Clinical characteristics of African- American men with hereditary prostate cancer: the AAHPC study

Transcription

1 Clinical characteristics of African- American men with hereditary prostate cancer: the AAHPC study (2004) 7, & 2004 Nature Publishing Group All rights reserved /04 $ ,2 *, A Baffoe-Bonnie 3,12, R Kittles 1,4, C Pettaway 5, I Powell 6, C Royal 1,7, H Wang 3, S Vijayakumar 8, J Bennett 9, G Hoke 10, T Mason 11, J Bailey-Wilson 12, W Boykin 9, K Berg 12, J Carpten 13, S Weinrich 14,JTrent 13, G Dunston 1,4 & F Collins 12 1 National Human Genome Center, Howard University, Washington, DC, USA; 2 Division of Urology, Howard University, Washington, DC, USA; 3 Fox Chase Cancer Center, Philadelphia, Pennsylvania, USA; 4 Department of Microbiology, Howard University, Washington, DC, USA; 5 MD Anderson Medical Center, Houston, Texas, USA; 6 Karmanos Cancer Institute, Wayne State University, Detroit, Michigan, USA; 7 Department of Pediatrics, Howard University, Washington, DC, USA; 8 University of Illinois, Chicago, Illinois, USA; 9 Midtown Urology, Atlanta, Georgia, USA; 10 Columbia-Presbyterian Medical Center, New York, New York, USA; 11 Michael Reese Hospital, Chicago, Illinois, USA; 12 National Human Genome Research Institute, NIH, Bethesda, Maryland, USA; 13 Translational Genomics Research Institute, Genetic Basis of Human Disease Research Division, Phoenix, Arizona, USA; and 14 University of South Carolina, Columbia, South Carolina, USA Introduction: The African-American Hereditary Prostate Cancer (AAHPC) Study was designed to recruit African-American families fulfilling very stringent criteria of four or more members diagnosed with prostate cancer at a combined age at diagnosis of 65 years or less. This report describes the clinical characteristics of a sample of affected AAHPC family members. Methods: In all, 92 African-American families were recruited into the study between 1998 and Complete clinical data including age and PSA at diagnosis, number of affected per family, stage, grade, and primary treatment were available on 154 affected males. Nonparametric Wilcoxon two-sample tests and Fisher s exact test (two-tailed), were performed to compare families with 4 6 and 46 affected males with respect to clinical characteristics. Results: The mean number of affected men per family was 5.5, with a mean age at diagnosis of 61.0 (78.4) years. Age at diagnosis, PSA and Gleason score did not show significant differences between the two groups of families. Based on the Gleason score, 77.2% of affected males had favorable histology. Significantly, there were marked differences between the two groups in the frequency of node-positive disease (P ¼ 0.01) and distant metastases (P ¼ ). Radical prostatectomy was the preferred primary therapy for 66.2% of all affected men followed by 20.8% who chose radiation therapy. Conclusions: Our findings suggest that affected males who carry the highest load of genetic factors are at the highest risk for early dissemination of disease, thus efforts at early diagnosis and aggressive therapeutic approaches may be warranted in these families. Since the primary therapy choices in our study favored definitive treatment (87.0%) *Correspondence:, Division of Urology, Howard University Hospital, 2041 Georgia Avenue Suite 4C02, Washington, DC 20060, USA. cahaghotu@howard.edu Received 7 October 2003; revised 10 March 2004; accepted 12 March 2004

2 166 when compared to the 1983 and 1995 SEER data in which 28 and 64% received definitive treatment, respectively, it appears that affected African-American men in multiplex families may be demonstrating the reported psycho-social impact of family history on screening practices and treatment decisions for prostate cancer. (2004) 7, doi: /sj.pcan Keywords: hereditary prostate cancer; African Americans Introduction Prostate cancer is the most common cancer in American menotherthanskincancer.intheunitedstatesinthe year 2003, it is estimated that about new cases of prostate cancer will be newly diagnosed, with about deaths from this disease. Multiple epidemiological studies have confirmed that prostate cancer occurs at a higher prevalence and with greater morbidity in African-American men than in most other populations. 1 5 The age-adjusted incidence rate of invasive prostate cancer for African-American men was per for the period from 1996 to compared to per in White males. Recently reported prostate cancer age-adjusted mortality rates between 1996 and 2000, are approximately two times worse among African-American than Caucasian men. 6 Agespecific mortality rates for prostate cancer are reportedly higher among African-American men compared to Caucasian men for all ages after age 40 y. 6 The reasons for these observations are still not known. Affected males from families with hereditary prostate cancer (HPC) account for 5 10% of all reported cases of prostate cancer in the United States and Europe. 