SEARCH

SEARCH BY CITATION

Keywords:

  • bone metabolism;
  • bone metastases;
  • adjuvant therapy;
  • bisphosphonates;
  • markers

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

OBJECTIVE

To evaluate and compare the value of several markers of bone turnover in different stages of prostate cancer, as bone metastases are a common feature in this disease, and for assessing bone metastases both bone formation and bone resorption markers are diagnostic.

PATIENTS AND METHODS

The prospective study included 219 men, i.e. 129 undergoing radical retropubic prostatectomy (RRP) and 25 with bone metastases due to prostate cancer, and 65 with benign urological disorders who served as controls. Before any treatment the concentrations of alkaline phosphatase (ALP), osteocalcin, serum C-terminal telopeptide of type I collagen (S-CTX) and tartrate-resistant acid phosphatase type 5b (TRACP5b) were determined.

RESULTS

Men undergoing RRP were divided into those with lymph node-negative, localized (pT3, 101) and lymph node-positive (28) disease, after histological examination. The controls had the lowest marker levels while patients with bone metastases due to prostate cancer had the highest levels, with significance for ALP, osteocalcin and TRACP5b. Patients with lymph node-positive cancer had significantly high serum levels of TRACP5b and ALP but not for osteocalcin and S-CTX.

CONCLUSIONS

Bone turnover markers represent a new diagnostic tool in prostate cancer; the present data show that both bone resorption and bone formation are crucial for detecting bone metastases in prostate cancer. The value of bone turnover markers in high-risk patients should be evaluated in a longitudinal study.


Abbreviations
RRP

radical retropubic prostatectomy

(b)ALP

(bone) alkaline phosphatase

S-CTX

serum C-terminal telopeptide of type I collagen

TRACP5b

tartrate-resistant acid phosphatase type 5b.

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Prostate cancer is the second most common cancer in men in North America and Western Europe. In the European Union the incidence of prostate cancer is ≈ 79/100 000 men/year and the mortality is ≈ 31/100 000 men/year. The mean age at diagnosis is 71 years [1]. During the clinical course of prostate cancer, patients are at high risk of skeletal morbidity due to bone loss because of advanced age at the time of diagnosis, androgen deprivation therapy and the development of bones metastases [2,3]. The skeleton is the preferred site of metastases and in advanced disease bone metastases will occur in 65–75% of patients. Bone manifestation of prostate cancer is associated with many complications and much discomfort for the patients. Such complications include severe bone pain, prolonged hospital stay, reduced mobility, hypercalcaemia and pathological fractures. Thus they are responsible for a decrease in quality of life in this palliative setting [4]. Furthermore, skeletal-related events have been correlated with reduced overall and median survival [5,6].

Predictors of future bone manifestations after initial therapy for prostate cancer are age, PSA level and its isoforms, Gleason score and the TNM classification [7]. Currently, these variables are routinely assessed to create nomograms that predict the risk of recurrence after initial therapy [8,9]. During the follow-up bone metastases can be detected by rising PSA levels, bone pain and pathological fractures. Prevention, early detection and adequate therapy of bone metastases due to prostate cancer represent both a major health problem and a therapeutic challenge.

