Natural History of Rising Serum Prostate-Specific Antigen in Men With Castrate Nonmetastatic Prostate Cancer
http://www.100md.com
《临床肿瘤学》
the Massachusetts General Hospital, Boston, MA
University of California at Los Angeles, Los Angeles
University of California at San Francisco, San Francisco
Western Clinical Research, Torrance, CA
University of Montreal, Montreal, Canada
University of Maryland, Baltimore, MD
S Florida Medical Research, Aventura, FL
Christchurch Hospital, Christchurch, New Zealand
Peter MacCallum Cancer Institute, Melbourne, Australia
Free University of Brussels, Brussels, Belgium
Novartis Oncology, East Hanover, NJ
University of Waterloo, Waterloo, Canada
University of Washington, Seattle, WA
ABSTRACT
PURPOSE: To describe the natural history of nonmetastatic prostate cancer and rising prostate-specific antigen (PSA) despite androgen deprivation therapy.
PATIENTS AND METHODS:: The 201 patients in this report were the placebo control group from an aborted randomized controlled trial to evaluate the effects of zoledronic acid on time to first bone metastasis in men with prostate cancer, no bone metastases, and rising PSA despite androgen deprivation therapy. Relationships between baseline covariates and clinical outcomes were assessed by Cox proportional hazard analyses. Covariates in the model were baseline PSA, Gleason sum, history of bilateral orchiectomies, regional lymph node metastases at diagnosis, prior prostatectomy, time from androgen deprivation therapy to random assignment, time from diagnosis to random assignment, and PSA velocity.
RESULTS: At 2 years, 33% of patients had developed bone metastases. Median bone metastasis–free survival was 30 months. Median time to first bone metastases and overall survival were not reached. Baseline PSA level greater than 10 ng/mL (relative risk, 3.18; 95% CI, 1.74 to 5.80; P < .001) and PSA velocity (4.34 for each 0.01 increase in PSA velocity; 95% CI, 2.30 to 8.21; P < .001) independently predicted shorter time to first bone metastasis. Baseline PSA and PSA velocity also independently predicted overall survival and metastasis-free survival. Other covariates did not consistently predict clinical outcomes.
CONCLUSION: Men with nonmetastatic prostate cancer and rising PSA despite androgen deprivation therapy have a relatively indolent natural history. Baseline PSA and PSA velocity independently predict time to first bone metastasis and survival.
INTRODUCTION
Patterns of diagnosis and treatment for prostate cancer have dramatically changed over the past decade. Prostate cancer screening with serum prostate-specific antigen (PSA) has been accompanied by a dramatic stage migration and now less than 20% of men in the United States have radiographic evidence of metastases at initial diagnosis. 1 Based on evidence that early androgen deprivation therapy improves outcomes in certain clinical settings, many men with nonmetastatic prostate cancer are treated with gonadotropin-releasing hormone agonists. Early primary androgen deprivation therapy improves survival for men with locally advanced prostate cancer. 2 Adjuvant androgen deprivation therapy improves survival for men with locally advanced prostate cancer treated with radiation therapy, 3 and men with lymph node-positive prostate cancer treated with radical prostatectomy and pelvic lymphadenectomy. 4 In addition, gonadotropin-releasing hormone agonists are commonly administered to men with rising PSA as the only indication of disease recurrence after surgery or radiation therapy for early-stage prostate cancer.
Men with metastatic prostate cancer and disease progression despite androgen deprivation therapy have a poor prognosis. In two recent randomized controlled trials for example, the median survival was only 16 to 18 months for men with progressive castrate metastatic prostate cancer. 5, 6 Recent changes in patterns of diagnosis and treatment have dramatically increased the number of men receiving gonadotropin-releasing hormone agonists for nonmetastatic prostate cancer. Although most of these men will eventually experience rising PSA despite castrate levels of testosterone, little is known about the natural history of men with androgen-independent prostate cancer and no radiographic evidence of metastases.
We conducted a randomized, double-blind, placebo-controlled study to evaluate the effects of zoledronic acid on time to first bone metastasis in men with prostate cancer, no radiographic evidence of bone metastases, and rising PSA despite androgen deprivation therapy. The study was terminated before completion of accrual after interim analyses demonstrated that the observed event rate was lower than expected. The low event rate and early termination of the study preclude evaluation of efficacy. We now report the outcomes for the 201 placebo-treated patients to describe the natural history of rising PSA in men with castrate nonmetastatic prostate cancer. The results of these analyses may facilitate the management of men with prostate cancer and inform the design of future clinical trials in this setting.
PATIENTS AND METHODS
A randomized, double-blind, placebo-controlled study was conducted to evaluate the effects of zoledronic acid on time to first bone metastases in men with prostate cancer, no radiographic evidence of bone metastases, and rising PSA despite androgen deprivation therapy (progressive castrate nonmetastatic prostate cancer). Three hundred ninety-eight patients were enrolled between September 1999 and September 2002. On December 6, 2001, the Data and Safety Monitoring Board placed the study on hold before reaching target accrual of 991 patients because the observed event rate was lower than expected. On September 27, 2002, the study was terminated.
Patients
All patients had prostate cancer, no radiographic evidence of metastases, and PSA progression, despite androgen deprivation therapy (bilateral orchiectomies or treatment with a gonadotropin-releasing hormone agonist). PSA progression was defined as three consecutive rises in serum PSA (measured at least 2 weeks apart), initial PSA rise within 10 months of study entry, and last PSA 150% nadir value. Patients with an intact prostate had a PSA level greater than 4 ng/mL. Patients with a history of prostatectomy had a PSA level greater than 0.8 ng/mL.
