety. Rather, we have
monitored a predefined set of safety parameters for treatment with TU:
Data base search
Identification of hypogonadal men being
treated with
testostosterone undecanoate 1000 mg i.m.
(n = 141)
Screening I
Exclusion of men with < 4 injections
or < 5 consecutive assessments
after initiation of substitution therapy
(n = 23)
Screening II
Exclusion of men receiving
intermediate injections
at other care centers or by home
physicians (n = 33)
Screening III
Missing data of >_ 1 safety parameter
_ 1 time-point (n = 19)
at >
Eligibility
for inclusion
(n = 66)
FIG. 1. Illustration of patient selection: all patients currently receiv-
ing im TU were identified from our patients’ database. Further
screening procedures excluded those men receiving less than four
injections (screening I), those with injections given intermediately at
other centers (screening II), and those with missing data (screening
III). Finally, 66 men were eligible for evaluation. Also see Patients and
Methods concerning inclusion and exclusion criteria for T treatment
and other details.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 May 2015. at 00:15 For personal use only. No other uses without permission. . All rights reserved.
J Clin Endocrinol Metab, October 2007, 92(10):3844 –3853 3845
hemoglobin content, hematocrit, weight, prostate parameters, lipid me-
tabolism, and circulation parameters (blood pressure and pulse) and
linked them analytically to baseline and undertreatment hormone con-
centrations, age, BMI, and the AR polymorphism.
No patient discontinued the treatment before receiving four injec-
tions. Further screening procedures excluded those men having yet
received less than four injections, those with injections given interme-
diately at other centers, and those with missing data. Finally, 66 men
were eligible for evaluation (Fig. 1). These 66 hypogonadal men received
a total of 383 injections of 1000 mg TU in 10- to 14-wk intervals. Eligibility
for evaluation was given after assessment of baseline values, a minimum
of four injections, and presentation to assess data before a putative
further (minimum fifth) injection, accumulating to 515 data time points
spanning 118 treatment-years. Thirty-five men had primary, 27 second-
ary, and four late-onset hypogonadism as previously defined (4, 19).
None of these patients has been reported in manuscripts concerning T
substitution and the modulation of androgen effects by the CAG repeat
polymorphism (citations 16 and 17 refer to patients who received other
preparations). Data of three of the 66 patients who are mentioned in this
report have been referred to in an interim report about the efficacy of im
TU in general, dating back to 2002 (9). Data of all other 63 patients have
not been evaluated in previous publications.
Biochemical analyses
All venous blood samples were obtained between 0800 and 1200 h
after a 30-min rest before the first or next injection of TU. Serum or
plasma were separated at 800 g. Samples were immediately stored at
⫺20 C. Serum T levels were measured by a commercial ELISA kit (DRG
Instruments GmbH, Marburg, Germany), and mean interassay coeffi-
cient of variation (CV) was less than 5%. Concentrations of SHBG were
determined by highly specific time-resolved fluoroimmunoassays (Au-
todelfia, Freiburg, Germany). Mean intraassay CVs were less than 5%
and mean interassay CVs less than 10%. Levels of free T were calculated
from levels of SHBG and total serum T according to previously pub-
lished calculations (20). PSA was determined with highly specific time-
resolved fluoroimmunoassay (Autodelfia), with a normal upper limit of
reference range of less than 4 ng/ml. Mean intra- and interassay CVs
were less than 2 and 5%, respectively. Sampling was performed before
prostate palpation and transrectal ultrasonography. Red blood count
was performed on a SE 9500 system (Sysmex Europe, Hamburg, Ger-
many). A Hitachi 917 autoanalyzer (Hitachi/ Roche Diagnostics, Mann-
heim, Germany) was used for the quantification of serum concentrations
of lipoproteins with enzymatic tests. Interassay CV was less than 5%.
Determination of AR CAG repeats
DNA was isolated from EDTA blood samples using the Nucleon kit
(Amersham Life Science, Freiburg, Germany) and analysis of the AR
gene microsatellite residues was performed as previously published
(16).
Measurement of prostate size
Transrectal ultrasonography was performed as previously described
using a high-resolution transrectal ultrasonography probe (7.5 MHz for
sagittal and transversal scans with an ultrasound scanner type 2002 ADI;
B-K Medical, Gentofte, Denmark) (16, 21).
Resting blood pressure and heart rate
Blood pressure and resting pulse were measured by trained physi-
cians using a standardized oscillometric device (M5 Professional; Om-
ron Medical Technics, Mannheim, Germany) with a cuff size appropriate
to individual body habitus. Measurements were taken after the subject
had rested in sitting position for at least 5 min in a quiet room.
Medication
In its injectable form, TU is a new preparation and has recently been
made available (2004 in Germany). The medication used before market
introduction was identical in preparation and content of TU and was
provided by Bayer Schering Pharma, (Berlin, Germany; former Schering
3846 J Clin Endocrinol Metab, October 2007, 92(10):3844 –3853 Zitzmann and Nieschlag • Safety of Testosterone Undecanoate

