BJUIB J U I N T E R N A T I O N A L

© 2 0 1 2 T H E A U T H O R S

B J U I N T E R N A T I O N A L © 2 0 1 2 B J U I N T E R N A T I O N A L | doi:10.1111/j.1464-410X.2011.10876.x 1

What ’ s known on the subject? and What does the study add? High grade prostatic intraepithelial neoplasia (HGPIN) is a risk factor for prostate cancer (PCa), but only multifocality is an indication for early rebiopsy. Other risk factors for PCa development from HGPIN remain unknown. PCa is related to testosterone. Testosterone has been proven to be linked to PCa detection and poor prognosis PCa.

This study shows that low free and bioavailable testosterone levels are associated with an increased risk of PCa in a rebiopsy after HGPIN diagnosis. Men with low testosterone levels and HGPIN could therefore be considered a high-risk cohort for developing PCa.

OBJECTIVE

• To determine the relevance of the hormonal profi le of patients with high grade prostatic intraepithelial neoplasia (HGPIN) and its relationship to prostate cancer (PCa) in rebiopsy.

PATIENTS AND METHODS

• We prospectively analysed 82 consecutive patients with a diagnosis of HGPIN without PCa in a prostate biopsy between September 2007 and December 2009. • Of these 82 patients, 45 underwent rebiopsy and their hormonal profi le was determined (testosterone and sex hormone-binding globulin [ SHBG ] ) as part of our clinical protocol. • Patient age, PSA level, prostate volume, PSA density, testosterone, free testosterone, bioavailable testosterone and SHBG were recorded prospectively. • A comparative study between those patients with a positive rebiopsy and

those with a negative rebiopsy was performed.

RESULTS

• We found that free testosterone ( P = 0.04), bioavailable testosterone ( P = 0.04) and SHBG ( P = 0.02) were signifi cantly associated with a positive rebiopsy. • Other variables such as age ( P = 0.745), PSA level ( P = 0.630), prostate volume ( P = 0.690), PSA density ( P = 0.950), testosterone ( P = 0.981) and prostatic intraepithelial neoplasia multifocality ( P = 0.777) were not associated with the presence of adenocarcinoma in the rebiopsy.

CONCLUSIONS

• Patients with adenocarcinoma of the prostate after a diagnosis of HGPIN have higher SHBG levels and lower calculated free testosterone levels than patients with a negative rebiopsy. • Testosterone levels might be a useful indication for rebiopsy after HGPIN diagnosis.

KEYWORDS

rebiopsy , prostatic intraepithelial neoplasia , testosterone

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

Low testosterone level predicts prostate cancer in re-biopsy in patients with high grade prostatic intraepithelial neoplasia Eduard Garc í a-Cruz , Marta Piqueras , Maria Jos é Ribal , Jorge Huguet , Rodrigo Serapiao , Lluis Peri , Laura Izquierdo and Antonio Alcaraz Urology Department, Hospital Clinic de Barcelona, Barcelona, Spain

INTRODUCTION

High grade prostatic intraepithelial neoplasia (HGPIN) is seen in 4 – 16% of males undergoing prostate biopsy [ 1,2 ] . HGPIN is related to the development of prostate cancer, as men with HGPIN are more likely to develop prostate cancer (PCa) when compared to those without [ 3 ] . It has been suggested that HGPIN might be a precursor

lesion in the development of PCa and this has led to recommendations for early rebiopsy after HGPIN. More recently, this has been modifi ed so that criteria for rebiopsy are based on prostatic intraepithelial neoplasia (PIN) multifocality [ 4,5 ] .

The current literature does not support the usefulness of clinical markers (e.g. DRE, PSA level, PSA density or free PSA) in predicting

which patients with HGPIN are more likely to progress to PCa [ 6 ] . Further investigation into the process of carcinogenetic transformation of HGPIN could be helpful in deciding which patients are at risk of progressing to PCa.

The role of testosterone is controversial; testosterone might be related to the development of PCa, as suggested by studies

Accepted for publication 9 November 2011

G A R C Í A - C R U Z E T A L .

