Clinical Cancer Research Targets Advances in Breast 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
Clinical Cancer Research 13, 6396, November 1, 2007. doi: 10.1158/1078-0432.CCR-07-1036
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
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 Armstrong, A. J.
Right arrow Articles by Eisenberger, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Armstrong, A. J.
Right arrow Articles by Eisenberger, M.

Cancer Therapy: Clinical

A Contemporary Prognostic Nomogram for Men with Hormone-Refractory Metastatic Prostate Cancer: A TAX327 Study Analysis

Andrew J. Armstrong1, Elizabeth S. Garrett-Mayer2, Yi-Chun Ou Yang2, Ronald de Wit3, Ian F. Tannock4 and Mario Eisenberger2

Authors' Affiliations: 1 Duke Comprehensive Cancer Center, Durham, North Carolina; 2 Sidney Kimmel Comprehensive Cancer Center, Baltimore, Maryland; 3 Department of Medical Oncology, Rotterdam Cancer Institute, Erasmus University Medical Center, Rotterdam, the Netherlands; and 4 Department of Medical Oncology, Princess Margaret Hospital and University of Toronto, Toronto, Ontario, Canada

Requests for reprints: Andrew J. Armstrong, Duke Comprehensive Cancer Center, 2424 Erwin Road, Hock Plaza, Suite 606, Room 6127, Durham, NC 27705. Phone: 919-668-8797; Fax: 919-668-7117; E-mail: andrew.armstrong{at}duke.edu.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Purpose: To develop a prognostic model and nomogram using baseline clinical variables to predict death among men with metastatic hormone-refractory prostate cancer (HRPC).

Experimental Design: TAX327 was a clinical trial that randomized 1,006 men with metastatic HRPC to receive every three week or weekly docetaxel or mitoxantrone, each with prednisone. We developed a multivariate Cox model and nomogram to predict survival at 1, 2, and 5 years.

Results: Ten independent prognostic factors other than treatment group were identified in multivariate analysis: (a) presence of liver metastases [hazard ratio (HR), 1.66; P = 0.019], (b) number of metastatic sites (HR, 1.63 if ≥2 sites; P = 0.001), (c) clinically significant pain (HR, 1.48; P < 0.0001), (d) Karnofsky performance status (HR, 1.39 if ≤70; P = 0.016), (e) type of progression (HR, 1.37 for measurable disease progression and 1.29 for bone scan progression; P = 0.005 and 0.01, respectively), (f) pretreatment prostate-specific antigen (PSA) doubling time (HR, 1.19 if <55 days; P = 0.066), (g) PSA (HR, 1.17 per log rise; P < 0.0001), (h) tumor grade (HR, 1.18 for high grade; P = 0.069), (i) alkaline phosphatase (HR, 1.27 per log rise; P < 0.0001), and (j) hemoglobin (HR, 1.11 per unit decline; P = 0.004). A nomogram was developed based on this multivariate model and validated internally using bootstrap methods, with a concordance index of 0.69.

Conclusions: This multivariate model identified several new independent prognostic factors in men with metastatic HRPC, including PSA doubling time, and led to the successful development of a clinically applicable nomogram. External prospective validation may support the wider use of this prognostic baseline model for men with HRPC treated with chemotherapy.

Despite the recent demonstration of palliative and survival benefits with docetaxel-based regimens, men with metastatic hormone-refractory prostate cancer (HRPC) have a poor prognosis, with a median survival of 16 to 20 months (1, 2). The biology of prostate cancer is heterogeneous, with expected survival depending on performance status, presence of visceral metastases, baseline prostate-specific antigen (PSA), advanced primary Gleason sum, baseline hemoglobin, lactate dehydrogenase, albumin, and alkaline phosphatase (3, 4). Nomograms have been developed as stratification tools for phase III clinical trials evaluating cytotoxic chemotherapy based on data from multiple studies that assessed prognostic factors (3–5).

Baseline PSA kinetics [PSA doubling time (PSADT) or PSA velocity] have not been shown conclusively to be an independent prognostic factor in HRPC, with most analyses based on retrospective reviews of relatively small numbers of patients (6–9). These analyses were limited by the number of controllable covariates, the duration of follow-up, potential confounders, and measurement bias. Furthermore, past prognostic analyses are limited by the inclusion of various types of noncytotoxic therapy and did not include patients treated with docetaxel-based therapy (3, 4).

In 2004, docetaxel was approved for the treatment of men with metastatic HRPC based on large multicenter randomized clinical trials (1, 2). The TAX327 study randomized 1,006 patients to receive one of two schedules of docetaxel or mitoxantrone, each given with low-dose prednisone, and showed an extension of overall survival, improvement in quality of life, pain control, PSA decline, and objective tumor response (2). A second phase III study using estramustine phosphate in combination with docetaxel confirmed the survival advantage of docetaxel-based chemotherapy (1).