AG, Berlin, Germany): 1000 mg of TU (since 2004: Nebido) are dissolved
in castor oil.
To achieve a fast steady-state after initiation of therapy, the second
injection has to be given after 6 –10 wk. Thereafter the dosing interval
can be prolonged, reaching 10 –14 wk. In general, patients were injected
according to the following regimen: wk 03wk 63wk 183wk 303fur-
ther injections, starting wk 40 – 44 on an individual basis titrated ac-
cording to nadir T levels at wk 30.
Statistics
All variables were checked for normal distribution by the Kolmog-
orov-Smirnov one-sample test for goodness of fit and log transformed
if necessary. All data are presented as mean ⫾ sd. In case of percent
values (hematocrit), arcsin transformation was performed. Because most
relevant changes of T treatment are likely to occur within the first year
of treatment (6, 22), we compared baseline and T steady-state conditions
before the fifth injection of TU (wk 40 – 44) by t tests for matched pairs.
In case of significant differences, analysis of covariance (ANCOVA)
models for repeated measurements including covariates of interest were
assessed (BMI, age, ␦-serum T concentrations, AR CAG repeats). Base-
line correlations were calculated as Spearman rank correlations using
nontransformed variables.
To assess long-term effects of safety, all data points except baseline
values (n ⫽ 449) were analyzed in terms of passing respective thresholds
into the pathological range. Such a condition was defined as yes or no
and entered in binomial regression models involving the above-named
covariates. Instead of ␦-T levels, nadir T concentrations were chosen
because this analysis applied to single time points. Because time under
treatment was variable and a pathological condition could occur more
than once in a single patient, duration of substitution was introduced as
inverse weight to stratify for possible confounding effects.
Because BMI turned out to be a major confounder of treatment and
safety, the relative risk was calculated for obese vs. nonobese men in ␹2
tests.
The association between T levels in combination with AR CAG re-
peats and safety parameters may not be linear, as previous publications
suggest (16). To this end, a nonlinear model to describe pharmacogenetic
effects on hematocrit was created, involving data points of persons with
a BMI less than 30 kg/m2 (n ⫽ 316) only because BMI is an additional
factor of influence (see Fig. 3 and Table 3). A three-dimensional model
to these data points was calculated, assuming a log-linear association of
T concentrations with androgen effects (19, 23), a quasilog-linear binding
of T to its receptor within the concentration range observed in humans
(reviewed in Ref. 24) and a linear effect of the AR polymorphism on the
transcription of androgen target genes (25). Thus, the underlying for-
mula is based on these principles: Androgen effect ⫽ Log (T) ⫻ [m ⫺
(n ⫻ log (T) ⫻ CAGn)] ⫹ k.
Constant m describes the general association of T with the androgen
effect, which is tissue specific and is mitigated by the length of the AR
CAGn polymorphism. Constant n relates to the AR CAGn polymor-
phism and is higher in target tissues with higher concentrations of AR
coactivators (26, 27) because these determine the effect of the polymor-
phism. The effect of the CAGn polymorphism depends on the amount
of T binding to the AR, which is, under physiological conditions, log
linear. Constant k describes the value of the androgen-dependent target
parameter, which is measurable at absolute T deficiency.
Computations were performed using the statistical software package
SPSS (Chicago, IL; release 14.0) and GraphPad Prism (San Diego, CA;
release 3.2).
Results
Overall, substitution treatment with TU was well toler-
ated. Patient characteristics and hormone levels at baseline as
well as under treatment are demonstrated in Table 1. Table
2 displays major changes of safety parameters and the sig-
nificant effectors of these changes after four injections of TU
1000 mg im at the time point before the fifth injection, i.e. after
40 – 44 wk of T substitution. Table 3 exhibits the frequency of
pathological observations during the whole treatment period
of 118 patient-years as well as predisposing factors. Figure 2
displays total serum T concentrations before and during
therapy (also see Table 1). Figure 3 displays the crude relative
risk of obese men vs. nonobese men to be diagnosed with

TABLE 1. General characteristics of 66 hypogonadal men treated by the long-acting, im formulation of TU 1000 mg (total of 383
injections)

Parameter Mean ⫾ SD, range

Age at start of treatment (yr) 38.1 ⫾ 9.9 (range 17– 66)
Cause of hypogonadism Primary: n ⫽ 35
Secondary: n ⫽ 27
Late onset: n ⫽ 4
Duration of treatment per patient (yr) 1.8 ⫾ 1.7 (range 0.8 –9.5)
AR CAG repeats (n) 21.9 ⫾ 3.2 (range 15–29)
Baseline BMI (kg/m2)a 28.1 ⫾ 4.9 (range 20.0 –38.4)
Caucasian ethnicity (n ⫽ 66) 66
Cigarette smoking (n ⫽ 6)a 19
Baseline serum concentrationsa
Total T (nmol/liter) 6.6 ⫾ 2.9 (range 1.3–11.8)a
SHBG (nmol/liter) 54 ⫾ 9 (range 35– 82)
Free T (pmol/liter) 99 ⫾ 52 (range 22–234)
Estradiol (pmol/liter) 54 ⫾ 10 (range 38 –72)
Concentrations after reaching steady state
(nadir levels prior to fifth injection)
Total T (nmol/liter) 16.2 ⫾ 6.2 (range 8.0 –24.4)
SHBG (nmol/liter) 38 ⫾ 7 (range 24 –53)
Free T (pmol/liter) 234 ⫾ 78 (range 143–509)
Estradiol (pmol/liter) 74 ⫾ 16 (range 43–120)
Concentrations during prolonged treatment
(nadir levels prior to injections, n ⱖ 8)
Total T (nmol/liter) 14.9 ⫾ 5.2 (range 8.2–27.5)
SHBG (nmol/liter) 43 ⫾ 9 (range 25– 63)
Free T (pmol/liter) 219 ⫾ 67 (range 142– 433)
Estradiol (pmol/liter) 73 ⫾ 15 (range 42–118)
For total serum T levels, see Fig. 2.
a
Defined as daily consumption of at least one cigarette.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 May 2015. at 00:15 For personal use only. No other uses without permission. . All rights reserved.
 TABLE 2. Changes of safety parameters during treatment with 1000 mg TU (im) in 66 hypogonadal men (assessments prior to putative fifth injection: 40 – 44 wk after
initiation of treatment)

Mean prolonged
Significance
treatment values
Pretreatment Intratreatment value of difference
Factor of influence (by (prior to
Parameter value (mean ⫾ after four injections (P by Direction of influence P
ANCOVA models) injections, n ⱖ 8,
SD) (mean ⫾ SD) matched
186 data time
pairs t test)
points)
Prostate volume, ml 16.9 ⫾ 5.5 19.2 ⫾ 5.4 0.009 Higher ␦-T Increment 0.001 20.0 ⫾ 5.3
Shorter AR CAG Increment 0.01
repeats
Higher age Increment 0.02
PSA level, ng/ml 0.7 ⫾ 0.5 0.7 ⫾ 0.4 0.95 No significant No significant No significant difference, 0.7 ⫾ 0.4
difference, analysis difference, analysis not applicable
not applicable analysis
not applicable
Hemoglobin, g/dl 14.9 ⫾ 1.3 15.9 ⫾ 1.0 ⬍ 0.001 Higher ␦-T Increment ⬍0.001 15.9 ⫾ 0.9
Higher BMI or Increment 0.001
higher E2a
Shorter AR CAG Increment 0.01
repeats
Zitzmann and Nieschlag • Safety of Testosterone Undecanoate

