Prostate cancer risk: The significance of differences in age related changes

in serum conjugated and unconjugated steroid hormone concentrations

between Arab and Caucasian men

E.O. Kehinde1, A.O. Akanji2, A. Memon3, A.A. Bashir4, A.S. Daar5, K.A. Al-Awadi1

& T. Fatinikun11Faculty of Medicine, Department of Surgery, Kuwait University, 24923, 13110 Safat, Kuwait; 2Faculty ofMedicine, Department of Pathology, Kuwait University, 24923, 13110 Safat, Kuwait; 3Faculty of Medicine,Department of Community Medicine, Kuwait University, 24923, 13110 Safat, Kuwait; 4Central Blood Bank,Jabriya, Kuwait; 5Sultan Qaboos University, Muscat, Oman

Abstract. Introduction: Factors responsible for the low incidence of clinical prostate cancer (3–8/100,000 men/year) in the Arab population remain unclear, but may be related to changes in steroidhormone metabolism. We compared the levels of serum conjugated and unconjugated steroids betweenArab and Caucasian populations, to determine if these can provide a rational explanation for differ-ences in incidence of prostate cancer between the two populations. Patients=Method: Venous bloodsamples were obtained from 329 unselected apparently healthy indigenous Arab men (Kuwaitis andOmanis) aged 15–80 years. Samples were also obtained from similar Arab men with newly diagnosedprostate cancer or benign prostatic hyperplasia (BPH). The samples were taken between 8:00 am and12:00 noon. Serum levels of total testosterone, (TT), sex hormone binding globulin (SHBG), freeandrogen index (FAI); adrenal C19-steroids, dehydroepiandrosterone sulphate (DHEAS) and andro-stenedione (ADT) were determined using Immulite kits (Diagnostic Systems Laboratories Inc, WebsterTexas, USA). The results obtained in Arab men were compared with those reported for similarly agedChinese, German and White USA men. Results: In all four ethnic groups, median TT and FAI declinedwith age, while SHBG increased with age. However, the mean TT and SHBG was significantly lower(p<0.01) and the FAI significantly higher in Arab men (p<0.01) compared to German men only in21–30 years age group. In the other age groups the levels of TT and SHBG were higher in the Germansbut the differences were not statistically significant. In all the racial groups serum levels of DHEAS andADT reached a peak by about 20 years of life, and then declined progressively. The mean DHEAS inAmerican Caucasians aged 20–29 years was 11.4 lmol/l compared to 6.22 lmol/l in the Arabs(p<0.001). The mean DHEAS in USA Caucasians aged 70–79 years was 2.5 lmol/l compared to1.8 lmol/l (p<0.03) in the Arabs. There was no significant difference in mean serum levels of DHEASbetween German and USA men. Similarly, there was no significant difference in the level of thehormones between Arab and Chinese men. Arab men with newly diagnosed prostate cancer had highserum TT, SHBG and DHEAS compared to those without the disease. Conclusions: The mean TT andSHBG was significantly lower in Arab men compared to Caucasian men especially in early adulthood.Caucasians have significantly higher serum levels of the precursor androgens DHEAS and ADTespecially in early adulthood compared to Arab men. These observations of low circulating androgensand their adrenal precursors in Arab men may partially account for the decreased risk for prostatecancer among Arab men.

Key words: Arabs, Caucasians, Epidemiology, Prostate cancer, Race, Steroids

International Urology and Nephrology (2006) 38:33–44 � Springer 2006DOI 10.1007/s11255-005-3619-1

Introduction

The incidence of clinical prostate cancer is low inAsia, intermediate in Africa and Eastern Europeand high in Western Europe and North Americaas illustrated in Figure 1 [1]. The factors respon-sible for the differences in incidence of prostatecancer in different parts of the world remainunclear, although it is well known that prostatecancer risk and male sex hormones are stronglyinterrelated. Significantly higher circulating tes-tosterone concentrations in young black mencompared with young adult white men have beenused to explain the differences in prostate cancerincidence between blacks and whites in the USA[2]. However, others have found no significantdifference in serum testosterone levels betweenblacks, whites or the Japanese even though theJapanese have a very low incidence of prostatecancer [3]. Steroid metabolism is affected by manyfactors. Hence to resolve the controversy, it mightbe worthwhile assessing other important factorswhich might explain the discrepancy. Correct elu-cidation of factors involved in prostate carcino-genesis is of importance in its chemoprevention [4].