The aim of the current analysis is to develop a multivariate prognostic model, based on the TAX327 trial, to investigate the independent prognostic significance of novel prognostic variables, such as pretreatment PSA kinetics, the presence of pain, number of metastatic sites, and the type of disease progression at baseline, in addition to traditional prognostic markers.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
TAX327 was a randomized, nonblinded, multinational phase III study involving 1,006 men with progressive metastatic HRPC. Full details of the eligibility and baseline characteristics are provided in the original report (2). Briefly, men were eligible if they had documented metastatic prostatic adenocarcinoma in the face of castrate levels of serum testosterone (<50 ng/mL) and had evidence of progression as defined by clinically or radiographically measurable disease or by PSA criteria. Baseline information collected on each individual included serum PSA, age, performance status, pain and analgesic score, hemoglobin, alkaline phosphatase, prior Gleason sum or tumor grade, presence of visceral metastases, location and number of metastases, prior therapies, and how progression was defined, in addition to other clinical variables. An approved and signed Institutional Review Board informed consent form was obtained for all participants at each participating institution.

In a subset of 686 men, three or more consecutive baseline PSA measurements each separated by at least 1 week were available for our prognostic evaluation of PSA kinetics. PSA velocity was calculated assuming first-order kinetics as the rise in serum PSA (ng/mL) over time based on individual linear regression before study initiation. PSADT was calculated from these data. PSADT was entered into the model using both continuous and log-transformed variables as well as categorical variables separated by median values.

The main end point of this analysis was duration of survival as defined by the time from randomization to death, with the survival data updated as of November 2006. Patients who were alive as of the cutoff date were administratively censored, whereas patients lost to follow-up were censored. The Kaplan-Meier product limit estimator was used to estimate the survival distribution (10). Univariate and multivariate Cox proportional hazards modeling was done using Stata 8.2 software (StataCorp LP). Cox proportional hazards assumptions were tested using Schoenfeld residual analysis for univariate and multivariate analysis using as a level of significance of 0.01. Comparisons of baseline characteristics were done using 95% confidence intervals (95% CI), t tests, and exploratory statistical analysis. Covariates of interest for univariate analysis included age, performance status, pain at baseline, baseline levels of hemoglobin, alkaline phosphatase and PSA, presence of visceral and/or liver metastases, pretreatment PSADT, type of progression (measurable, nonmeasurable, bone scan, and PSA-only progression), number of metastatic sites, Gleason score, presence of multiple bony metastases, and prior estramustine or second-line hormonal therapy. Variables (other than PSADT) were prospectively collected on case report forms. Tests for interaction were done for baseline PSA and PSADT. Colinearity and normality of variables were evaluated before inclusion. All variables with skewed distributions (PSA and alkaline phosphatase) were log transformed before incorporation. Internal validation of the model was done using the concordance index and internal bootstrap resampling methods, respectively.

In this data set, high-grade tumors were defined by a primary Gleason sum of ≥8 or a WHO grade of 3 to 4 for those tumors who did not have Gleason sums. This combined variable was included due to the omission of Gleason scoring in 288 subjects (28.6%) and the similarities in morphology in Gleason 8 to 10 compared with WHO 3 to 4.5 The similarities in the point estimates for overall survival on univariate analysis for each of these categories compared with the joint category further validate this assumption. Multiple sites of involvement on bone scan were defined as the presence of more than two regions of nuclear tracer uptake, such as spine and ribs or long bones and pelvis. Performance status was graded using the Karnofsky scale. Pain intensity at baseline was evaluated by the present pain intensity score from the McGill-Melzack questionnaire, and an analgesic score was calculated as the mean daily intake of analgesics, with a score of 4 assigned to a standard dose of narcotic (e.g., 10 mg morphine) and a score of 1 for a standard dose of anon-narcotic (e.g., acetylsalicylic acid or 325 mg acetaminophen; ref. 11). A present pain intensity of ≥2 and/or an analgesic score of ≥10 were defined in the original protocol as indicative of the presence of significant pain (2, 11).

The assumptions of the Cox model for our analysis was tested in two ways using binary versions of PSADT alone and also using PSADT and median baseline PSA. In both cases, there was not strong evidence of violation of the proportional hazards assumption based on graphical displays or by using P values (P = 0.07 and 0.04, respectively), and thus, this model design was appropriate for the current analysis.