7 9 It is against this background that the African-American Hereditary Prostate Cancer Study (AAHPC) was initiated to recruit African-American families to a hereditary prostate cancer study. This is the first large-scale study to describe the clinical characteristics of hereditary prostate cancer in affected African-American men from multiplex families. This elaborate multi-institutional network was designed to recruit African-American families that fulfill very stringent criteria for hereditary prostate cancer, exceeding those originally proposed by Carter et al. 10 The AAHPC inclusion criteria were as follows: (a) African-American families with four or more members diagnosed with prostate cancer, (b) the combined age at diagnosis of all affected members in the family should average 65 y or less, (c) at least three affected members must donate a sample of blood for genotyping. These criteria differ from the original HPC definition proposed by Carter et al, 7,10 in which the hereditary subgroup consisted of three successive generations with prostate cancer, a clustering of three or more affected individuals in a nuclear family and/or two relatives with early onset (less than 55 y) of prostate cancer. The details of the AAHPC network infrastructure and recruitment protocols have been previously reported. 3,11 Informed consent was obtained on all participants. We present a descriptive report of the clinical characteristics of a sample of affected AAHPC family members. Materials and methods In the AAHPC study, age at diagnosis was available for 191 ; therefore, the age parameters were calculated with this denominator. Of these individuals, 154 had complete clinical data that were retrospectively reviewed from medical and pathological records. The clinical parameters of interest included age at diagnosis, serum prostate-specific antigen (PSA) levels, tumor stage and grade, and primary therapy (which was obtained from questionnaires and patient medical records). Stage and grade data were standardized using the 1997 Tumor- Node-Metastasis staging classification 12 (local tumor growth (T), spread to regional lymph nodes (N), and distant metastases (M)), together with the Gleason score. 13 Distant metastases (M þ ) and node-positive disease without the presence of metastases (N þ, Mo) were also recorded. Pathologic stages were available for prostatectomy cases only. The families were dichotomized into those with 4 6 affected men vs those with more than six affected. We chose this stratification because we expected that families with more than six affected men might have certain unusual characteristics, such as more aggressive disease. The number of affected men with prostate cancer per family was calculated for each group (4 6, 46), with subsequent evaluation of all clinical parameters according to this stratification. For statistical analysis, nonparametric Wilcoxon two-sample tests were performed to compare these two groups in terms of the continuous variables such as age at diagnosis and PSA levels. Tumor stage was classified as localized (T1 T2/No/Mo), locally advanced (T3 T4/ No/Mo), locally advanced disease with regional lymph nodes (N þ /Mo), or distant metastases (M þ ). Fisher s exact test (two-tailed) was used to compare 4 6 and 46 two groups with respect to clinical characteristics such as tumor stage, PSA level at diagnosis (o10, 10 20, 420 ng/ml) and Gleason score (2 6, 7 10). Results In all, 92 families with 346 affected men were recruited from multiple sites within the United States between 1998 and Age at diagnosis information was available on 191 affected men, but complete clinical data were available on 154 men. This extent of missing information occurred due to the difficulty of obtaining clinical data on relatives long deceased for whom adequate records were no longer available. Clinical characteristics and primary therapy are shown in Tables 1 and 2, using all available

