; Urinary Prostate Specific Antigen Is Noninvasive Indicator of
Learning Center
Plans & pricing Sign in
Sign Out
Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Urinary Prostate Specific Antigen Is Noninvasive Indicator of


  • pg 1
									Journal of Andrology, Vol. 28, No. 1, January/February 2007
Copyright E American Society of Andrology

Urinary Prostate-Specific Antigen Is a Noninvasive Indicator of
Sexual Development in Male Children
From the *Scientific Crime Laboratory, Kanagawa Prefectural Police, Yokohama, Japan; the ÀDepartment of Legal
Medicine, St Marianna University School of Medicine, Kawasaki, Japan; the `Department of Microbiology and
Immunology, The Institute of Medical Science, Tokyo University, Tokyo, Japan; the 5Department of Pediatrics,
St Marianna University School of Medicine, Kawasaki, Japan; the "Department of Cellular and Molecular Biology,
Primate Research Institute, Kyoto University, Inuyama, Japan; and the #Department of Urology, St Marianna
University School of Medicine, Kawasaki, Japan.

ABSTRACT: Testicular androgen induces the synthesis of pros-                    children aged 0.3 to 9 years, after which a gradual increase in urinary
tate specific antigen (PSA) in acinar epithelial cells of the prostate.         PSA activity was observed after 10 years of age. Urinary PSA
We examined PSA activity in urine from 136 male children from birth             activity was markedly persistent after Tanner stage III pubertal
up to 17 years of age. We detected PSA at various intervals in early            development. To our knowledge, this is the first study to demonstrate
infant urine over a period of 1–4 months. During this period, urinary           an induction of PSA during early infancy by bioactive T in normally
secretion of testosterone (T) gradually declined, accompanied by 1              developing human males. We conclude that urinary PSA is a non-
or more surges of T prior to a transient increase in PSA in urine from          invasive, useful indicator for developmental studies from neonatal
full- and preterm infants (67%, n 5 6). Although mean urinary T                 and adolescent males, which can be measured with a confirmatory
concentrations during elevations of PSA in preterm infants were 3.1             semiquantitative PSA assay.
and 5.6 times greater than in full-term infants and adults, the overall           Key words: Urine, infant, puberty.
mean urinary PSA concentration of full and preterm infants was just               J Androl 2007;28:150–154
45% and 18% that of adults, respectively. PSA was not detected in

A     ndrogen is required for the development of
      secondary sexual characteristics as well as de-
velopment of the reproductive organs, and also influ-
                                                                                velopment of the external genitalia and sexual behavior,
                                                                                following administration of gonadotropin-releasing hor-
                                                                                mone (GnRH) agonists and antagonists in nonhuman
ences the development of social sexual behavior (Mann                           primates (Mann et al, 1989; Mann et al, 1993; Wallen et
and Fraser, 1996). Transitory testosterone (T) secretion                        al, 1995; Mann and Fraser, 1996; Brown et al, 1999).
occurs during early infancy following the appearance of                         Although T surges during the first 3 months of life may
luteinizing hormone (LH) and follicle-stimulating hor-                          influence the later onset of puberty in human males,
mone (FSH). Work by several investigators has shown                             direct evidence regarding the biological function of T
effects on testicular development, as well as on de-                            during early infancy has yet to be established (Forester
                                                                                et al, 1974).
                                                                                   Prostate-specific antigen (PSA; hK3 for a chymotryp-
   Supported by Grants-in-Aid for scientific research of IS (15922076
and 16922185) from the Japan Society for the Promotion Science as
                                                                                sin-like serine protease) is a member of the human
well as a 21st Century COE Research grant (Kyoto University, A14)               kallikrein (hK) family and might be a useful androgen
from the Ministry of Education, Science, Sports and Culture, Japan,             indicator of pubertal development in boys. Because
and in part by a Cooperation Research Program of Primate Research               testicular androgen induces the synthesis of PSA in
Institute, Kyoto University.
                                                                                acinar epithelial cells of the prostate (Stamey et al,
   § Present address: Research Institute, International Medical Center
of Japan.                                                                       1989), PSA can be detected in both semen and adult
   Correspondence to: Dr Itaru Sato, Forensic Biology Unit, Scientific          male urine (Sato et al, 2002). In addition, T elevation at
Crime Laboratory, Kanagawa Prefectural Police, 155-1, Naka-ku,                  the onset of puberty might induce permanent synthesis
Yamashita-cho, Yokohama, 231-0023, Japan (e-mail: itaru-s@m2.                   of PSA in the prostate (Kim et al, 1999). Urine can
   Received for publication April 7, 2006; accepted for publication
                                                                                usually be obtained noninvasively; therefore, measure-
August 28, 2006.                                                                ment of T and PSA in urine (Irani et al, 1996, 1997;
   DOI: 10.2164/jandrol.106.000190                                              Barret et al, 2002; Irani et al, 2005) may prove a useful

