Abstract
Aim: Although the impact of testosterone or obesity on the efficacy of androgen-deprivation therapy (ADT) has been reported, there exist few comprehensive analyses on the impact of these factors on ADT outcome. Therefore, in the present study, we investigated the relationship between serum testosterone or body mass index (BMI) and prognosis among men treated with primary ADT for metastatic prostate cancer. Patients and Methods: The study included fifty-six Japanese patients with prostate cancer treated at our Institution from 2000 through 2012. The relationship between serum testosterone or BMI and progression-free survival, cancer-specific survival, and overall survival among men with metastatic prostate cancer treated with primary ADT was examined. Results: The median of serum testosterone and BMI were 397 ng/dl (interquartile range (IQR), 278-464 ng/dl) and 21.9 kg/m2 (IQR, 19.2-23.6 kg/m2), respectively. Median progression-free survival, cancer-specific survival, and overall survival were 23.2 months, 68.9 months, and 68.1 months, respectively. Among clinicopathological parameters, the lowest-quartile group of serum testosterone level was a significant predictor of poor cancer-specific survival and overall survival as well as survival from castration resistance. However, BMI was not associated with prognosis. Conclusion: Serum testosterone level, but not obesity, is a prognostic factor for outcome including survival after getting castration-resistant prostate cancer in men with metastatic prostate cancer having undergone primary ADT.
Although the incidence rate of advanced prostate cancer has been decreasing in Western countries, likely due to early detection by widespread prostate-specific antigen (PSA) screening, approximately 5% of men diagnosed with prostate cancer in the United States are still diagnosed at advanced stages with nodal or distant metastatic disease (1). In Japan, as high as 10-20% of men with prostate cancer are diagnosed with metastatic disease, probably due to inadequate PSA screening (2, 3). In addition, Asian countries including Japan have been suffering from an increased prevalence rate of prostate cancer due to the expansion of the western-style diet and lifestyle and an increase of the aged population (4). Androgen-deprivation therapy (ADT), which reduces androgen production and inhibits androgen action in prostate cancer cells, has been a mainstay for the treatment of recurrent or advanced prostate cancer since the 1940s (5). So far, PSA level at diagnosis, pathological grade by Gleason score at biopsy and TNM stages are well-known predictors of prognosis in men with prostate cancer treated with primary ADT, and a predictive system using these parameters has been created and validated (6, 7).
Previous studies have shown that the androgen milieu before treatment was associated with prognosis among patients undergoing radical prostatectomy (8, 9) as well as re-classification among men under active surveillance (10). In particular, the pre-treatment androgen milieu level in serum (11-14) and tissues (15, 16) has been also associated with prognosis of patients treated with primary ADT. In contrast, recent studies failed to show a significant association between serum testosterone and prognosis (17). On the other hand, obesity represented by body mass index (BMI) has been also reported to associate with prognosis among patients treated with surgical therapy and radiotherapy (18). In addition, the prognosis among obese men that underwent salvage ADT for recurrence after radical prostatectomy was poor compared to non-obese men (19). However, investigations of the association between obesity and prognosis among men that underwent primary ADT have been limited.
Thus, both testosterone and obesity are possible predictors of prognosis in men treated with primary ADT. In addition, serum testosterone level has been reported to be associated with BMI among men (20). However, a comprehensive analysis using both parameters associated with ADT outcome is laeking. Therefore, in the present study, in order to establish their significance in primary ADT for metastatic prostate cancer, we examined the impact of serum testosterone level and BMI before treatment on prognosis in men with metastatic prostate cancer.
Patients and Methods
Japanese patients with prostate cancer treated with primary ADT at the Kyushu University Hospital (Fukuoka, Japan) from 2000 to 2012 and with serum testosterone measurement obtained before treatment, were enrolled in the study. Cases with regional lymph node or distant metastases were included. Blood from patients was obtained between 8:00 and 10:00 in the morning and serum testosterone was measured by the electrochemiluminescence immunoassay method. This study was approved by the institutional review board. Patients who received local treatments with ADT or other treatments before disease progression were excluded. Clinical TNM staging was determined in accordance with the unified TNM criteria based on the results of a digital rectal examination, transrectal ultrasound, magnetic resonance imaging, computed tomography and bone scan (21). All patients were primarily treated by ADT with surgical castration or medical castration using a luteinizing hormone-releasing hormone agonist (goserelin acetate or leuprorelin acetate) and/or an anti-androgen agent (bicalutamide, flutamide, or chlormadinone acetate). Progressive disease was defined as an increase in serum PSA of >2 ng/mL and a 25% increase over the nadir, or the appearance of a new lesion or the progression of one or more known lesions classified according to the Response Evaluation Criteria in Solid Tumours (RECIST) (22). Radiographic progression was defined as the progression of measurable disease or bone scan progression.
All statistical analyses were performed using the JMP9 software (SAS Institute, Cary, NC, USA). The correlations between parameters were examined by the Student's t-test, Pearson's test and Pearson product-moment correlation coefficient. Univariate and multivariate analyses were performed using the logistic regression model. p-Values <0.05 were considered significant.