3 Table 1 Clinical characteristics of affected men in 92 African-American hereditary prostate cancer families Number of affected per family Total (N ¼ 154) 4 6 (n ¼ 108) 46 (n ¼ 46) P-value (4 6 vs 46) 167 Mean number of affected per family a Range (4 11) (4 6) (7 11) Age at diagnosis (years) a Mean7s.d Median Range (32 80) (41 80) (32 74) Ager60 y (%) 79 (41.4) 60 (42.0) 19 (39.6) Age460 y (%) 112 (58.6) 83 (58.0) 29 (60.4) PSA (ng/ml) b Mean7s.d Median Range ( ) c ( ) c ( ) PSA (ng/ml) o (66.0) 74 (69.0) 28 (61.0) (25.0) 28 (26.0) 11 (24.0) (9.0) 6 (5.0) 7 (15.0) Gleason score (77.2) 95 (88.0) 24 (52.2) (11.0) 13 (12.0) 4 (8.7) Unknown 18 (11.8) 0 18 (47.8) Tumor identification T1 T2/No/Mo 105 (68.2) 81 (75.0) 24 (52.1) 1.0 T3 T4/No/Mo 16 (10.4) 13 (12.0) 3 (6.5) Unknown 43 (27.9) 14 (13.0) 19 (41.3) Node positive disease Present (N+/Mo) 19 (12.3) 9 (8.4) 10 (21.7) 0.01 Absent 121 (78.6) 94 (87.0) 27 (58.7) Unknown 14 (9.1) 5 (4.6) 9 (19.6) Distant metastases Present (M+) 14 (9.1) 3 (2.1) 11 (23.9) Absent 121 (78.6) 94 (87.0) 27 (58.7) Unknown 19 (12.3) 11 (10.9) 8 (17.4) a Based on 191 affected men with age at diagnosis. b PSA as a continuous variable. c Two outlier values were 210 and 253 ng/ml. Table 2 Primary treatment choices of 154 affected men in 92 African-American hereditary prostate cancer families Primary therapy Frequency Percent Watchful waiting Radical prostatectomy a Radiotherapy a Hormonal therapy a Definitive treatment. information. Overall 70% (64/92) of all families had four to six affected men with prostate cancer (n ¼ 108), while the remaining 30% (28/92) had more than six affected men per family with a range of 7 11 (n ¼ 46). For all families, the mean number of affected men per family was 5.5, with a mean age at diagnosis of 61.0 (78.4) y. For the 46 group, the mean number of affected per family was 8.5, with a mean age at diagnosis of 59.9 (79.4) y. The median age was 62 y across groups. The mean PSA level for affected males at diagnosis for all families was 19.3 (730.7) ng/ml, 21.2 (734.6) ng/ml for the 4 6 group and 13.9 (713.3) ng/ml for the 46 group. Without the two extreme high PSA values of 210 and 253 ng/ml in the 4 6 group, the mean PSA level was 17.3 (719.0) for that group, 13.9 (713.3) and 16.4 (717.7) for the 46 group, and for all families, respectively. PSA at diagnosis did not show significant difference between the two groups (P ¼ 0.578). PSA levels were also stratified into three categories similarly to those used in previous studies 14 and compared between the 4 6 and 46 groups. There was no significant difference between the two groups (P ¼ 0.158). Based on the Gleason score, 77.2% (119/154) had a range of 2 6 compared with 11.0% with higher grades 7 10 of histology. There was no difference between 4 6 and 46 group in terms of Gleason score (P ¼ ). In all families, 68.2% (105/154) of affected men presented with organ-confined disease (T1 T2/No/ Mo) at diagnosis compared with 10.4% (16/154) with more advanced disease (T3 T4/No/Mo). The families with 46 affected men had a significantly higher proportion of node-positive prostate cancer (21.7%) compared with 8.4% among families with 4 6 affected men (P ¼ 0.01). The presence of distant metastases at diagnosis also differed between these two groups, with 23.9%

4 168 (11/46) in the 46 group vs 2.1% (3/108) in the 4 6 group (Po0.0001). Radical prostatectomy was the preferred primary therapy for 66.2% of all affected men (102/154) followed by 20.8% (32/154) who chose radiation therapy (Table 2). The proportion of affected men receiving definitive therapy, 87.0% (134/154) was much higher compared to that of those choosing watchful waiting (2.0%). Discussion Few studies have evaluated the clinical characteristics of hereditary prostate cancer among African-American families only. The International Consortium for Prostate Cancer Genetics (ICPCG) in the year 2001, had a total of 1876 families from the US, Europe, and Australia. There were 146 African-American families, forming 7.8% of the total (William Issacs, personal communication). AAHPC contributed 63.0% (92/146) of the families in the African- American pool. Previous studies have included a few African-American HPC families as part of the enrolled non-white populations in these studies, 9,15 but the AAHPC study has the largest collection of African- American HPC families ascertained through probands. Besides having many affected males, these families are also notable for early onset of disease, with a mean age at diagnosis of 61.0 y. Following the Carter et al 10 categorization of prostate cancer into sporadic, familial, and hereditary subgroups, the clinical characteristics of HPC families, and how they differ from sporadic cases, have been the subject of considerable controversy in the literature A review of the clinical characteristics of 74 North American HPC families potentially linked to the HPC1 locus on chromosome 1 demonstrated three distinct clinical characteristics, including earlier age of onset, higher grade tumors, and more advanced disease at diagnosis compared to agematched sporadic cases. 