Biochemical markers of bone turnover are very specific for bone tissue compared with conventional markers. These new markers have been developed and improved during the last decade, and have augmented the analytical spectrum for assessing various skeletal pathologies. Several studies in postmenopausal osteoporosis showed that they have a high sensitivity in detecting the bone turnover rate, and they are widely used in clinical routine to assess the efficacy of therapy [10]. In general, markers of bone resorption are distinguished from markers of bone formation, and under normal conditions they are coupled to each other. Bone formation markers are direct or indirect products of osteoblast activity. Bone resorption markers physiologically result from bone collagen degradation [11]. Bone metastases deregulate the balance between bone formation and bone resorption. Therefore biochemical markers of bone turnover can provide an insight into tumours and their dynamic effects on bone interactions. With numerous markers described we selected four established serum markers, two each indicating bone formation and bone resorption, respectively, to evaluate their potential in patients with different stages of localized and metastatic prostate cancer. The markers are total serum alkaline phosphatase (ALP), the marker used most often for indicating bone formation, and that provides a good impression of osteoblast activity. Recent studies showed a preference for the use of bone-specific alkaline phosphatase (bALP) due to its higher specificity [12,13], but when starting the present study we determined ALP. Osteocalcin is also a marker of bone formation, produced by osteoblasts, and was recently described in prostate cancer [14]. Tartrate-resistant acid phosphatase type 5b (TRACP5b), a resorption marker, is a special isoform of acid phosphatases; recent studies showed that TRACP5b is characteristic for osteoclasts and allows their activity to be assessed [15]. The fourth marker used was serum C-terminal telopeptide of type I collagen (S-CTX), also an established and well described resorption marker [16].

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

This prospective study included 219 men (median age 66.1 years, range 43–88), with serum samples collected from July 2004 to June 2005; 65 men (median age 67 years, range 54–78) with different benign urological disorders and with no malignancy in their medical history served as controls. In all, 129 men (median age 66 years, range 47–84) with clinically localized prostate cancer had a radical retropubic prostatectomy (RRP) including regional lymphadenectomy in our department. After surgery the disease was pathologically staged according to the 2002 classification [17]. No patient in the control or RRP group received any medication influencing bone metabolism. Patients had normal values for hepatic and renal function. In addition, 25 patients (median age 70.5 years, range 43–88) with bone metastases were included; all had hormone-refractory prostate cancer and were scheduled to receive taxane-based chemotherapy. All patients with hormone-refractory prostate cancer received maintenance androgen deprivation therapy.

A physical examination, abdominal ultrasonography and bone scan were used for clinical staging, with staging completed by MRI or CT whenever appropriate. Blood samples were taken before any treatment, collected at 08:00 hours after an overnight fasting period in evacuated Monovette plastic tubes (Sarstedt, Nürnberg, Germany) and centrifuged at 2000 g for 10 min at 4 °C. Supernatants were frozen at − 80 °C.

ALP was measured in heparin plasma on a clinical chemistry analyser, with original test kits (Roche Diagnostics, Mannheim, Germany). Osteocalcin was measured in EDTA plasma on the Elecsys 2010 analyser (Elecsys N-MID osteocalcin ECLIA, Roche Diagnostics). The reference ranges (95th percentile) were 42 ng/mL (men 30–50 years) and 46 ng/mL (men 51–70 years) with interassay precision coefficients of variance of 2.2–8.6% in at 14.3–229 ng/mL [18].

TRACP5b was measured in serum (BoneTrap Assay, Medac, Wedel, Germany); the reference range, as the mean (2 sd) was 3.1 (1.6) U/L for men (>20 years) with an interassay precision coefficient of variation of 7.2–9.2% at 2.5–16.1 U/L. S-CTX was measured in EDTA plasma on the Elecsys 2010 analyser; the upper reference range (95th percentile) was 0.300 ng/mL (men up to 70 years) and 0.394 ng/mL (men 51–70 years) with an interassay precision coefficient of variation of 2.4–7.7% at 0.240–3.470 ng/mL [18].

The results were analysed statistically using the nonparametric Mann–Whitney U-test to compare the results of the different prostate cancer groups with the controls. The Kruskal–Wallis nonparametric anova was used to calculate the differences between the different groups, with P < 0.05 taken to indicate statistical significance.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

In all men undergoing surgery for prostate cancer, the bone scans showed no metastases or suspect lesions. After surgery and histopathological examination, the prostate cancer group was divided in 101 men (median age 66 years, range 47–78) with no lymph node metastases (pT3) and 28 (median age 65.5 years, range 48–84) with lymph node metastases. Table 1 summarizes the essential clinical data of all examined groups. The measured bone turnover marker levels are also shown in Table 1 and graphically in Fig. 1. Statistical analysis of all groups using the Kruskal–Wallis anova showed significant differences for ALP and TRACP5b (both P < 0.001) but not for osteocalcin (P = 0.22) and S-CTX (P = 0.5). The control group had the lowest bone turnover marker levels, and patients with bone metastases had the highest levels of ALP (P < 0.001), osteocalcin (P < 0.05) and TRACP5b (P < 0.001), but there was no statistically significant difference in S-CTX levels (P = 0.16) from the controls. Patients with localized prostate cancer had significantly higher levels of TRACP5b (P < 0.001) but not ALP (P = 0.1), osteocalcin (P = 0.58) and S-CTX (P = 0.35) than the controls. Patients with lymph node-positive prostate cancer had significantly higher levels of TRACP5b (P < 0.005) and ALP (P < 0.05), but not osteocalcin (P = 0.78) and S-CTX (P = 0.21) than the controls. Also, lymph node-positive patients had significantly higher levels of PSA before RRP than men with localized prostate cancer (P < 0.001), but there were no significant differences between these groups for bone turnover marker levels. There were no significant differences for bone turnover marker levels and T classification, Gleason score, PSA levels before RRP and prostate volume in patients with lymph node-positive disease.

Table 1.  The characteristics of the groups (219 men); values are the median (range) or median (sd), and the serum levels of bone resorption (TRACP5b, S-CTX) and bone formation (ALP, osteocalcin) for the different groups. Significant differences (vs controls) are shown in bold
VariableControlLocalized (pN0M0)Lymph node +ve (pN1M0)Bone metastases (M1)
  1. nd, not determined.

Number of patients65 10128 25
Age, years67 (54–78) 66 (47–78)65.5 (48–84) 70.5 (43–88)
PSA before RRP, ng/mL 4.7 (2.7)  7.9 (4.9)19.7 (21.9)108.5 (132)
Prostate volume, mLnd 40 (17.2)32 (13.7)nd
Tumour stage
 pT2  9112nd
 pT3  1014nd
 pT4   0 2nd
Gleason sum
 ≤7  9821nd
 >7   3 7nd
ALP, U/L63.5 (20.7) 73 (19.6)88 (23.4)381 (554.7)
Osteocalcin, ng/mL19 (15.3) 22 (7.9)22 (7.3)34.5 (20.5)
S-CTX, ng/mL 0.31 (0.27)  0.39 (0.2) 0.39 (0.26)  0.48 (0.42)
TRACP5b, U/L 2.65 (1.0)  3.3 (0.94)3.6 (0.84)  7.0 (3.6)
image

Figure 1. Box plots of the bone turnover marker levels, ALP, osteocalcin, S-CTX and TRACP5b, for the different patient groups. *P < 0.05 and **P < 0.005 vs controls; n.s., not significant.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The incidence of and mortality from prostate cancer continues to increase. The prevention and early detection of bone metastases from prostate cancer are of utmost importance for these patients, and represent a diagnostic dilemma. The bone scan is the reference standard for detecting and monitoring metastatic bone lesions [19] but for prostate cancer is it unsuitable for routine clinical use during the follow-up. However, it is recommended if patients report bone pain or present with increasing PSA and ALP levels, indicating recurrence and/or metastasis [20]. Bone turnover markers might be useful for filling this diagnostic deficiency.