Higher age Increment 0.03

Hematocrit, % 43.8 ⫾ 4.9 47.5 ⫾ 4.7 ⬍ 0.001 Higher ␦-T Increment ⬍0.001 47.1 ⫾ 5.0
Higher BMI or Increment 0.003
higher E2a
Shorter AR CAG Increment 0.005
repeats
Higher age Increment 0.04
LDL-cholesterol, 152 ⫾ 25 127 ⫾ 33 ⬍ 0.001 Higher ␦-T Decrement 0.002 130.0 ⫾ 32
mg/dl
Lower BMI Decrement 0.005
HDL-cholesterol, mg/dl 42.4 ⫾ 11.3 50.1 ⫾ 8.9 ⬍ 0.001 Lower age Increment 0.02 51.1 ⫾ 9.9
Lower BMI Increment 0.001
Resting RR systolic, 140 ⫾ 12 134 ⫾ 14 0.009 Lower age Decrement 0.001 133 ⫾ 13
mm Hg
Shorter AR CAG Decrement 0.01
repeats
Lower BMI Decrement 0.02
Resting RR diastolic, 90 ⫾ 10 83 ⫾ 11 ⬍ 0.001 Shorter AR CAG Decrement 0.03 81 ⫾ 11
mm Hg repeats
Lower BMI Decrement 0.001
Resting heart rate, 83 ⫾ 9 77 ⫾ 13 0.003 Higher ␦-T Decrement 0.02 77 ⫾ 14
bpm
Shorter AR CAG Decrement 0.01
repeats
Lower age Decrement 0.009
BMI, kg/m2 28.1 ⫾ 4.9 27.3 ⫾ 4.9 0.38 No significant No significant No significant difference, 27.1 ⫾ 4.7
difference, analysis difference, analysis not applicable
not applicable analysis
not applicable

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 May 2015. at 00:15 For personal use only. No other uses without permission. . All rights reserved.
Comparisons of pre- and intratreatment values were performed by t tests for matched pairs. In case of significant differences, ANCOVA models for repeated measurements
including ␦-serum T levels, AR CAG repeat length, age, and BMI were calculated. Estradiol (E2) levels under treatment were included in a second-step analysis because they were
strongly related to BMI. Replacement of total T by free T did not alter results. Note that, although ␦-T levels may not be included by all ANCOVA models, overall changes are most
J Clin Endocrinol Metab, October 2007, 92(10):3844 –3853

likely due to T therapy. Longer-term data (last right column) refer to 22 patients only and are not statistically compared with other data). RR, Blood pressure measured according
to the method of Riva and Rocci.
a
In case of hemoglobin and hematocrit, high estradiol levels replaced BMI as factor of influence.
3847
 3848

TABLE 3. Predictors of safety parameters that were assessed within the pathological range during 118 patient-years of treatment with TU im in 66 hypogonadal men
(maximum individual duration: 494 wk)

Threshold to Cases Factor of significant influence and Mean ⫾ SD of confounder at Mean ⫾ SD of confounder at
Parameter P
pathological range (n ⫽ 449) dimension time points of incident time points of nonincident
Prostate volume ⬎30 ml 3 No. of data points not sufficient No. of data points not No. of data points not
for analysis sufficient for analysis sufficient for analysis
PSA level ⬎4 ng/ml 0 No data observed No data observed No data observed
Hemoglobin ⬎18 g/dl 12 Higher BMI (kg/m2) or higher 36.8 ⫾ 1.9 27.0 ⫾ 4.6 ⬍ 0.001
E2 level (pmol/liter)a 86 ⫾ 21 69 ⫾ 15 ⬍ 0.001
J Clin Endocrinol Metab, October 2007, 92(10):3844 –3853

Hematocrit ⬎50% 47 Higher age (yr) 42.5 ⫾ 8.7 41.0 ⫾ 4.8 0.08 (trend)
Higher BMI (kg/m2) or higher 31.8 ⫾ 6.2 26.8 ⫾ 4.5 ⬍ 0.001
E2 level (pmol/liter)a 89 ⫾ 15 65 ⫾ 13 ⬍ 0.001
Shorter AR CAG repeats (n) 19.5 ⫾ 1.2 22.1 ⫾ 1.9 ⬍ 0.001
Higher nadir T level 18.9 ⫾ 6.2 13.5 ⫾ 6.3 ⬍ 0.001
(nmol/liter)
LDL-cholesterol ⬎150 mg/dl 56 Higher BMI (kg/m2) 29.6 ⫾ 4.8 27.5 ⫾ 5.0 0.01
HDL-cholesterol ⬍35 mg/dl 9 Higher BMI (kg/m2) 36.1 ⫾ 2.4 28.2 ⫾ 5.1 ⬍ 0.001
Resting RR systolic ⬎140 mm Hg 89 Higher BMI (kg/m2) 30.1 ⫾ 4.8 26.9 ⫾ 4.9 ⬍ 0.001
Lower nadir T level (nmol/liter) 13.0 ⫾ 5.6 14.5 ⫾ 6.8 0.06 (trend)
Resting RR diastolic ⬎90 mm Hg 63 Higher BMI (kg/m2) 29.8 ⫾ 4.9 27.4 ⫾ 4.9 0.001
Longer AR CAG repeats (n) 23.1 ⫾ 2.9 21.8 ⫾ 3.0 0.01
Lower nadir T level (nmol/liter) 11.5 ⫾ 5.6 14.6 ⫾ 6.7 0.001
Resting heart rate ⬎95 bpm 19 Higher BMI (kg/m2) 29.5 ⫾ 3.7 27.7 ⫾ 5.1 0.05
Longer AR CAG repeats (n) 23.8 ⫾ 3.3 21.9 ⫾ 3.0 0.05
Lower nadir T level (nmol/liter) 13.7 ⫾ 6.6 18.8 ⫾ 5.3 0.001
Higher age (yr) 44.1 ⫾ 5.0 41.9 ⫾ 8.8 0.08 (trend)
Binomial regression analysis including time under treatment as covariate (see Statistics) as well as nadir serum T levels (nanomoles per liter) before next injection, age (years),
BMI (kilograms per meter2) and AR CAG repeats (n). Estradiol (E2) levels under treatment were included in a second-step analysis because they were strongly related to BMI.
Replacement of total T by free T did not alter results. All data time points except baseline examinations (n ⫽ 449).
a
In case of hemoglobin and hematocrit, high estradiol levels replaced BMI as factor of influence.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 May 2015. at 00:15 For personal use only. No other uses without permission. . All rights reserved.
Zitzmann and Nieschlag • Safety of Testosterone Undecanoate
Zitzmann and Nieschlag • Safety of Testosterone Undecanoate J Clin Endocrinol Metab, October 2007, 92(10):3844 –3853 3849

50 different results in statistical evaluations. Serum concentra-

Total Testosterone nmol/L

45 tions of estradiol under treatment were strongly related to

40
35
30
25
20
15
10
5 Besides the direct relevance of elevation of T levels by sub-
0 stitution therapy, we demonstrate the influence of BMI and
0.0 0.2 0.4 0.6 0.8 1 22
1.0
years
FIG. 2. Total serum T concentrations displaying 515 data points of 66
hypogonadal men before (circles) and during substitution therapy
with TU im (diamonds: nadir levels). The smoothing curve for mean
nadir T concentrations during therapy was created from sliding av-
erage data (lag time 40 d). Note that the scale of the x-axis was broken
for better display.
safety parameters within the pathological range during T
substitution. Figure 4 represents a nonlinear model of phar-
macogenetics of T treatment in regard to hematocrit as a
safety parameter and allows individual application of thresh-
old levels for nadir T levels under substitution therapy.
Correlations of BMI with nadir levels of sex hormones and
SHBG under treatment (at the time point after four injections
of TU 1000 mg im before the fifth injection, i.e. after 40 – 44
wk of T substitution) exhibited the following picture: T con-
centrations were not related to BMI (rs ⫽ 0.03, P ⫽ 0.62).
Serum SHBG levels were lower (rs ⫽ ⫺0.26, P ⫽ 0.001) in men
with higher BMI. Correspondingly, serum concentrations of
free T under treatment exhibited a positive association with
BMI, albeit as nonsignificant trend (rs ⫽ 0.16, P ⫽ 0.06).
Concentrations of free T under treatment closely reflected
total T levels (rs ⫽ 0.96, P ⬍ 0.001) and did not exhibit
BMI (rs ⫽ 0.44, P ⬍ 0.001). Estradiol levels were used as
additional independent variables for analyses in second-step
approaches (Tables 2 and 3).
Discussion
This open one-arm study presents novel data on safety
issues related to T substitution therapy in hypogonadal men.
3 4 5 6 7 8 910
age as well as pharmacogenetics exerted by the AR CAG
repeat polymorphism. The nonexistence of longer placebo-
controlled trials of T substitution is quite understandable
because ethics in general and the physicians’ responsibility
for his hypogonadal male patients in particular do not allow
long-term placebo-controlled trials: the adverse effects of
hypogonadism on well-being, health, and life expectancy are
known (1, 4, 19, 28 –30). Nevertheless, this study provides an
advantage over metaanalyses by involving a single T prep-
aration being used over a considerable period of time in a
single center applying standardized and well comparable
procedures. Notwithstanding, comparisons between sub-
groups according to BMI or other confounders are feasible.
In agreement with other long-term surveillance ap-
proaches that involve transdermal T preparations (22, 31), we
observed major changes within the first year of treatment
(Table 2). After reaching a plateau of androgenization, hy-
pogonadal men seem to maintain a stable level of androgen-
related effects; correspondingly, nadir total T levels are per-
petuated within the normal range when hypogonadal men
are regularly treated with injectable TU (Table 1).
Prostate safety is an issue of major concern among phy-
sicians treating hypogonadal men with T substitution. This
study demonstrates that PSA levels are not necessarily af-

Parameter p - value
Hemoglobin
> 18 g/dL <0.001
Hematocrit
> 50 % <0.001
LDL cholesterol
> 150 mg/dL 0.001
HDL cholesterol
< 35 mg/dL 0.007
Resting RR systolic
> 140 mmHg <0.001
Resting RR diastolic
> 90 mmHg
0.001
Resting heart rate
> 95 bpm 0.02

1,0 10,0 100,0

0 1 2 3 4 5 6 7 8 9 10 15 20 25

Relative Risk and 95 % CI

FIG. 3. Crude relative risk (RR) for hypogonadal men with a BMI 30 kg/m2 or greater (n ⫽ 16) involving 133 data time points in comparison
with nonobese men (n ⫽ 50) involving 316 data time points to be observed with pathological safety parameters during androgen substitution
with TU 1000 mg im (data obtained during 118 treatment-years, involved are all data time points except baseline examinations: n ⫽ 449). Results
according to ␹2 tests, logarithmic scale.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 May 2015. at 00:15 For personal use only. No other uses without permission. . All rights reserved.
3850 J Clin Endocrinol Metab, October 2007, 92(10):3844 –3853 Zitzmann and Nieschlag • Safety of Testosterone Undecanoate

FIG. 4. Nonlinear pharmacogenetic model to describe the effects of T levels and the AR polymorphism on hematocrit based on the 316
intratreatment observations in nonobese men. The best-fitting model to predict hematocrit as androgen target parameter under substitution
therapy with TU 1000 mg im was found for constants m ⫽20, n ⫽ 0.2, k ⫽ 31 (for the formula outlined in Statistics; R2 ⫽ 0.75 for the nonlinear
model and R2 ⫽ 0.40 for linear associations; superiority of the nonlinear model by f test: P ⬍ 0.001); x-axis: length of the AR CAG repeat
polymorphism; y-axis: nadir of total T concentration before the next injection of TU 1000 mg im; z-axis: hematocrit. Light gray plane is the safety
limit of hematocrit (50%). Note that this applies to nonobese persons only. Obese men will require lower nadir T levels (Fig. 3 and Table 3).
The table inset outlines the combination of nadir total testosterone (TT) levels achieved under substitution therapy with TU and AR CAG repeat
length at which a hematocrit of 50% is reached (these data represent results of the nonlinear formula outlined in Statistics). AR CAG repeat
lengths above 25 require supraphysiological T levels to reach this safety threshold. Because this model is based on data obtained during the
observation period, it would be hypothetical to extend the table to present such high T levels: there are no real data present to support such
assumptions. The same applies to AR CAG repeat lengths less than 15 in terms of very low nadir levels. Note that obese men with an AR CAG
repeat length 26 or greater might reach hematocrit levels greater than 50% at physiological T levels (see above). Also note that the TT levels
displayed represent nadir T levels under substitution therapy by TU 1000 mg im and should not be mistaken for physiological T concentrations
in eugonadal men.