20 Bratt et al 21 recently reported earlier age of onset in Swedish HPC families compared to sporadic cases, but found no differences in clinical characteristics or survival between the two groups. Other reports have reported no differences between biological aggressiveness of hereditary prostate cancer and sporadic cases. Bauer et al 22 reviewed 474 radical prostatectomy patients, 78 of whom had at least one affected first-degree family member. The mean age at diagnosis for this radical prostatectomy series was 62 y. Bauer et al 22 found no differences in histologic grade, margin positive status, capsular penetration, or disease-free survival in patients with positive family history vs sporadic cases. Bova et al 17 reported similar findings. However, neither study 17,22 reported the numbers of affected per family. Hence, these families may represent familial cases vs hereditary cases as defined by Carter et al. 10 Schleutker et al 18 reported late age of onset in Finnish HPC families linked to the HPCX susceptibility locus. The important findings from the current study include the considerably lower mean age at diagnosis of 61.0 y for all affected men in the AAHPC study, compared to the overall HPC family collections in the ICPCG. Additionally, the mean number of 5.5 cases per family exceeds the already stringent criteria for defining hereditary prostate cancer. Both of these observations support the conclusion that the AAHPC families may be enriched for cases in which genetic factors are playing a particularly prominent role. This circumstance might be expected to be most apparent in those families with the largest number of affected cases, and so we stratified the sample into those families with 4 6 affected males, and those with 46 affected males. Gleason scores in affected males were not different between these two groups. Overall, 77.2% of the affected men in this study demonstrated relatively favorable histology, with Gleason scores of 2 6 compared to 11% with scores of Keetch et al 19 reported mean Gleason scores that were lower in HPC families compared to sporadic cases, although their findings did not reach statistical significance. Nippon et al 23 reported similar findings but also demonstrated that their younger patients presented with earlier stage disease. Bastacky et al 24 noted that hereditary tumors had a significantly lower mean Gleason score of 4.9 vs 5.6 in sporadic cases. In a retropubic prostatectomy series of 321 African- Americans (29% of whom had a positive family history), Narain et al 5 found no differences in age or PSA at diagnosis, distribution of histologic grade or stage between familial and sporadic cases. Further investigation of associated prostate pathology, particularly highgrade PIN and prostatitis, may provide more insight into the natural history of prostate cancer in HPC. Significantly, however, there was a marked difference between the 4 6 and 46 groups in the frequency of nodepositive disease (8.4 vs 21.7%) and distant metastases (2.1 vs 23.9%). This supports the conclusion that affected males who carry the highest load of genetic factors are at the highest risk for early dissemination of disease. That would suggest in turn that efforts at early diagnosis and aggressive therapeutic approaches may be particularly warranted in these families. Before considering that conclusion as proven, however, it is noteworthy that other researchers have reported no association between familial predisposition and clinical features or clinical course of the disease. Thus, whether hereditary prostate cancer is truly distinct from sporadic forms will probably await more accurate stratification based on discovery of the precise underlying heritable alterations. 25 Finally, it is of note that primary therapy choices in our study favored definitive treatment (87.0%) when compared to the 1983 and 1995 SEER data in which 28 and 64% received definitive treatment, respectively, 26,27 thus suggesting that affected African-American men in multiplex families may be demonstrating the reported psychosocial impact of family history on screening practices and treatment decisions for prostate, breast, and ovarian cancer These men may be modifying their treatment choices based on past experience with certain practices and outcomes among other affected family members. Watchful waiting may be less attractive in men with hereditary prostate cancer, who may have had the experience of close relatives dying of metastatic disease. 21 Acknowledgements The African-American Hereditary Prostate Cancer Study (AAHPC) is a national collaboration initially funded by the National Center for Minority Health and Health