We excluded patients with disease-related symptoms, Karnofsky performance status less than 90, life expectancy less than 6 months, second malignancy diagnosed within 5 years, prior chemotherapy or other systemic anticancer therapy other than a gonadotropin releasing-hormone agonist or nonsteroidal antiandrogen, radiation therapy within 6 weeks, or investigational drugs within 28 days. We also excluded patients with serum total testosterone 30 ng/dL, WBC count less than 3.0 x 109 cells/L, absolute neutrophil count less than 1.5 x 109 cells/L, hemoglobin less than 8.0 g/dL, platelet count less than 75 x 109/L, liver function tests more than 2.5x the upper limit of normal, or serum creatinine more than 1.5x the upper limit of normal.
The institutional review board at each participating institution approved the study. All patients gave written informed consent.
Study Design
Patients were randomly assigned to receive either zoledronic acid (Zometa; Novartis Pharama AG, Basel, Switzerland; Novartis Pharmaceuticals Corp, East Hanover, NJ) or placebo intravenously every 4 weeks for 49 treatments. All patients were prescribed supplemental calcium (500 mg/d) and vitamin D (400 to 500 U/d). Men receiving androgen deprivation therapy with a gonadotropin-releasing hormone agonist continued gonadotropin-releasing hormone agonist treatment throughout the study. Concurrent treatment with other bisphosphonates was prohibited. Treatment with secondary hormonal therapy or chemotherapy was at the discretion of the treating physician.
Patients were evaluated monthly for 48 months. Symptom assessment and physical examination were performed at each visit. Serum PSA was measured at baseline and then every 4 months. Radionuclide bone scans were performed every 4 months. Additional radionuclide bone scans were performed to evaluate patients with symptoms suggestive of bone metastases. Additional bone scans were not performed based on changes in serum PSA. For all regions of new or worsening uptake by bone scan, plain radiographs, magnetic resonance imaging, or computed tomography were performed to confirm or exclude the presence of bone metastases.
Outcomes
The primary outcomes for these analyses were time to first bone metastasis, overall survival, and bone metastasis–free survival.
Statistical Analyses
The time to first bone metastasis was defined as the time to first positive bone scan/radiograph. Patients not observed to develop bone metastases were censored at the end of the blinded follow-up phase. Standard competing risks methodology justifies Cox regression analyses based on these times as cause-specific (bone metastases) failure time models. 7 Univariate Cox regression models were fit for the time to first bone metastasis using each covariate as a single explanatory variable. Next, multivariate models were fit using all explanatory variables to identify those that were independently predictive. Stepwise backward elimination was carried out to determine the simplest multivariate model that contained only explanatory variables that were independently significant and predictive of time to first bone metastasis.
The covariates considered in this and other Cox regression analyses included baseline PSA (> 10 ng/mL v 10 ng/mL), PSA velocity, prior prostatectomy (yes v no), Gleason sum ( 7 v < 7), bilateral orchiectomies (yes v no), regional lymph node, metastases at diagnosis (yes v no), time from androgen deprivation therapy to random assignment (> 2 years v 2 years), and time from diagnosis to random assignment (years). PSA velocity was defined as the slope of the regression line of the natural log PSA by time over the first 8 months of follow-up. Specifically, we carried out linear regression of log PSA values at baseline, 4 months, and 8 months, by time. Patients without at least two PSA assessments over the first 8 months were excluded from the PSA velocity analyses. If a baseline assessment was not available, the PSA value at the screening assessment was used. PSA doubling time (PSADT) was calculated by natural log of 2 (0.693) divided by the PSA velocity. 8 Cumulative incidence plots were created to characterize the proportion of patients having experienced at least one new bone metastasis.
Cox regression analyses were also conducted for death, where the times were taken as the number of days from random assignment to death, or the time from random assignment to last contact for those who were not observed to die, with the latter again being indicated as a right censoring time. Cox models for metastases-free survival were also fit in a similar way. Here, however, the censoring time was the minimum censoring time for bone metastases (end of the double-blind follow-up phase) and the censoring time for death (last contact). This was necessary since data on new bone metastases following the end of the double-blind phase were not available, and so patients were not strictly at risk for this component of the composite end point in this period. Kaplan-Meier estimates for the survival distribution and the metastases-free survival distribution we constructed for descriptive purposes.
In the Cox regression analyses described above, some clinical events (development of bone metastases, for example) may have occurred during the 8-month assessment of PSA velocity. To account for temporal ordering of PSA data and clinical events, additional Cox regression analyses were conducted. In this second set of analyses, patients were not considered to be at risk for the clinical events until month 8. Data from the first 8 months or PSA were used to compute the PSA velocity, but no events were counted during this period. While this excludes some events, it ensures a temporal ordering of the PSA data over the first 8 months, and the subsequent clinical events.
RESULTS
Patient Characteristics
Baseline characteristics are summarized in Table 1. Mean (± standard deviation) age was 73 ± 7 years. Thirty-four percent of men had a history of radical prostatectomy. Eighty-two percent initiated androgen deprivation therapy more than 2 years before study entry. Median serum PSA at study entry was 13.8 ng/mL (range, 0.9 to 630 ng/mL).
Time to First Bone Metastasis and Survival
Figure 1 provides a cumulative incidence plot of the proportion of subjects with at least one bone metastasis, and the Kaplan-Meier estimates of the proportion of subjects alive and the proportion of subjects with at least one metastasis or death, respectively. At 2 years, 33% of subjects had developed at least one bone metastasis, 21% had died, and 42% had experienced a bone metastasis or had died. Median bone metastasis–free survival was 907 days. Median overall survival was not reached.