In prostate cancer, bone turnover markers in urine and serum have been extensively studied as indicators of bone metastases, and they correlate with the number of lesions [13,21,22]. Whether bone turnover markers have predictive value for the development of osseous metastases in patients with prostate cancer after initial therapy has not yet been clarified. The many varying and inconsistent results can be explained by different assays used, with diverse antibodies recognizing different epitopes, and different handling techniques [11,23]. Also, diagnostic stability in urine and serum after collection is different. We determined bone turnover markers only in serum and EDTA-plasma collected during a defined period in the morning. The highest bone turnover marker levels were in men with bone metastases. Both bone formation (ALP, osteocalcin) and bone resorption (TRACP5b) markers were significantly higher than in controls. In line with our results, other groups also found significantly higher levels of ALP, bALP and TRACP5b, underlining the usefulness of bone turnover markers as a predictor of skeletal complications [13,22]. However, Jung et al.[13] found no significant differences for osteocalcin in their study population. This differing result for osteocalcin levels might reflect the heterogeneity of this patient group, with different therapies reflected by different bone turnover marker levels. However, the present data underline that both osteoblast and osteoclast activity might have a substantial role in bone metastases due to prostate cancer. Bone histology examinations support our findings, showing both osteoblastic and resorptive processes [24]. One group reported significantly greater values of the resorption markers S-CTX and TRACP5b in androgen-deprived patients with lymph node-positive prostate cancer [13]. However, there were no significant differences of these variables in lymph node-positive disease with no hormonal treatment. Hormonal therapy itself can change bone metabolism [24–26] and might be responsible for significantly increased levels of TRACP5b and S-CTX [13].

To our knowledge we present the first prospective study showing significantly elevated levels of bone formation and bone resorption markers at initial diagnosis in patients with lymph node-positive disease. Not unexpectedly, PSA levels in these patients were significantly higher than in men with localized stages, confirming PSA level as a predictor of advanced disease [8]. Unlike PSA, bone turnover markers did not seem to predict nodal disease in prostate cancer.

Patients with node-positive disease diagnosed by radical surgery represent a high-risk population for recurrence. Early adjuvant hormonal therapy is recommended by many groups and seems to be coupled to better survival, hence patients often have a decrease in quality of life due to several side-effects [27,28]. However, markers of bone formation and resorption might be helpful tools to detect bone disease and to initiate adequate therapy in time. TRACP5b and ALP indicated activated bone metabolism despite an inconspicuous bone scan. These markers could be useful to indicate occult bone metastases in high-risk patients earlier than the other clinical variables and a bone scan. A longitudinal study is needed to determine if and when patients with lymph node-positive disease and elevated TRACP5b or ALP levels will develop bone metastases. In this setting, adjuvant bisphosphonates might be useful to delay or even prevent the manifestation of clinically significant bone disease, by inhibiting both osteoclast and osteoblast activity. In animal and cell-culture studies zoledronic acid has shown inhibitory effects both on osteoblastic and osteolytic metastases of prostate cancer [29]. Furthermore, anti-neoplastic and apoptotic effects were reported [30]. In the present study all patients with prostate cancer, even those with no lymph node metastases, had significantly high TRACP5b (P < 0.001) serum levels, which could indicate bone resorption. As in men with lymph node-positive disease, a longitudinal study is needed to clarify the potential value and therapeutic consequences of this phenomenon.

In summary, the present data showed that both bone resorption (TRACP5b) and bone formation (ALP, osteocalcin) might be crucial for detecting bone metastases from prostate cancer, and useful in therapeutic decision making in terms of hormonal or bisphosphonate therapy. Bone turnover markers offer a new diagnostic tool for the clinician in high-risk patients. Adjuvant application of bisphosphonates might be helpful to delay or prevent bone manifestation of prostate cancer in the presence of impaired bone metabolism. An ongoing longitudinal study will define the clinical value of serum bone turnover markers in the follow-up of high-risk patients. Currently, TRAPC5b and ALP appear to be the most promising markers.

ACKNOWLEDGEMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Parts of this work were presented and highlighted at the annual meeting of the AUA 2005 in San Antonio (Abstract 1142). This article includes parts of the doctoral thesis of L. Koliva. B. Kosche is to be acknowledged for her excellent technical assistance.

CONFLICT OF INTEREST

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Source of funding: Research grant of Novartis Pharma GmbH, BU Oncology (Nürnberg, Germany).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES