- Email: [email protected]
Comparison of serum prostate-specific antigen levels and PSA density in african-american, white, and hispanic men without prostate cancer
ADULT U R O L O G Y
ELSEVIER
COMPARISON OF SERUM PROSTATE-SPECIFIC ANTIGEN LEVELS AND PSA DENSITY IN AFRICAN-AMERICAN, WHITE, AND HISPANIC MEN WITHOUT PROSTATE CANCER IBRAHIM ABDALLA, PAUL RAY, VERA RAY, FLORIN VAIDA, AND SRINIVASAN VIJAYAKUMAR
ABSTRACT Objectives. To determine the influence of race or ethnicity on serum prostate-specific antigen (PSA) levels and PSA density (PSAD) in a population of healthy men without clinically evident prostate cancer. Methods. This retrospective study was conducted between January 1988 and January 1993. The serum PSA levels were measured in 859 men (586 African Americans, 142 whites, and 131 Hispanics) who were participants in a prostate cancer screening program or had urinary symptoms suggestive of prostate gland pathology. All men underwent a detailed clinical examination, including digital rectal examination, serum PSA determination, and transrectal ultrasound (TRUS). None of the subjects included had clinical or TRUS evidence of prostate cancer (furthermore, 283 men were pathologically proved to be cancer-free by prostate biopsies). Serum PSA levels and PSA densities as a function of each individual's ethnic background were determined. Results. The mean serum PSA level in African Americans was 2.1 ng/mL, which was significantly higher than that of whites (mean PSA of 1.53 ng/mL) and Hispanics (mean PSA of 1.83 ng/mL) (P = 0.003). Similar differences among the three groups were observed in PSA density (the mean PSAD was 0.078, 0.057, and 0.065 for African Americans, whites, and Hispanics, respectively). A separate analysis for the biopsynegative men was performed, and the findings were consistent with the observations for the entire study group. After adjustment for age and prostate volume, the differences remained statistically significant. Conclusions. Among men without evidence of prostate cancer, African Americans have higher serum PSA levels and PSA densities than do whites or Hispanics. Race or ethnicity was an independent factor that affected serum PSA levels even after adjustment for age and prostate volume. UROLOGY 51: 300-305, 1998. © 1998, Elsevier Science Inc. All rights reserved.
cancer (PC) is the most common nonp rostate skin cancer and the second leading cause of cancer-related deaths in American men. It is estb mated that 334,500 men will be diagnosed with PC and 41,800 will die of this disease in 1997.1 PC is a greater health problem in African-American men than in whites. African Americans have one of the highest incidences of PC in the world. 2 African Americans also have more advanced stages of PC at From the The University of Chicago~Michael Reese/University of Illinois Center for Radiation Oncology, Chicago, Illinois; Department of Urology, Cook County Hospital, Chicago, Illinois; Department of Pathology, Provident Hospital of Cook County, Chicago, Illinois; and Department of Statistics, University of Chicago, Chicago, Illinois Reprint requests: Srinivasan Vijayakumar, M.D., Department of Radiation Oncology, 2929 South Ellis Avenue, Room 149 MR, Chicago, IL 606I 6 Submitted: April 4, 1997, accepted (with revisions): August 20, 1997
300
O 1998, ELSEVIERSCIENCEINC. ALLRIGHTSRESERVED
diagnosis and higher mortality rates from PC than whites. Moreover, a greater increase in the PC incidence and in mortality rates has been noted over the past few decades in African Americans in comparison with whites. 3-s The use of serum levels of prostate-specific antigen (PSA) has revolutionized the early detection of and screening for PC. 6-9 Since the use of serum PSA levels for PC screening became widespread, several studies have suggested that this method has increased the percentage of patients who have organ-confined PC at diagnosis, and therefore it may improve the outcome of PC. l°-13 Despite the compelling data about the gravity of PC as a health issue in African Americans and the convincing evidence for the role of PSA in early diagnosis of PC, except for a recent Walter Reed Army Medical Center study, 14 there has been no large-scale study for evaluation of the baseline PSA levels in African 0090-4295/98/$19.00 PII S0090-4295(97)00617-1
Americans. We previously showed that, among patients with PC, African Americans had higher mean PSA levels than whites, and this racial difference remained statistically significant even after adjustment for tumor stage and grade.i5@ The Patterns of Care Study reported by Teshima et a1.,17 the Walter Reed studies reported by Morgan et al. l4 and Moul et ul.18, and data from Albert Einstein Medical Center in Philadelphia19 have all independently confirmed our initial observation. More recently, a prospective national sample similarly showed higher mean PSA levels in African-American patients with PC than whites, which prevailed on multivariate analysis. 2o Because of these findings in patients with PC, it is even more important to study PSA baseline ranges separately for African Americans; this is the primary aim of our study. Many attempts have been made to improve the sensitivity, specificity, and accuracy of serum PSA in screening for PC. These attempts include agespecific ranges for PSA, PSA density (PSAD), PSA velocity, and, more recently, measurement of free PSA level. Although the prognostic value of PSAD over the serum PSA level is controversial, PSAD adjusts for prostate gland volume. In our current study, we analyzed PSAD and age-adjusted PSA levels, in addition to serum PSA levels, to make certain that the racial differences in serum PSA, if present, are not due to differences in age or prostate volume. PSA velocity or free PSA information was not available on our study population. Our objectives in this study were (1) to determine the ranges of serum PSA levels and PSAD in a large group of African Americans without PC and (2) to compare their PSA levels and PSAD with those in contemporary groups of whites and Hispanics. MATERIAL
AND METHODS
PATIENTS A total of 859 adult male patients were seen between January 1988 and January 1993 in the Urology Clinic at Cook County Hospital because they had urinary symptoms suggestive of prostate pathology or because they were participants in a PC screening program. Of these men, 586 were African Americans, 142 were whites, and 131 were Hispanics. Twenty-five percent of the African Americans, 33.8% of whites, and 29.7% of Hispanics were 60 years old or younger. Table I lists the patient characteristics. All men were evaluated with a complete medical history and physician examination, including digital rectal examination (DRE), serum PSA level, and transrectal ultrasound (TRUS). Biopsy specimens were obtained only if the DRE findings were suspicious for malignancy or the serum PSA levels were elevated (greater than 4 ng/mL); no routine specimens were obtained. When TRUS-guided, systematic, and lesion-specific biopsies were performed, several specimens were obtained. All specimens were reviewed by the same pathologist (V.R.). Of the men who underwent prostate biopsy, only the ones whose specimens showed no pathologic evidence of cancer were inUROLOGY
51 (21, 1998
TABLE 1. Characteristics of the study population Characteristic Race (No. of men) AAS Whites Hispanics
586 142 131
Age (yr) Median Range PSA (ng/mL) Median Range PSA density Median Range Prostate volume [cc) Median Range Prostate biopsy Not done Negative Positive
64 36-90 1.9 0.2-33 0.068 0.005-2.6 26.2 12.5-200 576 283 None included here
KEY: AAs= African-American men; PSA = prostate-specific antigen
eluded in this study. Excluded from the study were patients with specimens positive for PC, patients with a history of PC, or those who had prior hormonal therapy or orchiectomy. Of the 859 men in the study, 283 (33%) had undergone prostate gland biopsies showing no PC (196 African Americans, 34 whites, and 53 Hispanics). PSA levels were measured by the Hybritech technique. TRUS was performed with an 1846 model Bruel and Kjaer scanner and a ~-MHZ multiplane probe. PSAD1s.lg was measured by dividing the serum PSA by the prostate volume as determined by applying the ellipsoid formula: Volume = 8 A’/ 3nW, where A is the area of the prostate cross section and W is the width of the prostate. PSA and PSAD means were calculated for each age group (see Table V).