Cox Proportional Hazard Analyses
The relationships between baseline characteristics and time to first bone metastases, overall survival, and metastasis-free survival were assessed using univariate and multivariate Cox proportional hazard analyses. Descriptive statistics for subject demographics and covariates used in the Cox regression models are provided in Table 1.
Time to First Bone Metastasis
In univariate analyses, baseline PSA level greater than 10 ng/mL was associated with shorter time to first bone metastases (relative risk [RR], 2.96; 95% CI, 1.63 to 5.38; P < .001; Table 2). High PSA velocity was also associated with shorter time to bone metastases (RR, 1.47 for each year increase in PSA velocity; 95% CI, 1.23 to 1.75; P < .001). Baseline PSA and PSA velocity remained statistically significant in the full and reduced multivariate Cox regression model, with RRS of 3.18 (95% CI, 1.74 to 5.80; P < .001) and 1.50 (95% CI, 1.26 to 1.78; P < .001), respectively (Table 2). In analyses in which subjects were not considered at risk until month 8, baseline PSA and PSA velocity were also significantly associated with shorter time to first bone metastasis (data not shown).
Survival
In univariate analyses, baseline PSA level greater than 10 ng/mL (RR, 3.10; 95% CI, 1.47 to 6.54; P = .003) and PSA velocity (RR, 1.38 for each year increase in PSA velocity; 95% CI, 1.14 to 1.68; P = .001) were associated significantly with shorter survival (Table 3). History of bilateral orchiectomies (RR, 0.44; 95% CI, 0.19 to 1.06; P = .07) and time from androgen deprivation therapy to randomization greater than 2 years (RR, 0.53; 95% CI, 0.25 to 1.12; P = .09) were associated with longer overall survival, though these were not statistically significant. In multivariate Cox regression analyses, only baseline PSA level greater than 10 ng/mL and PSA velocity independently predicted overall survival, with RRs of 3.19 (95% CI, 1.51 to 6.73; P = .002) and 1.39 (95% CI, 1.15 to 1.69; P < .001), respectively (Table 3).
Bone Metastasis–Free Survival
In univariate analyses, baseline PSA level greater than 10 ng/mL (RR, 2.95; 95% CI, 1.71 to 5.09; P < .001) and PSA velocity (RR, 1.44 for each year increase in PSA velocity; 95% CI, 1.23 to 1.70; P < .001) were significantly associated with shorter bone metastasis–free survival (Table 4). Gleason score greater than 7 had a RR of 1.62, though this was not statistically significant (95% CI, 0.96 to 2.75; P = .07). Multivariate Cox regression analyses demonstrated that only baseline PSA levels greater than 10 ng/mL and PSA velocity were independently predictive for bone metastasis–free survival, with RRs of 3.19 (95% CI, 1.84 to 5.53; P < .001) and 1.48 (95% CI, 1.25 to 1.74; P < .001), respectively. Similar results were observed when subjects were not considered at risk for clinical events until month eight (data not shown). Tertiles of PSA (< 7.7, 7.7 to 24, > 24 ng/mL) and PSADT (< 6.3, 6.3 to 18.8, > 18.8 months) were associated with different bone metastasis-free survival (P < .001; Fig 2).
DISCUSSION
We found that men with prostate cancer, no bone metastases, and rising PSA despite androgen deprivation therapy have a relatively indolent natural history. Only one third of men developed bone metastases at 2 years, and the median metastasis-free survival was approximately 30 months. Baseline PSA and PSA velocity independently predicted time to first bone metastasis, overall survival, and bone metastasis–free survival. No other covariate consistently predicted clinical outcomes.
To the best of our knowledge, this is the first analysis to prospectively evaluate the natural history of nonmetastatic prostate cancer and rising PSA despite androgen deprivation therapy. Notable features of the study design include confirmation of a castrate testosterone level at study entry, documentation of PSA progression with three serial rises, and radiographic screening to exclude men with bone metastases at study entry. In addition, bone scans were obtained every 4 months in order to minimize PSA bias in defining the primary study outcome of time to first bone metastases.
Elevation in serum PSA levels after surgery or radiation therapy for early-stage prostate cancer typically predates clinically or radiographically detectable metastatic disease by several years, without additional treatment. In a retrospective series of 315 men with a rising serum PSA after radical prostatectomy for example, the median time to first bone metastasis was 8 years from the time of PSA elevation. 8 Three years following initial postoperative PSA elevation, only 27% of patients had radiographically detectable metastases. The results of our study suggest that rates of disease progression remain low for men with "PSA only" prostate cancer, even after androgen deprivation therapy.
These analyses extend other evidence linking PSA velocity to clinical outcomes. In men with rising PSA after surgery or radiation therapy for early-stage disease, PSA velocity predicts time to metastases 8- 12 and survival. 13- 15 PSA velocity also predicts survival in men with hormone-refractory metastatic prostate cancer. 16
The role of bisphosphonates for prevention of bone metastases in men with prostate cancer remains undefined. Mason et al recently reported the preliminary results of Medical Research Council Pr04, a randomized controlled trial of clodronate to prevent symptomatic bone metastases in men with nonmetastatic prostate cancer. 17 The study included 508 men receiving standard treatment for clinical stage T2 to T4 prostate cancer with no evidence of bone metastases. Men were assigned randomly to either oral clodronate (2,080 mg daily) or placebo for 5 years. Most of the subjects received either external beam radiation therapy, external beam radiation therapy with hormone therapy, or primary hormone therapy as standard treatment. The primary end point was time to development of symptomatic bone metastases or prostate cancer death. There were no scheduled radiographic studies to diagnose new asymptomatic metastases. At a median follow-up of 7 years, there were no significant differences between the groups in either time to symptomatic bone metastases or survival.
