Clinical Cancer Research The Science of Cancer Health Disparities Stand Up to Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cook, R. J.
Right arrow Articles by Smith, M. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cook, R. J.
Right arrow Articles by Smith, M. R.
Clinical Cancer Research Vol. 12, 3361-3367, June 1, 2006
© 2006 American Association for Cancer Research

Imaging, Diagnosis, Prognosis

Markers of Bone Metabolism and Survival in Men with Hormone-Refractory Metastatic Prostate Cancer

Richard J. Cook1, Robert Coleman2, Janet Brown2, Allan Lipton3, Pierre Major4, Yong Jiang Hei5, Fred Saad6 and Matthew R. Smith7

Authors' Affiliations: 1 Department of Statistics and Actuarial Science, University of Waterloo, Waterloo, Ontario, Canada; 2 Academic Unit of Clinical Oncology, Weston Park Hospital, Sheffield, United Kingdom; 3 Milton S. Hershey Medical Center, Hershey, Pennsylvania; 4 Hamilton Regional Cancer Center and Department of Medicine, McMaster University, Hamilton, Ontario, Canada; 5 Novartis Pharmaceuticals Corp., East Hanover, New Jersey; 6 Centre Hospitalier de l'Université de Montréal, Hôpital Notre-Dame, Montreal, Quebec, Canada; and 7 Massachusetts General Hospital, Boston, Massachusetts

Requests for reprints: Matthew R. Smith, Massachusetts General Hospital, Cox 640, 100 Blossom Street, Boston, MA 02114. Phone: 617-724-5257; Fax: 617-726-4899; E-mail: smith.matthew{at}mgh.harvard.edu.


   Abstract
Top
 Abstract
Patients and Methods
 Results
 Discussion
 References
 
Purpose: To evaluate the relative prognostic value for specific markers of osteoblast and osteoclast activity while controlling for previously reported prognostic variables among men with hormone-refractory metastatic prostate cancer.

Experimental Design: The 643 subjects in this report were participants in multicenter randomized controlled trial of zoledronic acid in men with metastatic prostate cancer. All subjects had bone metastases and disease progression despite medical or surgical castration. Relationships between baseline covariates and overall survival were examined by Cox proportional hazard model. Serum bone-specific alkaline phosphatase (BAP) and urinary N-telopeptide were assessed as representative specific markers of osteoblast and osteoclast activity, respectively. Other covariates in the model were age, log prostate-specific antigen, hemoglobin, lactate dehydrogenase, albumin, analgesic use, and Eastern Cooperative Oncology Group performance status.

Results: Serum BAP was significantly correlated with urinary N-telopeptide (correlation coefficient = 0.674; 95% confidence interval, 0.628-0.715; P < 0.0001). In univariate analyses, higher levels of serum BAP and urinary N-telopeptide levels were significantly associated with shorter overall survival. After controlling for the other variables, including N-telopeptide, in multivariate models, higher serum BAP levels were consistently associated with shorter survival. In contrast, urinary N-telopeptide levels were not significantly associated with survival in multivariate analyses. Variables retained in the reduced multivariate model were age, log prostate-specific antigen, hemoglobin, lactate dehydrogenase, analgesic use, and BAP.

Conclusions: Serum BAP significantly correlates with urinary N-telopeptide in men with androgen-independent prostate cancer and bone metastases. In multivariate models, higher levels of serum BAP but not urinary N-telopeptide are associated with shorter overall survival.

Most bone metastases in men with prostate cancer seem osteoblastic by radiographic imaging. Osteolytic and osteoblastic lesions represent two extremes of a spectrum, however, and bone metastases from prostate cancer are characterized by both excess bone formation and bone resorption. Number and activity of both osteoblasts and osteoclasts are increased in typical bone metastases from prostate cancer (1, 2). Specific biochemical markers of both osteoblast and osteoclast activity are markedly elevated in men with metastatic prostate cancer, and levels correspond to extent of skeletal involvement (3, 4).