A stepwise procedure including a bootstrap was used for determining the final model for the nomogram. All variables that were significant in univariate analyses based on an {alpha} level of 0.10 were included in a Cox multiple regression model. The covariate with the largest P value was removed, and the model was refit iteratively until all covariates in the model had P values of <0.10. This model was considered the final model for the data set. To prevent against overfitting, this procedure was repeated 500 times using bootstrapped data sets. Covariates that were included in more than half of the final models in the 500 bootstrapped samples were used to define the final set of covariates. The final set of covariates was included in a Cox regression using the original data set. This final Cox proportional hazards regression model was used to create a nomogram for prediction of median and 1-, 2-, and 5-year overall survival, and our approach for development of the nomogram and validation was similar to that described by Kattan et al. (12). The R Design library was used to produce the nomogram and to estimate the c-indices for validation (13, 14). The concordance index is the probability that given two randomly drawn patients that the patient who dies first has the higher predicted mortality based on the model. A higher concordance index on a 0 to 1 scale indicates a higher predictive ability.


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
Of 1,006 men with HRPC accrued between March 2000 and June 2002, 686 men had three or more baseline PSA measurements separated each by more than 1 week for calculation of pretreatment PSA kinetics. At the time of this analysis on November 6, 2006, there were 800 mortality events of 1,006 subjects and 18,886 person-months of follow-up, with the longest time to event being 70.8 months. Men in the PSA kinetic subset of the overall database had slightly different baseline characteristics than those patients not included in this subset of 686 men. For example, pain, poor performance status, and visceral metastases were more common in those subjects excluded from this analysis, whereas the proportion of men with PSA-only progression at entry were more likely to be in the PSA kinetic subgroup (Supplementary Table S1). However, the median survival of men in this subgroup did not differ from those that were excluded due to lack of sufficient PSA data, which indicates that these two groups have similar clinical outcomes. PSADTs did not differ according to treatment group at baseline (Supplementary Fig. S1). In addition, the covariates of interest were well distributed across treatment groups within the PSA kinetic subset, and the median survival for each treatment arm reflects that seen in the overall updated survival analysis (Supplementary Table S2; ref. 15).

The median baseline PSADT in this cohort of men was 55.8 days (mean, 79 days; range, 5-1245 days; SD, 92 days). As a secondary analysis, men with a more rapid PSADT (<55 days) were more likely to have pain at baseline, progression on bone scan, multiple hotspots on the baseline bone scan, a lower baseline hemoglobin, and higher baseline alkaline phosphatase compared with those men with a slower PSADT (≥55 days; Table 1 ).


View this table:
[in this window]
[in a new window]

  		Table 1. Baseline characteristic of patients according to more rapid (<55 d) or slower (≥55 d) PSA Doubling Time (PSADT) at baseline

 
In univariate Cox proportional hazards analysis (Table 2 ), all traditional prognostic markers showed a significant association with overall survival, confirming prior reports and adding validity to the Halabi nomogram (3). These variables included performance status, baseline hemoglobin and alkaline phosphatase, tumor grade (Gleason sum or by WHO criteria), presence of visceral metastases, and baseline serum PSA. Neither lactate dehydrogenase nor albumin was uniformly measured in the TAX327 database due to the high variability of normal values used in this multinational study and, thus, could not be included in this analysis. The variables with the greatest risk of death included liver metastases [hazard ratio (HR), 2.28], more than two sites involved (HR, 1.99), poor performance status (HR, 1.90), and presence of pain at baseline (HR, 1.72).


View this table:
[in this window]
[in a new window]

  		Table 2. Univariate Cox proportional hazards analysis

 
In univariate analysis, a shorter pretreatment PSADT (<55 versus ≥55 days) was associated with a 46% increase in the risk of death (HR, 1.46; 95% CI, 1.24-1.73; P < 0.001). Other variables with prognostic significance included extent of involvement on bone scan (more than two regions involved) and type of progression at baseline. Progression by measurable disease or by bone scan was associated with a 26% to 36% increase in the risk of death (HR, 1.26 and 1.36, respectively; P = 0.003 and P < 0.001, respectively), whereas a PSA-only progression at baseline was associated with a 25% decrease in the risk of death (HR, 0.75; 95% CI, 0.61-0.92; P = 0.003). Non-Caucasian race was not significantly associated with better survival but this was based on only 70 non-Caucasian subjects (HR, 0.80; P = 0.13).

