Comparison of the clinical outcome after hormonal therapy for prostate cancer between Japanese and Caucasian men

Authors


Takashi Fukagai, Department of Urology, Showa University School of Medicine, Tokyo, Japan. e-mail: fukagai@attglobal.net

Abstract

OBJECTIVE

To investigate the impact of race on the effectiveness of hormonal therapy in patients with prostate cancer, by comparing the outcomes of Caucasian men (CM) and Japanese-American men (JAM) treated with hormonal therapy at one institution.

PATIENTS AND METHODS

Fifty-nine CM and 105 JAM with prostate cancer were treated with hormonal therapy at The Queen’s Medical Center in Honolulu. Age, stage, Gleason score, race, and pretreatment PSA levels were abstracted. The Kaplan–Meier method was used to construct overall and cause-specific survival curves, which were compared using log-rank statistics. These factors were assessed as to their interdependence and correlation with the clinical course using a Cox proportional hazards regression model.

RESULTS

Although there were no statistical differences in patient background, JAM who had received hormonal treatment had a better outcome than CM for overall and cause-specific survival rate (P = 0.001 and 0.036, respectively). Race was one of the significant prognostic factors in the multivariate analysis (P = 0.03). The findings suggest a difference in the effectiveness of hormonal therapy for prostate cancer in JAM living in Hawaii compared to CM.

CONCLUSIONS

There were marked racial differences in clinical outcome after hormonal therapy between JAM and CM. A prospective study with more patients might be necessary to elucidate the differential effectiveness of hormonal therapy for prostate cancer in different races, especially between Japanese and Caucasians.

Abbreviations
JAM

Japanese-American men

CM

Caucasian men

CAB

combined androgen blockade

DHT

dihydrotestosterone.

INTRODUCTION

Prostate cancer is the most common noncutaneous cancer in American men and the second most common cause of cancer death. It is estimated that, in 2005, 232 090 new cases will be diagnosed and 30 350 men will die from the disease [1]. However, the incidence of prostate cancer is markedly different among various races and countries. Japan has one of the lowest rates of prostate cancer [2]. Interestingly, the incidence in Japanese-American men (JAM) is intermediate between Japanese living in Japan and Caucasian men (CM) in the USA [3]. The reasons for the risk differential are unknown. Prostate cancer is well known as an androgen-dependent cancer and some studies indicate that the different character of prostate cancer in each race might be associated with the different hormonal environments [4,5]. This might influence the effectiveness of hormonal therapy in different races. To investigate the impact of race on the effectiveness of hormonal therapy in patients with prostate cancer, we compared the outcomes of CM and JAM treated by hormonal therapy for prostate cancer.

PATIENTS AND METHODS

Between January 1992 and December 2001, 59 CM and 105 JAM with prostate cancer treated with hormonal therapy at The Queen’s Medical Center in Honolulu were identified. Clinical information was obtained from the hospital tumour registry. All living patients were followed to the end of 2001 and none were lost to follow-up. Age, stage, Gleason score, race and pretreatment PSA levels were abstracted. The specific type of hormonal therapy each patient received (conventional surgical, medical castration alone or combined androgen blockade, CAB) was not available. Details of patient background of the two ethnic groups are given in Table 1. None of the patients received definitive surgical or radiation therapy. Differences in patient background between the groups were estimated by using Student’s t-test and the chi-square test. The clinical endpoint was survival; the Kaplan–Meier method was used to construct overall and cause-specific survival curves, which were compared using log-rank statistics. Age, stage, Gleason score, race, and pretreatment PSA levels were assessed as to their interdependence and correlation with the clinical course using a Cox proportional hazards regression model. The hospital’s Institutional Research Board approved the study.

Table 1.  Baseline characteristics of the two ethnic groups
VariableJAMCMP
  • *

    Clinical stage defined by 1997 TNM system;

  • †chi-square test.