Sato et al     N   Urinary PSA Assay and Sexual Development                                                                    151

Table 1. Urinary PSA in relation to pubertal stage in 59
healthy boys*

Tanner             .1       1–11      12          13         ,14
  I          0 (6) 1+ (2)   0 (35)   0 (4)       0 (1)
  II                         0 (2)            0 (1) 4+ (1)
  III                                4+ (1)      4+ (2)      4+ (2)
  IV                                             4+ (1)
  V                                                          4+ (1)
 * Urinary PSA activity from 59 individuals was screened using
SMITEST PSA card with a detection range of 0 to 4+.

indicator by which to follow sexual development in male
  We therefore investigated whether male sexual de-
velopment from birth to adolescence can be evaluated
using urinary PSA levels.

                                                                      Figure 1. Urinary PSA activity from birth to adolescence was
Materials and Methods                                                 screened using the immunochromatographic membrane test for
                                                                      PSA on the SMITEST PSA card. The line is a polynominal
We obtained morning (0930–1200 hours) spot urine samples
from 92 boys between the ages of 2 months and 14 years with
                                                                      regression fit to the data points ( ).

nonendocrinological conditions attending the Pediatrics Unit          changes in total PSA concentrations in 6 male infants. The
at St Marianna University School of Medicine, Kawasaki,               analytical sensitivity of this system is 0.003 ng/mL. Total
Japan, as well as from 44 healthy male volunteers between 6           urinary T levels were measured using a commercially available
and 18 years of age. Clinical data was obtained for these boys        solid-phase radioimmunoassay (Diagnostic Products Corp)
by their pediatricians, including height, body weight, testicular     according to the manufacturer’s instructions, with slight
mass, and development of pubic hair, after which they were            modifications according to the method of Huhtaniemi et al
classified into 5 groups, spanning Tanner stages I to V               (1986). Prior to analysis, urine samples were treated with
(Table 1). We obtained morning urine samples from 6 male              12 N NaOH to ensure hydrolysis and then boiled for
infants from birth up to 18 weeks of age at weekly intervals.         15 minutes. PSA and T levels were expressed as micrograms
We obtained the history of each individual, including his date        (mg) per millimole (mmol) of urinary creatinine (Irani et al,
of birth, expected date of confinement, birth weight, and             1997).
feeding manner, as well as the parity of the mother. We also
obtained spot urine samples from 6 female infants (aged 0.3–
1.2 years) and 6 pubertal girls (aged 8–12) as negative controls
in this study. A local research ethics committee approved the
study, and the parents or guardians of the children provided
written informed consent to all procedures associated with the        We measured urinary PSA activity in normal urine
study. We measured PSA activity in spot urine samples using           samples from birth (age: 0 days) up to adolescence (age:
a SMITEST PSA card (Seratec Diagnostica, Gottingen,                   17 years) using an immunochromatographic membrane
Germany) and used an immunochromatographic membrane                   PSA test, which can visually identify the presence of
method for semiquantitative screening. Although the SMIT-             PSA (negative: 0; positive range: from 1+ to 4+), as
EST PSA card has a limit of detection of 4 ng/mL PSA, a faint         shown in Figure 1. Within 4 months of birth, faint PSA
immunoreaction still appears in the range of 1–3 ng/mL                activity appeared in 20 (25%) of 80 urine samples (n 5
according to the manufacturer. Therefore, all responses were          14, PSA range: 1–3 ng/mL). Although undetectable in
compared with the reference sample (4 ng/mL) and quantified
                                                                      urine from children between the ages of 4 months and
as follows: PSA 4+ (4 ng/mL), PSA 3+ (3 ng/mL), PSA 2+
                                                                      9 years (n 5 64, 73 urine samples), urinary PSA activity
(2 ng/mL), PSA + (1 ng/mL), or PSA negative (2) (Sato et al,
2002). In addition, we measured PSA activity in spot urine            was identified in 11 (32%) of 34 boys (34 urine samples)
samples using an IMMULITE third-generation PSA enzyme-                aged between 10 and 12 years (PSA range: 1–3 ng/mL).
amplified luminescence system (Diagnostic Products Corp,              Thereafter, urinary PSA was typically detected in
Los Angeles, Calif) according to the method of Randell et al          individuals over 13 years of age (n 5 24, 24 urine
(1996) as an ultrasensitive quantitative assay for monitoring         samples, PSA range: greater than 4 ng/mL). When the
152                                                                                Journal of Andrology         N    January ÙFebruary 2007