In multivariate Cox proportional hazards analysis, 635 men were available for analysis after exclusion of missing data from tumor grade and PSADT. Table 3 describes the relative hazards for each covariate in the final multivariate model. Supplementary Table S3 shows the results of the stepwise model fitting for the 500 boostrapped samples. Independent prognostic factors in order of importance included liver metastases (HR, 1.66; P = 0.019), more than two regions of metastatic disease (HR, 1.63; P = 0.001), chemotherapy type (weekly docetaxel versus q3w docetaxel: HR, 1.12; P = 0.32; mitoxantrone versus q3w docetaxel: HR, 1.43; P = 0.001), significant baseline pain (HR, 1.48; P < 0.001), Karnofsky performance status ≤70 (HR, 1.39; P = 0.016), bone scan progression (HR, 1.29; P = 0.010) or measurable disease progression (HR, 1.37; P = 0.005), PSADT ≤55 days (HR, 1.19; P = 0.066), PSA (per log rise, HR, 1.17; P < 0.001), high tumor grade (HR, 1.18; P = 0.069), alkaline phosphatase (per unit log rise, HR, 1.27; P < 0.001), and baseline hemoglobin (per unit decline, HR, 1.11; P = 0.004). In this analysis, each progression category was compared with a reference group without the corresponding progression type (i.e., measurable disease progression compared with no measurable disease progression). Due to colinearity with the presence of liver metastasis at baseline and a weaker prognostic ability, visceral metastasis was dropped from the multivariate model. Additional factors that were not included due to lack of a statistically significant prognostic effect included race, age, number of hotspots on bone scan, prior estramustine, nonmeasurable disease progression, prior radiotherapy, prior surgery, and baseline testosterone (data not shown).


View this table:
[in this window]
[in a new window]

  		Table 3. Multivariate Cox proportional hazards analysis

 
In an analysis of PSADT as a predictor of mortality in HRPC, Kaplan-Meier survival curves were constructed according to PSADT, categorized by month groupings: <1 month, 1 to 2 months, 2 to 3 months, 3 to 6 months, and >6 months, representing 109, 264, 151, 118, and 44 subjects, respectively. Using a PSADT <1 month as a reference, we found a decreasing risk of death as PSADT lengthened, with a relative hazard of 0.79, 0.69, 0.53, and 0.37 for those with a PSADT of 1 to 2 months, 2 to 3 months, 3 to 6 months, and >6 months (P < 0.001 for trend; Fig. 1 ). Although the categorical separation of PSADT at the median (55 days) provided a significant prognostic separation, the monthly categorical separation provided a more refined stratification for the prognostic nomogram. In bivariate analysis, baseline PSA and PSADT contained independent prognostic information, and a simple two-variable model with these terms had a moderate concordance index of 0.60. Survival curves based on PSA and PSADT categories are shown in Fig. 2 . The test for interaction between baseline PSADT and PSA was nonsignificant (P = 0.18).


Figure 1
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. Kaplan-Meier estimates of overall survival according to PSADT in TAX327 cohort (n = 686, 518 events). Median PSADT is 55 d in this study. PSADT is separated into five cohorts: >6 mo (n = 44; median overall survival not reached; mean overall survival, 25 mo), 3 to 6 mo (n = 118; median overall survival not reached; mean overall survival, 22.5 mo), 2 to 3 mo (n = 151; median overall survival, 20.7 mo), 1 to 2 mo (n = 264; median overall survival, 18.6 mo), and <1 mo (n = 109; median overall survival, 13.3 mo).

 

Figure 2
View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 2. Kaplan-Meier estimates of overall survival according to PSADT and baseline PSA values in TAX327 cohort (n = 686, 518 events). Test for interaction is nonsignificant (HR, 0.79; P = 0.18) for PSADT and baseline PSA.

 
The internal validity of the multivariate model was evaluated by the nonparametric bootstrap resampling method. The concordance index of the final multivariate model was 0.69 and the index-corrected c-index was 0.68, indicating a moderately high level of predictive discrimination. A nomogram was developed using this multivariate model that may be used to predict overall survival at 1, 2, and 5 years and is shown in Fig. 3 with instructions. Note that PSADT is represented using an ordinal scale in months rather than a separation at the median.


Figure 3
View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 3. Nomogram for survival of patients with progressive HRPC, including data derived from 686 patients and 518 mortality events. Note: a present pain intensity of ≥2 and/or an analgesic score of ≥10 were defined in the original protocol as indicative of the presence of significant pain. Instructions for physician: Locate the liver metastasis axis. Draw a straight line upward to the points axis to determine how many points toward survival the patient receives for the presence or absence of liver metastases. Repeat this process for each predictor variable and sum the points for each predictor. Locate this sum on the total points axis. Draw a straight line downward from the total points axis to identify the predicted median survival and the predicted 1-, 2-, and 5-years predicted overall survival probabilities. Instructions to patient: "Mr. X, if we had 100 men exactly like you, we would expect <nomogram prediction x 100> to be alive in 1, 2, and 5 y, respectively, and we expect 50 of those to be alive after <median survival prediction> months."