Number of patients105 59 
Age, years
 median (range) 77 (53–92) 77 (51–90) 
 mean (sd) 75.8 (8.1) 76.5 (8.2)0.55
PSA level, ng/mL
 n (%) with PSA level of:
  ≤10 10 (10) 16 (27) 
  10.1–20 15 (14) 13 (22) 
  20.1–100 39 (37)  9 (15) 
  >100 26 (25)  11 (19) 
  unknown 15 (14) 10 (17) 
 median (range) 46.1 (3.9–9783) 16.2 (0.6–4000) 
 mean (sd) 351.1 (1437.4)152.3 (583.9)0.355
Gleason score
 Gleason score, 2–6/7/8–10/unknown, total n (%), at clinical stage*:
  I  0/0/0/0, 0  0/0/0/0, 0 
  II  7/23/22/1, 53 (50)  4/16/16/1, 37 (63) 
  III  2/5/2/1, 10 (10)  0/0/1/0, 1 (2)0.1
  IV  5/5/28/4, 42 (40)  4/2/14/1, 21 (36) 
  Total, n (%) 14 (13)/33 (31)/52 (50)/6 (6)  8 (14)/18 (31)/ 31 (53)/ 2 (3) 
 median (range)  8 (4–10)  8 (4–10) 
 mean (sd)  7.7 (1.2)  7.7 (1.4)0.883

RESULTS

The median (range) follow-up was 51 (2–126) months for JAM and 36 (2–88) months for CM. There were no statistically significant differences in the patients’ background between the groups (Table 1). Forty-five JAM and 32 CM died during the follow-up. At 60 months the overall survival rate was 65.5% for JAM and 41.7% for CM (P = 0.01). JAM who received hormonal treatment had a better overall and cause-specific survival rate than CM (P = 0.001 and 0.036, respectively) (Fig 1A,B). For patients with metastasis (CM 15, JAM 37) and with no metastasis (CM 44, JAM 68), the overall survival rate was higher in JAM (P = 0.032 and 0.007, respectively). To clarify in more detail the relationship between tumour progression and ethnic differences in survival after hormone therapy, we compared the survival curves between the ethnic groups according to PSA levels. Among patients whose PSA level was >100 ng/mL (CM 11, JAM 26), there was no statistically significant difference in the survival curves between CM and JAM (Fig. 2). We examined the simultaneous influence of five covariates (age, clinical stage, race, Gleason score, and pretreatment PSA level) on the time to death. Race was a significant prognostic factor in the multivariate analysis (P = 0.03), along with pretreatment PSA level (P = 0.03). These findings suggest a difference in the effectiveness of hormonal therapy for prostate cancer in JAM living in Hawaii compared to CM.

Figure 1.

The overall (A) and cause-specific survival (B) rate according to race.

Figure 2.

Overall survival rate according to race by PSA level, i.e. ≤ 100 ng/mL or >100 ng/mL.

DISCUSSION

The incidence of prostate cancer differs among the various races; Asians, especially Japanese and Chinese, have a lower incidence than Caucasians. The incidence in JAM is intermediate between Japanese living in Japan and CM in the USA [3]. The lower incidence among Japanese born in the USA than in whites in the USA might mean that Japanese immigrants retain some genetic and/or lifestyle characteristics that make their risk for prostate cancer less than that of USA whites.

Few publications have compared prostate cancer treatment and patient survival among different ethnic groups. Several studies indicate that race is a prognostic factor for African-American and White men. Eisenberger et al.[6], in a review of the National Cancer Institute Intergroup Study ♯0036 for prognostic factors in Stage D2 prostate cancer, noted that the median survival of the 107 African-American men was 26.4 months, compared with 33.6 months in the 442 CM, and concluded that African-American race was an adverse prognostic factor in patients with stage D2 prostate cancer treated with hormonal therapy. Similar results were found in other studies [7,8], but other authors did not find race to be an adverse prognostic factor in patients who had their prostate cancer treated with hormonal therapy [9,10]. McLeod et al.[9] reported no difference in the outcome of CAB for treating prostate cancer between African-American men and White men. Young et al.[11] reported, from the Surveillance Epidemiology and End Results programme data, that Japanese men had a higher survival from prostate cancer, but the reason for this was not established and survival was not examined by treatment received.