Androgens are responsible for the differentia-tion and maturation of the male sexual organs aswell as the male secondary sexual characteristics.Approximately 90% of the androgens are pro-duced by the Leydig cells in the testes are secretedas testosterone, and the remainder are secreted bythe adrenal cortex as dehydroepiandrosterone(DHEA), which can be converted to testosteronein other tissues. Testosterone is the most impor-tant circulating androgen and its production bythe testis is regulated by a negative feedback

regulated by the luteinizing hormone (LH) and theluteinizing hormone releasing hormone (LHRH)via the gonad-hypothalamus – pituitary axis. Cir-culating testosterone is bound mostly to albumin(54%) in a low affinity fashion that will dissociateeasily, while 44% is bound to a sex hormonebinding globulin (SHBG) and only 1–2% is free.

Testosterone may be converted to more potentforms, either to estradiol by aromatization (lessthan 0.5% of the daily production of testosterone)or to the more potent androgen dihydrotestoster-one (DHT) by the enzyme 5a-reductase, which ispresent in many androgen target tissues such as theprostate. DHT is much more potent than testos-terone in promoting growth of the prostate; how-ever, testosterone is better in regulating itsdifferentiation. Testosterone enters the prostate cellby passive or active diffusion. Passively, testoster-one can enter the cell either from free testosteroneof through the disassociation of albumin near themembrane. Testosterone bound to SHBG requiresa membrane receptor in order to enter the cell.Once in the cytoplasm it will remain as testosteroneor transform to DHT by 5a-reductase and willbe attached predominantly to a cytoplasmicreceptor or androgen receptor (AR) or to a nuclearreceptor (estrogen receptor). The activated steroid-receptor complex binds to specific DNA segmentscalled hormone responsive elements, which arelocated in promoters of hormone-regulated genes.The receptor – DNA complex will associate withtranscriptional components and co-activators topromote gene transcription.

The aims of this study were to compare serumlevels of conjugated steroids (total testosterone(TT), SHBG, free androgen index (FA1)) andunconjugated steroids (dehydroepiandrosteronesulphate, DHEAS, androstenedione, ADT)between apparently healthy Arabs (Kuwaiti andOmani) with a reportedly very low incidence ofprostate cancer and German men with a reportedlyhigh incidence of the disease. Kuwaiti and OmaniArabs hereafter are referred to as Arabs. In addi-tion, the findings were compared with reportedvalues obtained from other racial/ethnic groups.

Patients and methods

Venous blood samples were obtained from 329unselected apparently healthy indigenous Arab

Figure 1. Incidence (ASR) of prostate cancer by World region(Parkins et al. [1]) (Figures not produced by Parkins et al. 1997but by us suing data Published by Parkins et al. 1997).

34

men aged 15–80 years, spread across the socialspectrum. Samples were collected between 8 amand 12 noon. None of the volunteers had anyknown serious systemic disease or was on medi-cations known to affect normal steroid hormoneproduction [5]. Other demographic data collectedfrom each volunteer included social habits, weight,height and if married, whether they had maleinfertility. Patients with male infertility wereexcluded from this analysis.