STATISTICAL METHODS Separate univariate and multivariate regression analyses were performed for serum PSA level, PSAD, and prostate gland volume. Race and age were included as variables in the multivariate analyses. Because PSA values and PSAD have a skewed distribution, they were analyzed on a logarithmic scale that provides a normal distribution. We performed a chi-square test on the racial distribution of the PSA, grouped according to the thresholds that are currently in clinical use (0 to 4,4 to 10, and more than 10 ng/mL for serum PSA; less than 0.1 and greater than 0.1 for PSAD). The prostate volumes were compared among races by the one-way analysis of variance (ANOVA) F test.
RESULTS Table II shows the distribution of serum PSA levels for the entire study population of 859 men. Of the African Americans, 27.7Oh had PSA values greater than 4.0 ng/mL, whereas only 13.4% of whites and 19.9% of Hispanics had these PSA levels (P = 0.005). The mean PSA level in African Amer301
TABLE II. Distribution of serum PSA levels of the entire population by race * PSA (ng/mL)
AAs (016) Whites (%)
o-4 4-10 More than 10
72.4 19.5 8.2
86.6 10.6 2.8
Hispanics (%) 80.2 13.0 6.9
Abbreviations as in Table I. * overall P value = 0.005.
TABLE III.
Distribution of PSA density of the entire study population *
PSAD
AAs (%)
Whites (%)
Hispanics (%)
64 36
78 22
70 30
Less than 0.1 More than 0.1 KEY:AAs = African-American * Gverall P value = 0.007.
mm; PSAD = prostate-specijc
antigen density.
TABLE IV. Comparison of prostate gland volume among the three racial groups Mean prostate volume 95% Cl
AAs
Whites
26.6 25.3-28.0
27.3 24.7-30.1
Hispanics 27.8 25.1-30.9
KEY:AAs = African-American men; CI = confidence interval.
icans was 2.1 ng/mL, which was significantly higher than that of the white patients (mean PSA 1.53 ng/mL) and the Hispanics (mean PSA 1.83 ng/mL) (P = 0.003). PSAD was significantly higher for our African-American cohort than for whites (P = 0.002) and Hispanics (P = 0.05). The mean PSA density for African Americans was 0.078, compared with 0.057 for whites and 0.065 for Hispanics. Table III shows the PSAD distribution. Among African Americans, 36% had PSAD value greater than 0.1, compared with 22% of whites and 30% of Hispanics (P = 0.05). There was no difference in prostate gland volumes among the three ethnic groups (Table IV). Even after adjustment for age and prostate volume, on multivariate regression analysis for PSAD and PSA levels, the racial differences remained significant (P cO.05). PSA and PSAD increased with older age as shown in Table V. For the 283 men who had negative TRUSguided biopsies (196 African Americans, 34 whites, and 53 Hispanics), a separate analysis was performed, and similar results were obtained. Among African Americans, 40.3% had PSA levels greater than 4.0 mg/mL, compared with 17.6% of whites and 16.9% of Hispanics (P = 0.065), as shown in Table VI. The mean PSA value for the African Americans who had undergone biopsies was 2.82 ng/mL, versus 2.0 mg/mL for whites and 1.7 ng/mL for Hispanics (P = 0.003). Table VII 302
lists the PSAD distribution in the biopsied group. The mean PSAD for African Americans was 0.093, which was higher than that for whites (mean PSAD 0.07) and Hispanics (mean PSAD 0.064) (P = 0.02). However, on multivariate analysis, the difference in PSAD between African Americans and whites did not reach statistical significance (P = 0.3), whereas it was significant between African Americans and Hispanics (P = 0.03). COMMENT Our study suggests that, among men without PC, African Americans have higher serum PSA levels and PSADs than do whites and Hispanics, even after adjustment for age. However, our results should be interpreted carefully because only 33% of the men in our study population were pathologically proved to be cancer-free by biopsy. For this reason, we separately analyzed the subgroup of men who had undergone biopsies of the prostate. The findings for this subgroup were consistent with the overall observation. Within the same race, the mean PSA and PSAD were similar for the entire population and for the biopsied group, thus validating our findings for the entire group (Table VIII). Because many studies have shown higher serum PSA levels in African Americans with PC than in whites with PC, it is important to study the racial differences in the general population. Very few studies have addressed this issue. Whittmore et a1.,21 at the Kaiser Permanente Medical Care Program, showed no difference in serum PSA levels between African Americans and whites. However, the number of patients in that study was much smaller (100 whites and 84 African Americans) than ours, and only men who had at least three serum measurements were chosen for the analysis; thus, this was a highly selected cohort. In a recent study, Morgan et al., l+ at Walter Reed Army Medical Center, reported a mean serum PSA value of 2.6 ng/mL in African Americans and 1.9 ng/mL in whites. Our findings of higher mean PSA levels and PSAD among African-American men without PC may be attributable to one or more of the following: (1) The incidence of high-grade prostatic intraepithelial neoplasia (PIN) has been reported to be higher in African Americans than in whites.2*J3 Whether the PIN contributes to higher PSA levels is not well documented.24,25 However, Brawer et ~11.24 compared the mean PSA levels among patients with PIN, benign prostate conditions, and PC and reported a higher mean value in men with PIN than in those with benign conditions (7.8 ng/mL for men with PIN, 5.8 ng/mL in patients with benign UROLOGY 51 (21, 1998
TABLE V.
Mean PSA and PSAD as a function of age PSAIPSAD
Me group
c50 yr
Entire study population AAS Whites Hispanics Biopsied
1.0/0.055 0.75/O. 123 0.6OlO.053
yr
1.0/0.055
Whites Hispanics
None 1.0/0.046
yr
1.4/0.067 1.5/0.056 1.210.048
2.0/0.07 1 1.6/0.052 2.OlO.062
1.6/0.069 2.910.081 1.1/0.048
2.610.086 1.7/0.060 2.0/0.050
70-79
yr
80 yr and older
2.2510.084 2.610.072 2.3lO.076
3.710.083 4.0/0.043 3.210.180
3.210.097 1.9/0.075 2.210.076
4.410.086 None 2.010.147
antigen; PSAD = prostate-specific antigen density; AAs = African-American men.
= prostate-specijc
TABLE VI. Distribution of serum PSA levels of the 283 men with negative biopsy * PSA (ng/mL) O-4 4-10 More than
10
AAs (96)
Whites (%)
59.7 26.0 14.3
82.4 14.7 2.9
Hispanics (016) 83.0 9.4 7.5
Abbreviations as in Table V. * Overall P value = 0.05.