 

   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
The importance of a prognostic model rests on its ability to capture clinically relevant and measurable variables for routine use by clinicians to inform patients, improve palliation and treatment decisions, and create homogeneous prognostic strata for randomized comparative trials of therapeutic agents. The prognostic nomogram developed by Halabi et al. (3) collects readily available clinical information, such as primary tumor Gleason sum, performance status, PSA, lactate dehydrogenase, alkaline phosphatase, hemoglobin, and presence of visceral metastases: it divides patients into quartiles of risk with good discriminatory ability for median overall survival in men with HRPC. However, this nomogram and several prognostic models before it were based on data from many smaller and older trials where the majority of patients were not treated with currently used chemotherapeutic regimens, such as docetaxel or mitoxantrone with prednisone (3, 4, 16, 17). Although PSA kinetics are thought to carry prognostic information in HRPC, only recently have they been investigated together with traditional prognostic factors in a systematic way (6–9, 17–19). Given the large and multinational scope of the current study, and the availability of multiple clinical variables including pretreatment PSA kinetics in a large subset of subjects, we were able to develop a comprehensive baseline prognostic model for men with HRPC.

We found baseline PSADT to be independently associated with overall survival, regardless of baseline PSA. We also found that patients with relatively fast PSA kinetics (PSADT <55 days) were more likely to have pain at baseline, higher serum PSA and alkaline phosphatase, lower hemoglobin concentration, more hotspots on the baseline bone scan, and be of younger age. Despite these confounding associations, PSADT retained independent prognostic significance in our multivariate model, and a nomogram was developed that incorporates PSA kinetics along with other significant prognostic variables.

Our study was based on a large phase III trial of men with progressive metastatic HRPC, but PSA kinetics were not calculated nor stratified for nor considered in the primary analysis. Our analysis is thus retrospective and based on a subset of these individuals who had sufficient baseline data to allow an estimation of PSADT and included only men with three or more PSA values separated by more than 1 week. Given the short PSADTs in this cohort of patients with a median of 55 days, this separation of measurements in time seems reasonable. We found that this subgroup of patients with sufficient data on PSA kinetics had a similar median survival to those excluded from the nomogram development and thus should not limit the generalizability of this analysis. These modest differences should not influence the generalization or significance of our findings, and external validation of this model using prospective data sets that capture PSA kinetics is planned.

PSADT has emerged as an easily obtainable and clinically relevant prognostic marker in several stages of prostate cancer, including pre-prostatectomy, pre-radiation therapy, rising PSA after local therapy, nonmetastatic HRPC, and, most recently, metastatic HRPC (6–9, 20–24). After primary therapy, observations across different studies suggest that the PSADT may shorten as the disease progresses, although longitudinal studies are needed to address PSA acceleration within the same groups of patients. Despite this shortening in median doubling time in metastatic HRPC, heterogeneity still exists in overall survival, as indicated by our multivariate model and nomogram.

In addition to PSA and PSADT, we identified several important prognostic markers from our data set, including number of metastatic sites (three or more), type of chemotherapy, presence of pain, presence of liver metastases, and type of progression at baseline (measurable disease or bone scan compared with PSA only). Baseline pain and pain response have been evaluated as prognostic markers in numerous data sets and should be considered a validated prognostic factors based on this and prior analyses (16, 25–28). Liver metastases were found to carry greater influence than overall visceral metastases: they may be more common with neuroendocrine differentiation, which may itself have prognostic import 7(16, 28–30). An additional novel independent factor in this analysis was the mode of disease progression at baseline, wherein men who had progression by bone scan or measurable soft tissue disease had a 1.28- and 1.40-fold increased risk of death, respectively, compared with those men without these modes of progression, such as PSA-only or nonmeasurable disease progression (P = 0.014 and 0.002, respectively).

The nomogram and multivariate model was found to have a bootstrap concordance index of 0.69, indicating that, for 31% of patient pairs, the patient predicted to have a better prognosis died first. This predictive ability compares favorably with the concordance indices (0.67-0.68) seen with prior nomograms that have not been validated in the current era of docetaxel-based chemotherapy (3, 4, 31). Prior nomograms have not included patient symptoms, such as pain, nor have they examined the independent role of disease burden or PSA kinetics. The lack of complete predictive ability illustrates that, despite the inclusion of 10 independently significant variables, additional factors may contribute to overall survival. Other potential factors that were not examined in this model include lactate dehydrogenase, albumin, serum biomarkers for vascular endothelial growth factor and other cytokines, circulating tumor cells, type of treatment beyond progression, and other unmeasured prognostic factors (3, 4, 32–37). The addition of these biomarkers in future studies may increase the predictive ability of this nomogram, whereas the addition of PSA kinetics, pain, mode of progression, and number of sites of metastases may add predictive ability to other nomograms that did not contain these variables.

In conclusion, we have developed a baseline prognostic model and nomogram for men with HRPC using the largest clinical trial data set available for this disease. This model shows internal validity to a similar degree compared with historic models in the pre-docetaxel era and includes simple and easily obtainable clinical variables that may be useful for clinical prognostication or stratification of subjects in clinical trials in this population. In addition, we have shown that PSADT, baseline pain, mode of progression, and the number of metastatic disease sites are independently associated with risk of death in men with HRPC, despite accounting for traditional risk factors. Prospective external validation of this model is planned and will be essential for more widespread clinical application.


   Acknowledgments
 
We thank the many patients who volunteered to participate in the TAX327 study and the investigators who recruited them and managed their care, listed below. M.A. Rosenthal (Australia); D. Campos (Argentina); H. Gurney and M. Stockler (Australia); M. Rauchenwald (Austria); T. Gil, Y. Humblet, and A. van Oosterom (Belgium); D. Herchenhorn (Brazil); S. Ernst, L. Lacombe, M. Moore, F. Saad, D. Soulieres, P. Venner, and E. Winquist (Canada); M. Urban (Czech Republic); P. Kellokumpu-Lehtinen (Finland); G. Deplanque, B. Duclos, I. Krakowski, L. Mignot, S Oudard, and C. Theodore (France); J. Breul and R. Paul (Germany); I. Bodrogi (Hungary); S. Bracarda (Italy); G. Chahine (Lebanon); J.L. Bruins and J.A. Witjes (the Netherlands); T. Demkow and K. Bar (Poland); O. Kariakine and M. Matveev (Russia); I. Andrasina (Slovak Republic); D. Vorobiof (South Africa); J. Bellmunt and J-R. Germa (Spain); I. Turesson and A. Widmark (Sweden); A. Horwich, T. Roberts, N.D. James, and J. Wylie (United Kingdom); and L. Baez, W. Dugan, and N. Tirumali (United States).


   Footnotes
 
Grant support: Sanofi-Aventis, Celgene, and Cytogen (M. Eisenberger); Sanofi-Aventis (R. de Wit); and Immunex (I.F. Tannock).

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: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).

M. Eisenberger is a consultant to Sanofi-Aventis, Celgene, Cytogen, Bristol-Myers Squibb, Merck, and ImClone. R. de Wit is a consultant to Sanofi-Aventis.

5 I. Sesterhenn, personal communication. Back

Received 5/ 1/07; revised 7/17/07; accepted 8/ 2/07.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Petrylak DP, Tangen CM, Hussain MH, et al. Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N Engl J Med 2004;351:1513–20.[Abstract/Free Full Text]
  2. Tannock IF, de Wit R, Berry WR, et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004;351:1502–12.[Abstract/Free Full Text]
  3. 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]
  4. 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]
  5. Ryan CJ, Lin AM, Small EJ. Angiogenesis inhibition plus chemotherapy for metastatic hormone refractory prostate cancer: history and rationale. Urol Oncol 2006;24:250–3.[Medline]
  6. Rozhansky F, Chen MH, Cox MC, Dahut W, Figg WD, D'Amico AV. Prostate-specific antigen velocity and survival for patients with hormone-refractory metastatic prostate carcinoma. Cancer 2006;106:63–7.[CrossRef][Medline]
  7. Scotte F, Banu E, Oudard S, et al. Prostate-specific antigen doubling-time (PSA DT) before onset of chemotherapy as survival predictor for hormone refractory prostate cancer (HRPC) patients. J Clin Oncol (Meet Abstr) 2005;23:4551.
  8. Semeniuk RC, Venner PM, Hanson J, North SA. A retrospective review of prostate specific antigen doubling time (PSA-DT) in men with hormone refractory prostate cancer (HRPC). J Clin Oncol (Meet Abstr) 2005;23:4666.
  9. Srinivas S, Balu V, King C. Utility of serum prostate specific antigen doubling time (PSADT) in patients with androgen independent prostate cancer. J Clin Oncol (Meet Abstr) 2005;23:4631.
  10. Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–81.[CrossRef]
  11. Melzack R. The McGill Pain Questionnaire: major properties and scoring methods. Pain 1975;1:277–99.[CrossRef][Medline]
  12. Kattan MW, Eastham JA, Stapleton AM, Wheeler TM, Scardino PT. A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst 1998;90:766–71.[Abstract/Free Full Text]
  13. Harrell FE, Jr., Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med 1996;15:361–87.[CrossRef][Medline]
  14. Harrell FE, Jr. Regression modeling strategies with applications to linear models, logistic regression, and survival analysis. New York: Springer-Verlag; 2001.
  15. Berthold DR, Pond G, De Wit R, Eisenberger M, Tannock IF. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer: updated survival of the TAX 327 study. J Clin Oncol (Meet Abstr) 2007;25:5005.
  16. Berry WR, Laszlo J, Cox E, Walker A, Paulson D. Prognostic factors in metastatic and hormonally unresponsive carcinoma of the prostate. Cancer 1979;44:763–75.[CrossRef][Medline]
  17. Vollmer RT, Kantoff PW, Dawson NA, Vogelzang NJ. A prognostic score for hormone-refractory prostate cancer: analysis of two cancer and leukemia group B studies. Clin Cancer Res 1999;5:831–7.[Abstract/Free Full Text]
  18. Loberg RD, Fielhauer JR, Pienta BA, et al. Prostate-specific antigen doubling time and survival in patients with advanced metastatic prostate cancer. Urology 2003;62 Suppl 1:128–33.[CrossRef][Medline]
  19. Vollmer RT, Dawson NA, Vogelzang NJ. The dynamics of prostate specific antigen in hormone refractory prostate carcinoma: an analysis of cancer and leukemia group B study 9181 of megestrol acetate. Cancer 1998;83:1989–94.[CrossRef][Medline]
  20. Akimoto S, Masai M, Akakura K, Shimazaki J. Tumor marker doubling time in patients with prostate cancer: determination of prostate-specific antigen and prostatic acid phosphatase doubling time. Eur Urol 1995;27:207–12.[Medline]
  21. D'Amico AV, Chen MH, Roehl KA, Catalona WJ. Preoperative PSA velocity and the risk of death from prostate cancer after radical prostatectomy. N Engl J Med 2004;351:125–35.[Abstract/Free Full Text]
  22. Freedland SJ, Humphreys EB, Mangold LA, et al. Risk of prostate cancer-specific mortality following biochemical recurrence after radical prostatectomy. JAMA 2005;294:433–9.[Abstract/Free Full Text]
  23. Smith MR, Kabbinavar F, Saad F, et al. Natural history of rising serum prostate-specific antigen in men with castrate nonmetastatic prostate cancer. J Clin Oncol 2005;23:2918–25.[Abstract/Free Full Text]
  24. Zhou P, Chen MH, McLeod D, Carroll PR, Moul JW, D'Amico AV. Predictors of prostate cancer-specific mortality after radical prostatectomy or radiation therapy. J Clin Oncol 2005;23:6992–8.[Abstract/Free Full Text]
  25. Berthold DR, Pond G, De Wit R, Eisenberger MA, Tannock IF. Association of pain and quality of life (QOL) response with PSA response and survival of patients (pts) with metastatic hormone refractory prostate cancer (mHRPC) treated with docetaxel or mitoxantrone in the TAX-327 study. J Clin Oncol (Meet Abstr) 2006;24:4516.
  26. Emrich LJ, Priore RL, Murphy GP, Brady MF. Prognostic factors in patients with advanced stage prostate cancer. Cancer Res 1985;45:5173–9.[Abstract/Free Full Text]
  27. Fossa SD, Paus E, Lindegaard M, Newling DW. Prostate-specific antigen and other prognostic factors in patients with hormone-resistant prostatic cancer undergoing experimental treatment. Br J Urol 1992;69:175–9.[Medline]
  28. Petrylak DP, Scher HI, Li Z, Myers CE, Geller NL. Prognostic factors for survival of patients with bidimensionally measurable metastatic hormone-refractory prostatic cancer treated with single-agent chemotherapy. Cancer 1992;70:2870–8.[CrossRef][Medline]
  29. Kelly WK, Scher HI, Mazumdar M, Vlamis V, Schwartz M, Fossa SD. Prostate-specific antigen as a measure of disease outcome in metastatic hormone-refractory prostate cancer. J Clin Oncol 1993;11:607–15.[Abstract]
  30. Taplin ME, George DJ, Halabi S, et al. Prognostic significance of plasma chromogranin A levels in patients with hormone-refractory prostate cancer treated in Cancer and Leukemia Group B 9480 study. Urology 2005;66:386–91.[CrossRef][Medline]
  31. Small EJ, Meyer M, Marshall ME, et al. Suramin therapy for patients with symptomatic hormone-refractory prostate cancer: results of a randomized phase III trial comparing suramin plus hydrocortisone to placebo plus hydrocortisone. J Clin Oncol 2000;18:1440–50.[Abstract/Free Full Text]
  32. Armstrong AJ, Garrett-Mayer ES, Ou Yang Y, De wit R, Tannock I, Eisenberger M. Limitations of the current progression-free survival (PFS) definition in hormone-refractory prostate cancer (HRPC): benefit associated with continuation of docetaxel beyond TAX327 protocol-defined progression. ASCO Aastate Cancer Symposium 2007, abstract 223, p. 183.
  33. George DJ, Halabi S, Shepard TF, et al. Prognostic significance of plasma vascular endothelial growth factor levels in patients with hormone-refractory prostate cancer treated on Cancer and Leukemia Group B 9480. Clin Cancer Res 2001;7:1932–6.[Abstract/Free Full Text]
  34. George DJ, Halabi S, Shepard TF, et al. The prognostic significance of plasma interleukin-6 levels in patients with metastatic hormone-refractory prostate cancer: results from cancer and leukemia group B 9480. Clin Cancer Res 2005;11:1815–20.[Abstract/Free Full Text]
  35. Ghossein RA, Rosai J, Scher HI, et al. Prognostic significance of detection of prostate-specific antigen transcripts in the peripheral blood of patients with metastatic androgen-independent prostatic carcinoma. Urology 1997;50:100–5.[Medline]
  36. Halabi S, Small EJ, Hayes DF, Vogelzang NJ, Kantoff PW. Prognostic significance of reverse transcriptase polymerase chain reaction for prostate-specific antigen in metastatic prostate cancer: a nested study within CALGB 9583. J Clin Oncol 2003;21:490–5.[Abstract/Free Full Text]
  37. Moreno JG, Miller MC, Gross S, Allard WJ, Gomella LG, Terstappen LW. Circulating tumor cells predict survival in patients with metastatic prostate cancer. Urology 2005;65:713–8.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 JCOHome page
S. Halabi, N. J. Vogelzang, S.-S. Ou, K. Owzar, L. Archer, and E. J. Small
Progression-Free Survival as a Predictor of Overall Survival in Men With Castrate-Resistant Prostate Cancer
J. Clin. Oncol., June 10, 2009; 27(17): 2766 - 2771.
[Abstract] [Full Text] [PDF]