In large multivariate analyses of men with androgen-independent metastatic prostate cancer, elevated levels of serum total alkaline phosphatase, a nonspecific marker of osteoblast activity, were independently associated with shorter overall survival (5, 6). Age, performance status, hemoglobin, albumin, lactate dehydrogenase, and log prostate-specific antigen (PSA) were also associated with overall survival in multivariate models. In recent univariate analyses, elevated levels of either urinary N-telopeptide, a specific marker of osteoclast activity, or serum bone-specific alkaline phosphatase (BAP), a specific marker of osteoblast activity, were associated with shorter overall survival in with androgen-independent prostate cancer and bone metastases (7, 8). It is not known whether specific markers of osteoclast or osteoclast activity are associated with survival after controlling for established prognostic variables.

The objective of this study was to evaluate the prognostic value for specific markers of osteoblast and osteoclast activity in men with hormone-refractory metastatic prostate cancer while controlling for established prognostic factors. Data from a multicenter randomized controlled trial (9) were used to examine the relationships between baseline variables, including serum BAP and urinary N-telopeptide, and overall survival.


   Patients and Methods
Top
 Abstract
 Patients and Methods
Results
 Discussion
 References
 
Study population. We used data from an international randomized controlled trial of zoledronic acid in men with metastatic prostate cancer (9). The study included 643 men with histologically confirmed prostate cancer, bone metastases, and disease progression despite medical or surgical castration. All subjects had Eastern Cooperative Oncology Group performance status of ≤2, serum testosterone <50 ng/dL, corrected serum calcium level ≤8.0 mg/dL, and serum creatinine ≤3.0 mg/dL. Men with prior cytotoxic chemotherapy, radiation therapy within 3 months, or severe cardiovascular disease were excluded. All subjects were enrolled between June 1998 and January 2001. Subjects were followed for up to 24 months.

At a baseline visit, blood was collected for complete blood count and measurement of serum levels of total alkaline phosphatase, BAP, testosterone, albumin, lactate dehydrogenase, and PSA. A morning second-void urine was collected for measurement of urinary N-telopeptide. A central laboratory (Mayo Medical Laboratories, Rochester, MN) did the measurements of serum and urine markers of bone metabolism.

Statistical analyses. The primary outcome for these analyses was overall survival. The survival time was defined as interval from study entry to death. The survival time for a subject not observed to death was censored at the minimum of time to study completion or study withdrawal.

Univariate stratified Cox regression models were fit for overall survival using each covariate as a single explanatory variable. Next, multivariate models were fit using all explanatory variables to identify those that were independently predictive (10). Stepwise backward elimination was carried out to determine the simplest multivariate model; only terms remaining significant at the 5% level were retained in the reduced models. To assess the sensitivity of the findings to the coding of the bone marker variables, three sets of models were considered with BAP and N-telopeptide summarized as dichotomous (<median or ≥median) variables, quartiles, or continuous variables. All Cox regression models were stratified by cancer duration (<5 or ≥5 years), presence of bone metastases at initial prostate cancer diagnosis (yes or no), and the treatment assignment (placebo group, zoledronic acid 4 mg group, or zoledronic acid 8 mg group). Kaplan-Meier plots were used to estimate the probability of survival as a function of baseline marker values.

Based on the results of large multivariate analyses of men with hormone-refractory prostate cancer (5, 6), we included baseline age, log PSA, Eastern Cooperative Group, hemoglobin, albumin, lactate dehydrogenase, and Eastern Cooperative Oncology Group performance status as covariates in the Cox regression analyses. We also included requirement for analgesics as a measure of symptomatic metastatic disease. Serum BAP and urinary N-telopeptide were included in the analyses as representative specific markers of osteoblast and osteoclast activity, respectively. The Cox regression analyses were based on a complete data set obtained by restricting attention to subjects (n = 592) with available data for all covariates in the model.


   Results
Top
 Abstract
 Patients and Methods
 Results
Discussion
 References
 
Subject characteristics. Baseline characteristics are summarized in Table 1 . All subjects had bone metastases at study entry; 278 subjects (43.2%) had bone metastases at initial prostate cancer diagnosis. Three hundred ninety-six subjects (61.9%) required analgesics for bone pain.