Serum TT and SHBG levels were determined byimmunoassay with chemiluminescent detection onan immulite analyser using appropriate kits(Diagnostic Systems Laboratories Inc. Webster,Texas, USA). FAI was derived from the formula100TT/SHBG. The results obtained from Arabmen were compared with those reported using thesame assay kits for 300 similarly aged German men[6]. The assay reproducibility data was: inter-assaycoefficient of variation (CV) 4.6% and intra-assayCV was 3.1%. Similarly, we compared serum tes-tosterone levels obtained in Arab and German menwith values obtained from other racial/ethnicgroups and correlated the TT levels with the inci-dence rates of prostate cancer in each ethnic group.Serum level of adrenal C19-steroids, DHEAS andADT were also determined using Immulite kits(Diagnostic Systems Laboratories Inc, WebsterTexas, USA). The results obtained in Arab menwere compared with those reported for similarlyaged Chinese, German and USA Caucasian men.

The levels of these hormones were also deter-mined in Arab men with newly diagnosed benignprostatic hyperplasia (BPH) and prostate cancer.The levels of the hormones were comparedbetween Arab men with and without BPH orprostate cancer.

Statistical analysis

Descriptive statistical methods were applied for theentire study. For parameters with normal (Gauss-ian) distribution, the reference range wasmean±2SD, while for those with non-normal(non-Gaussian) distribution, the values were log-transformed for analysis and the referencerange was 2.5–97.5 percentiles. Pearson product-correlation coefficients were calculated to explorethe association between serum testosterone andage, SHBG and age, FAI and age, DHEAS and ageand ADT and age.

Normograms portraying the distribution ofserum testosterone, SHBG, FAI, DHEAS andADT values as a function of age were generatedbased on the least squares regression models.Comparisons were then made between parametersobtained in Arabs, with those reported previouslyin USA Caucasians, German Caucasians and theChinese [5, 6]. These analyses were made using thecomputer programme SPSS (version 10.0). For allanalyses a p-value <0.05 was considered statisti-cally significant.

Result

Table 1 summarizes the age distribution and cor-responding mean±SD of serum TT, SHBG,DHEAS and ADT in healthy Arab men. The tableshows that testosterone decreases with age (Fig-ure 2), SHBG increases with age (Figure 3)DHEAS and ADT both decrease with age Fig-ures 4 and 5. Thus, there were significant negativerelationship between age and TT (r=)0.11,p<0.05), DHEAS (r=)0.54, p<0.0001) and ADT(r=)0.28, p<0.001). On the other hand, SHBGsignificantly increased with age (r=0.46,p<0.0001). The data for German men (Table 2)while quantitatively higher than in Arab men forTT and SHBG (P>0.06), showed a similar trend.Levels for FA1 were similar for both groups andshowed similar age trend. The percentage differ-ence in TT levels between the Germans and Arabmen decreased from 17.2% in the 21–30 years agegroup to 6% in the 61–70 year age group (Table 3).As also shown in Table 2, the difference in serumTT levels was significantly smaller after the age of40 years compared to below the age of 40. Fig-ures 6–11 show the frequency distribution patternof DHEA-S and ADT respectively in healthy Arabpopulation.

In the three ethnic groups serum levelsof DHEAS and ADT reached a peak by about20 years, and then declined progressively(Figures 4 and 5). The mean DHEAS in AmericanCaucasians aged 20–29 years was 11.4 lmol/lcompared to 6.22 lmol/l in the Arabs (p<0.001).The mean DHEAS in USA Caucasians aged70–79 years was 2.5 lmol/l compared to1.85 lmol/l (p<0.03) in the Arabs. There was nosignificant difference in mean serum levelsof DHEAS between German and USA men.

35

Similarly, there was no significant difference in thelevel of the hormones between Arabs and Chinesemen.

As shown in Table 1 Arab men with newlydiagnosed prostate cancer and BPH had higherserum TT and SHBG compared to age matchedcontrols without prostate cancer or BPH. Table 1also shows the calculated reference ranges forserum TT, SHBG, DHEAS and ADT in healthyArab men.