                                                                       Table 2. Comparison of reproductive parameters between
                                                                       full-term and preterm delivery during early infancy*

                                                                       Delivery                Testosterone                    PSA
                                                                       Full-term        6.11   6   4.62 (n 5 42)        1.31 6 1.65 (n 5 7)
                                                                                        4.71   6   2.56 (n 5 7)4
                                                                       Preterm         19.16   6   18.29 (n 5 29)§      0.53 6 0.40 (n 5 9)
                                                                                       14.44   6   7.94 (n 5 7)
                                                                       Adult3           3.42   6   1.85 (n 5 5)         2.92 6 2.63 (n 5 5)
                                                                         * Values are expressed as micrograms per millimole of urinary
                                                                       creatinine (mean 6 SD).
                                                                         3 Adult urine samples were obtained from 5 healthy individuals
                                                                       aged in their thirties.
                                                                         4 Testosterone levels (the lower values) selected from the data
                                                                       points at each PSA peak.
                                                                         § For statistical differences, testosterone levels in early infancy
                                                                       (the upper values) were compared with full-term and preterm delivery
                                                                       results using Student’s paired t test (P , .01).
Figure 2. Trends in urinary PSA and testosterone concentrations in 6
neonates. Urinary PSA and T levels were measured using an
enzyme-amplified luminescence method known as IMMULITE Third
Generation (Randell et al, 1996), and a solid-phase radioimmuno-       Discussion
assay, respectively. PSA and T levels were expressed as micro-
grams per millimole of urinary creatinine. Symbols indicate testos-
terone ( ), PSA (#), and expected date of confinement (m). Values
within parentheses indicate birth weight. Arrows indicate T surge.
                                                                       Urinary PSA activity was examined during infancy and
                                                                       adolescence, as shown in Figure 1. Our results support
                                                                       those of an immunohistochemical study demonstrating
                                                                       PSA activity in the lumina of primary prostatic ducts, as
relationship between urinary PSA activity and puber-                   well as the transitional epithelium of infants aged
tal stage was examined (Table 1), PSA activity was                     between 0 and 6 months (Goldfarb et al, 1986), with
detected in pubertal boys meeting the criteria for Tanner              detectable changes in PSA activity in neonatal serum
stages II and III (aged 12–13 years), as well as boys                  (Randell et al, 1996). This prior study also suggests
meeting the criteria for Tanner stage I development                    a lack of PSA expression in prostate tissue between
(aged less than 1 year). PSA activity was not detected in              6 months and 10 years of age, after which increased
urine from 6 female infants and 6 pubertal girls (data                 levels are observed until puberty (Goldfarb et al, 1986).
not shown).                                                            Furthermore, a polynomial regression line of PSA
   Following this, we quantitatively monitored urinary T               production parallels T production from birth until
and PSA levels in 6 male infants from birth to 18 weeks                puberty (Mann and Fraser, 1996).
at weekly intervals (Figure 2). Although T levels in full-                PSA reappears at the onset of development of
term infants (Figure 2, numbers 1, 2, and 3) remained                  secondary sexual characteristics in boys aged between
low throughout early infancy, high levels were observed                10 and 12 years (Randell et al, 1996; Sato et al, 2002),
in preterm infants (Figure 2, numbers 4, 5, and 6), after              a period corresponding to testicular growth and the
which levels were noted to decline with age. In all                    development of pubic hair (Rosenfield, 1990). Although
infants, a peak in PSA occurred once (Figure 2, numbers                increased serum PSA activity corresponds to increased
1 and 5), 2–4 times (Figure 2, numbers 2 and 6), or                    serum T levels in healthy pubertal boys that have
continuously (Figure 2, number 4) in each individual                   reached Tanner stages III (T-III) development, PSA
prior to age 9 weeks.                                                  activity in boys with precocious puberty exceeds that of
   We compared urinary T and PSA levels during early                   healthy boys (T-II), even in subjects under 10 years of
infancy among full-term and preterm infants (Table 2).                 age (Viera et al, 1994; Juul et al, 1997; Kim et al, 1999).
Urinary T levels in preterm infants were significantly                 The present data of boys aged 12 indicates greater
higher during early infancy than in full-term infants.                 urinary PSA activity in taller boys with increased
Although mean urinary T concentrations during each                     testicular mass and body weight, compared to shorter
PSA peak in preterm infants were 3.1- and 4.2-fold                     boys of similar (Table 1, Tanner stage I–III). In shorter
those in full-term infants and adults, respectively, mean              boys of lesser weight, weak or absent urinary PSA
PSA concentrations were only 45% and 18% those of                      activity was detected (Table 1, Tanner stage I at
adults.                                                                13 years of age). The onset of puberty is well correlated
Sato et al   N   Urinary PSA Assay and Sexual Development                                                                 153

with bone density, and T plays a role in skeletal               Although T levels in full-term infants remained low
development (Rosenfield, 1990; Mann et al, 1993).            throughout early infancy in the present study, preterm
Therefore, abnormal androgen production in boys              infants showed elevated levels, which declined with age.
with a delayed mean age of onset of puberty might            When T levels in full-term infants were compared with T
influence PSA production as well as sexual behavior          levels in preterm infants, mean levels were 3.1-fold lower
(Rosenfield, 1990; Mann et al, 1993; Wallen et al, 1995).    throughout early infancy (Table 2). Significant differ-
Measuring urinary PSA activity and urinary T with            ences in T levels according to mode of delivery were also
respect to pubertal development might be useful in           observed (t test; P , .01). However, T concentrations
examining their relationship with the various Tanner         did not appear to be related to PSA activity following
stages. Lack of detection of PSA in the urine of boys        either mode of delivery (Figure 2 and Table 2). There-
aged 4 months to 9 years might reflect a decreased           fore, T levels may not modulate PSA activity. In
production of sex steroid hormones during this period.       addition, all individuals initially exhibited urinary PSA
As of yet, the physiological significance of PSA             within a few weeks after the first T surge (arrows in
production between 1 and 4 months of age remains             Figure 2). This indicates that PSA synthesis by acinar
unclear. The present findings suggest that synthesis of      epithelial cells of the prostate may be promoted and/or
PSA in acinar epithelial cells of the prostate might be      enhanced by the T surge (Goldfarb et al, 1986).
promoted and enhanced by peaks in T (Goldfarb et al,            Although the mature prostate of the adult does not
1986).                                                       require abundant T to synthesize PSA, increased stimu-
   Recently, Barrett et al (2002) reported similar changes   lation by T might be required for synthesis in the imma-
in urinary and serum T levels in nonhuman primates. To       ture prostate in early infancy (Table 2). In full-term
better understand the relationship between T and PSA,        infants (Table 2, numbers 1, 2, and 3), PSA appeared
we quantitatively monitored urinary T and PSA levels in      uniformly in urine from 5 weeks after birth, while
male infants from birth up to 18 weeks of age                detection of urinary PSA was delayed in preterm infants
(Figure 2). Although a direct comparison between             (Table 2, numbers 4, 5, and 6). However, PSA activity
urinary T concentrations (Figure 2) and previous             appeared in urine at 5 weeks of age, around the normal
reported salivary T concentrations (Huhtanieme et al,        time of birth.
1986) could not be made to give differences in reporting        We first demonstrated that an adult reproductive