TABLE VII. Distribution of PSAD of the 283 men with negative biopsies * PSAD
AAs (96)
Whites (%)
Hispanics (96)
59 41
65 35
66 34
Less than 0.1 More than 0.1 Abbreviations as in Table V. * Overall P value = 0.01.
conditions, and 18.3 ng/mL for patients with PC). We are currently comparing the incidence of PIN in our population. (2) Most African-American men generally have a lower socioeconomic status and less access to medical care than whites and Hispanics and therefore might have a higher incidence of undiagnosed PC, with an associated higher PSA level. However, this cannot explain our findings because all men in our study population were seen at a county hospital (Cook County Hospital) with an equal-access health care environment.16J8 Our results among “biopsied men” with no evidence of cancer are similar to the entire population. (3) A few studies have shown that college-age African Americans have higher testosterone levels than other racial groups. 26,27It is not clear whether testosterone levels have an impact on serum PSA levels. However, in a previous study, we addressed the relationship between serum male hormone levels and PSA values in patients with localized PC and showed that there were no racial differences in androgen levels in PC patients and that androgens had no influence on serum PSA levels.28 In a recent UROLOGY
60-69
group
AAS
KEY:PSA
50-59
51 (Z), 1998
study, exogenous testosterone administration had no effect on serum PSA leve1.29 Serum PSA is a tumor marker which is organspecific, but not cancer-specific. Various benign conditions (benign prostatic hyperplasia [BPH] , prostate infection, or inflammation) can affect PSA levels. Many authors agree that, for PC screening, a serum PSA level of greater than 4.0 ng/mL has sensitivity and specificity ranges between 70% and 90%, with a positive predictive value of 35% to 45% and a negative predictive value of 90% to 98°h.7,25J0 Many attempts have been made to enhance the accuracy of the use of PSA in the detection of PC. The parameters studied include PSAD, PSA velocity, and age-specific PSA ranges. The PSA velocity (ie, the rate of PSA increase) is not available for our study population (all patients had only one PSA level measurement). Oesterling et ~1.~~advocated the establishment of age-specific PSA reference ranges for improved specificity and sensitivity of PSA. Their study included 471 men who were randomly chosen from 2119 healthy men aged 40 to 79 years who had no evidence of PC by DRE or TRUS. The serum PSA levels were correlated with the patients’ age (P
TABLE VIII.
Comparison between the entire study population and the group with negative biopsies AAs
Entire study population Biopsied group
2.10 2.82
Mean PSA Whites 1.53 2.0
Hispanics
AAs
1.83 1.70
0.078 0.093
Mean PSAD Whites 0.057 0.07
Hispanics 0.065 0.064
Abbreviations as in Table V.
included age, and the above-mentioned findings of Morgan et al. 14 suggest that age-adjusted PSA ranges have to be prospectively studied in African Americans. PSAD (also known as PSA index) is defined as a patient's serum PSA level divided by the prostate gland volume. It was developed so that one could adjust the variations in the PSA level due to differences in the size of the prostate gland and the extent of BPH. The generally accepted cutoff value for a normal PSAD is 0.1 units. Benson et al. 34'35 measured the prostate volume by magnetic resonance imaging (MRI) or from surgical specimens and determined the PSAD. The mean PSAD for PC patents was 0.58, whereas that for BPH cases was 0.044 (P ~0.002). No patient with BPH had a PSAD of more than 0.117, whereas 97% of PC patients had PSAD values of more than 0.1. Mettlin et al. 33 studied 2011 men without and 171 with PC. The average PSAD for the noncancer population was 0.06 versus 0.35 for the PC group. The sensitivity and specificity of PSAD were 74% and 85%, respectively, and the investigators concluded that the PSAD has no advantage over serum PSA (0 to 4 ng/mL) for early detection of PC. It should be noted that the PSAD of 0.057 for whites in our study is similar to the values for the noncancer population in the above-mentioned studies, which included predominantly white populations. We used the PSAD to determine whether serum PSA values per unit volume of prostate gland differed between ethnic groups. Whether PSAD will have a practical role in PC detection does not detract from the significance of our findings; our observations and those of Morgan et al. 14 suggest that, among men without PC, there are racial differences in PSA levels even after adjustment for age and/or prostate volume (ie, race appears to be an independent factor influencing the serum PSA level). In conclusion, our study is one of the first attempts to determine mean serum PSA levels and PSAD among a large group of African Americans without PC in comparison with other ethnic groups, including Hispanics, and it shows that such differences do exist. However, neither our study nor that of Morgan et al. 14 explain why such differences exist. Whether these differences are due to biologic or sociologic reasons, or a combi304
nation of both, is not known at this time and needs further study. 16,36-39 Although our study population is relatively large, the study is not populationbased and hence might have a selection bias. In addition, only one third of our patients underwent prostate biopsies. Population-based studies confirming our observation are needed; the causes of such differences need to be elucidated. 36 REFERENCES 1. Parker SL, Tong T, Bolden S, and Wingo PA: Cancer Statistics, 1996. CA CanJ Clin 47: 5-27, 1997. 2. Muir C, NectouxJ, and StaszewskiJ: The epidemiology of prostatic cancer. Acta Oncol 30: 133-140, 1991. 3. AustinJP, Aziz H, Potters L, Thelmo W, Chen P, Choi K, Brandys M, Macchia RJ, and Rotman M: Diminished survival of young blacks with adenocarcinoma of the prostate. AmJ Clin Oncol 13: 465-469, 1990. 4. Pienta KJ, Demers R, Hoff M, Kau TY, Montie JE, and Severson RK: Effect of age and race on the survival of men with prostate cancer in the Metropolitan Detroit tricounty area, 1973 to 1987. Urology 45: 93-102, 1995. 5. MebaneC, GibbsT, andHormJ: Currentstatusofprostate cancer in North American black males. J Natl Med Assoc 82: 782-788, 1990. 6. Stamey TA, Yang N, Hay AR, McNealJE, Freiha FS, and Redwine E: PSA as a serum marker for prostate adenocarcinoma. N EnglJ Med 317: 909-916, 1987. 7. Catatona WJ, Smith DS, Ratliff TL, Dodds KM, Coplen DE, Yuan JJ, PetrosJA, and Andriole GL: The measurement of PSA as a screening test for prostate cancer. N EnglJ Med 24: 1156-1161, 1991. 