The early termination of our study bars conclusions about the efficacy of zoledronic to prevent bone metastases in men with castrate nonmetastatic prostate cancer. The results of this study, however, inform the design of future clinical trials to prevent bone metastases in men with prostate cancer. The relatively indolent natural history of castrate nonmetastatic prostate cancer suggests that future clinical trials in this setting should be very large or should select subjects at particularly high risk in order to have adequate statistical power to detect a clinically meaningful treatment effect. The independent predictive value of PSA velocity suggests that PSA kinetics may provide an effective strategy to identify men at high risk for bone metastases or death.
Our study has potential limitations. The study allowed treatment with secondary hormonal therapy and/or chemotherapy at physician discretion. This part of the study design increases the generalizability of the results. Because detailed information about secondary hormonal therapy and chemotherapy was not collected, however, we cannot assess the impact of these salvage therapies on clinical outcomes. Less than one third of subjects had a Gleason sum greater than 7 at initial prostate cancer diagnosis. The relatively low proportion of men with a Gleason sum greater than 7 suggests that our study may have selected subjects with lower Gleason grades. Alternatively, men with a Gleason sum greater than 7 may be more likely to develop bone metastases early (at diagnosis or first PSA failure) and may be underrepresented in studies of men with a rising PSA as the only indication of disease progression. Notably, the proportion of men with high-grade tumors in our study (29%) is similar to the proportion of men with Gleason sum more than 7 in a study of men with rising PSA after prostatectomy (32%). 8 Lastly, our analyses evaluated a limited number of variables, and additional studies are needed to identify additional predictors of clinical outcomes in this setting.
In summary, men with castrate nonmetastatic prostate cancer and rising PSA have an indolent natural history, with a median metastasis-free survival of approximately 30 months. Baseline PSA and PSA velocity independently predict time to first bone metastasis, overall survival, and metastasis-free survival.
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Employment: Ronald Linnartz, Novartis Oncology; Ming Zheng, Novartis Oncology; Carsten Goessl, Novartis Oncology; Yong-Jiang Hei, Novartis Oncology. Stock Ownership: Ming Zheng, Novartis Oncology; Yong-Jiang Hei, Novartis Oncology. For a detailed description of these categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
Acknowledgment
We thank the following investigators who participated in the study: A. Avila (Santa Fe de Bogota, Colombia), A. Cajigas (Santa Fe de Bogota, Colombia), A. Torres (Mexico DF, Mexico), A. Malzyner (Sao Paulo, Brazil), B. Blank (Portland, OR), B. Aragao (Belo Horizonte, Brazil), C. Dzik (Sao Paulo, Brazil), C. Tosello (Sao Paulo, Brazil), C. McMurtry (Richmond, VA), D. Culkin (Oklahoma City, OK), D. Kim (Bay Shore, NY), D. Shevrin (Evanston, IL), D. Shevrin (Evanston, IL), D. Fiske (Stuart, FL), D. Bell (Halifax, Nova Scotia, Canada), D. Vaughn (Philadelphia, PA), D. Gleason (Tucson, AZ), D. Locke (Summerfield, FL), E.B. Alentorn (Vic, Spain), F. Boeminghaus (Neuss, Germany), F.H. de Mendoza (Lima, Peru), F. Desiderio (Rimini, Italy), F. Yunus (Memphis, TN), G. Comeri (Como, Italy), G. Greene (Little Rock, AK), H. Carloss (Paducah, KY), J. Lechner (Olympia, WA), J. McMurray (Hunstville, AL), J. Breza (Bratislava, Slovak Republic), J. Morales (San Juan, Puerto Rico), J. Metts III (Charleston, SC), J. Kaufman (Aurora, IL), J. Forrest (Tulsa, OK), J. Trachtenberg (Toronto, Ontario, Canada), J.C. Olmo (Granada, Spain), J. Camps (Tallahassee, FL), J. Chin (London, Ontario, Canada), K. Slawin (Houston, TX), L. Valansky (Kosice, Slovak Republic), L. Klotz (Toronto, Ontario, Canada), L. Kalman (Miami, FL), L. Reyno (Halifax, Nova Scotia, Canada), L. Balducci (Tampa, FL), L. Lacombe (Quebec, Montreal, Canada), L. Baez (San Juan, Puerto Rico), M. Colombel (Lyon, France), M. Ratner (Rockville, MD), M. Bondhus (Pinecrest, FL), M. Boyer (Concord, NH), M. McCleod (Ft Meyers, FL), M. Fodor (Santiago, Chile), M. Wiatrak (Milwaukee, WI), M. Spaulding (Buffalo, NY), N. Clarke (Manchester, UK), O. Bertetto (Torino, Italy), P. Donovan (Scottsdale, AZ), P. Miller (Lakewood, CO), P. Clingan (Wollongong, New South Wales, Australia), P. Nickers (Liege, Belgium), R. Bukowski (Cleveland, OH), A. Corica (Mendoza, Argentina), A. Dalul (Sante Fe, NM), A. Mandressi (Busto Arsizio, Italy), A. Muzio (Capital Federal, Argentina), A. Villaronga (Rem de Escalada, Argentina), A. Macan (Martinez, France), A. Comandone (Torino, Italy), A. Vasconcelos (Salvador, Brazil), A. DiStefano (Arlington, VA), B. Mellado (Barcelona, Spain), B. Paule (Creteil, France), B. Chern (Redcliff, Australia), B. Corser (Cincinnati, OH), B. Woodworth (Columbus, OH), B. Aragao (Belo Horizonte, Brazil), B. Hodge (Cumming, GA), C. Nolasco (Capital Federal, Argentina), C. Cordova (Hermosillo, Spain), C. Theodossiou (New Orleans, LA), D. Campos (San Isidro, Argentina), D. Rukstalis (Philadelphia, PA), D. Toledo (Santa Fe de Bogota, Colombia), D. Varcasia (Capital Federal, Argentina), D. Dearnaley (Surrey, UK), D. Gillatt (Bristol, UK), D. Joseph (Nedlands, Australia), D. Sahasharabude (Rochester, NY), D. Jocham (Lübeck, Germany), D. Leusmann (Kln, Germany), D. Riquet (Valenciennes, France), F. Boccardo (Genova, Italy), F.J. Marx (Kln, Germany), F. Hamdy (Sheffield, UK), G. Frenkel (Dallas, TX), G. Manikhas (St Petersburg, Russia), G. Porcile (Alba, Italy), H.U. Eickenberg (Bielefeld, Germany), H.J. Melchior (Kassel, Germany), I. Roussakov (Moscow, Russia), I. Klimberg (Ocala, FL), J.M. Wolff (Rostock, Germany), J. Kliment (Martin, Slovak Republic), J. Vinholes (Porto Alegre, Brazil), J. Gingrich (Memphis, TN), J. Picus (St Louis, MO), J. Grygiel (Darlinghurst, Australia), J. Hulbert (Minneapolis, MN), J. Ramsey (Indianapolis, IN), J. Esterellas (Rosario, Argentina), J.M. De Marco (Buenos Aires, Argentina), J. Premoli (Rosario, Argentina), J. Kish (Detroit, MI), K. Stockamp (Ludwigshafen, Germany), K.F. Klippel (Celle, Germany), Ken Pittman (Adelaide, SA), K. Clarke (Wodonga, Australia), K. Miller (Berlin, Germany), L. Weissbach (Berlin, Germany), L. Dogliotti (Orbassano, Italy), L. Montes de Oca (Capital Federal, Argentina), M. Siegsmund (Mannheim, Germany), M. Merlano (Cuneo, Italy), M. Gleave (Vancouver, British Columbia, Canada), M. Boyer (Camperdown, New South Wales, Australia), M. Lill (Los Angeles, CA), M. Carmel (Sherbrooke, Canada), M. Lopez (Capital Federal, Argentina), M. Moises (Tucuman, Argentina), N. Fredotovich (Capital Federal, Argentina), OA Dr Zimmermann (Greifswald, Germany), O. Damia (Capital Federal, Argentina), O.R. Tutor (Mendoza, Argentina), O. Aren (Uruguay), P. Benedetto (Miami, FL), P. Keane (Belfast, Ireland), P. McInerney (Plymouth, MA), P. Schelhammer (Norfolk, VA), P. Pizao (Campinas, Brazil), P. Venner (Edmonton, Alberta, Canada), P. Grise (Rouen, France), P.F. Conte (Pisa, Italy), P. Major (Hamilton, Ontario, Canada), P. Teillac (Paris, France), P. Lara (Sacramento, CA), R. Persad (Bristol, UK), R. Gerson (Mexico DF CP, Mexico), R. Anchelerguez (Mendoza, Argentina), R. Wainstein (Villa Sarmiento, Argentina), R. Costabile (Tacoma, WA), R. Kahnoski (Grand Rapids, MI), R. Mansour (Shreveport, LA), R. Middleton (Salt Lake City, UT), R. Feldman (Waterbury, CT), R. Smith (Columbia, SC), R. Quintana (Capital Federal, Argentina), R. Vega (Colonia La Raza, Mexico), R. Dmochowski (Rockville, MD), R. Carroll (Scarborough, UK), R. Lewis (Augusta, GA), R. Yanagihara (Gilroy, CA), R. Shah (Lake Bluff, IL), Roy MacKintosh (Reno, NV), R. Bengio (Cordoba, Argentina), S. Ernst (Calgary, Alberta, Canada), S. Metrebian (Cordoba, Argentina), S. Tchekmedyian (Long Beach, CA), S. Childs (Cheyenne, WY), S.C. Mueller (Bonn, Germany), S. Rosenberg (Des Moines, IA), T. Beck (Springdale, AR), T. Szlaby (Husum, Germany), T. Hlavinka (San Antonio, TX), T. Roberts (Newcastle, UK), U.W. Tunn (Offenbach, Germany), V. Benavente (Lima, Peru), V. Gorbunova (Moscow, Russia), W. Jacobsen (Kiel, Russia), W. Bohnert (Phoenix, AZ), W. Monnig (Cincinnati, OH), W. Moseley (San Diego, CA), W. Schubach (Seattle, WA), Y. Rahim (Toronto, Ontario, Canada), Z. Wajsman (Gainesville, FL). Dr. Smith is supported by the John and Claire Bertucci Center for Genitourinary Malignancies at Massachusetts General Hospital.