Discussion

There is ample circumstantial evidence from theliterature that androgens are implicated in thepathogenesis of prostate cancer [7]. Androgens areessential for the growth of the prostate gland, caninduce malignant proliferation of prostate cancercells in vitro and do produce prostate cancer inexperimental animals in vivo [7]. Furthermore,androgen deficiency before adulthood prevents the

g g p y

Table 1. Mean±SD of serum testosterone, SHBG, androstenedione, and DHEAS in healthy Arab males; In Arabs with BPH andnewly diagnosed prostate cancer and reference ranges for each analyte

Patient group/

age in years

Testosterone

(nmol/l)

SHBG

(nmol/l)

Androstenedione

(ng/ml)

DHEAS (umol/l)

Group Age range n Mean ±SD n Mean ±SD n Mean ±SD n Mean ±SD

A 15–19 49 13.61±6.75 36 19.88±11.39 26 2.66±1.07 36 6.60±3.50

B 20–29 59 15.73±8.76 60 21.13±9.75 25 2.82±1.12 26 6.22±2.54

C 30–39 53 12.83±5.55 50 21.31±8.99 22 2.38±0.68 27 4.52±2.54

D 40–49 59 13.49±6.10 56 24.72±10.89 16 2.09±0.61 26 4.17±2.28

E 50–59 66 12.45±5.94 51 33.38±19.39 22 2.32±0.82 44 3.00±1.86

F 60–69 34 10.8±8.44 35 33.51±15.15 19 2.07±0.91 25 2.65±1.73

G 70–79 7 9.63±6.36 7 58.83±19.97 3 1.80±0.23 6 1.85±0.87

P P=Prostate Cancer* 30 13.24±5.03 30 37.83±12.50 30 2.46±1.07 30 1.83±1.28

Z Z=BPH* 40 16.31±11.47 40 39.64±18.28 40 1.92±0.51 40 3.14±1.76

RR 3.0–31.0 6.5–65.7 0.5–4.3 2.2–15.2

*Mean age of patients with prostate cancer and BPH was 69.1 and 61.1 years respectively. RR=Reference ranges.

Figure 2. Serum testosterone as a function of age in healthy Arab men.

36

development of prostate cancer and andro-gen ablation is associated with prostate cancerremission in the majority of patients for about

18–24 months [8, 9]. A hormonal basis has beensuggested to explain the higher rate of prostatecancer among African-American men compared to

Figure 3. Serum SHBG as a function of age in healthy Arab men.

Figure 4. Serum FAI as a function of age in healthy Arab men.

37

Caucasians [10–12]. However, several studies havesought racial–ethnic differences in circulatingandrogen levels with conflicting results [2, 13, 14].Thus, while Ross et al. [2] found a statisticallysignificant higher TT (19%) and free testosterone(21%) levels in young African-American mencompared to Caucasians, Wright et al. [13], Ellisand Nyberg [14] found higher but statisti-cally insignificant testosterone, free testosteroneand SHBG levels in comparable groups ofAfrican-American men and Caucasians living invarious parts of the USA. These studies mayhowever been flawed by small sample size, forexample, Ross et al. [2] studied only 50 young men(20–30 years old) from each racial group and

Wright et al. [13] studied a total of 33 men in bothracial groups who were 20–40 years old. Otherflaws of some of the earlier studies include inap-propriate sample collection and storage, forexample while Ellis and Nyborg [14] studied 525African-American and 3654 Caucasian men,unfortunately the blood samples were collected atvarious times during the day in this study. It isnow well established that diurnal variation existsin circulating testosterone concentrations in menwith the highest levels found at 8.00 am and lowestlevels at 8.00 pm [15, 16]. All of these flaws may beresponsible for the conflicts in the associationbetween steroid and steroid metabolites levels andaetiology of prostate cancer. We avoided these

Figure 5. Serum androstenedione as a function of age in healthy Arab men.