View this table:
[in this window]
[in a new window]
  		Table 1. Subject characteristics

 
Relationships between markers. Scatter plots of serum BAP versus serum total alkaline phosphatase and serum BAP versus urinary N-telopeptide are shown in Fig. 1A and B . Serum BAP was significantly correlated with serum total alkaline phosphatase [Pearson's correlation coefficient = 0.986; 95% confidence interval (95% CI), 0.982-0.987; P < 0.0001]. Serum BAP was also significantly correlated with urinary N-telopeptide (correlation coefficient = 0.674; 95% CI, 0.628-0.715; P < 0.0001). Log PSA correlated weakly with serum BAP (correlation coefficient = 0.311; 95% CI, 0.239-0.380; P < 0.0001) and urinary N-telopeptide (correlation coefficient = 0.280; 95% CI, 0.206-0.352; P < 0.0001).


Figure 1
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 1. Scatter plots of baseline serum total alkaline phosphatase versus BAP (A) and serum BAP versus urinary N-telopeptide (NTX; B).

 
Cox proportional hazard analyses. Median overall survival was 478 days (95% CI, 443-531). The Kaplan-Meier estimates for survival at 12 and 24 months were 0.653 (95% CI, 0.616-0.690) and 0.385 (95% CI, 0.346-0.424), respectively. The relationships between baseline characteristics and overall survival were assessed by univariate and multivariate Cox proportional hazard model. Descriptive statistics for the subject demographics and covariates used in the Cox regression models are summarized in Table 1. Five-hundred ninety-two subjects had complete data for the covariates in the model.

Table 2 displays the results of univariate stratified Cox regression analyses for overall survival. Higher levels of serum BAP and urinary N-telopeptide levels were associated with shorter overall survival whether these markers were summarized as dichotomous (<median or ≥median) variables, quartiles, or continuous variables. In the analyses using marker quartiles, overall survival tended to progressively decrease with higher levels of serum BAP and urinary N-telopeptide. Kaplan-Meier plots of overall survival according to quartiles of urinary N-telopeptide and serum BAP reveal consistently greater mortality rates with higher levels of serum BAP and urinary N-telopeptide (Fig. 2A and B ).


View this table:
[in this window]
[in a new window]
  		Table 2. Univariate Cox regression analyses for overall survival

 

Figure 2
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 2. Kaplan-Meier estimates of overall survival by quartiles of urinary N-telopeptide (NTX; A) and serum BAP (B).

 
Table 3 shows the results of multivariate stratified Cox regression models with the different approaches for summarizing the marker data and the respective reduced models obtained by backwards elimination. After controlling for the other variables, including urinary N-telopeptide, in the full multivariate models, higher levels of serum BAP were significantly associated with shorter survival when bone markers were summarized as dichotomous variables, quartiles, or continuous variables. In contrast, N-telopeptide levels were not significantly associated with survival in multivariate analyses. Serum BAP was retained as a significant prognostic marker in reduced multivariate models. Other variables retained in the reduced multivariate model were age, log PSA, hemoglobin, lactate dehydrogenase, and analgesic use.


View this table:
[in this window]
[in a new window]
  		Table 3. Multivariate Cox regression analyses for overall survival

 
In full and reduced multivariate analyses, the third and fourth quartiles for BAP were significantly associated with greater relative risk of death than the first quartile. The similarity for relative risk of death between the first and second quartiles and between the third and fourth quartiles suggests that dichotomizing BAP (<median or ≥median) is a reasonable method for summarizing BAP in multivariate analyses.

Figure 3 provides Kaplan-Meier estimates for overall survival for subjects grouped according to baseline BAP (<median or ≥median) and N-telopeptide (<median or ≥median). Consistent with results of multivariate analyses, subjects with high BAP and low N-telopeptide had shorter survival than subjects with low BAP and high N-telopeptide. As expected, subjects with high BAP and high N-telopeptide had the shortest survival.


Figure 3
View larger version (13K):
[in this window]
[in a new window]
  		Fig. 3. Kaplan-Meier estimates for overall survival for subjects grouped by low (<median) and high (≥median) levels of serum BAP and urinary N-telopeptide (NTX).