results, similar trends were observed because urinary T      characteristic is evident in male neonates by measuring
was expressed as micrograms per millimole of urinary         urinary T and PSA. Although PSA is also detected in
creatinine. Transient PSA peaks were observed in all         neonatal serum (Sato et al, 2002), it can be difficult to
subjects prior to 9 weeks of age, indicating fluctuations    obtain from infants and children; thus, the presence of
in urinary PSA during this period. During the neonatal       PSA in urine is of considerable interest. Urine can
period, urinary PSA levels reached adult levels (Ta-         usually be directly obtained from individuals with
ble 2). The highest (H) and lowest (L) urinary PSA           minimal damage or stress. The PSA assay used in this
values prior to correction for creatinine were 9.1- and      study might be used to monitor PSA levels in
600-fold greater than serum PSA levels during this           accordance with various stages of sexual development
period (0.007 mg/L; Randell et al, 1996). Urinary PSA        in the future. However, PSA activity within spot urine
reached a high of 4.25 ng/mL 5 weeks after birth in          samples is difficult to detect and may fluctuate during
subject number 1, and a low of 0.064 ng/mL 4 weeks           the neonate period (eg, a full-term infant; Figure 2,
after birth in subject number 4. This indicates that         number 3). Therefore, functional studies of the imma-
almost all of the PSA synthesized by acinar epithelial       ture prostate, specifically hormonal and protein assays,
cells of the prostate was concentrated in urine. In          should be used performed on pooled urine rather than
addition, we did not observe any PSA activity in the         spot urine samples.
urine of females during the neonatal period using               Further investigation into the responsiveness of the
a SMITEST PSA card. These results suggest that               prostate to androgen, as well as the behavioral effects of
androgen activity in male neonates might stimulate the       PSA production, during the neonatal period, is required.
prostate to produce PSA, and that PSA might modulate
the interaction between male infants and their mothers.
During the first year of life in rhesus monkeys, females
spend more time close to their mothers than males, and
                                                             This study was performed by the Cooperation Research Program of
neonatal T-suppressed males show more maternal
                                                             Primate Research Institute, Kyoto University. The authors wish to
dependence, while T-augmented males exhibit less             thank Miss Mieko Askaham, M Sci, Department of Pediatrics, St
maternal interaction and more independence (Wallen           Marianna University School of Medicine, for providing excellent
et al, 1995).                                                secretarial support during the preparation of the manuscript.
154                                                                                 Journal of Andrology          N   January ÙFebruary 2007