fected to a clinically relevant degree, if at all, by external viewed in Ref. 34). This effect is generally considered ben-
application of T. Nevertheless, to find a significant treatment eficial because anemia is frequently encountered in male
effect on prostate volume without effect on serum PSA may hypogonadism and contributes to overall complaints of
seem rather unexpected. However, prostate volumes did not weakness and fatigue (22). However, overstimulation of
change to a major degree (Table 2). In addition, T treatment erythropoiesis may cause an elevated hematocrit. Hemato-
may not necessarily significantly alter PSA levels, as has been crit values over 50 –52% are related to increased blood vis-
demonstrated by the largest metaanalysis of placebo-con- cosity, and there is evidence from larger neurological studies
trolled trials regarding T substitution (6). Another large unrelated to T that elevated hematocrit can result in cerebral
study involving transdermal application of T gel reported a ischemia: erythrocyte aggregation may facilitate platelet ac-
significant increase in PSA levels. Nevertheless, these tivation and aggregation (35–37). Identifiable risk factors in
changes were subtle (baseline: 0.85 ⫾ 0.06 ng/ml, 1.11 ⫾ 0.08 a prospective study involving 1000 stroke patients were ar-
ng/ml at month 6) and remained well within the normal terial hypertension, smoking, diabetes mellitus, hypercho-
range (22). lesterolemia, and high hematocrit (ⱖ50%) (38). However, it
Such findings may be explained by low intraprostatic remains unclear whether high hematocrit caused by T treat-
changes of hormone concentrations during T substitution ment has the same relevance as described in the stroke stud-
(32). ies because T exerts direct counterregulatory effects on he-
Prostate size of hypogonadal men being substituted with mostasis (39). Although hematocrit was observed to be
TU im increases to values within the normal range, corrob- higher in a recent metaanalysis of T substitution, the inci-
orating earlier findings with other T preparations (33). This dence of cardiovascular events was not increased in com-
effect is most visible within the first year of treatment and is parison with placebo-treated men (6). It is also known that
more pronounced in older men and those with short AR the T effect on hematocrit is dose as well as age related (40,
CAG repeats and/or larger changes of T levels (Table 2). We 41). In agreement, we demonstrate that effects of injectable
have previously demonstrated such effects in a cohort of 131 TU on hematocrit and erythropoiesis are augmented by high
men being treated with various androgen preparations (16). nadir T concentrations as well as advanced age (Tables 2 and
Erythropoiesis is stimulated by T supplementation (re- 3). In addition, we are able to report that short AR CAG

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 May 2015. at 00:15 For personal use only. No other uses without permission. . All rights reserved.
Zitzmann and Nieschlag • Safety of Testosterone Undecanoate J Clin Endocrinol Metab, October 2007, 92(10):3844 –3853 3851