Table 2. Mean testosterone, SHBG, FAI in various age groups in Kuwaiti and German men aged 20–70 years

Age groups (years) Testosterone (nmol/l) SHBG (nmol/l) FAI

Kuwaiti (n) German (n)+ Kuwaiti German+ Kuwaiti German+

21–30 15.73(59)* 19.0 (50)* 22.1 27.7 71.2 68.6

31–40 12.83 (53) 17.2 (50) 24.1 31.6 53.2 54.4

41–50 13.4 (59) 15.3 (50) 25.2 35.5 53.5 43.1

51–60 12.45 (66) 13.4 (50) 28.3 39.4 43 34

61–70 10.8 (34)** 11.49 (50)** 37.5 43.3 28.8 26.5

+Data from Elmlinger et al. [6]. Differences in serum total testosterone using paired t-test: Kuwait vs. German 21–30 years(p<0.001)*, 61–70 years (p=0.08)**.

38

problems by using rather large sample size andcollected samples in the fasting state between8:00 am and 12:00 mid-day. This method of col-lection was used in Germany as well [5]. Identicalanalytical kits were used in Germany and Kuwait.By correcting for these factors, Winters et al. [15],found a statistically significant higher levels of TTand SHBG in twenty three 18–24 years old Afri-can-American compared to 23 Caucasian men – afinding in keeping with the original observation ofRoss et al. [2].

Our finding of a statistically significant lowerserum TT and SHBG in Arab men aged20–30 years old may also provide a possible

explanation for the low incidence of prostatecancer in the Arab population compared toCaucasians of German origin. However, it ispertinent to note that the higher levels of testos-terone in German men compared to Arab menwas of statistical significance in the young agegroup (20–30 years old) where the percentagedifference in TT was 20.5% compared to 6% inthe 61–70 years age group. It had been suggestedthat, prolonged exposure (i.e. from <30 years ofage) to high androgen levels in African-Americanand Caucasians compared to the Chinese is thebasis for the observed difference in the effect ofandrogens in various racial/ethnic groups al-though other environmental, metabolic, dietaryand genetic factors may also have importantcontributory factors [17]. Based on the percentagedifference in mean serum testosterone in 21–30 year old German, their calculated risk forprostate cancer is expected to be about 2.5 timesthat of the Arabs [18]. From published series asindicated in Table 4, there would appear to be avery good correlation between the incidence ofprostate cancer in a racial group and serum tes-tosterone in young men from the particular racialgroup. A 15% difference in serum TT has beencalculated to increase the risk of prostate cancerby 1.9 times [18]. In a meta analysis of high

Table 3. The percentage difference in testosterone levelsbetween Kuwaiti and German men

Age group

(years)

Mean testosterone

(nmol/l)

% Difference

in testosterone

p

Kuwaitis Germans*

21–30 15.73 19.80 25.9 0.001

31–40 12.83 17.20 25 0.001

41–50 13.49 15.30 11.83 0.04

51–60 12.45 13.40 7.08 0.12

61–70 10.8 11.49 6 0.26

*Data from Elmlinger et al. [6].

Figure 6. Serum DHEAS as a function of age in healthy Arab men.

39

quality studies examining the hormonal predictorsof risk for prostate cancer in Caucasians, Gannet al. [11], Shaneyfelt et al. [21] concluded thatmen with either serum testosterone or IGF-1 inthe upper quartile of the population distributionhave an approximately twofold higher risk for

developing prostate cancer. As our study hasshown in the Arab population, the mean TTlevels for all age groups (i.e. age specific levels)are in the lower quartile for those of age-matchedGerman men. Consequently, Arab men would beexpected to have a reduced risk for developing

Figure 7. Androstenedione profile in healthy Arab males.

Figure 8. DHEAS profile in healthy Arab males.

40

prostate cancer. Interestingly enough, evenamong Arabs, the mean serum testosterone andSHBG for men with newly diagnosed prostatecancer was slightly higher (13.24±5.03 nmol/l and37.83±12.50 respectively compared to 12.45 nmol/l and 33.38±19.39 nmol/l) in age matched menwithout prostate cancer. Similarly Arab men whodeveloped BPH had higher testosterone andSHBG compared to Arab men without the disease.