 

   Discussion
Top
 Abstract
 Patients and Methods
 Results
 Discussion
References
 
We observed that serum BAP levels correlated significantly with urinary N-telopeptide levels in men with hormone-refractory prostate cancer and bone metastases. Elevated serum BAP levels were independently associated with shorter overall survival after controlling for other variables, including urinary N-telopeptide. In contrast, baseline N-telopeptide levels were not significantly associated with survival in multivariate analyses. Thus, whereas both serum BAP and urinary N-telopeptide are markedly elevated in men with metastatic prostate cancer and are associated with shorter survival, only BAP remains independently associated with shorter overall survival in multivariate models.

Several previous studies have reported that elevated levels of serum total alkaline phosphatase were independently associated with shorter survival (5, 6). Tissue-specific isoenzymes of alkaline phosphatases are present in bone, liver, intestine, kidney, and leukocytes. In normal adults, serum total alkaline phosphatase mainly reflects liver and bone isoenzymes (11). We found that serum levels of total and BAP are highly correlated, showing that, as expected, total alkaline phosphatase primarily reflects the bone-specific isoenzyme in men with hormone-refractory prostate cancer and bone metastases. We also found that elevated serum levels of either total alkaline phosphatase or BAP are independently associated with shorter survival (data not shown). Taken together, these observations suggest that total alkaline phosphatase is an appropriate surrogate for the bone-specific isoform for most clinical and research purposes in men with metastatic prostate cancer.

In our univariate analyses of all 643 subjects, elevated levels of urinary N-telopeptide were associated with shorter overall survival. In contrast, baseline levels of urinary N-telopeptide were not associated with overall survival in the entire study population when controlling for other baseline variables, including BAP. In addition, baseline levels of urinary N-telopeptide levels were not independently associated with survival in other multivariate analyses restricted to the placebo group (data not shown). In the prospective clinical trial, serum BAP and N-telopeptide were measured at baseline, 1 month, 3 months, and every 3 months thereafter. Our analyses were restricted to baseline marker levels. We limited the analyses to the baseline BAP and N-telopeptide values to allow unbiased comparison with other variables that were not serially assessed on study. Because zoledronic acid significantly decreases urinary N-telopeptide, we also limited the analyses to baseline values to allow inclusion of subjects in the zoledronic acid groups without the potential confounding effects of treatment on the analyses. Taken together, these results suggest that other covariates in the multivariate model explain the previously reported association between N-telopeptide and overall survival. Our observations do not diminish the value of urinary N-telopeptide for monitoring the efficacy of bisphosphonate treatment in patients with cancer.

Consistent with recent large multivariate analyses of men with hormone-refractory metastatic prostate cancer (5, 6), low hemoglobin concentrations and elevated levels of PSA and lactate dehydrogenase were independently associated with shorter survival in our multivariate model. We also found that analgesic use, a variable not assessed in previous studies, was independently associated with shorter survival. Additional studies are needed to confirm this observation. Because our analyses were restricted to men with bone metastases, additional studies are also needed to determine whether a requirement for analgesics is associated with shorter survival in men without skeletal involvement.

As previously reported, overall survival was similar for men in the placebo and zoledronic acid groups (9). Nevertheless, our analyses were stratified according to treatment assignment in both the univariate and multivariate models to reduce any potential effect of study treatment on survival. We used serum BAP and urinary N-telopeptide as representative markers of osteoblast and osteoclast activity. Although other specific markers of bone metabolism have similar operating characteristics, different results may be observed using other markers of osteoblast and osteoclast activity. Subjects in the study cohort received additional treatment, including secondary hormonal therapy and/or chemotherapy at physician discretion. This feature of the study cohort increases the generalizability of our observations. Because detailed information about secondary hormone therapy and chemotherapy was not collected, however, we cannot assess the effects of these other treatments on survival.

In summary, elevated serum BAP levels are independently associated with shorter overall survival in men with androgen-independent prostate cancer and bone metastases. Although significantly correlated with serum BAP, urinary N-telopeptide is not significantly associated with survival after controlling for other variables.


   Footnotes
 
Grant support: Novartis Oncology and the John and Claire Bertucci Center for Genitourinary Malignancies at Massachusetts General Hospital.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: This work has not been published or presented previously.

Received 2/ 7/06; revised 3/20/06; accepted 4/ 4/06.