5 Disparities, the National Human Genome Research Institute, and the National Cancer Institute of the National Institutes of Health, under NIH Contract HG This work was partially supported by USPHS Grant CA and an appropriation from the Commonwealth of Pennsylvania. Source of Funding: N01-ES /1/01-9/30/02 NHGRI Supplement for African American Hereditary Prostate Cancer Study Network References 1 Clegg L et al. Cancer survival among US whites and minorities: a SEER (Surveillance, Epidemiology, and End Results) Program population-based study. Arch Intern Med 2002; 162: Zeigler-Johnson C. CYP3A4: a potential prostate cancer risk factor for high-risk groups. Clin J Oncol Nurs 2001; 5: Powell IJ, Meyskens Jr FL. African-American men and hereditary/familial prostate cancer: Intermediate-risk populations for chemoprevention trials. Urology 2001; 57 (Suppl 1): Cunningham GR et al. Familial aggregation of prostate cancer in African-Americans and white Americans. Prostate 2003; 56: Narain V, Cher ML, Wood Jr DP. Prostate cancer diagnosis, staging and survival. Cancer Metastasis Rev 2002; 21: Ries L et al. (eds). SEER Cancer Statistics Review, National Cancer Institute: Bethesda, MD Carter BS et al. Mendelian inheritance of familial prostate cancer. Proc Natl Acad Sci USA 1992; 89: Cooney KA et al. Prostate cancer susceptibility locus on chromosome 1q: a confirmatory study. J Natl Cancer Inst 1997; 89: Cooney KA. Hereditary prostate cancer in African-American families. Semin Urol Oncol 1998; 16: Carter BS et al. Hereditary prostate cancer: epidemiologic and clinical features. JUrol1993; 150: Royal C et al. Recruitment experience in the first phase of the African American Hereditary Prostate Cancer (AAHPC) study. Ann Epidemiol 2000; 10 (Suppl 8): S68 S AJCC. American Joint Committee on Cancer AJCC Cancer Staging Manual, 5th Edn. Lippincott-Raven: Philadelphia Gleason D. Histologic grading and clinical staging of prostatic carcinoma. In: Tannenbaum (ed), Urologic Pathology: The Prostate. Lea and Febiger: Philadelphia 1987; pp Goode EL et al. Clinical characteristics of prostate cancer in an analysis of linkage to four putative susceptibility loci. Clin Cancer Res 2001; 7: Smith JR et al. Major susceptibility locus for prostate cancer on chromosome 1 suggested by a genome-wide search. Science 1996; 274: Paiss T et al. Association between the clinical presentation and epidemiological features of familial prostate cancer in patients selected for radical prostatectomy. Eur Urol 2003; 43: Bova GS et al. Biological aggressiveness of hereditary prostate cancer: long-term evaluation following radical prostatectomy. J Urol 1998; 160 (Part 1): Schleutker J et al. A genetic epidemiological study of hereditary prostate cancer (HPC) in Finland: frequent HPCX linkage in families with late-onset disease. Clin Cancer Res 2000; 6: Keetch DW et al. Clinical and pathological features of hereditary prostate cancer. JUrol1996; 155: Gronberg H et al. Characteristics of prostate cancer in families potentially linked to the hereditary prostate cancer 1 (HPC1) locus. JAMA 1997; 278: Bratt O, Damber JE, Emanuelsson M, Gronberg H. Hereditary prostate cancer: clinical characteristics and survival. JUrol2002; 167: Bauer JJ et al. Significance of familial history of prostate cancer to traditional prognostic variables, genetic biomarkers, and recurrence after radical prostatectomy. Urology 1998; 51: Nippon S et al. An analysis of familial prostate cancer. Nippon Hinyokika Gakkai Zasshi. 1995; 86: Bastacky SI et al. Pathological features of hereditary prostate cancer. JUrol1995; 153 (Part 2): Paiss T et al. Familial versus sporadic prostate cancer in the German population. Clinical and pathological characteristics in patients after radical prostatectomy. Urologe A 2003; 42: Stanford J et al. Surveiilance, epidemiology and end results. (eds). In: Prostate Cancer Trends, NCI: Bethesda, MD Harlan L et al. Geographic, age, and racial variation in the treatment of local/regional carcinoma of the prostate. J Clin Oncol 1995; 13: Reis-Starr C, Weinrich SP, Creanga D, Weinrich M. The association between family history and participation in free prostate cancer screening. Am J Health Studies 1998; 14: Bratt O, Kristofferson U, Lundgren R, Olsson H. Sons of men with prostate cancer: their attitudes regarding possible inheritance of prostate cancer, screening and genetic testing. Urology 1997; 50: Calle E, Flanders WD, Thun MJ, Martin LM. Demographic predictors of mammography and pap smear screening in U. S. women. Am J Public Health 1993; 83: