Low pretreatment total testosterone (<3 ng/mL) predicts extraprostatic disease in prostatectomy specimens from patients with preoperative localized prostate cancer

Authors


Professor Alexandre de la Taille, Department of Urology, Henri Mondor Hospital, 51 Avenue du Maréchal de Lattre de Tassigny, 94010 Créteil, France.
e-mail: adelataille@hotmail.com

Abstract

Study Type – Therapy (case series)
Level of Evidence 4

OBJECTIVE

  • • To investigate the relationship between pretreatment testosterone levels and pathological specimen characteristics, by prospectively examining serum androgen concentrations in a well-studied cohort of patients who underwent radical prostatectomy (RP) for localized prostate cancer.

PATIENTS AND METHODS

  • • A total of 107 patients with clinically localized prostate cancer had an assay of total testosterone before laparoscopic RP at our institution.
  • • The results were classified into two groups based on the total serum testosterone: group1, <3 ng/mL; group 2, ≥3 ng/mL.
  • • Student’s t-test was used to compare continuous variables, and Fisher’s exact test or the chi-squared test was used to compare categorical variables.
  • • Survival curves were established using the Kaplan–Meier method and compared using the log-rank test. In all tests, P < 0.05 was considered to indicate statistical significance.

RESULTS

  • • All patients had localized prostate cancer based on digital rectal examination (DRE) and preoperative magnetic resonance imaging (MRI). Groups 1 and 2 were similar in terms of age, body mass index, preoperative co-morbidities (cardiovascular and diabetes mellitus), clinical stage of prostate cancer and preoperative PSA levels.
  • • In pathological specimens, low total testosterone (<3 ng/mL) was an independent risk factor for high Gleason score (>7) and for locally advanced pathological stage (pT3 and pT4).
  • • Higher preoperative testosterone correlated with disease confined to the gland.
  • • There was no association between serum testosterone levels and surgical margin status, on the one hand, and biochemical recurrence on the other.

CONCLUSION

  • • Low serum testosterone appears to be predictive of aggressive disease (Gleason score >7 and extraprostatic disease, pathological stage >pT2) in patients who underwent RP for localized prostate cancer.
Abbreviations
BMI

body mass index

RP

radical prostatectomy.

INTRODUCTION

Several studies have attempted to determine a relationship between androgens and prostate cancer risk with varied results. Thus, the relationship between testosterone and prostate cancer is not fully elucidated. Lower pretreatment testosterone is associated with a significantly poorer treatment response and worse prognosis in metastatic cases. Specifically, pretreatment testosterone level <3 ng/mL is associated with shorter survival [1], suggesting more aggressive disease [2]. However, this relationship has been more difficult to establish in patients with clinically localized prostate cancer. To date, clinical stage, biopsy Gleason score and serum PSA level are the established preoperative prognostic markers for pathological stage [3]. Yet it has been documented that 30–40% of patients who undergo radical prostatectomy (RP) for clinically organ-confined localized carcinoma of the prostate will have extraprostatic disease or experience disease recurrence [4]. Finding reliable preoperative markers that will allow us to understand the disease more accurately would certainly be useful in therapeutic strategy and disease management.

We explored the relationship between preoperative total testosterone and pathological stage and progression in patients with clinically localized prostate cancer treated with RP.

PATIENTS AND METHODS

From January 2006 to January 2008, 107 patients with clinically localized prostate cancer had an assay of total testosterone before laparoscopic RP at our institution. All patients had been assessed by diagnostic prostatic biopsies before RP. No patient had clinical evidence of locally advanced prostate cancer. Clinical stage was assigned by the attending surgeon according to preoperative DRE findings and confirmed during the preoperative MRI. None of the patients received neoadjuvant androgens. We prospectively collected the following data: patient’s age at the time of surgery, clinical presentation, preoperative PSA level, body mass index (BMI), co-morbidities (cardiovascular and diabetes mellitus). All patients were evaluated at 3 and 6 months after RP and at least once a year thereafter. Each visit included a clinical examination, a DRE and measurement of PSA. Biochemical recurrence of the disease was defined as two consecutive PSA levels >0.2 ng/mL after RP.

For endocrine assessment, the assay was done between 7 am and 10 am fasting in the same laboratory. The results were classified into two groups based on the total testosterone: group 1, <3 ng/mL; group 2, ≥3 ng/mL.

Gleason score was determined in prostatic biopsies and RP specimens by the same uropathologist. Patients were divided into those with low (2–6) and those with high Gleason score (7–10). Pathological TNM staging, capsular perforation, seminal vesicle involvement and surgical margin status were determined in all surgical specimens.

Student’s t-test was used to compare continuous variables, and Fisher’s exact test or the chi-squared test was used to compare categorical variables. PSA-free survival was defined from the date of surgery to the last follow-up visit. Prostate cancer-specific survival was defined as the time of death from prostate cancer in any patient from the date of surgery. Survival curves were established using the Kaplan–Meier method and compared using the log-rank test. Prognostic factors were established by univariable analysis, and those that were significant were entered into the multivariable analysis using a Cox stepwise regression method. In all tests, P < 0.05 was considered to indicate statistical significance.

RESULTS

The clinical and biological characteristics of the patients of the two groups are reported in Table 1. Groups 1 and 2 were similar in terms of age, BMI, preoperative co-morbidities (cardiovascular and diabetes mellitus), clinical stage prostate cancer and preoperative PSA levels. All patients had clinically localized prostate cancer. Neither extracapsular extension nor seminal vesicle invasion was notified in preoperative MRI. There was no difference in prostate biopsies, especially Gleason score, between the two groups.

Table 1.  Clinical and biological patient characteristics
 Group 1
(total testosterone <3 ng/mL)
(n= 21)
Group 2
(total testosterone ≥3 ng/mL)
(n= 86)
P
  1. ns, non significant.

Mean (range) age, years63.8 (52–75)61.9 (52–73)ns
Caucasian, %9493ns
Mean (range) BMI, kg/m225.9 (21.7–33.8)25.8 (20.1–32)ns
Cardiovascular co-morbidities, % 6 5ns
Diabetes mellitus, % 3 3.5ns
PSA, ng/mL 9.6410.88ns
Clinical staging, %   
 T11687ns
 T28413ns

In pathological specimens, low testosterone (<3 ng/mL) was an independent risk factor for high Gleason score (≥7) and for locally advanced pathological stage (pT3 and pT4). Higher preoperative testosterone correlated with disease confined to the gland. There was no association between serum testosterone levels and the surgical margin status (Table 2).

Table 2.  Prostatic biopsies and pathological specimen results according to serum total testosterone levels
 Group 1
(total testosterone <3 ng/mL)
(n= 21)
Group 2
(total testosterone ≥3 ng/mL)
(n= 86)
P
  1. ns, non-significant.

Biopsy   
Gleason score   
 <759%62%ns
 ≥741%38% 
Pathological   
Gleason score   
 <717%31%0.01
 ≥783%69% 
Pathological stage   
 pT1–pT257%75%0.04
 pT3–pT443%25% 
Margin status   
 Positive41%39%ns
 Negative59%61% 

Finally, we compared the Kaplan–Meier PSA-free survival curves between the two groups. With a median follow-up of 35 months, there was no difference in terms of biochemical recurrence (Fig. 1).

Figure 1.

PSA-free survival stratified by pretreatment total testosterone.

DISCUSSION

Searching for preoperative markers that could identify more aggressive prostate cancer and influence patient survival is one of the main objectives of prostate oncology research.