8. Brawer MK, and Lange PH: PSA in the screening staging and follow-up of prostate cancer. WorldJ Urol 7: 7-11, 1989. 9. Brawer MK, Beatie J, Werner MH, Vessella RL, Preston SD, and Lange PH: Screening for prostate carcinoma with PSA. Results of the second year. J Urol 150: 106-109, 1993. 10. Catalona WJ, Smith DS, Ratliff TL, and Basler JW: Detection of organ-confined prostate cancer is increased through prostate-specific antigen-based screening. J Am Med Assoc 271: 192-193, 1993. 11. Mettlin C, Murphy GP, Lee F, Littrup PJ, Chesley A, Babaian R, Badalament R, Kane RA, and Mostofi FK: Characteristics of prostate cancer detected in the American Cancer Society-National Prostate Cancer Detection Project. J Urol 152(5 Pt 2): 1737-1740, 1994. 12. Trapasso JG, deKernionJB, Smith RB, and Dorey F: The incidence and significance of detectable levels of serum prostate specific antigen after radical prostatectomy. J Urol 152(5 Pt 2): 1821-1825, 1994. 13. Murphy G, Natarajan N, PontesJE, Schmitz RL, Smart CR, Schmidt JD, and Mettlin C: The national survey of prostate cancer in the United States by the American College of Surgeons. J Urol 127: 928-934, 1982. 14. Morgan TO, Jacobsen SJ, McCarthy WF, Jacobson DJ, UROLOGY 51 (2), 1998
McLean DG, and Mot.11JW: Age-specific reference ranges for serum prostate-specific antigen in black men. N Engl J Med 335: 304-310,1996. 15. Vijayakumar S, Karrison T, Weichselbaum RR, Chan S, Quadri SF, and Awan AM: Racial differences in prostate-specific antigen levels in patients with local-regional prostate cancer. Cancer Epidemiol Biomark Prev 1: 541-554, 1991. 16. Vijayakumar S, Weichselbaum R, Vaida F, Dale W, and Hellman S: Prostate-specific antigen levels in Afiican-Americans correlate with insurance status as an indicator of socioeconomic status. Cancer J Sci Am 2: 225-233,1996. 17. Teshima T, Hanlon AM, and Hanks GE: Pretreatment prostate-specific antigen values in patients with prostate cancer: 1989 patterns of care study process survey. lnt J Radiat Oncol Biol Phys 33: 809-814, 1995. 18. Moul JW, Sesterhenn IA, Connelly RR, Douglas T, Srivastava S, Mostofi FK, and McLeod DG: Prostate-specific antigen values at the time of prostate cancer diagnosis in AfiicanAmerican men. J Am Med Assoc 274: 1277-1281,1995. 19. Asbell SO, and Vijayakumar S: Racial differences in prostate-specific antigen levels in patients with local-regional prostate cancer. Prostate 31: 42-46, 1997. 20. Vijayakumar S, Winter K, Sause W, Gallagher MJ, Perez C, and Bondy M: African-American (AA) men with local-regional prostate cancer (PC) present with higher prostate specific antigen (PSA) levels than whites: results of RTOG 94-12. Int J Radiat Oncol Biol Phys Suppl36: 305,1996. 21. Whittemore A, Lele C, Friedman GD, Stamey T, Vogelman JH, and Orentreich N: Prostate-specific antigen as predictor of prostate cancer in black men and white men. J Nat1 Cancer Inst 87: 354-360, 1995. 22. Sakr WA, Haas GP, Cassin BF, Pontes JE, and Crissman JD: The frequency of carcinoma and intraepithelial neoplasia of the prostate in young male patients. J Urol 150: 379-385, 1993. 23. Sakr WA, Grignon DJ, Haas GP, Shomer KL, Heilbum LK, Cassin BJ, Powell J, Montie JA, Pontes JE, and Crissman JD: Epidemiology of high grade prostatic intraepithelial neoplasia. Path01 Res Pratt 191: 838-841, 1995. 24. Brawer M, Rennels MA, Nagle RB, Schifman R, and Gaines JA: Serum prostate-specific antigen and prostate pathology in men having simple prostatectomy. Am J Clin Path01 92: 760-764, 1989. 25. Cooner WI-I, Mosley BR, Rutherford CL Jr, Beard JH, Pond HS, Terry WJ, Igel TC, and Kidd DD: Prostate cancer detection in a clinical urological practice by ultrasonography, digital rectal examination and prostate specific antigen. J Urol 143: 1146-1152,199O. 26. Ross R, Bernstein L, Judd H, Hanisch R, Pike M, and Henderson B: Serum testosterone levels in healthy young black and white men. J Nat1 Cancer Inst 76: 45-48, 1986. 27. Ross RK, Bernstein L, Lobo RA, Shimizu H, Stanczyk
UROLOGY 51 (21, 1998
FZ, Pike MC, and Henderson BF: 5-alpha-reductase activity and risk of prostate cancer among Japanese and US white and black males. Lancet 339: 887-889, 1992. 28. Vijayakumar S, Quadri SF, Dong L, Ignacio L, Kathuria IN, Sutton H, and Halpren H: Results of a study to correlate serum prostate specific antigen and reproductive hormone levels in patients with localized prostate cancer. J Nat1 Med Assoc 87: 813-819, 1995. 29. Cooper CS, MacIndoe JH, Perry PJ, Yates WR, and Williams RD: The effect of exogenous testosterone on total and free prostate specific antigen levels in healthy young men. J Urol156: 438-441,1996. 30. Babaian RJ, Miyashita H, Evans RB, von Eschenbach AC, and Ramirez EI: Early detection program for prostate cancer: results and identification of high-risk patient population. Urology 37: 193-197, 1991. 31. Oesterling JE, Jacobsen SJ, Chute CG, Guess HA, Girman CJ, Panser LA, and Lieber MM: Serum prostate-specific antigen in a community-based population of healthy men. J Am Med Assoc 270: 860-864,1993. 32. Speights VO Jr, Brawn PN, Foster DM, Spiekerman AM, Kuhl D, and Riggs MW: Evaluation of age-specific normal ranges for prostate-specific antigen. Urology 45: 454-457, 1995. 33. Mettlin C, Littrup PJ, Kane RA, Murphy GP, Lee F, Chesley A, Badalament R, and Mostofi FK: Relative sensitivity and specificity of serum prostate specific antigen (PSA) level compared with age-referenced PSA, PSA density, and PSA change. Cancer 74: 1615-1620,1994. 34. Benson MC, Whang IS, Olsson CA, McMahon DJ, and Cooner WI-I: The use of prostate specific antigen density to enhance the predictive value of intermediate levels of serum prostate specific antigen. J Urol 147(3 Pt 2): 817-821, 1992. 35. Benson MC, Whang IS, Pantuck A, Ring K, Kaplan SA, Olsson CA, and Cooner WH: Prostate specific antigen density: a means of distinguishing benign prostatic hypertrophy and prostate cancer. J Uroll47(3 Pt 2): 815-816, 1992. 36. Albertsen PC: African-Americans and prostate cancer (commentary). Cancer J Sci Am 2: 200-201, 1996. 37. Vijayakumar S, Winter K, Sause W, Gallagher MJ, Michalski J, Raoch M, Porter A, and Bondy M: Prostate specific antigen levels are higher in African-American patients than whites in a multicenter registration study: results of RTOG 94-12. Int J Radiat Oncol Biol Phys (in press). 38. Vijayakumar S, Vaida F, Weichselbaum RR, and Hellman S: Race and Will Rogers phenomenon in prostate cancer. Cancer J Sci Am (in press). 39. Dale W, Vijayakumar S, Lawlor EF, and Merrell K: Prostate cancer, race, and socioeconomic status: inadequate adjustment for social factors in assessing racial differences. Prostate 29: 271-281, 1996.