SHBG is a circulating glycoprotein that is pro-duced by the liver and binds testosterone and othersteroids with high affinity and is thereby animportant predictor of the circulating TT con-centration in men. Thus, it follows from the higherlevel of SHGB in Caucasians that the TT level washigher also in Caucasian men. It is now believedthat increased testosterone levels affect risk fordeveloping prostate cancer by either increasing

Figure 9. Testosterone profile in healthy Arab males.

Figure 10. SHBG profile in healthy Arab males.

41

substrate availability for intra prostate DHT pro-duction and/or resulting in direct androgenic ac-tion on prostate epithelial cells [15, 21, 22]. This isalso in keeping with recent experimental data thatindicate that differences in levels of androgenicprecussors like testosterone levels rather than dif-ferences in 5a-reductase activity per se, areresponsible for the observed differences in prostatecancer incidence between White and Chinese sub-jects [17, 20].

In an attempt to further elucidate the mecha-nism by which androgens contribute to theinduction of prostate cancer, previous reports hadevaluated adrenal C19 steroid precursors which aretransformed into active androgens and oestrogens

in the peripheral steroid tissues [23]. To that extentwe measured DHEAS and ADT in this study, asthese are respectively the main circulating adrenalC19-steroids while ADT and androstane – 3a, 17b-diol (3a-DIOL) and androstane – 3b, 17b diol (3b-DIOL) are the final steroid metabolites in serum.In the Arab population as in other populationstudies there is a marked decline with age inDHEAS and ADT with the highest serum levelsobtained in the 15–19 year age group. The markedreduction in the formation of DHEAS by theadrenals with ageing results in reduced androgensynthesis in peripheral target tissues, which is alsothought to be associated with such disorders asinsulin resistance, cardiovascular diseases and

Figure 11. Free androgen index profile in healthy Arab males.

Table 4. Mean serum testosterone levels in young men (21–30 years) in various racial groups and the incidence of prostate cancer

Racial group Mean testosterone

(nmol/l)

% Difference

(compared to Kuwait)

n Incidence of

prostate cancer/100,000 men/year

Kuwaiti Arabs 15.73 59 6.5

German (Tubingen) 19.00 20 50 39.5

Africans (Nigerians)a 22.00 39.9 70 32–137a

USA Whitesb 18.60 18.2 23 108

USA Blacks 23.30 48.1 23 137

Chinese (Shanghai) 15.30c )2.7 50 2.3

aData from Lagos, Nigeria (Osegbe 1997 [19]). Probably an over estimation. bWinters et al. [15]. cJin et al. [20].

42

obesity [24, 25]. Moreover, low circulating levels ofDHEAS and DHEA have been found in patientswith breast and prostate cancer [26]. In this study,Arab patients with prostate cancer had lowerDHEAS (1.83±1.28 lmol/l) compared to simi-larly aged matched controls or those without thedisease (2.65±1.73) lmol/l. Furthermore, inanother study by Lookingbill et al. [27], youngAmerican Caucasian men aged 19–35 years, hadsignificantly higher levels compared to young Chi-nese (DHEAS (9.5±0.6 vs. 6.5±0.4 lmol/l,p<0.001) and ADT levels (4.1±0.2 vs.3.1±0.2 nmol/l, p<0.01). A recent analysisshowed that German men aged 20–29 years hadDHEA-S level of 9.28±0.78 compared to6.22±2.54 lmol/l for Arab men (p<0.001) [5]. Inthe study from Germany and Kuwait, the sameimmulite kits were used for analysis. The meanDHEAS in 70–79 years old American Caucasianswere higher than those of Arabs, although notstatistically significant. Therefore Caucasianswhether Americans or Germans had significantlyhigher DHEAS levels in comparison to Chineseand Arabs. There was no significant differencehowever in levels of DHEAS between American(9.5±0.6 lmol/l) and German (9.28±0.28 lmol/l)Caucasians, nor between Arabs (6.22±2.54 lmol/l) and Chinese (6.5±0.4 lmol/l). The reason forthis observation is not known, although it con-firms previous observations that urinary 17-ketosteroid levels are higher in Caucasians than inChinese and suggests additional metabolic differ-ences between the Chinese and Caucasians [27, 28].The differences observed between these racialgroups may therefore represent genetic differencesin the production of these hormones. Previouswork has shown that serum levels of 3a andro-stanediol glucuronide (3a–DIOL G) and DHEASare 65–84% determined by genetic factors [29, 30].These observations support the thesis that differ-ences between racial groups that we and othershave observed in DHEAS levels are most likelydue to genetic factors [27, 29, 30]. Other factors,including high fat diet and environmental factorsmay also be involved in the high levels of testos-terone, SHBG, DHEAS and ADT levels in Cau-casians compared to Arab and the Chinese. Howall these factors interplay and ways of manipulat-ing them to reduce the risk for induction of pros-tate cancer, that is, chemoprevention, remains aworthwhile research endeavour [4].