Since Huggins et al.’s laboratory work in 1941 [5], a clear association has been established between androgens and prostate cancer, showing that androgens are important for the growth and maintenance of the prostatic gland, stimulating proliferation of human prostate cancer in vitro and producing prostate cancer in rodents [6]. Furthermore, serum PSA production has also been shown to be androgen-dependent at the cellular level [7]. Pretreatment total testosterone in patients with metastatic prostate disease (stage D2) has been investigated by numerous groups during the last two decades. Studies have consistently shown more aggressive disease, a worse prognosis and worse treatment response in patients with low serum total testosterone [1,8,9]. More recent works claim that low testosterone levels might be related to worse clinical and pathological determinants of prostate cancer, including an increased risk of prostate cancer [10,11]; a worse 5-year biochemical relapse-free survival [11]; higher Gleason sums on biopsy [12,13]; an increased percentage positive-core rate at biopsy [12]; worse pathological stage [14–16]; and an increased risk of positive surgical margins [17].

The major finding of the present study is that there is a statistically significant correlation between pretreatment total testosterone levels and pathological stage in patients with preoperative (clinical and MRI confirmed) localized prostate cancer treated with laparoscopic RP. This correlation held up in multivariate analysis when pretreatment total testosterone emerged as an independent predictor of extraprostatic disease. As serum testosterone decreased, patients had an increased likelihood of non-organ-confined disease (pT3–T4). Furthermore, a low total testosterone was also correlated with a higher Gleason score in pathological specimens. These results indicate that low pretreatment total testosterone could be a marker for more aggressive disease in localized prostate cancer at preoperative evaluation. Recently, similar results were found in two retrospective studies [9,16]. The present study validates these results with prospectively collected findings.

Other studies have described the relationship of lower testosterone with more advanced disease. Imamoto et al.[16] reported that their patients with localized cancer had higher testosterone levels than those with D2 disease [16], showing more advanced disease with lower amounts of androgen. Hoffman et al.[12] showed that patients with lower testosterone were more likely to have a Gleason score ≥8 as well as a higher percentage of positive cores on biopsy [12]. It was also recently reported that patients on finasteride, a 5α-reductase inhibitor that inhibits the conversion of testosterone to the androgen dihydrotestosterone, had a lower overall rate of cancer but were at increased risk for high-grade tumours [18]. By decreasing intraprostatic testosterone, finasteride might have created an environment for high-grade cancers that were less dependent on androgens for growth [19,20]. In another study of interest, low serum testosterone in patients with newly diagnosed prostate cancer was associated with higher tumour microvessel and androgen receptor density as well as with higher Gleason score [13].

Other investigators have reported an association between low testosterone levels and the likehood of positive surgical margins in pathological specimens after RP [17]. This association was not elucidated in the present study. Also, despite the fact that higher Gleason score and worse pathological stage are recognized predictive factors of biochemical recurrence of clinical prostate cancer after RP, in the present study, there was no significant difference between the PSA- free survival curves of the two groups (low and high testosterone levels). This can be explained by the small number of recurrences that we observed, possibly due to the small follow-up and the resulting lower statistical power.

The reasons for the association between high-risk prostate cancer and low serum testosterone are still under investigation. This association could be secondary to hormonal changes in chronic diseases [21], to the inhibition of testosterone levels by high-grade tumours [14], to the central inhibition of the hypothalamic–pituitary axis [9], to a selection of poorly differentiated cancer cells due to low androgen levels, or purely a combination of other factors related to the pathological stage [22].

The strength of the present study is the prospectively collected database. The limitations include its small follow-up, which does not permit a sufficient number of recurrences and means that correlation between disease recurrence is yet to be seen. It would also be valuable to have information on the postoperative testosterone level after RP and during follow-up/progression. Indeed, some groups, such as that in a study by Miller, have shown interesting changes in testosterone after RP in patients with high-grade vs low-grade Gleason score and an impact on progression [22].

In conclusion, we found that lower testosterone in patients with clinically localized prostate cancer treated with RP was associated with a worsening Gleason score and pathological stage. However, whether this correlates with disease recurrence is yet to be determined, a limitation of the present study that was possibly due to the small number of recurrences that we observed and the resulting lower statistical power. This information could be useful in treatment decision-making before operation as well as surgical planning for the control of prostate cancer. The implications for the therapeutic attitude and prognosis need further studies.

CONFLICT OF INTEREST

None declared.

Ancillary