NOTES
Supported by Novartis Oncology. M.R.S. is supported by the John and Claire Bertucci Center for Genitourinary Malignancies at Massachusetts General Hospital.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Jemal A, Tiwari RC, Murray T, et al: Cancer statistics, 2004. CA Cancer J Clin 54:8-29, 2004
Anonymous: Immediate versus deferred treatment for advanced prostatic cancer: Initial results of the Medical Research Council Trial—The Medical Research Council Prostate Cancer Working Party Investigators Group. Br J Urol 79: 235-246, 1997
Bolla M, Gonzalez D, Warde P, et al: Improved survival in patients with locally advanced prostate cancer treated with radiotherapy and goserelin. N Engl J Med 337:295-300, 1997
Messing EM, Manola J, Sarosdy M, et al: Immediate hormonal therapy compared with observation after radical prostatectomy and pelvic lymphadenectomy in men with node-positive prostate cancer. N Engl J Med 341:1781-1788, 1999
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 351:1513-1520, 2004
Tannock IF, de Wit R, Berry WR, et al: Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 351:1502-1512, 2004
Kalbfleisch J, Prentice R: The Statistical Analysis of Failure Time Data (ed 2). Indianapolis, IN, John Wiley and Sons, 2002
Pound CR, Partin AW, Eisenberger MA, et al: Natural history of progression after PSA elevation following radical prostatectomy. JAMA 281:1591-1597, 1999
Lee WR, Hanks GE, Hanlon A: Increasing prostate-specific antigen profile following definitive radiation therapy for localized prostate cancer: Clinical observations. J Clin Oncol 15:230-238, 1997
Patel A, Dorey F, Franklin J, et al: Recurrence patterns after radical retropubic prostatectomy: Clinical usefulness of prostate specific antigen doubling times and log slope prostate specific antigen. J Urol 158:1441-1445, 1997
Roberts SG, Blute ML, Bergstralh EJ, et al: PSA doubling time as a predictor of clinical progression after biochemical failure following radical prostatectomy for prostate cancer. Mayo Clin Proc 76:576-581, 2001
Sartor CI, Strawderman MH, Lin XH, et al: Rate of PSA rise predicts metastatic versus local recurrence after definitive radiotherapy. Int J Radiat Oncol Biol Phys 38:941-947, 1997
Sandler HM, Dunn RL, McLaughlin PW, et al: Overall survival after prostate-specific-antigen-detected recurrence following conformal radiation therapy. Int J Radiat Oncol Biol Phys 48:629-633, 2000
D'Amico AV, Cote K, Loffredo M, et al: Determinants of prostate cancer-specific survival after radiation therapy for patients with clinically localized prostate cancer. J Clin Oncol 20:4567-4573, 2002
D'Amico AV, Moul JW, Carroll PR, et al: Surrogate end point for prostate cancer-specific mortality after radical prostatectomy or radiation therapy. J Natl Cancer Inst 95:1376-1383, 2003
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 83:1989-1994, 1998
Mason MD, Collaborators MP: Development of bone metastases from prostate cancer: First results of the MRC PR04 trial (ISCRTN 61384873). Proc Am Soc Clin Oncol 22: 384S, 2004 (abstr 4511)(Matthew R. Smith, Fairooz)
University of California at Los Angeles, Los Angeles
University of California at San Francisco, San Francisco
Western Clinical Research, Torrance, CA
University of Montreal, Montreal, Canada
University of Maryland, Baltimore, MD
S Florida Medical Research, Aventura, FL
Christchurch Hospital, Christchurch, New Zealand
Peter MacCallum Cancer Institute, Melbourne, Australia
Free University of Brussels, Brussels, Belgium
Novartis Oncology, East Hanover, NJ
University of Waterloo, Waterloo, Canada
University of Washington, Seattle, WA
ABSTRACT
PURPOSE: To describe the natural history of nonmetastatic prostate cancer and rising prostate-specific antigen (PSA) despite androgen deprivation therapy.
PATIENTS AND METHODS:: The 201 patients in this report were the placebo control group from an aborted randomized controlled trial to evaluate the effects of zoledronic acid on time to first bone metastasis in men with prostate cancer, no bone metastases, and rising PSA despite androgen deprivation therapy. Relationships between baseline covariates and clinical outcomes were assessed by Cox proportional hazard analyses. Covariates in the model were baseline PSA, Gleason sum, history of bilateral orchiectomies, regional lymph node metastases at diagnosis, prior prostatectomy, time from androgen deprivation therapy to random assignment, time from diagnosis to random assignment, and PSA velocity.
RESULTS: At 2 years, 33% of patients had developed bone metastases. Median bone metastasis–free survival was 30 months. Median time to first bone metastases and overall survival were not reached. Baseline PSA level greater than 10 ng/mL (relative risk, 3.18; 95% CI, 1.74 to 5.80; P < .001) and PSA velocity (4.34 for each 0.01 increase in PSA velocity; 95% CI, 2.30 to 8.21; P < .001) independently predicted shorter time to first bone metastasis. Baseline PSA and PSA velocity also independently predicted overall survival and metastasis-free survival. Other covariates did not consistently predict clinical outcomes.
CONCLUSION: Men with nonmetastatic prostate cancer and rising PSA despite androgen deprivation therapy have a relatively indolent natural history. Baseline PSA and PSA velocity independently predict time to first bone metastasis and survival.
INTRODUCTION
Patterns of diagnosis and treatment for prostate cancer have dramatically changed over the past decade. Prostate cancer screening with serum prostate-specific antigen (PSA) has been accompanied by a dramatic stage migration and now less than 20% of men in the United States have radiographic evidence of metastases at initial diagnosis. 1 Based on evidence that early androgen deprivation therapy improves outcomes in certain clinical settings, many men with nonmetastatic prostate cancer are treated with gonadotropin-releasing hormone agonists. Early primary androgen deprivation therapy improves survival for men with locally advanced prostate cancer. 2 Adjuvant androgen deprivation therapy improves survival for men with locally advanced prostate cancer treated with radiation therapy, 3 and men with lymph node-positive prostate cancer treated with radical prostatectomy and pelvic lymphadenectomy. 4 In addition, gonadotropin-releasing hormone agonists are commonly administered to men with rising PSA as the only indication of disease recurrence after surgery or radiation therapy for early-stage prostate cancer.