   References
Top
 Abstract
 Patients and Methods
 Results
 Discussion
 References
 

  1. Clarke NW, McClure J, George NJ. Morphometric evidence for bone resorption and replacement in prostate cancer. Br J Urol 1991;68:74–80.[Medline]
  2. Clarke NW, McClure J, George NJ. Osteoblast function and osteomalacia in metastatic prostate cancer. Eur Urol 1993;24:286–90.[Medline]
  3. Garnero P. Markers of bone turnover in prostate cancer. Cancer Treat Rev 2001;27:187–92.[CrossRef][Medline]
  4. Fontana A, Delmas PD. Markers of bone turnover in bone metastases. Cancer 2000;88:2952–60.[CrossRef][Medline]
  5. Smaletz O, Scher HI, Small EJ, et al. Nomogram for overall survival of patients with progressive metastatic prostate cancer after castration. J Clin Oncol 2002;20:3972–82.[Abstract/Free Full Text]
  6. Halabi S, Small EJ, Kantoff PW, et al. Prognostic model for predicting survival in men with hormone-refractory metastatic prostate cancer. J Clin Oncol 2003;21:1232–7.[Abstract/Free Full Text]
  7. Brown JE, Cook RJ, Major P, et al. Bone turnover markers as predictors of skeletal complications in prostate cancer, lung cancer, and other solid tumors. J Natl Cancer Inst 2005;97:59–69.[Abstract/Free Full Text]
  8. Coleman RE, Major P, Lipton A, et al. Predictive value of bone resorption and formation markers in cancer patients with bone metastases receiving the bisphosphonate zoledronic acid. J Clin Oncol 2005;23:4925–35.[Abstract/Free Full Text]
  9. Saad F, Gleason DM, Murray R, et al. A randomized, placebo-controlled trial of zoledronic acid in patients with hormone-refractory metastatic prostate carcinoma. J Natl Cancer Inst 2002;94:1458–68.[Abstract/Free Full Text]
  10. Kalbfleisch J, Prentice R. The statistical analysis of failure time data. 2nd ed. John Wiley and Sons; 2002.
  11. Posen S, Doherty E. The measurement of serum alkaline phosphatase in clinical medicine. Adv Clin Chem 1981;22:163–245.[Medline]



This article has been cited by other articles:


Home page
 Cleveland Clinic Journal of MedicineHome page
F. R. SINGER and D. R. EYRE
Using biochemical markers of bone turnover in clinical practice.
Cleveland Clinic Journal of Medicine, October 1, 2008; 75(10): 739 - 750.
[Abstract] [Full Text] [PDF]

Home page
 CA Cancer J ClinHome page
M. D. Michaelson, S. E. Cotter, P. C. Gargollo, A. L. Zietman, D. M. Dahl, and M. R. Smith
Management of Complications of Prostate Cancer Treatment
CA Cancer J Clin, July 1, 2008; 58(4): 196 - 213.
[Abstract] [Full Text] [PDF]

Home page
 Ann OncolHome page
M. Aapro, P. A. Abrahamsson, J. J. Body, R. E. Coleman, R. Colomer, L. Costa, L. Crino, L. Dirix, M. Gnant, J. Gralow, et al.
Guidance on the use of bisphosphonates in solid tumours: recommendations of an international expert panel
Ann. Onc., March 1, 2008; 19(3): 420 - 432.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
Y. Zheng, H. Zhou, J. R.K. Modzelewski, R. Kalak, J. M. Blair, M. J. Seibel, and C. R. Dunstan
Accelerated Bone Resorption, Due to Dietary Calcium Deficiency, Promotes Breast Cancer Tumor Growth in Bone
Cancer Res., October 1, 2007; 67(19): 9542 - 9548.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
J. S. Johansen, K. Brasso, P. Iversen, B. Teisner, P. Garnero, P. A. Price, and I. J. Christensen
Changes of Biochemical Markers of Bone Turnover and YKL-40 Following Hormonal Treatment for Metastatic Prostate Cancer Are Related to Survival
Clin. Cancer Res., June 1, 2007; 13(11): 3244 - 3249.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cook, R. J.
Right arrow Articles by Smith, M. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cook, R. J.
Right arrow Articles by Smith, M. R.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online