© 2 0 1 2 T H E A U T H O R S

2 B J U I N T E R N A T I O N A L © 2 0 1 2 B J U I N T E R N A T I O N A L

on castration and PCa regression by Huggins and Hodges [ 7 ] and Huggins et al . [ 8 ] , but this hypothesis is being reconsidered [ 9,10 ] because over the last decade some data have shown that patients with PCa do not have higher testosterone levels [ 11,12 ] .

Furthermore, there is a growing body of evidence that low testosterone has been a useful clinical predictor of PCa before prostate biopsy [ 13,14 ] . In our experience, high testosterone levels are not related to PCa development. On the contrary, our group showed an association between lower testosterone levels and a higher rate of positive prostate biopsies [ 15 ] . It is therefore important to establish whether in the early stages of PCa carcinogenesis there is a relationship between testosterone and PCa.

The aim of the present study was to analyse the relationship between hormonal levels and the diagnosis of PCa after a biopsy in patients with HGPIN, to determine whether testosterone levels are related to the risk of PCa in patients with HGPIN.

MATERIAL AND METHODS

We prospectively analysed 1000 prostate biopsies carried out in our centre from February 2007 to December 2009. Indications for prostate biopsy were suspicious DRE and/or elevated PSA levels. All prostate biopsies were 5 + 5-core and were TRUS-guided. From these 1000 patients we selected those with HGPIN with no associated PCa ( n = 82 [ 8.2% ] ).

We prospectively recorded age, PSA level, prostate volume, PSA density, DRE, hormonal status and the pathological report.

As part of our clinical protocol, we determined testosterone and sex hormone-binding globulin (SHBG) levels using an automated immunometric chemiluminescent method (Cobas ® ; Roche Molecular Diagnostics, Germany), in all patients undergoing prostate biopsy. Serum samples were obtained by puncture of fasting patients between 7:00 and 11:00 AM. Using Vermeulen × s formula, free testosterone and bioavailable testosterone were calculated [ 16 ] . Patients who were receiving LHRH analogues or testosterone replacement therapy were excluded from the analysis.

Patient age, body mass index, PSA level, prostate volume, PSA density and testosterone, free testosterone, bioavailable testosterone and SHBG levels were analysed as continuous variables. PSA level was analysed both as a continuous and a categorical variable (PSA < 10 ng/mL, PSA 10 – 20 ng/mL and PSA > 20 ng/mL). DRE was analysed as a categorical variable (normal or abnormal).

Clinical, biochemical and hormonal variables were analysed. We performed a comparative study in the subgroup of patients undergoing rebiopsy, between those with a diagnosis of PCa and those with a negative rebiopsy result. Shapiro – Wilk and Student t -tests were used to assess the normal distribution of the variables and their correlation.

RESULTS

Of the 82 patients included in the study, 45 had available rebiopsy data and were suitable for further analysis. The clinical

characteristics of these 82 patients are shown in Table 1 . The hormonal pattern distribution is shown in Fig. 1 .

Indications for rebiopsy were based on clinical criteria (a rise in PSA level, clinical suspicion or patient preference). No differences between the cohort with and that without biopsy were observed. Rebiopsy was carried out 3 – 6 months after the fi rst diagnosis of HGPIN.

Table 2 shows the results of the analysis in the subset of patients submitted to rebiopsy, comparing men with and without PCa in the rebiopsy.

DISCUSSION

High grade prostatic intraepithelial neoplasia is associated with the development of PCa; men with a diagnosis of HGPIN are more likely to develop PCa in the future than those without HGPIN (36.3% vs 25%) [ 17 ] . Nevertheless, no clinical factors seem to be useful in identifying which patients with

TABLE 1 Comparative analysis of patients with and without rebiopsy after HGPIN diagnosis

Rebiopsy cohort, n = 45 No rebiopsy cohort, n = 37 P Mean ( SD ) age, years 70 (8) 68 (7) 0.638Mean ( SD ) PSA, ng/dL 10 (7) 8 (3) 0.107Mean ( SD ) PSA density, ng/dL * g 0.39 (0.4) 0.3 (0.5) 0.553Mean ( SD ) prostate volume, mL 53 (23) 52 (23) 0.850Normal DRE, yes/no (%) 37/40 (92.5) 32/36 (88.9%) 0.587

FIG. 1. Graphic representation of the distribution of the hormonal pattern variables. Units from the variables in the x-axis correspond to those in the y-axis. A Shapiro – Wilk test was used to assess the normal distribution of the variables. In all cases, variables followed a normal distribution (testosterone, P = 0.248; SHBG, P = 0.299; bioavailable testosterone, P = 0.317; free calculated testosterone, P = 0.321).

1000.0

800.0

400.0

600.0

200.0

0.0Testosterone, ng/dL

BioavailableTestosterone, ng/dL

SHBG, nmol/L

Free CalculatedTestosterone, ng/dL

50100150200250300350

120100

6080

2040

0

12.5

10

5

7.5

2.5

T E S T O S T E R O N E A S P R E D I C T O R O F P C A I N P A T I E N T S W I T H H G P I N

© 2 0 1 2 T H E A U T H O R S

B J U I N T E R N A T I O N A L © 2 0 1 2 B J U I N T E R N A T I O N A L 3

HGPIN are at risk of PCa progression [ 17 ] . Only HGPIN multifocality seems to be consistently related to a higher risk of a positive rebiopsy [ 17 – 19 ] . Our data are consistent with previous literature as no clinical variable (DRE, PSA level and its surrogates or prostate volume) was able to predict a positive result in an early rebiopsy. In contrast to reported data, the present study found that multifocality had no predictive value in PCa progression.

Our group has experience in the analysis of hormonal pattern and its association with PCa, so it was possible to determine that, in men with low testosterone, the lower the testosterone, the higher the PCa risk. Our data also confi rm the hypothesis that men with low testosterone undergoing prostate biopsy are more likely to harbour PCa. Previous evidence suggests that testosterone might be helpful in prostate cancer screening [ 13 – 15 ] . Yano et al . [ 13 ] found that testosterone levels were lower in patients with PCa than in those with BPH and low PSA levels. Moreover, low testosterone was related to high grade PCa (Gleason score ≥ 7). In a population of 211 men undergoing prostate biopsy, Sofi kerim et al . [ 14 ] found similar results, with men with PCa having lower testosterone levels than men with a negative biopsy. Previous data [ 15 ] from our group showed that low bioavailable testosterone levels are related to a higher rate of prostate cancer in men undergoing prostate biopsy.

The relationship between PCa and testosterone has been controversial since 1941 when Huggins and Hodges [ 7 ]

established PCa regression after testosterone deprivation. Since then attempts have been made to link testosterone and the process of PCa carcinogenesis. Published data are sometimes paradoxical given that, on the one hand, testosterone defi ciency is related to a poor prognosis in PCa and, on the other hand, the experience of supplementation with testosterone in men at high risk of PCa shows that those with HGPIN and testosterone supplementation were not found to have an increased PCa risk. In a retrospective analysis of 75 men with testosterone supplementation, 20 of them had previous HGPIN. In the group of 20 men with previous HGPIN, only one patient was diagnosed with PCa after 1-year follow-up. No difference was found between the HGPIN and the control group in terms of de novo PCa after testosterone supplementation. Contrary to the general belief that testosterone might induce PCa in this high-risk population, Rhoden and Morgentaler [ 20 ] found that testosterone supplementation was not related to PCa development in HGPIN patients. Hypothesizing that this relationship might be established from the initiation of PCa carcinogenesis we decided to perform a preliminary study to analyse whether testosterone levels might be considered a risk factor for PCa in those men with a prior diagnosis of HGPIN. To our knowledge, this relationship has never been published before.

Our data show that patients with a positive rebiopsy after HGPIN diagnosis have signifi cantly lower free testosterone/bioavailable testosterone and higher SHBG

levels than those men with a negative rebiopsy. Determining testosterone levels in men where PCa is suspected might useful in subsequent clinical decisions.

There are some limitations to the present study. First, it was a single-centre, small series so, although our data are new and surprising, our fi ndings must be interpreted with care. Second, time is crucial in the progression to PCa from HGPIN [ 21,22 ] . As previously noted, the risk of PCa after HGPIN increases with time, from 2.3 – 4.4% [ 17,23 ] in 1-year rebiopsies, to 14.7 – 25.8% [ 17,23 ] in the 3-year biopsies, and to 26.6% at 5-year biopsies in patients with unifocal HGPIN. The follow-up period in the present study was too short to determine the importance of time and its impact on the rebiopsy result.

This very preliminary experience suggests that low testosterone levels may be a clinical risk factor for the diagnosis of PCa on rebiopsy after HGPIN. Further analysis with a larger group and longer follow-up are warranted.

ACKNOWLEDGEMENTS

We thank Nizam Mamode, MD PhD from Guy ’ s and St. Thomas ’ Hospital and Andrea Sallent for the reviewing of the manuscript.

CONFLICT OF INTEREST

None declared.

REFERENCES

1 Bostwick DG , Qian J . High-grade prostatic intraepithelial neoplasia . Mod Pathol 2004 ; 17 : 360 – 79

2 Bostwick DG , Liu L , Brawer MK , Qian J . High-grade prostatic intraepithelial neoplasia: the only accepted prostate cancer precursor lesion . Rev Urol 2004 ; 6 : 171 – 9

3 Park S , Shinohara K , Grossfeld G , Carroll P . Prostate cancer detection in men with prior high grade prostatic intraepithelial neoplasia or atypical prostate biopsy . J Urol 2001 ; 165 : 1409 – 14

4 Moore CK , Karikehalli S , Nazeer T , Fisher HA , Kaufman RP Jr , Mian BM .

TABLE 2 Comparative analysis between patients with a positive and those with a negative rebiopsy

Positive rebiopsy, n = 10

Negative rebiopsy, n = 35 P

Mean ( SD ) age, years 68 (8) 76 (7) 0.192Mean ( SD ) PSA level, ng/dL 10.9 (7.0) 9.5 (7.1) 0.630Mean ( SD ) PSA density, ng/dL * g 0.37 (0.23) 0.39 (43) 0.950Mean ( SD ) prostate volume, mL 50 (22) 54 (24) 0.690Normal DRE (%) 10/10 (100) 27/30 (90) 0.298Multifocality (%) 5/8 (62.5) 19/29 (65.5) 0.874Mean ( SD ) testosterone level, ng/dL 490 (150) 488 (176) 0.981Mean ( SD ) free calculated testosterone level, ng/dL 6.9 (1.2) 9.3 (3.2) 0.041Mean ( SD ) bioavailable testosterone level, ng/dL 162 (28) 217 (74) 0.04Mean ( SD ) SHBG level, nmol/L 58 (19) 39 (16) 0.020

G A R C Í A - C R U Z E T A L .

© 2 0 1 2 T H E A U T H O R S

4 B J U I N T E R N A T I O N A L © 2 0 1 2 B J U I N T E R N A T I O N A L

Prognostic signifi cance of high grade prostatic intraepithelial neoplasia and atypical small acinar proliferation in the contemporary era . J Urol 2005 ; 173 : 70 – 2

5 Heidenreich A , Bolla M , Joniau S et al . Guidelines on Prostate Cancer . Arnhem : European Association of Urology , 2009

6 Epstein JI , Srigley J , Grignon D et al . Recommendations for the reporting of prostate carcinoma . Hum Pathol 2007 ; 38 : 1305 – 9

7 Huggins C , Hodges C . Studies on prostatic cancer. I. The effect of castration, of estrogen, and of androgen injection on serum phos- phatases in metastatic carcinoma of the prostate . Cancer Res 1941 ; 1 : 293 – 7