Men with metastatic prostate cancer and disease progression despite androgen deprivation therapy have a poor prognosis. In two recent randomized controlled trials for example, the median survival was only 16 to 18 months for men with progressive castrate metastatic prostate cancer. 5, 6 Recent changes in patterns of diagnosis and treatment have dramatically increased the number of men receiving gonadotropin-releasing hormone agonists for nonmetastatic prostate cancer. Although most of these men will eventually experience rising PSA despite castrate levels of testosterone, little is known about the natural history of men with androgen-independent prostate cancer and no radiographic evidence of metastases.
We conducted a randomized, double-blind, placebo-controlled study to evaluate the effects of zoledronic acid on time to first bone metastasis in men with prostate cancer, no radiographic evidence of bone metastases, and rising PSA despite androgen deprivation therapy. The study was terminated before completion of accrual after interim analyses demonstrated that the observed event rate was lower than expected. The low event rate and early termination of the study preclude evaluation of efficacy. We now report the outcomes for the 201 placebo-treated patients to describe the natural history of rising PSA in men with castrate nonmetastatic prostate cancer. The results of these analyses may facilitate the management of men with prostate cancer and inform the design of future clinical trials in this setting.
PATIENTS AND METHODS
A randomized, double-blind, placebo-controlled study was conducted to evaluate the effects of zoledronic acid on time to first bone metastases in men with prostate cancer, no radiographic evidence of bone metastases, and rising PSA despite androgen deprivation therapy (progressive castrate nonmetastatic prostate cancer). Three hundred ninety-eight patients were enrolled between September 1999 and September 2002. On December 6, 2001, the Data and Safety Monitoring Board placed the study on hold before reaching target accrual of 991 patients because the observed event rate was lower than expected. On September 27, 2002, the study was terminated.
Patients
All patients had prostate cancer, no radiographic evidence of metastases, and PSA progression, despite androgen deprivation therapy (bilateral orchiectomies or treatment with a gonadotropin-releasing hormone agonist). PSA progression was defined as three consecutive rises in serum PSA (measured at least 2 weeks apart), initial PSA rise within 10 months of study entry, and last PSA 150% nadir value. Patients with an intact prostate had a PSA level greater than 4 ng/mL. Patients with a history of prostatectomy had a PSA level greater than 0.8 ng/mL.
We excluded patients with disease-related symptoms, Karnofsky performance status less than 90, life expectancy less than 6 months, second malignancy diagnosed within 5 years, prior chemotherapy or other systemic anticancer therapy other than a gonadotropin releasing-hormone agonist or nonsteroidal antiandrogen, radiation therapy within 6 weeks, or investigational drugs within 28 days. We also excluded patients with serum total testosterone 30 ng/dL, WBC count less than 3.0 x 109 cells/L, absolute neutrophil count less than 1.5 x 109 cells/L, hemoglobin less than 8.0 g/dL, platelet count less than 75 x 109/L, liver function tests more than 2.5x the upper limit of normal, or serum creatinine more than 1.5x the upper limit of normal.
The institutional review board at each participating institution approved the study. All patients gave written informed consent.
Study Design
Patients were randomly assigned to receive either zoledronic acid (Zometa; Novartis Pharama AG, Basel, Switzerland; Novartis Pharmaceuticals Corp, East Hanover, NJ) or placebo intravenously every 4 weeks for 49 treatments. All patients were prescribed supplemental calcium (500 mg/d) and vitamin D (400 to 500 U/d). Men receiving androgen deprivation therapy with a gonadotropin-releasing hormone agonist continued gonadotropin-releasing hormone agonist treatment throughout the study. Concurrent treatment with other bisphosphonates was prohibited. Treatment with secondary hormonal therapy or chemotherapy was at the discretion of the treating physician.
Patients were evaluated monthly for 48 months. Symptom assessment and physical examination were performed at each visit. Serum PSA was measured at baseline and then every 4 months. Radionuclide bone scans were performed every 4 months. Additional radionuclide bone scans were performed to evaluate patients with symptoms suggestive of bone metastases. Additional bone scans were not performed based on changes in serum PSA. For all regions of new or worsening uptake by bone scan, plain radiographs, magnetic resonance imaging, or computed tomography were performed to confirm or exclude the presence of bone metastases.
Outcomes
The primary outcomes for these analyses were time to first bone metastasis, overall survival, and bone metastasis–free survival.
Statistical Analyses
The time to first bone metastasis was defined as the time to first positive bone scan/radiograph. Patients not observed to develop bone metastases were censored at the end of the blinded follow-up phase. Standard competing risks methodology justifies Cox regression analyses based on these times as cause-specific (bone metastases) failure time models. 7 Univariate Cox regression models were fit for the time to first bone metastasis using each covariate as a single explanatory variable. Next, multivariate models were fit using all explanatory variables to identify those that were independently predictive. Stepwise backward elimination was carried out to determine the simplest multivariate model that contained only explanatory variables that were independently significant and predictive of time to first bone metastasis.
The covariates considered in this and other Cox regression analyses included baseline PSA (> 10 ng/mL v 10 ng/mL), PSA velocity, prior prostatectomy (yes v no), Gleason sum ( 7 v < 7), bilateral orchiectomies (yes v no), regional lymph node, metastases at diagnosis (yes v no), time from androgen deprivation therapy to random assignment (> 2 years v 2 years), and time from diagnosis to random assignment (years). PSA velocity was defined as the slope of the regression line of the natural log PSA by time over the first 8 months of follow-up. Specifically, we carried out linear regression of log PSA values at baseline, 4 months, and 8 months, by time. Patients without at least two PSA assessments over the first 8 months were excluded from the PSA velocity analyses. If a baseline assessment was not available, the PSA value at the screening assessment was used. PSA doubling time (PSADT) was calculated by natural log of 2 (0.693) divided by the PSA velocity. 8 Cumulative incidence plots were created to characterize the proportion of patients having experienced at least one new bone metastasis.