repeats and high BMI contributed to the elevation of hemat- systolic blood pressure as well as resting heart rate decreased
ocrit during T substitution. Table 2 displays general, signif- significantly during treatment with TU. This new and im-
icant changes in hemoglobin content and hematocrit after portant observation is especially notable in the case of
initiation of therapy, and Table 3 exhibits risk markers for younger, slimmer men with higher androgen activity due to
elevation of these parameters above safety thresholds. Re- the CAG repeat AR polymorphism and/or higher ␦-T con-
markably, men with a BMI 30 kg/m2 or greater and/or CAG centration under treatment (Table 2). However, and similar
repeats less than 20 seem to have a markedly increased risk to the effects described in lipid metabolism, some patients
of developing blood hyperviscosity (Table 2 and Fig. 3). We exhibited safety parameters regarding blood pressure and
present a nonlinear pharmacogenetic model to estimate an- heart rate above the normal range and thus of inverse nature
drogen effects on hematocrit (Fig. 4). to the general treatment effect (Table 3). Blood pressure val-
We also considered estradiol concentrations as pivotal in ues (systolic and/or diastolic) or resting heart rates were
regard to obesity. Aromatase activity may be increased due elevated in men with higher BMI, lower nadir T concentra-
to augmented fat tissue in obese men and may thus cause tions, and attenuated androgen action due to the AR poly-
higher estradiol concentrations on T substitution therapy. morphism (Table 3).
Indeed, serum concentrations of estradiol under treatment In summary, a marked tendency to augment the effects of
were strongly related to BMI (see Results). Estradiol levels T substitution (in case of hematocrit) or inverse to the fa-
were used as additional independent variables for analyses vorable androgenic response (in case of lipid metabolism and
in second-step approaches and replaced BMI as confounder circulation) can be observed in obese hypogonadal men. The
of hemoglobin levels and hematocrit (Tables 2 and 3), sug- crude relative risk for hypogonadal men with a BMI 30
gesting an independent effect of this sex steroid. This finding kg/m2 or greater to be diagnosed with adverse safety pa-
supports a previous, similar observation (41). Preexisting rameters during treatment with TU is quite relevant (Fig. 3).
lung disease may present an additional factor in regard to the In principle, this finding reflects the generally increased car-
increase of hematocrit but was not diagnosed in our patients. diovascular risk of obese men and may not be directly related
Cigarette smoking did not exhibit a significant effect on the to T substitution (47).
red blood picture under T substitution. Because the effect of BMI on cardiovascular risk might not
In regard to lipid metabolism, the general effect induced be related to serum concentrations of total T, this might still
by androgen substitution within our study is a shift toward be the case for free T because BMI reduces SHBG. However,
favorable profiles with lower low-density lipoprotein (LDL)- overall levels of free T rather followed concentrations of total
cholesterol and higher high-density lipoprotein (HDL)-cho- T and results of analyses did not change (see Results and
lesterol, which seems to be especially the case in younger, Tables 2 and 3). Thus, observation of factors that can be
slimmer men receiving higher doses of T (Table 2). These considered adverse in regard to cardiovascular risk (high
changes in LDL-cholesterol concentrations are in agreement LDL-cholesterol, low HDL-cholesterol) were not related to
with a recent metaanalysis (5). Alterations in HDL-choles- total or free T but high BMI as such (Table 3). Rather, the
terol under T therapy are usually described as a slight de- general treatment trend was reduction of these parameters
cline. For example, a study conducted in younger hypogo- (Table 2) and related to higher T concentrations and lower
nadal men showed a decrease of HDL levels with T treatment BMI. In addition, insufficient T concentrations seemed to
(im and transdermal) relative to the untreated state (e.g. Refs. contribute to the observation of high blood pressure (Table
5, 42), whereas we report a significant increase, especially in 3). Free T levels tended to be higher in obese men (see Re-
younger, slimmer men. Such findings are in agreement with sults). Thus, it is, from a clinically intuitive point, unlikely
cross-sectional, noninterventional observations (reviewed in that free T itself contributed to the generation of unfavorable
Ref. 43). It can be speculated that our findings are due to the observations (Table 3). These observations are most likely
longer observation period of therapy, which may induce due to other factors inherent to obesity (47). However, a
favorable alterations in body composition. This aspect was statistical dissociation of obesity from levels of free T is not
not measured here but is known from previous observations possible within this setting. Hence, we cannot exclude that
in T-treated hypogonadal men (44, 45). In opposition to this in some obese men free T levels might have contributed to
general treatment effect within the total cohort (Table 2), we adverse effects, which may have been facilitated via lower
also observed some hypogonadal men with elevated LDL- SHBG and hence high levels of free T.
cholesterol and decreased HDL-cholesterol during androgen In addition, the relevance of the AR CAG repeat poly-
substitution: their specific characteristics are detailed in Ta- morphism in pharmacogenetics of T treatment seems to be
ble 3 and are due to a higher BMI (also see Fig. 3). Thus, a of clinical significance, as previously indicated (16, 17). The
higher body mass in hypogonadal men may attenuate or length of the repeat tract (normal range 9 –37) is inversely
even reverse general effects of T treatment in regard to lipid associated with the transcriptional and physiological activity
metabolism. induced by T in vitro (25), in mouse models (48), and in
An association between hypogonadism and arterial blood humans (15–17, 49, metaanalysis in Ref. 50). The modulatory
pressure in men has been suggested by a cross-sectional effect on androgen-dependent gene transcription is linear,
approach: inverse correlations between endogenous T levels given a fixed concentration of T over a range from 0 to 200
and systolic as well as diastolic blood pressure were de- repeats as seen in in vitro studies (51), and is probably me-
scribed in 356 healthy elderly men (46). Here we report a diated by a differential affinity of coactivator proteins to the
systematic approach to treatment effects of T substitution in encoded polyglutamine stretch of the actual AR protein, such
men and demonstrate favorable effects because diastolic and as ARA24 and p160 (26, 27). Because these proteins are ubiq-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 May 2015. at 00:15 For personal use only. No other uses without permission. . All rights reserved.
3852 J Clin Endocrinol Metab, October 2007, 92(10):3844 –3853
uitously but nevertheless nonuniformly expressed, the mod-