8 Huggins C , Stevens R , Hodges C . Studies on prostatic cancer. II. The effects of castration on advanced carcinoma of the prostate gland . Arch Surg 1941 ; 43 : 209 – 23

9 Isbarn H , Pinthus J , Marks L et al . Testosterone and prostate cancer: revisiting old paradigms . Eur Urol 2009 ; 56 : 48 – 56

10 Morgentaler A , Traish AM . Shifting the paradigm of testosterone and prostate cancer: the saturation model and the limits of androgen-dependent growth . Eur Urol 2009 ; 55 : 310 – 21

11 Statin P , Lumme S , Tenkanen L et al . High levels of circulating testosterone are not associated with increased prostate cancer risk: a pooled prospective study . Int J Cancer 2004 ; 108 : 418 – 24

12 Morote J , Planas J , Ramirez C et al . Evaluation of the serum testosterone to prostate-specifi c antigen ratio as a predictor of prostate cancer risk . BJU Int 2009 ; 104 : 486 – 9

13 Yano M , Imamoto T , Suzuki H et al . The clinical potential of pretreatment serum testosterone level to improve the effi ciency of prostate cancer screening . Eur Urol 2007 ; 51 : 375 – 80

14 Sofi kerim M , Eskicorapci S , Oru ç O , Ozen H . Hormonal predictors of prostate cancer . Urol Int 2007 ; 79 : 13 – 8

15 Garc í a-Cruz E , Huguet J , Piqueras M et al . Low testosterone bioavailability is related to prostate cancer diagnose in patients submitted to prostate biopsy . World J Urol 2011 ; [ Epub ahead of print ]

16 Vermeulen A , Kaufman JM . Diagnosis of hypogonadism in the aging male . Aging Male 2002 ; 5 : 170 – 6

17 Lee MC , Moussa AS , Yu C , Kattan MW , Magi-Galluzzi C , Jones JS . Multifocal high grade prostatic intraepithelial neoplasia is a risk factor for subsequent prostate cancer . J Urol 2010 ; 184 : 1958 – 62

18 Merrimen JL , Jones G , Walker D , Leung CS , Kapusta LR , Srigley JR . Multifocal high grade prostatic intraepithelial neoplasia is a signifi cant risk factor for prostatic adenocarcinoma . J Urol 2009 ; 182 : 485 – 90

19 Akhavan A , Keith JD , Bastacky SI , Cai C , Wang Y , Nelson JB . The proportion of cores with high-grade prostatic intraepithelial neoplasia on extended-pattern needle biopsy is signifi cantly

associated with prostate cancer on site-directed repeat biopsy . BJU Int 2007 ; 99 : 765 – 9

20 Rhoden EL , Morgentaler A . Testosterone replacement therapy in hypogonadal men at high risk for prostate cancer: results of 1 year of treatment in men with prostatic intraepithelial neoplasia . J Urol 2003 ; 170 : 2348 – 51

21 Bostwick DG . Prostatic intraepithelial neoplasia (PIN): current concepts . J Cell Biochem Suppl 1992 ; 16H : 10 – 9

22 Lefkowitz GK , Taneja SS , Brown J et al . Follow-up interval prostate biopsy 3 years after diagnosis of high grade prostatic intraepithelial neoplasia is associated with high likelihood of prostate cancer, independent of change in prostate specifi c antigen levels . J Urol 2002 ; 168 : 1415 – 8

23 Lefkowitz GK , Sidhu GP , Torre P et al . Is repeat prostate biopsy for high-grade prostatic intraepithelial neoplasia necessary after routine 12-core sampling? Urology 2001 ; 58 : 999 – 1003

Correspondence: Eduard Garc í a-Cruz, Urology Department, Hospital Clinic de Barcelona, 170 Villarroel St., 08036 Barcelona, Spain. e-mail: edu_garcia_cruz@yahoo.com

Abbreviations : HGPIN , high grade prostatic intraepithelial neoplasia ; PCa , prostate cancer ; SHBG , sex hormone-binding globulin.