305
ELSEVIER
COMPARISON OF SERUM PROSTATE-SPECIFIC ANTIGEN LEVELS AND PSA DENSITY IN AFRICAN-AMERICAN, WHITE, AND HISPANIC MEN WITHOUT PROSTATE CANCER IBRAHIM ABDALLA, PAUL RAY, VERA RAY, FLORIN VAIDA, AND SRINIVASAN VIJAYAKUMAR
ABSTRACT Objectives. To determine the influence of race or ethnicity on serum prostate-specific antigen (PSA) levels and PSA density (PSAD) in a population of healthy men without clinically evident prostate cancer. Methods. This retrospective study was conducted between January 1988 and January 1993. The serum PSA levels were measured in 859 men (586 African Americans, 142 whites, and 131 Hispanics) who were participants in a prostate cancer screening program or had urinary symptoms suggestive of prostate gland pathology. All men underwent a detailed clinical examination, including digital rectal examination, serum PSA determination, and transrectal ultrasound (TRUS). None of the subjects included had clinical or TRUS evidence of prostate cancer (furthermore, 283 men were pathologically proved to be cancer-free by prostate biopsies). Serum PSA levels and PSA densities as a function of each individual's ethnic background were determined. Results. The mean serum PSA level in African Americans was 2.1 ng/mL, which was significantly higher than that of whites (mean PSA of 1.53 ng/mL) and Hispanics (mean PSA of 1.83 ng/mL) (P = 0.003). Similar differences among the three groups were observed in PSA density (the mean PSAD was 0.078, 0.057, and 0.065 for African Americans, whites, and Hispanics, respectively). A separate analysis for the biopsynegative men was performed, and the findings were consistent with the observations for the entire study group. After adjustment for age and prostate volume, the differences remained statistically significant. Conclusions. Among men without evidence of prostate cancer, African Americans have higher serum PSA levels and PSA densities than do whites or Hispanics. Race or ethnicity was an independent factor that affected serum PSA levels even after adjustment for age and prostate volume. UROLOGY 51: 300-305, 1998. © 1998, Elsevier Science Inc. All rights reserved.
cancer (PC) is the most common nonp rostate skin cancer and the second leading cause of cancer-related deaths in American men. It is estb mated that 334,500 men will be diagnosed with PC and 41,800 will die of this disease in 1997.1 PC is a greater health problem in African-American men than in whites. African Americans have one of the highest incidences of PC in the world. 2 African Americans also have more advanced stages of PC at From the The University of Chicago~Michael Reese/University of Illinois Center for Radiation Oncology, Chicago, Illinois; Department of Urology, Cook County Hospital, Chicago, Illinois; Department of Pathology, Provident Hospital of Cook County, Chicago, Illinois; and Department of Statistics, University of Chicago, Chicago, Illinois Reprint requests: Srinivasan Vijayakumar, M.D., Department of Radiation Oncology, 2929 South Ellis Avenue, Room 149 MR, Chicago, IL 606I 6 Submitted: April 4, 1997, accepted (with revisions): August 20, 1997
300
O 1998, ELSEVIERSCIENCEINC. ALLRIGHTSRESERVED
diagnosis and higher mortality rates from PC than whites. Moreover, a greater increase in the PC incidence and in mortality rates has been noted over the past few decades in African Americans in comparison with whites. 3-s The use of serum levels of prostate-specific antigen (PSA) has revolutionized the early detection of and screening for PC. 6-9 Since the use of serum PSA levels for PC screening became widespread, several studies have suggested that this method has increased the percentage of patients who have organ-confined PC at diagnosis, and therefore it may improve the outcome of PC. l°-13 Despite the compelling data about the gravity of PC as a health issue in African Americans and the convincing evidence for the role of PSA in early diagnosis of PC, except for a recent Walter Reed Army Medical Center study, 14 there has been no large-scale study for evaluation of the baseline PSA levels in African 0090-4295/98/$19.00 PII S0090-4295(97)00617-1
Americans. We previously showed that, among patients with PC, African Americans had higher mean PSA levels than whites, and this racial difference remained statistically significant even after adjustment for tumor stage and grade.i5@ The Patterns of Care Study reported by Teshima et a1.,17 the Walter Reed studies reported by Morgan et al. l4 and Moul et ul.18, and data from Albert Einstein Medical Center in Philadelphia19 have all independently confirmed our initial observation. More recently, a prospective national sample similarly showed higher mean PSA levels in African-American patients with PC than whites, which prevailed on multivariate analysis. 2o Because of these findings in patients with PC, it is even more important to study PSA baseline ranges separately for African Americans; this is the primary aim of our study. Many attempts have been made to improve the sensitivity, specificity, and accuracy of serum PSA in screening for PC. These attempts include agespecific ranges for PSA, PSA density (PSAD), PSA velocity, and, more recently, measurement of free PSA level. Although the prognostic value of PSAD over the serum PSA level is controversial, PSAD adjusts for prostate gland volume. In our current study, we analyzed PSAD and age-adjusted PSA levels, in addition to serum PSA levels, to make certain that the racial differences in serum PSA, if present, are not due to differences in age or prostate volume. PSA velocity or free PSA information was not available on our study population. Our objectives in this study were (1) to determine the ranges of serum PSA levels and PSAD in a large group of African Americans without PC and (2) to compare their PSA levels and PSAD with those in contemporary groups of whites and Hispanics. MATERIAL
AND METHODS
PATIENTS A total of 859 adult male patients were seen between January 1988 and January 1993 in the Urology Clinic at Cook County Hospital because they had urinary symptoms suggestive of prostate pathology or because they were participants in a PC screening program. Of these men, 586 were African Americans, 142 were whites, and 131 were Hispanics. Twenty-five percent of the African Americans, 33.8% of whites, and 29.7% of Hispanics were 60 years old or younger. Table I lists the patient characteristics. All men were evaluated with a complete medical history and physician examination, including digital rectal examination (DRE), serum PSA level, and transrectal ultrasound (TRUS). Biopsy specimens were obtained only if the DRE findings were suspicious for malignancy or the serum PSA levels were elevated (greater than 4 ng/mL); no routine specimens were obtained. When TRUS-guided, systematic, and lesion-specific biopsies were performed, several specimens were obtained. All specimens were reviewed by the same pathologist (V.R.). Of the men who underwent prostate biopsy, only the ones whose specimens showed no pathologic evidence of cancer were inUROLOGY
51 (21, 1998
TABLE 1. Characteristics of the study population Characteristic Race (No. of men) AAS Whites Hispanics
586 142 131
Age (yr) Median Range PSA (ng/mL) Median Range PSA density Median Range Prostate volume [cc) Median Range Prostate biopsy Not done Negative Positive
64 36-90 1.9 0.2-33 0.068 0.005-2.6 26.2 12.5-200 576 283 None included here
KEY: AAs= African-American men; PSA = prostate-specific antigen
eluded in this study. Excluded from the study were patients with specimens positive for PC, patients with a history of PC, or those who had prior hormonal therapy or orchiectomy. Of the 859 men in the study, 283 (33%) had undergone prostate gland biopsies showing no PC (196 African Americans, 34 whites, and 53 Hispanics). PSA levels were measured by the Hybritech technique. TRUS was performed with an 1846 model Bruel and Kjaer scanner and a ~-MHZ multiplane probe. PSAD1s.lg was measured by dividing the serum PSA by the prostate volume as determined by applying the ellipsoid formula: Volume = 8 A’/ 3nW, where A is the area of the prostate cross section and W is the width of the prostate. PSA and PSAD means were calculated for each age group (see Table V).
STATISTICAL METHODS Separate univariate and multivariate regression analyses were performed for serum PSA level, PSAD, and prostate gland volume. Race and age were included as variables in the multivariate analyses. Because PSA values and PSAD have a skewed distribution, they were analyzed on a logarithmic scale that provides a normal distribution. We performed a chi-square test on the racial distribution of the PSA, grouped according to the thresholds that are currently in clinical use (0 to 4,4 to 10, and more than 10 ng/mL for serum PSA; less than 0.1 and greater than 0.1 for PSAD). The prostate volumes were compared among races by the one-way analysis of variance (ANOVA) F test.
RESULTS Table II shows the distribution of serum PSA levels for the entire study population of 859 men. Of the African Americans, 27.7Oh had PSA values greater than 4.0 ng/mL, whereas only 13.4% of whites and 19.9% of Hispanics had these PSA levels (P = 0.005). The mean PSA level in African Amer301
TABLE II. Distribution of serum PSA levels of the entire population by race * PSA (ng/mL)
AAs (016) Whites (%)
o-4 4-10 More than 10
72.4 19.5 8.2
86.6 10.6 2.8
Hispanics (%) 80.2 13.0 6.9
Abbreviations as in Table I. * overall P value = 0.005.
TABLE III.
Distribution of PSA density of the entire study population *
PSAD
AAs (%)
Whites (%)
Hispanics (%)
64 36
78 22
70 30
Less than 0.1 More than 0.1 KEY:AAs = African-American * Gverall P value = 0.007.
mm; PSAD = prostate-specijc
antigen density.
TABLE IV. Comparison of prostate gland volume among the three racial groups Mean prostate volume 95% Cl
AAs
Whites
26.6 25.3-28.0
27.3 24.7-30.1
Hispanics 27.8 25.1-30.9
KEY:AAs = African-American men; CI = confidence interval.
icans was 2.1 ng/mL, which was significantly higher than that of the white patients (mean PSA 1.53 ng/mL) and the Hispanics (mean PSA 1.83 ng/mL) (P = 0.003). PSAD was significantly higher for our African-American cohort than for whites (P = 0.002) and Hispanics (P = 0.05). The mean PSA density for African Americans was 0.078, compared with 0.057 for whites and 0.065 for Hispanics. Table III shows the PSAD distribution. Among African Americans, 36% had PSAD value greater than 0.1, compared with 22% of whites and 30% of Hispanics (P = 0.05). There was no difference in prostate gland volumes among the three ethnic groups (Table IV). Even after adjustment for age and prostate volume, on multivariate regression analysis for PSAD and PSA levels, the racial differences remained significant (P cO.05). PSA and PSAD increased with older age as shown in Table V. For the 283 men who had negative TRUSguided biopsies (196 African Americans, 34 whites, and 53 Hispanics), a separate analysis was performed, and similar results were obtained. Among African Americans, 40.3% had PSA levels greater than 4.0 mg/mL, compared with 17.6% of whites and 16.9% of Hispanics (P = 0.065), as shown in Table VI. The mean PSA value for the African Americans who had undergone biopsies was 2.82 ng/mL, versus 2.0 mg/mL for whites and 1.7 ng/mL for Hispanics (P = 0.003). Table VII 302
lists the PSAD distribution in the biopsied group. The mean PSAD for African Americans was 0.093, which was higher than that for whites (mean PSAD 0.07) and Hispanics (mean PSAD 0.064) (P = 0.02). However, on multivariate analysis, the difference in PSAD between African Americans and whites did not reach statistical significance (P = 0.3), whereas it was significant between African Americans and Hispanics (P = 0.03). COMMENT Our study suggests that, among men without PC, African Americans have higher serum PSA levels and PSADs than do whites and Hispanics, even after adjustment for age. However, our results should be interpreted carefully because only 33% of the men in our study population were pathologically proved to be cancer-free by biopsy. For this reason, we separately analyzed the subgroup of men who had undergone biopsies of the prostate. The findings for this subgroup were consistent with the overall observation. Within the same race, the mean PSA and PSAD were similar for the entire population and for the biopsied group, thus validating our findings for the entire group (Table VIII). Because many studies have shown higher serum PSA levels in African Americans with PC than in whites with PC, it is important to study the racial differences in the general population. Very few studies have addressed this issue. Whittmore et a1.,21 at the Kaiser Permanente Medical Care Program, showed no difference in serum PSA levels between African Americans and whites. However, the number of patients in that study was much smaller (100 whites and 84 African Americans) than ours, and only men who had at least three serum measurements were chosen for the analysis; thus, this was a highly selected cohort. In a recent study, Morgan et al., l+ at Walter Reed Army Medical Center, reported a mean serum PSA value of 2.6 ng/mL in African Americans and 1.9 ng/mL in whites. Our findings of higher mean PSA levels and PSAD among African-American men without PC may be attributable to one or more of the following: (1) The incidence of high-grade prostatic intraepithelial neoplasia (PIN) has been reported to be higher in African Americans than in whites.2*J3 Whether the PIN contributes to higher PSA levels is not well documented.24,25 However, Brawer et ~11.24 compared the mean PSA levels among patients with PIN, benign prostate conditions, and PC and reported a higher mean value in men with PIN than in those with benign conditions (7.8 ng/mL for men with PIN, 5.8 ng/mL in patients with benign UROLOGY 51 (21, 1998
TABLE V.
Mean PSA and PSAD as a function of age PSAIPSAD
Me group
c50 yr
Entire study population AAS Whites Hispanics Biopsied
1.0/0.055 0.75/O. 123 0.6OlO.053
yr
1.0/0.055
Whites Hispanics
None 1.0/0.046
yr
1.4/0.067 1.5/0.056 1.210.048
2.0/0.07 1 1.6/0.052 2.OlO.062
1.6/0.069 2.910.081 1.1/0.048
2.610.086 1.7/0.060 2.0/0.050
70-79
yr
80 yr and older
2.2510.084 2.610.072 2.3lO.076
3.710.083 4.0/0.043 3.210.180
3.210.097 1.9/0.075 2.210.076
4.410.086 None 2.010.147
antigen; PSAD = prostate-specific antigen density; AAs = African-American men.
= prostate-specijc
TABLE VI. Distribution of serum PSA levels of the 283 men with negative biopsy * PSA (ng/mL) O-4 4-10 More than
10
AAs (96)
Whites (%)
59.7 26.0 14.3
82.4 14.7 2.9
Hispanics (016) 83.0 9.4 7.5
Abbreviations as in Table V. * Overall P value = 0.05.