ulatory effect of the CAG repeat chain on AR target genes is
most likely not only dependent on androgenic saturation and
AR expression but also varies from tissue to tissue.
However, androgen effects seem to follow a nonlinear
pattern, when nadir total T levels under substitution therapy
with TU and AR CAG repeat length are simultaneously taken
into consideration: we present a nonlinear model derived
from general hormone kinetics to describe the effects of T
treatment on hematocrit (see Statistics and Fig. 4; also see
review in Ref. 52). Quite in agreement with previous clinical
findings (50), an AR CAG repeat length greater than 25 seems
to require T levels quite above the normal range and may
cause a picture of hypogonadism in the presence of normal
T levels.
According to the number of AR CAG repeats, the nadir T
level (respectively dosing interval of TU im) may be indi-
vidually modulated to achieve beneficial effects (Table 2) yet
simultaneously avoid an elevated hematocrit. According to
our data and a recent metaanalysis (6), this safety parameter
may deserve even greater recognition in comparison with
prostate issues when treating hypogonadal men.
In conclusion, this long-term observation demonstrates
that the use of the novel modality of T substitution by in-
jectable, long-acting TU is generally well tolerated. Special
attendance during T treatment in general is required in obese
men because adverse effects are more prevalent in this sub-
group. Pharmacogenetics exerted by the AR will be a future
option to tailor individual T dosage.
Acknowledgments
The authors thank Susan Nieschlag, M.A., for language editing and
Farid Saad (Bayer-Schering Health Care, Berlin, Germany) for providing
the medication.
Received March 20, 2007. Accepted July 9, 2007.
Address all correspondence and requests for reprints to: Eberhard
Nieschlag, Institute of Reproductive Medicine of the University, Muen-
ster D-48149, Germany. E-mail: eberhard.nieschlag@ukmuenster.de.
The trial has been retroactively entered into the www.clinicaltrials.
gov online database. Study ID nos.: IRM 96/17, EK 78a/97Nie1; last
updated: March 26, 2007; record first received, March 26, 2007;
ClinicalTrials.gov Identifier: NCT00452322; Health Authority: Ger-
many, Federal Institute for Drugs and Medical Devices.
Disclosures: Both authors received lecture fees of less than $10,000 in
regard to this topic. Both authors received consulting fees of less than
$10,000 in regard to this topic.
References
1. Bhasin S, Cunningham GR, Hayes FJ, Matsumoto AM, Snyder PJ, Swerdloff
RS, Montori VM 2006 Testosterone therapy in adult men with androgen
deficiency syndromes: an endocrine society clinical practice guideline. J Clin
Endocrinol Metab 91:1995–2010
2. Nieschlag E, Behre HM, Bouchard P, Corrales JJ, Jones TH, Stalla GK, Webb
SM, Wu FC 2004 Testosterone replacement therapy: current trends and future
directions. Hum Reprod Update 10:409 – 419
3. Nieschlag E 2006 Testosterone treatment comes of age: new options for hy-
pogonadal men. Clin Endocrinol (Oxf) 65:275–281
4. Nieschlag E, Swerdloff R, Behre HM, Gooren LJ, Kaufman JM, Legros JJ,
Lunenfeld B, Morley JE, Schulman C, Wang C, Weidner W, Wu FC 2005
Investigation, treatment and monitoring of late-onset hypogonadism in males:
ISA, ISSAM, and EAU recommendations. Int J Androl 28:125–127
5. Isidori AM, Giannetta E, Greco EA, Gianfrilli D, Bonifacio V, Isidori A,
Lenzi A, Fabbri A 2005 Effects of testosterone on body composition, bone
metabolism and serum lipid profile in middle-aged men: a meta-analysis. Clin
Endocrinol (Oxf) 63:280 –293
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 May 2015. at 00:15 For personal use only. No other uses without permission. . All rights reserved.
Zitzmann and Nieschlag • Safety of Testosterone Undecanoate
6. Calof OM, Singh AB, Lee ML, Kenny AM, Urban RJ, Tenover JL, Bhasin S
2005 Adverse events associated with testosterone replacement in middle-aged
and older men: a meta-analysis of randomized, placebo-controlled trials. J
Gerontol A Biol Sci Med Sci 60:1451–1457
7. Behre HM, Abshagen K, Oettel M, Hübler D, Nieschlag E 1999 Intramuscular
injection of testosterone undecanoate for the treatment of male hypogonadism:
phase I studies. Eur J Endocrinol 140:414 – 419
8. Nieschlag E, Büchter D, von Eckardstein S, Abshagen K, Simoni M, Behre
HM 1999 Repeated intramuscular injections of testosterone undecanoate for
substitution therapy in hypogonadal men. Clin Endocrinol (Oxf) 51:757–763
9. von Eckardstein S, Nieschlag E 2002 Treatment of male hypogonadism with
testosterone undecanoate injected at extended intervals of 12 weeks: a phase
II study. J Androl 23:419 – 425
10. Weinbauer GF, Partsch CJ, Zitzmann M, Schlatt S, Nieschlag E 2003 Phar-
macokinetics and degree of aromatization rather than total dose of different
preparations determine the effects of testosterone: a nonhuman primate study
in Macaca fascicularis. J Androl 24:765–774
11. Wistuba J, Marc Luetjens C, Kamischke A, Gu YQ, Schlatt S, Simoni M,
Nieschlag E 2005 Pharmacokinetics and pharmacodynamics of injectable tes-
tosterone undecanoate in castrated cynomolgus monkeys (Macaca fascicularis)
are independent of different oil vehicles. J Med Primatol 34:178 –187
12. Schubert M, Minnemann T, Hübler D, Rouskova D, Christoph A, Oettel M,
Ernst M, Mellinger U, Krone W, Jockenhövel F 2004 Intramuscular testos-
terone undecanoate: pharmacokinetic aspects of a novel testosterone formu-
lation during long-term treatment of men with hypogonadism. J Clin Endo-
crinol Metab 89:5429 –5434
13. Harle L, Basaria S, Dobs AS 2005 Nebido: a long-acting injectable testosterone
for the treatment of male hypogonadism. Expert Opin Pharmacother 6:1751–
1759
14. Coviello AD, Lakshman K, Mazer NA, Bhasin S 2006 Differences in the
apparent metabolic clearance rate of testosterone in young and older men with
gonadotropin suppression receiving graded doses of testosterone. J Clin En-
docrinol Metab 91:4669 – 4675
15. Zitzmann M, Nieschlag E 2003 The CAG repeat polymorphism within the
androgen receptor gene and maleness. Int J Androl 26:76 – 83
16. Zitzmann M, Depenbusch M, Gromoll J, Nieschlag E 2003 Prostate volume
and growth in testosterone-substituted hypogonadal men are dependent on
the CAG repeat polymorphism of the androgen receptor gene: a longitudinal
pharmacogenetic study. J Clin Endocrinol Metab 88:2049 –2054
17. Zitzmann M, Depenbusch M, Gromoll J, Nieschlag E 2004 X-chromosome
inactivation patterns and androgen receptor functionality influence phenotype
and social characteristics as well as pharmacogenetics of testosterone therapy
in Klinefelter patients. J Clin Endocrinol Metab 89:6208 – 6217
18. Wakisaka N, Taira Y, Ishikawa M, Nakamizo Y, Kobayashi K, Uwabu M,
Fukuda Y, Taguchi Y, Hama T, Kawakami M 2005 Effectiveness of finasteride
on patients with male pattern baldness who have different androgen receptor
gene polymorphism. J Investig Dermatol Symp Proc 10:293–294
19. Zitzmann M, Faber S, Nieschlag E 2006 Association of specific symptoms and
metabolic risks with serum testosterone in older men. J Clin Endocrinol Metab
91:4335– 4343