To our knowledge, the present is the first study to compare the prognosis after hormonal therapy between CM and JAM in the USA. The study included relatively few patients and the information on progression-free survival rate was not available. All patients were treated by the same group of urologists in one institution, and with the same follow-up management in case of relapse. The number of patients who received CAB was also unavailable in the study, but the survival benefit of CAB is limited [12]. It is also unlikely that only one ethnic group, the JAM, had a large proportion of CAB treatment, because both ethnic groups had almost the same clinical background and were treated by the same urologists. Therefore, we think that the study is significant even though we did not have this information. The study showed a better overall and disease-specific survival for JAM with hormone treatment for prostate cancer. In contrast, there was no racial difference in overall survival in the group of patients who received no treatment for prostate cancer during the same period (JAM 73, CM 79, P = 0.57). Shiraishi et al.[13] reported differences in the pattern of metastasis of prostatic carcinoma between Japanese in Japan, and JAM and CM in Hawaii. The present study showed that overall survival rates were higher for JAM with and with no metastasis than in CM. The findings indicate that JAM obtained more survival benefit with hormonal therapy than CM of all stages. However, the present analysis showed that for patients’ whose PSA level was >100 ng/mL (more advanced stage) there was no statistically significant difference in survival between CM and JAM. PSA level is a strong prognostic factor that correlates with the extension of prostate cancer. It is well known that prostate cancer shows various genetic changes associated with progression [14]. Genetic changes in the more advanced stages of cancer might reduce the effects of ethnic differences in prognosis after hormone treatment, as noted in the present study. One of the putative reasons for the difference might be fewer side-effects, especially cardiovascular disease, associated with hormonal therapy in JAM. Although the present outcome data did not include the cause of death, except cancer death, the JAM had a better prognosis in cause-specific survival and overall survival rate. Also, recent hormonal treatment methods, e.g. LHRH agonists, antiandrogens and orchidectomy, do not have the cardiovascular side-effects associated with previously used oestrogens. The findings might indicate a different sensitivity for hormonal therapy in JAM.

Several studies have shown racial differences in CAG repeat alleles. Irvine et al.[15] found that the prevalence of short CAG alleles was highest in African-American men with the highest risk of prostate cancer, intermediate in intermediate-risk non-Hispanic whites, and lowest in Asians, who are at very low risk of prostate cancer. Bratt et al.[16] found an association between long CAG repeat length and a good response to hormonal therapy. These findings might explain the present results showing a better prognosis in JAM after hormonal therapy, corresponding to the low prevalence of short CAG and high prevalence of long CAG repeats in Asians. However, other studies report conflicting results; Suzuki et al.[17] found that shorter CAG repeat length was correlated with a better response to hormonal therapy and prognosis in Japanese men with prostate cancer. Others also noted a higher frequency of short CAG repeat lengths in African-American men than in whites and Asians, but no difference in CAG repeat length between whites and Asians [18–20].

Another racial difference in association with androgen levels is 5α-reductase activity. Ross et al.[4] measured serum testosterone, dihydrotestosterone (DHT) and metabolites of DHT among young American black men, white men and native Japanese men. In that study, both black and white men had significantly higher serum levels of DHT metabolites than did native Japanese men. The influence of the different serum levels of DHT metabolites on the effectiveness of hormone treatment is unclear, but might potentially have an influence on the different effectiveness of hormone treatment between Japanese and CM. Further studies are needed to clarify the relation between CAG repeat alleles, 5α-reductase activity and response to hormonal therapy.

An ongoing study of other ethnic groups in Hawaii has also shown a different prognosis after hormonal therapy (unpublished data). Chinese men show almost the same prognosis as JAM and a better prognosis than CM. Filipinos show a worse prognosis after hormonal therapy than JAM but better than CM. Interestingly, the prognosis after hormonal therapy is inversely correlated with the prostate cancer incidence of the ethnic group. As previously noted, the incidence of prostate cancer within ethnic groups might relate to the exogenous or endogenous hormonal environment, which in turn might also relate to the different effectiveness of hormonal therapy.

In conclusion, there was a marked racial difference in clinical outcome after hormonal therapy between JAM and CM. A prospective study with more patients might be necessary to elucidate the differential effectiveness of hormonal therapy of prostate cancer in different races, especially between Japanese and Caucasians.

ACKNOWLEDGEMENTS

We thank Dr Fernand Labrie (Oncology and Molecular Endocrinology Research Center, Laval University Medical Center, Quebec City, Quebec, Canada) for his advice and critical reading of this manuscript. We also thank the data management staff of Oncology Data Registry at The Queen’s Medical Center.

CONFLICT OF INTEREST

None declared.

Ancillary