TABLE VII. Distribution of PSAD of the 283 men with negative biopsies * PSAD
AAs (96)
Whites (%)
Hispanics (96)
59 41
65 35
66 34
Less than 0.1 More than 0.1 Abbreviations as in Table V. * Overall P value = 0.01.
conditions, and 18.3 ng/mL for patients with PC). We are currently comparing the incidence of PIN in our population. (2) Most African-American men generally have a lower socioeconomic status and less access to medical care than whites and Hispanics and therefore might have a higher incidence of undiagnosed PC, with an associated higher PSA level. However, this cannot explain our findings because all men in our study population were seen at a county hospital (Cook County Hospital) with an equal-access health care environment.16J8 Our results among “biopsied men” with no evidence of cancer are similar to the entire population. (3) A few studies have shown that college-age African Americans have higher testosterone levels than other racial groups. 26,27It is not clear whether testosterone levels have an impact on serum PSA levels. However, in a previous study, we addressed the relationship between serum male hormone levels and PSA values in patients with localized PC and showed that there were no racial differences in androgen levels in PC patients and that androgens had no influence on serum PSA levels.28 In a recent UROLOGY
60-69
group
AAS
KEY:PSA
50-59
51 (Z), 1998
study, exogenous testosterone administration had no effect on serum PSA leve1.29 Serum PSA is a tumor marker which is organspecific, but not cancer-specific. Various benign conditions (benign prostatic hyperplasia [BPH] , prostate infection, or inflammation) can affect PSA levels. Many authors agree that, for PC screening, a serum PSA level of greater than 4.0 ng/mL has sensitivity and specificity ranges between 70% and 90%, with a positive predictive value of 35% to 45% and a negative predictive value of 90% to 98°h.7,25J0 Many attempts have been made to enhance the accuracy of the use of PSA in the detection of PC. The parameters studied include PSAD, PSA velocity, and age-specific PSA ranges. The PSA velocity (ie, the rate of PSA increase) is not available for our study population (all patients had only one PSA level measurement). Oesterling et ~1.~~advocated the establishment of age-specific PSA reference ranges for improved specificity and sensitivity of PSA. Their study included 471 men who were randomly chosen from 2119 healthy men aged 40 to 79 years who had no evidence of PC by DRE or TRUS. The serum PSA levels were correlated with the patients’ age (P
TABLE VIII.
Comparison between the entire study population and the group with negative biopsies AAs
Entire study population Biopsied group
2.10 2.82
Mean PSA Whites 1.53 2.0
Hispanics
AAs
1.83 1.70
0.078 0.093
Mean PSAD Whites 0.057 0.07
Hispanics 0.065 0.064
Abbreviations as in Table V.
included age, and the above-mentioned findings of Morgan et al. 14 suggest that age-adjusted PSA ranges have to be prospectively studied in African Americans. PSAD (also known as PSA index) is defined as a patient's serum PSA level divided by the prostate gland volume. It was developed so that one could adjust the variations in the PSA level due to differences in the size of the prostate gland and the extent of BPH. The generally accepted cutoff value for a normal PSAD is 0.1 units. Benson et al. 34'35 measured the prostate volume by magnetic resonance imaging (MRI) or from surgical specimens and determined the PSAD. The mean PSAD for PC patents was 0.58, whereas that for BPH cases was 0.044 (P ~0.002). No patient with BPH had a PSAD of more than 0.117, whereas 97% of PC patients had PSAD values of more than 0.1. Mettlin et al. 33 studied 2011 men without and 171 with PC. The average PSAD for the noncancer population was 0.06 versus 0.35 for the PC group. The sensitivity and specificity of PSAD were 74% and 85%, respectively, and the investigators concluded that the PSAD has no advantage over serum PSA (0 to 4 ng/mL) for early detection of PC. It should be noted that the PSAD of 0.057 for whites in our study is similar to the values for the noncancer population in the above-mentioned studies, which included predominantly white populations. We used the PSAD to determine whether serum PSA values per unit volume of prostate gland differed between ethnic groups. Whether PSAD will have a practical role in PC detection does not detract from the significance of our findings; our observations and those of Morgan et al. 14 suggest that, among men without PC, there are racial differences in PSA levels even after adjustment for age and/or prostate volume (ie, race appears to be an independent factor influencing the serum PSA level). In conclusion, our study is one of the first attempts to determine mean serum PSA levels and PSAD among a large group of African Americans without PC in comparison with other ethnic groups, including Hispanics, and it shows that such differences do exist. However, neither our study nor that of Morgan et al. 14 explain why such differences exist. Whether these differences are due to biologic or sociologic reasons, or a combi304
nation of both, is not known at this time and needs further study. 16,36-39 Although our study population is relatively large, the study is not populationbased and hence might have a selection bias. In addition, only one third of our patients underwent prostate biopsies. Population-based studies confirming our observation are needed; the causes of such differences need to be elucidated. 36 REFERENCES 1. Parker SL, Tong T, Bolden S, and Wingo PA: Cancer Statistics, 1996. CA CanJ Clin 47: 5-27, 1997. 2. Muir C, NectouxJ, and StaszewskiJ: The epidemiology of prostatic cancer. Acta Oncol 30: 133-140, 1991. 3. AustinJP, Aziz H, Potters L, Thelmo W, Chen P, Choi K, Brandys M, Macchia RJ, and Rotman M: Diminished survival of young blacks with adenocarcinoma of the prostate. AmJ Clin Oncol 13: 465-469, 1990. 4. Pienta KJ, Demers R, Hoff M, Kau TY, Montie JE, and Severson RK: Effect of age and race on the survival of men with prostate cancer in the Metropolitan Detroit tricounty area, 1973 to 1987. Urology 45: 93-102, 1995. 5. MebaneC, GibbsT, andHormJ: Currentstatusofprostate cancer in North American black males. J Natl Med Assoc 82: 782-788, 1990. 6. Stamey TA, Yang N, Hay AR, McNealJE, Freiha FS, and Redwine E: PSA as a serum marker for prostate adenocarcinoma. N EnglJ Med 317: 909-916, 1987. 7. Catatona WJ, Smith DS, Ratliff TL, Dodds KM, Coplen DE, Yuan JJ, PetrosJA, and Andriole GL: The measurement of PSA as a screening test for prostate cancer. N EnglJ Med 24: 1156-1161, 1991. 