Home page
 Endocr Relat CancerHome page
C. Festuccia, G. L. Gravina, A. M. D'Alessandro, P. Muzi, D. Millimaggi, V. Dolo, E. Ricevuto, C. Vicentini, and M. Bologna
Azacitidine improves antitumor effects of docetaxel and cisplatin in aggressive prostate cancer models
Endocr. Relat. Cancer, June 1, 2009; 16(2): 401 - 413.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
A. J. Armstrong, P. Creel, J. Turnbull, C. Moore, T. A. Jaffe, S. Haley, W. Petros, S. Yenser, J. P. Gockerman, D. Sleep, et al.
A Phase I-II Study of Docetaxel and Atrasentan in Men with Castration-Resistant Metastatic Prostate Cancer
Clin. Cancer Res., October 1, 2008; 14(19): 6270 - 6276.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
G. Sonpavde, M. T. Fleming, T. E. Hutson, and M. D. Galsky
Trial Design for Metastatic Castration-Resistant Prostate Cancer
J. Clin. Oncol., July 20, 2008; 26(21): 3647 - 3648.
[Full Text] [PDF]

Home page
 JCOHome page
H. I. Scher, S. Halabi, I. F. Tannock, M. Morris, C. N. Sternberg, M. A. Carducci, M. A. Eisenberger, C. Higano, G. J. Bubley, R. Dreicer, et al.
In Reply
J. Clin. Oncol., July 20, 2008; 26(21): 3648 - 3649.
[Full Text] [PDF]

Home page
 JCOHome page
S. Halabi, N. J. Vogelzang, A. B. Kornblith, S.-S. Ou, P. W. Kantoff, N. A. Dawson, and E. J. Small
Pain Predicts Overall Survival in Men With Metastatic Castration-Refractory Prostate Cancer
J. Clin. Oncol., May 20, 2008; 26(15): 2544 - 2549.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
C. Nabhan
Is Chemotherapy the Standard for Asymptomatic Androgen-Independent Prostate Cancer?
J. Clin. Oncol., May 10, 2008; 26(14): 2413 - 2414.
[Full Text] [PDF]

Home page
 JNCI J Natl Cancer InstHome page
A. J. Armstrong, E. L. Garrett-Mayer, and M. Eisenberger
Re: Adaptive Therapy for Androgen-Independent Prostate Cancer: A Randomized Selection Trial of Four Regimens
J Natl Cancer Inst, May 7, 2008; 100(9): 681 - 682.
[Full Text] [PDF]

Home page
 JNCI J Natl Cancer InstHome page
R. E. Millikan, C. J. Logothetis, and P. F. Thall
Response:Re: Adaptive Therapy for Androgen-Independent Prostate Cancer: A Randomized Selection Trial of Four Regimens
J Natl Cancer Inst, May 7, 2008; 100(9): 682 - 683.
[Full Text] [PDF]

Home page
 JCOHome page
H. I. Scher, S. Halabi, I. Tannock, M. Morris, C. N. Sternberg, M. A. Carducci, M. A. Eisenberger, C. Higano, G. J. Bubley, R. Dreicer, et al.
Design and End Points of Clinical Trials for Patients With Progressive Prostate Cancer and Castrate Levels of Testosterone: Recommendations of the Prostate Cancer Clinical Trials Working Group
J. Clin. Oncol., March 1, 2008; 26(7): 1148 - 1159.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
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 Armstrong, A. J.
Right arrow Articles by Eisenberger, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Armstrong, A. J.
Right arrow Articles by Eisenberger, M.

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