Conclusion

These data indicate that racial differences in menare established in early adulthood and influencecirculating SHBG and also TT levels. The findingof relatively low TT levels provides a possibleexplanation for the decreased risk for prostatecancer among Arab men. The further reports thatCaucasians have significantly higher serum levelsof the precursor androgens DHEAS and ADTespecially in early adulthood compared to Araband Chinese men, may further underlay the rela-tively low incidence of clinical prostate cancer inthese latter populations.

Acknowledgements

This work was supported by Kuwait UniversityResearch Grant MS01/99. We will like to thankMrs. Ramani Varghese, Mr. Mathew Abrahamand Dr. Mohammed Maher for specimen pro-cessing and analysis of data.

References

1. Parkins DM, Whelan SL, Ferlay J et al. Cancer incidence

in five continents. IARC Scientific Publications No. 143,

Lyons, France. IARC Scientific Publications 1997, 5–9.

2. Ross RK, Bernstein L, Judd H et al. Serum testosterone

levels in healthy young black and white men. JCNI 1986;

76: 45–48.

3. Ross RK, Bernstein L, Lobo RA et al. 5 alpha reductase

activity and risk of prostate cancer among Japanese and

US white and Black males. Lancet 1992; 339: 887–889.

4. Lieberman R. Evolving strategies for prostate cancer

chemoprevention trials. World J Urol 2003; 21: 3–8.

5. Petitclere C. Approval recommendation on the theory of

reference values. Part 2. Selection of individuals for the

production of reference values. J Clin Biol Chem 1987; 25:

639–644.

6. Elmlinger MW, Dengler T, Weinstock C, Kuhnel W.

Endocrine alterations in the ageing male. Clin Chem Lab

Med 2003; 41: 934–941.

7. Smolev JK, Coffey DS, Scott WW. Experimental models

for study of prostatic adenocarcinormas. J Urol 1997; 118:

216–220.

8. Hsing AW. Hormones and prostate cancer. Where do we

go from here? JCNI 1996; 88: 1093–1095.

9. Normura AM, Kolonel LN. Prostate cancer: a current

perspective. Epidemiol Rev 1991; 13: 200–207.

10. Ross RK, Coetzee GA, Reichavolt J et al. Does the racial–

ethnic variation in prostate cancer risk have a hormonal

basis? Cancer 1995; 75: 1778–1782.

43

11. Gann PH, Hennekens CH, Ma J, Stampfer MJ, Longcope

C. Prospective study of sex hormone levels and risk of

prostate cancer. JCNI 1996; 88: 1118–1126.

12. Pettaway CA. Racial differences in the androgen/andro-

gen receptor pathway in prostate cancer. J Natl Med

Assoc 1999; 91: 653–660.

13. Wright NM, Renault J, Willi S et al. Greater secretion of

growth hormone in black than in white men: possible

factor in greater bone mineral density. A clinical research

study. J Clin Endocrinol Metab 1995; 80: 2291–2292.

14. Ellis L, Nyberg H. Racial/ethnic variation in male tes-

tosterone levels; a probable contributor to group differ-

ences in health. Steroids 1992; 57: 72–75.

15. Winters SJ, Brufsky A, Weissfeld J et al. Testosterone, sex

hormone binding globulin and body composition in young

adult African American and Caucasian men. Metabolism

2001; 50: 1242–1247.

16. Veldhius JD, King J, Urban RJ et al. Operating charac-

teristics of the male hypothalamo-pituitary-gonadal axis:

pulsatile release of testosterone and FSH and their tem-

poral coupling with lutenising hormone. J Clin Encocrinol

Metab 1987; 65: 929–941.

17. Santner SJ, Albertson B, Zhang GY et al. Comparative

rates of androgen production and metabolism in Cauca-

sian and Chinese subjects. J Clin Endocrinol Metab 1998;

83: 2104–2109.

18. Pike MC, Krailo MD, Hendersan BE et al. Hormonal risk

factors, breast tissue age and the age incidence of breast

cancer. Nature 1983; 303: 767–770.

19. Osegbe DN. Prostate Cancer in Nigerians: facts and

nonfacts. J Urol 1997; 157: 1340–1343.

20. Jin B, Beilin J, Zajac J, Handelsman DJ. Androgen

receptor gene polymorphism and prostate zonal volumes

in Australian and Chinese men. J Androl 2000; 21: 91–98.

21. Shaneyfelt T, Husein R, Bubley G, Mantzoros CS. Hor-

monal predictors of prostate cancer: a meta-analysis.

J Clin Oncol. 2000; 18: 847–853.

22. Montie JE, Pienta K. Review of the role of androgenic

hormones in the epidemiology of prostate cancer and

benign prostatic hyperplasia. Urology 1994; 43: 892–

898.

23. Belanger A, Candas B, Dupomt A, Cusan L et al. Changes

in serum concentrations of conjugated and unconjugated

steroids in 40 to 80 years old men. J Clin Endocrinol

Metab 1994; 76: 1080–1090.

24. Schriok ED, Buffington CK, Hubert GD et al. Divergent

correlations of circulating dehydroepiandrosterone sul-

phate and testosterone with insulin levels and insulin

receptor binding. J Clin Endocrinol Metab 1988; 66:

1329–1331.

25. MacEwen EG, Kurzman ID. Obesity in the dog: role of

the adrenal steroid dehydroepiandrosterone. J Nutr 1991;

121: 551–555.

26. Stahl F, Schnorr D, Pilz C, Dorner G. Dehydroepiand-

rosterone levels in patients with prostatic cancer, heart

diseases and under surgery stress. Exp Clin Endocrinol

1992; 99: 68–70.

27. Lockingbill DP, Demers LM, Wang C, Leung A, Ritt-

master RS, Santen RJ. Clinical and biochemical parame-

ters of androgen action in normal healthy Caucasian

verses Chinese subjects. J Clini Endocrinol Metab 1991;

72: 1242–1248.

28. Kobayashi T, Lobotsky J, Lloyd CW. Plasma testosterone

and urinary 17-keto steroids in Japanese and occidentals.

J Clin Endocrinol Metab 1966; 26: 610–614.

29. Meikle AW, Bishop DT, Stringham JD, West DW.

Quantitating genetic and non genetic factors that deter-

mine plasma sex steroid variation in normal male twins.

Metabolism 1987; 35: 1090–1095.

30. Potter JI, Wong FI, Lifrak ET, Parker LN. A genetic

compment to the variation of DHEA SO4. Metabolites

1985; 34: 514–517.

Address for correspondence: Dr. E.O. Kehinde, Faculty of

Medicine, Department of Surgery, Kuwait University, P.O. Box

24923, 13110 Safat, Kuwait

Phone: +965-531-9475; Fax: +965-531-9597/+965-533-8638

E-mail: ekehinde@hsc.edu.kw

44