8. Brawer MK, and Lange PH: PSA in the screening staging and follow-up of prostate cancer. WorldJ Urol 7: 7-11, 1989. 9. Brawer MK, Beatie J, Werner MH, Vessella RL, Preston SD, and Lange PH: Screening for prostate carcinoma with PSA. Results of the second year. J Urol 150: 106-109, 1993. 10. Catalona WJ, Smith DS, Ratliff TL, and Basler JW: Detection of organ-confined prostate cancer is increased through prostate-specific antigen-based screening. J Am Med Assoc 271: 192-193, 1993. 11. Mettlin C, Murphy GP, Lee F, Littrup PJ, Chesley A, Babaian R, Badalament R, Kane RA, and Mostofi FK: Characteristics of prostate cancer detected in the American Cancer Society-National Prostate Cancer Detection Project. J Urol 152(5 Pt 2): 1737-1740, 1994. 12. Trapasso JG, deKernionJB, Smith RB, and Dorey F: The incidence and significance of detectable levels of serum prostate specific antigen after radical prostatectomy. J Urol 152(5 Pt 2): 1821-1825, 1994. 13. Murphy G, Natarajan N, PontesJE, Schmitz RL, Smart CR, Schmidt JD, and Mettlin C: The national survey of prostate cancer in the United States by the American College of Surgeons. J Urol 127: 928-934, 1982. 14. Morgan TO, Jacobsen SJ, McCarthy WF, Jacobson DJ, UROLOGY 51 (2), 1998
McLean DG, and Mot.11JW: Age-specific reference ranges for serum prostate-specific antigen in black men. N Engl J Med 335: 304-310,1996. 15. Vijayakumar S, Karrison T, Weichselbaum RR, Chan S, Quadri SF, and Awan AM: Racial differences in prostate-specific antigen levels in patients with local-regional prostate cancer. Cancer Epidemiol Biomark Prev 1: 541-554, 1991. 16. Vijayakumar S, Weichselbaum R, Vaida F, Dale W, and Hellman S: Prostate-specific antigen levels in Afiican-Americans correlate with insurance status as an indicator of socioeconomic status. Cancer J Sci Am 2: 225-233,1996. 17. Teshima T, Hanlon AM, and Hanks GE: Pretreatment prostate-specific antigen values in patients with prostate cancer: 1989 patterns of care study process survey. lnt J Radiat Oncol Biol Phys 33: 809-814, 1995. 18. Moul JW, Sesterhenn IA, Connelly RR, Douglas T, Srivastava S, Mostofi FK, and McLeod DG: Prostate-specific antigen values at the time of prostate cancer diagnosis in AfiicanAmerican men. J Am Med Assoc 274: 1277-1281,1995. 19. Asbell SO, and Vijayakumar S: Racial differences in prostate-specific antigen levels in patients with local-regional prostate cancer. Prostate 31: 42-46, 1997. 20. Vijayakumar S, Winter K, Sause W, Gallagher MJ, Perez C, and Bondy M: African-American (AA) men with local-regional prostate cancer (PC) present with higher prostate specific antigen (PSA) levels than whites: results of RTOG 94-12. Int J Radiat Oncol Biol Phys Suppl36: 305,1996. 21. Whittemore A, Lele C, Friedman GD, Stamey T, Vogelman JH, and Orentreich N: Prostate-specific antigen as predictor of prostate cancer in black men and white men. J Nat1 Cancer Inst 87: 354-360, 1995. 22. Sakr WA, Haas GP, Cassin BF, Pontes JE, and Crissman JD: The frequency of carcinoma and intraepithelial neoplasia of the prostate in young male patients. J Urol 150: 379-385, 1993. 23. Sakr WA, Grignon DJ, Haas GP, Shomer KL, Heilbum LK, Cassin BJ, Powell J, Montie JA, Pontes JE, and Crissman JD: Epidemiology of high grade prostatic intraepithelial neoplasia. Path01 Res Pratt 191: 838-841, 1995. 24. Brawer M, Rennels MA, Nagle RB, Schifman R, and Gaines JA: Serum prostate-specific antigen and prostate pathology in men having simple prostatectomy. Am J Clin Path01 92: 760-764, 1989. 25. Cooner WI-I, Mosley BR, Rutherford CL Jr, Beard JH, Pond HS, Terry WJ, Igel TC, and Kidd DD: Prostate cancer detection in a clinical urological practice by ultrasonography, digital rectal examination and prostate specific antigen. J Urol 143: 1146-1152,199O. 26. Ross R, Bernstein L, Judd H, Hanisch R, Pike M, and Henderson B: Serum testosterone levels in healthy young black and white men. J Nat1 Cancer Inst 76: 45-48, 1986. 27. Ross RK, Bernstein L, Lobo RA, Shimizu H, Stanczyk
UROLOGY 51 (21, 1998
FZ, Pike MC, and Henderson BF: 5-alpha-reductase activity and risk of prostate cancer among Japanese and US white and black males. Lancet 339: 887-889, 1992. 28. Vijayakumar S, Quadri SF, Dong L, Ignacio L, Kathuria IN, Sutton H, and Halpren H: Results of a study to correlate serum prostate specific antigen and reproductive hormone levels in patients with localized prostate cancer. J Nat1 Med Assoc 87: 813-819, 1995. 29. Cooper CS, MacIndoe JH, Perry PJ, Yates WR, and Williams RD: The effect of exogenous testosterone on total and free prostate specific antigen levels in healthy young men. J Urol156: 438-441,1996. 30. Babaian RJ, Miyashita H, Evans RB, von Eschenbach AC, and Ramirez EI: Early detection program for prostate cancer: results and identification of high-risk patient population. Urology 37: 193-197, 1991. 31. Oesterling JE, Jacobsen SJ, Chute CG, Guess HA, Girman CJ, Panser LA, and Lieber MM: Serum prostate-specific antigen in a community-based population of healthy men. J Am Med Assoc 270: 860-864,1993. 32. Speights VO Jr, Brawn PN, Foster DM, Spiekerman AM, Kuhl D, and Riggs MW: Evaluation of age-specific normal ranges for prostate-specific antigen. Urology 45: 454-457, 1995. 33. Mettlin C, Littrup PJ, Kane RA, Murphy GP, Lee F, Chesley A, Badalament R, and Mostofi FK: Relative sensitivity and specificity of serum prostate specific antigen (PSA) level compared with age-referenced PSA, PSA density, and PSA change. Cancer 74: 1615-1620,1994. 34. Benson MC, Whang IS, Olsson CA, McMahon DJ, and Cooner WI-I: The use of prostate specific antigen density to enhance the predictive value of intermediate levels of serum prostate specific antigen. J Urol 147(3 Pt 2): 817-821, 1992. 35. Benson MC, Whang IS, Pantuck A, Ring K, Kaplan SA, Olsson CA, and Cooner WH: Prostate specific antigen density: a means of distinguishing benign prostatic hypertrophy and prostate cancer. J Uroll47(3 Pt 2): 815-816, 1992. 36. Albertsen PC: African-Americans and prostate cancer (commentary). Cancer J Sci Am 2: 200-201, 1996. 37. Vijayakumar S, Winter K, Sause W, Gallagher MJ, Michalski J, Raoch M, Porter A, and Bondy M: Prostate specific antigen levels are higher in African-American patients than whites in a multicenter registration study: results of RTOG 94-12. Int J Radiat Oncol Biol Phys (in press). 38. Vijayakumar S, Vaida F, Weichselbaum RR, and Hellman S: Race and Will Rogers phenomenon in prostate cancer. Cancer J Sci Am (in press). 39. Dale W, Vijayakumar S, Lawlor EF, and Merrell K: Prostate cancer, race, and socioeconomic status: inadequate adjustment for social factors in assessing racial differences. Prostate 29: 271-281, 1996.
305