                                                                           Kim MR, Gupta MK, Travers SH, Rogers DG, Van Lente F, Faiman
References                                                                    C. Serum prostate specific antigen, sex hormone binding globulin
Barrett GM, Shimizu K, Bardi M, Mori A. Fecal testosterone                    and free androgen index as markers of pubertal development in
   immunoreactivity as a non-invasive index of functional testoster-          boys. Clin Endocrinol. 1999;50:203–210.
   one dynamics in male Japanese macaques (Macaca fuscata).                Mann DR, Akinbami MA, Gould KG, Tanner JM, Wallen K.
   Primates. 2002;43:29–39.                                                   Neonatal treatment of male monkeys with a gonadotropin-re-
Brown GR, Nevison CM, Fraser HM, Dixson AF. Manipulation of                   leasing hormone agonist alters differentiation of central nervous
   postnatal testosterone levels affects phallic and clitoral develop-        system centers that regulate sexual and skeketal development. J Clin
   ment in infant rhesus monkeys. Int J Androl. 1999;22:119–128.              Endocrinol Metab. 1993;76:1319–1324.
Forest MG, Sizonenko PC, Cathiard AM, Bertrand J. Hypophyso-               Mann DR, Fraser HM. The neonatal period: a critical interval in male
   gonadal function in humans during the first year of life. J Clin           primate development. J Endocrinol. 1996;149:191–197.
   Invest. 1974;53:819–828.                                                Mann DR, Gould KG, Collins DC, Wallen K. Blockade of neonatal
Goldfarb DA, Stein BS, Shamszadeh M, Petersen RO. Age-related                 activation of the pituitary-testicular axis: effect on prepubertal
   changes in tissue levels of prostatic acid phosphatase and prostate        luteinzing hormone and testosterone secretion and testicular
   specific antigen. J Urol. 1986;136:1266–1269.                              development in male monkeys. J Clin Endocrinol Metab. 1989;68:
Huhtanieme I, Dunkel L, Perheentupa J. Transient increase in                  600–607.
   postnatal testicular activity is not revealed by longitudinal           Randell EW, Diamandis EP, Ellis G. Serum prostate-specific
   measurements of salivary testosterone. Pediatric Res. 1986;20:             antigen measured in children from birth to age 18 years. Clin
   1324–1327.                                                                 Chem. 1996;42:420–423.
Irani J, Millet C, Levillain P, Dore B, Begon F, Aubert J. Serum-to-
                                    ´                                      Rosenfield RL. Clinical review 6: diagnosis and management of
   urinary prostate-specific antigen ratio: a potential means of              delayed puberty. J Clin Endocrinol Metab. 1990;70:559–562.
   distinguishing benign prostatic hyperplasia from prostate cancer.       Sato I, Sagi M, Ishiwari A, Nishijima H, Ito E, Mukai T. Use of the
   Eur Urol. 1996;29:407–412.                                                 ‘‘SMITEST’’ PSA card to identify the presence of prostate-specific
Irani J, Millet C, Levillain P, Dore B, Begon F, Aubert J. Serum-to-
                                    ´                                         antigen in semen and male urine. Forensic Sci Int. 2002;127:71–74.
   urinary prostate specific antigen ratio: its impact in distinguishing   Stamey TA, Kabalin JN, Ferrari M, Yang N. Prostate specific antigen
   prostate cancer when serum prostate specific antigen level is 4 to         in the diagnosis and treatment of adenocarcinoma of the prostate.
   10 ng./ml. J Urol. 1997;157:185–188.                                       IV. Anti-androgen treated patients. J Urol. 1989;141:1088–1090.
Irani J, Salomon L, Soulie M, Zlotta A, de la Taille A, Dore B, Millet
                                                              ´            Vieira JGH, Nishida SK, Pereira AB, Pereira AB, Arraes RF,
   C. Urinary/serum prostate-specific antigen ratio: comparison with          Verreschi ITN. Serum levels of prostate-specific antigen in normal
   free/total serum prostate-specific antigen ratio in improving              boys through puberty. J Clin Endocrinol Metab. 1994;78:1185–
   prostate cancer. Urology. 2005;65:533–537.                                 1187.
Juul A, Muller J, Skakkebæk NE. Prostate specific antigen in boys          Wallen K, Maestripieri D, Mann DR. Effects of neonatal testicular
   with precocious puberty before and during gonadal suppression by           suppression with a GnRH antagonist on social behavior in group-
   GnRH agonist treatment. Eur J Endcrinol. 1997;136:401–405.                 living juvenile rhesus monkeys. Horm Behav. 1995;29:322–337.

To top