Results
The clinical and pathological characteristics of 56 Japanese patients whose serum testosterone level was measured before primary ADT are shown in Table I. The median age of the patients was 74 years (interquartile range [IQR], 67–80 years) and the median PSA was 172.3 ng/mL (IQR, 82.5-687.0 ng/mL) at diagnosis. Gleason scores of biopsy specimens from 9 patients (17.3%) and 43 patients (82.7%) were ≤7 and >7, respectively. Clinical T-stage was cT2, cT3 and cT4 in 17 patients (30.9%), 22 patients (40.0%) and 16 patients (29.1%), respectively. Among 56 patients in total, 36 patients (65.5%) had metastasis to regional lymph nodes (N1). Fifty patients (89.3%) and 6 patients (10.7%) were primarily treated with combined-androgen blockade and castration alone, respectively. During the median follow-up period of 47.0 months, disease progression, cancer-specific death, and all-cause death occurred in 37 cases (66.1%), 22 cases (39.3%), and 25 cases (44.6%), respectively. Median progression-free survival (PFS), cancer-specific survival (CSS), and overall survival (OS) were 23.2 months, 68.9 months, and 68.1 months, respectively. The 5-year PFS rate was 26.3%, while the 5-year CSS and OS rates were 68.1% and 63.7%, respectively.
Patients' characteristics.
Using this cohort, we performed a univariate analysis for PFS, CSS and OS based on clinicopathological factors including age, Gleason score, PSA at diagnosis, and clinical stage as well as serum testosterone level and BMI before treatment. In univariate analysis on PFS, no clinicopathological factors, including serum testosterone and BMI before treatment, were significant in the examined cohort (Table II). On the other hand, as shown in Table II, clinical T-stage as well as serum testosterone level before treatment was associated with CSS. Furthermore, serum testosterone before treatment was also associated with OS (Table II). Accordingly, we focused on serum testosterone level at pre-treatment and analyzed its impact on prognosis. When serum testosterone was divided into quartiles (1st quartile, Q1; 2nd quartile, Q2; 3rd quartile, Q3; 4th quartile, Q4), the lowest quartile of serum testosterone group showed the worst PFS (p=0.19; Figure 1A), CSS (p=0.061; Figure 1B) and OS (p=0.026; Figure 1C). In addition, when serum testosterone was divided into lowest quartile (Q1) and 2nd to 4th quartiles (Q2-Q4), the lowest quartile of serum testosterone group showed worse prognosis with regard to PFS (p=0.036; Figure 2A), CSS (p=0.0076; Figure 2B) and OS (p=0.0024; Figure 2C). On the other hand, BMI was not associated with PFS, CSS and OS. When survival rate from the progression was analyzed, the lowest quartile (Q1) of serum testosterone group showed the worst prognosis, compared with the other quartile (Q2, Q3, Q4) groups (p=0.10; Figure 3A) and the 2nd to 4th quartiles (Q2-Q4) group (p=0.014; Figure 3B). However, there was no difference in clinicopathological parameters among quartile testosterone groups except that the 4th quartile (Q4) testosterone group showed a higher frequency of a Gleason score less than 7 (Table III).
The association beween clinicopathological characteristics and prognosis.
Characteristics of 56 patients according to serum testosterone level.
Prognostic stratification in patients with metastatic prostate cancer treated with primary ADT according to serum testosterone level before treatment divided into quartiles. Progression-free survival (PFS) (A), cancer-specific survival (CSS) (B), and overall survival (OS) (C) in patients stratified by quartile (Q1, Q2, Q3, Q4) of serum testosterone level.
Prognostic stratification in patients with metastatic prostate cancer treated with primary ADT according to serum testosterone level before treatment divided into 1st quartile and 2nd to 4th quartiles. PFS (A), CSS (B), and OS (C) in patients stratified by 1st quartile (Q1) and 2nd to 4th quartiles (Q2-Q4) of serum testosterone level.
Stratification of overall survival from the progression to castration resistance in patients with metastatic prostate cancer treated with primary ADT according to pre-treatment serum testosterone level. Overall survival from the progression to castration resistance in patients stratified by serum testosterone level divided into (A) quartiles (Q1, Q2, Q3, Q4) or (B) into 1st quartile (Q1) and 2nd to 4th quartiles (Q2-Q4).
The associations between BMI or age and serum testosterone level. The plot graph of BMI (A) or age (B) and serum testosterone level are indicated.
In contrast, although serum testosterone level was previously reported to associate with BMI, there was no correlation between serum testosterone level and clinicopathological parameters (data not shown) as well as BMI in this cohort (r=0.18, p=0.18; Figure 4A). However, only age was associated with serum testosterone level before treatment (r=−0.44, p=0.0006; Figure 4B).
Discussion
Previous studies have suggested that pathological grade, serum PSA level and disease extension could well predict prognosis among men that underwent primary ADT (6, 7). In addition, several clinicopathological parameters including serum testosterone and obesity represented by BMI suggested an association of these parameters with the prognosis of primary ADT. Low serum male hormone level before treatment was shown to be associated with poor prognosis in US (11, 12) and Asian cohorts (13, 14). In line with these findings, our study showed the worse prognosis in the lowest quartile group of serum testosterone level, although the prognosis among men in the 2nd to 4th quartile groups of serum testosterone level was similar.
Several hypotheses on the relationship between serum testosterone and prostate cancer characteristics have been suggested. In addition to the classical linear model between testosterone and prostate cancer characteristics, both the U-shaped model (23) and saturation model (24, 25) have been proposed as non-linear models. Our findings presented in this study support the saturation model as indicated in Table II.
BMI has been also reported to associate with the prognosis of primary ADT (19). However, the present study could not show a significant correlation between the prognosis of primary ADT and BMI, indicating there are controversies in the relationship between obesity and outcome of primary ADT. Since the prevalence of obesity has been known to differ among ethnic groups, with a higher rate in the US, intermediate rate in Europe and lower rate in Asian countries (26), ethnic differences in the obese prevalence rates may account for the differential impact of BMI on the prognosis of primary ADT. In addition, the varying conditions among studies might also be a cause of controversial results.
J-CAPRA score employing serum PSA level at diagnosis, Gleason score at biopsy and TNM-stage was developed using a large database from Japanese patients undergoing primary ADT, and is valuable to predict the prognosis in patients with not only localized disease, but advanced cancer when treated primarily with ADT. In addition to other clinicopathological parameters such as age (27) and PSA kinetics (28), the usefulness of several biomarkers such as expression, amplification, mutation, and polymorphisms in the androgen receptor gene (AR) and AR-related genes as well as serum markers have been reported for the prediction of prognosis among men that underwent primary ADT (30). Furthermore, several studies, including the current study, showed a great prognostic impact of serum testosterone level before treatment. In the future, the incorporation of these biomarkers including serum testosterone level before treatment into the J-CAPRA score system may improve its accuracy. In addition, the present study showed worse prognosis from the time of development into castration-resistant prostate cancer (CRPC) in men with low serum testosterone level. The COU-AA-301 trial recently explored the significance of serum testosterone during ADT on OS in men with metastatic CRPC treated with abiraterone acetate in a post-chemotherapy setting (30). However, to the best of our knowledge, this study showed for the first time the significance of serum testosterone before ADT on survival from CRPC.
It is well-known that prognosis of high-risk prostate cancer patients is poor and an improvement of outcome in high-risk patients has been required. A previous study reported that inadequate suppression of androgen in serum may be a poor prognostic marker among patients treated with ADT (31), suggesting that ADT-resistant prostate cancer requires a more complete form of ADT than conventional ADT by castration and anti-androgens. Recently, several novel agents targeting both AR signaling and non-AR signaling have been and are being developed (32). Therefore, novel therapeutic strategies such as up-front use of chemotherapeutic agents or novel agents combined with conventional ADT may be promising (33). The Chemohormonal Therapy versus Androgen Ablation Randomized Trial for Extensive Disease in Prostate Cancer, CHAARTED study recently showed superior OS among patients with hormone-naïve metastatic prostate cancer treated with docetaxel chemotherapy and primary ADT, compared to primary ADT alone (34). Therefore, risk stratification using several factors including serum testosterone level at pre-treatment could contribute to appropriate selection of patients suitable for more intense multimodality therapeutics.
Previous studies have established that prostate cancer growth and survival are dependent on androgens. However, the dependency on androgens varies among prostate cancer because of their heterogeneity between patients and within a patient. Therefore, pre-treatment testosterone may reflect the nature of how much the tumor is dependent on androgens. Accordingly, low serum testosterone may mean less dependency on androgen. In addition, surgical resection of aggressive prostate cancer was demonstrated to restore serum luteinizing hormone (35) as well as testosterone level (36), suggesting that prostate cancer may suppress serum testosterone level through the hypothalamic pituitary axis (37). Also, the discrepancy between serum and tissue androgen levels has been reported (38). Furthermore, men with high testosterone levels in tissues easily become refractory to ADT and develop CRPC (16). Thus, low serum testosterone associated with prostate cancer might be either a cause or a result of progressed prostate cancer, and this should be examined in future studies.
The present study had several limitations. The sample size was relatively small, and the study design was retrospective. Nevertheless, this study was able to present a significant impact of serum testosterone level before treatment on the prognosis of primary ADT. Additionally, the present study included Japanese men only, which may limit findings to Japanese populations. Furthermore, serum testosterone level in this study, which is known to vary diurnally, was measured at various times from 8:00 to 10:00. However, serum testosterone in aged people has been shown not to fluctuate greatly (39), justifying our results in the study.
Conclusion
Serum testosterone level at pre-treatment, but not obesity, may be a predictive factor of outcome including survival after CRPC in Japanese men treated with primary ADT for metastatic prostate cancer.
Acknowledgements
This work was supported by Kakenhi grants (25462484 and 26861273) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan. We would like to thank Edanz Group Japan for editorial assistance.
- Received August 14, 2015.
- Revision received September 11, 2015.
- Accepted September 14, 2015.
- Copyright© 2015 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