20. Vermeulen A, Verdonck L, Kaufman JM 1999 A critical evaluation of simple
methods for the estimation of free testosterone in serum. J Clin Endocrinol
Metab 84:3666 –3672
21. Behre HM, Yeung CH, Holstein AF, Weinbauer GF, Gassner P, Nieschlag E
2000 Diagnosis of male infertility and hypogonadism: transrectal ultrasonog-
raphy of the prostate gland and the seminal vesicles. In: Nieschlag E, Behre
HM, eds. Andrology: male reproductive health and dysfunction. 2nd ed.
Berlin, Heidelberg, Germany: Springer; 98
22. Wang C, Cunningham G, Dobs A, Iranmanesh A, Matsumoto AM, Snyder
PJ, Weber T, Berman N, Hull L, Swerdloff RS 2004 Long-term testosterone
gel (AndroGel) treatment maintains beneficial effects on sexual function and
mood, lean and fat mass, and bone mineral density in hypogonadal men. J Clin
Endocrinol Metab 89:2085–2098
23. Bhasin S, Woodhouse L, Casaburi R, Singh AB, Bhasin D, Berman N, Chen
X, Yarasheski KE, Magliano L, Dzekov C, Dzekov J, Bross R, Phillips J,
Sinha-Hikim I, Shen R, Storer TW 2001 Testosterone dose-response relation-
ships in healthy young men. Am J Physiol Endocrinol Metab 281:E1172–E1181
24. Klocker H, Gromoll J, Cato ACB 2004 The androgen receptor: molecular
biology. In: Nieschlag E, Behre HM, eds. Testosterone: action, deficiency,
substitution. 3rd ed. Cambridge, UK: Cambridge University Press; 39 –92
25. Mhatre AN, Trifiro MA, Kaufman M, Kazemi-Esfarjani P, Figlewicz D,
Rouleau G, Pinsky L 1993 Reduced transcriptional regulatory competence of
the androgen receptor in X-linked spinal and bulbar muscular atrophy. Nat
Genet 5:184 –188
26. Hsiao PW, Lin DL, Nakao R, Chang C 1999 The linkage of Kennedy’s neuron
disease to ARA24, the first identified androgen receptor polyglutamine region-
associated coactivator. J Biol Chem 274:20229 –20234
27. Irvine RA, Ma H, Yu MC, Ross RK, Stallcup MR, Coetzee GA 2000 Inhibition
of p160-mediated coactivation with increasing androgen receptor polyglu-
tamine length. Hum Mol Gen 9:267–274
28. Shores MM, Matsumoto AM, Sloan KL, Kivlahan DR 2006 Low serum
testosterone and mortality in male veterans. Arch Intern Med 166:1660 –1665
Zitzmann and Nieschlag • Safety of Testosterone Undecanoate
29. Keating NL, O’Malley AJ, Smith MR 2006 Diabetes and cardiovascular dis-
ease during androgen deprivation therapy for prostate cancer. J Clin Oncol
24:4448 – 4456
30. Amin S, Zhang Y, Felson DT, Sawin CT, Hannan MT, Wilson PW, Kiel DP
2006 Estradiol, testosterone, and the risk for hip fractures in elderly men from
the Framingham Study. Am J Med 119:426 – 433
31. Behre HM, von Eckardstein S, Kliesch S, Nieschlag E 1999 Long-term sub-
stitution therapy of hypogonadal men with transscrotal testosterone over 7–10
years. Clin Endocrinol (Oxf) 50:629 – 635
32. Marks LS, Mazer NA, Mostaghel E, Hess DL, Dorey FJ, Epstein JI, Veltri RW,
Makarov DV, Partin AW, Bostwick DG, Macairan ML, Nelson PS 2006 Effect
of testosterone replacement therapy on prostate tissue in men with late-onset
hypogonadism: a randomized controlled trial. JAMA 296:2351–2361
33. Behre HM, Bohmeyer J, Nieschlag E 1994 Prostate volume in testosterone-
treated and untreated hypogonadal men in comparison to age-matched nor-
mal controls. Clin Endocrinol (Oxf) 40:341–349
34. Zitzmann M, Nieschlag E 2004. Androgens and erythropoiesis. In: Nieschlag
E, Behre HM, eds. Testosterone: action, deficiency, substitution. 3rd ed. Cam-
bridge, UK: Cambridge University Press; 283–296
35. Lowe GD 1999 Rheological influences on thrombosis. Baillieres Best Pract Res
Clin Haematol 12:435– 449
36. Lowe GD, Forbes CD 1985 Platelet aggregation, haematocrit, and fibrinogen.
Lancet 16:395–396
37. Wannamethee G, Perry IJ, Shaper AG 1994 Haematocrit, hypertension and
risk of stroke. J Intern Med 235:163–168
38. Lee BI, Nam HS, Heo JH, Kim DI; Yonsei Stroke Team 2001 Yonsei Stroke
Registry. Analysis of 1,000 patients with acute cerebral infarctions. Cerebro-
vasc Dis 12:145–151
39. Zitzmann M, Junker R, Kamischke A, Nieschlag E 2002 Contraceptive ste-
roids influence the hemostatic activation state in healthy men. J Androl 23:
503–511
40. Bhasin S, Woodhouse L, Casaburi R, Singh AB, Mac RP, Lee M, Yarasheski
KE, Sinha-Hikim I, Dzekov C, Dzekov J, Magliano L, Storer TW 2005 Older
men are as responsive as young men to the anabolic effects of graded doses
of testosterone on the skeletal muscle. J Clin Endocrinol Metab 90:678 – 688
41. Dobs AS, Meikle AW, Arver S, Sanders SW, Caramelli KE, Mazer NA 1999
Pharmacokinetics, efficacy, and safety of a permeation-enhanced testosterone
transdermal system in comparison with bi-weekly injections of testosterone
JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the
endocrine community.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 May 2015. at 00:15 For personal use only. No other uses without permission. . All rights reserved.
J Clin Endocrinol Metab, October 2007, 92(10):3844 –3853 3853
enanthate for the treatment of hypogonadal men. J Clin Endocrinol Metab
84:3469 –3478
42. Dobs AS, Bachorik PS, Arver S, Meikle AW, Sanders SW, Caramelli KE,
Mazer NA 2001 Interrelationships among lipoprotein levels, sex hormones,
anthropometric parameters, and age in hypogonadal men treated for 1 year
with a permeation-enhanced testosterone transdermal system. J Clin Endo-
crinol Metab 86:1026 –1033
43. Barrett-Connor EL 1995 Testosterone and risk factors for cardiovascular dis-
ease in men. Diabetes Metab 21:156 –161
44. Snyder PJ, Peachey H, Hannoush P, Berlin JA, Loh L, Lenrow DA, Holmes
JH, Dlewati A, Santanna J, Rosen CJ, Strom BL 1999 Effect of testosterone
treatment on body composition and muscle strength in men over 65 years of
age. J Clin Endocrinol Metab 84:2647–2653
45. Woodhouse LJ, Gupta N, Bhasin M, Singh AB, Ross R, Phillips J, Bhasin S
2004 Dose-dependent effects of testosterone on regional adipose tissue dis-
tribution in healthy young men. J Clin Endocrinol Metab 89:718 –726
46. Fogari R, Preti P, Zoppi A, Fogari E, Rinaldi A, Corradi L, Mugellini A 2005
Serum testosterone levels and arterial blood pressure in the elderly. Hypertens
Res 28:625– 630
47. Schulte H, Cullen P, Assmann G 1999 Obesity, mortality and cardiovascular
disease in the Munster Heart Study (PROCAM). Atherosclerosis 144:199 –209
48. Albertelli MA, Scheller A, Brogley M, Robins DM 2006 Replacing the mouse
androgen receptor with human alleles demonstrates glutamine tract length-
dependent effects on physiology and tumorigenesis in mice. Mol Endocrinol
20:1248 –1260
49. Canale D, Caglieresi C, Moschini C, Liberati CD, Macchia E, Pinchera A,
Martino E 2005 Androgen receptor polymorphism (CAG repeats) and andro-
genicity. Clin Endocrinol (Oxf) 63:356 –361
50. Zeegers MP, Kiemeney LA, Nieder AM, Ostrer H 2004 How strong is the
association between CAG and GGN repeat length polymorphisms in the
androgen receptor gene and prostate cancer risk? Cancer Epidemiol Biomar-
kers Prev 13:1765–1771
51. Nakajima H, Kimura F, Nakagawa T, Furutama D, Shinoda K, Shimizu A,
Ohsawa N 1996 Transcriptional activation by the androgen receptor in X-
linked spinal and bulbar muscular atrophy. J Neurol Sci 142:12–16
52. Zitzmann M 2007 Mechanisms of disease: pharmacogenetics of testosterone
therapy in hypogonadal men. Nat Clin Pract Urol 4:161–166