Cox regression analyses were also conducted for death, where the times were taken as the number of days from random assignment to death, or the time from random assignment to last contact for those who were not observed to die, with the latter again being indicated as a right censoring time. Cox models for metastases-free survival were also fit in a similar way. Here, however, the censoring time was the minimum censoring time for bone metastases (end of the double-blind follow-up phase) and the censoring time for death (last contact). This was necessary since data on new bone metastases following the end of the double-blind phase were not available, and so patients were not strictly at risk for this component of the composite end point in this period. Kaplan-Meier estimates for the survival distribution and the metastases-free survival distribution we constructed for descriptive purposes.
In the Cox regression analyses described above, some clinical events (development of bone metastases, for example) may have occurred during the 8-month assessment of PSA velocity. To account for temporal ordering of PSA data and clinical events, additional Cox regression analyses were conducted. In this second set of analyses, patients were not considered to be at risk for the clinical events until month 8. Data from the first 8 months or PSA were used to compute the PSA velocity, but no events were counted during this period. While this excludes some events, it ensures a temporal ordering of the PSA data over the first 8 months, and the subsequent clinical events.
RESULTS
Patient Characteristics
Baseline characteristics are summarized in Table 1. Mean (± standard deviation) age was 73 ± 7 years. Thirty-four percent of men had a history of radical prostatectomy. Eighty-two percent initiated androgen deprivation therapy more than 2 years before study entry. Median serum PSA at study entry was 13.8 ng/mL (range, 0.9 to 630 ng/mL).
Time to First Bone Metastasis and Survival
Figure 1 provides a cumulative incidence plot of the proportion of subjects with at least one bone metastasis, and the Kaplan-Meier estimates of the proportion of subjects alive and the proportion of subjects with at least one metastasis or death, respectively. At 2 years, 33% of subjects had developed at least one bone metastasis, 21% had died, and 42% had experienced a bone metastasis or had died. Median bone metastasis–free survival was 907 days. Median overall survival was not reached.
Cox Proportional Hazard Analyses
The relationships between baseline characteristics and time to first bone metastases, overall survival, and metastasis-free survival were assessed using univariate and multivariate Cox proportional hazard analyses. Descriptive statistics for subject demographics and covariates used in the Cox regression models are provided in Table 1.
Time to First Bone Metastasis
In univariate analyses, baseline PSA level greater than 10 ng/mL was associated with shorter time to first bone metastases (relative risk [RR], 2.96; 95% CI, 1.63 to 5.38; P < .001; Table 2). High PSA velocity was also associated with shorter time to bone metastases (RR, 1.47 for each year increase in PSA velocity; 95% CI, 1.23 to 1.75; P < .001). Baseline PSA and PSA velocity remained statistically significant in the full and reduced multivariate Cox regression model, with RRS of 3.18 (95% CI, 1.74 to 5.80; P < .001) and 1.50 (95% CI, 1.26 to 1.78; P < .001), respectively (Table 2). In analyses in which subjects were not considered at risk until month 8, baseline PSA and PSA velocity were also significantly associated with shorter time to first bone metastasis (data not shown).
Survival
In univariate analyses, baseline PSA level greater than 10 ng/mL (RR, 3.10; 95% CI, 1.47 to 6.54; P = .003) and PSA velocity (RR, 1.38 for each year increase in PSA velocity; 95% CI, 1.14 to 1.68; P = .001) were associated significantly with shorter survival (Table 3). History of bilateral orchiectomies (RR, 0.44; 95% CI, 0.19 to 1.06; P = .07) and time from androgen deprivation therapy to randomization greater than 2 years (RR, 0.53; 95% CI, 0.25 to 1.12; P = .09) were associated with longer overall survival, though these were not statistically significant. In multivariate Cox regression analyses, only baseline PSA level greater than 10 ng/mL and PSA velocity independently predicted overall survival, with RRs of 3.19 (95% CI, 1.51 to 6.73; P = .002) and 1.39 (95% CI, 1.15 to 1.69; P < .001), respectively (Table 3).
Bone Metastasis–Free Survival
In univariate analyses, baseline PSA level greater than 10 ng/mL (RR, 2.95; 95% CI, 1.71 to 5.09; P < .001) and PSA velocity (RR, 1.44 for each year increase in PSA velocity; 95% CI, 1.23 to 1.70; P < .001) were significantly associated with shorter bone metastasis–free survival (Table 4). Gleason score greater than 7 had a RR of 1.62, though this was not statistically significant (95% CI, 0.96 to 2.75; P = .07). Multivariate Cox regression analyses demonstrated that only baseline PSA levels greater than 10 ng/mL and PSA velocity were independently predictive for bone metastasis–free survival, with RRs of 3.19 (95% CI, 1.84 to 5.53; P < .001) and 1.48 (95% CI, 1.25 to 1.74; P < .001), respectively. Similar results were observed when subjects were not considered at risk for clinical events until month eight (data not shown). Tertiles of PSA (< 7.7, 7.7 to 24, > 24 ng/mL) and PSADT (< 6.3, 6.3 to 18.8, > 18.8 months) were associated with different bone metastasis-free survival (P < .001; Fig 2).
DISCUSSION
We found that men with prostate cancer, no bone metastases, and rising PSA despite androgen deprivation therapy have a relatively indolent natural history. Only one third of men developed bone metastases at 2 years, and the median metastasis-free survival was approximately 30 months. Baseline PSA and PSA velocity independently predicted time to first bone metastasis, overall survival, and bone metastasis–free survival. No other covariate consistently predicted clinical outcomes.
To the best of our knowledge, this is the first analysis to prospectively evaluate the natural history of nonmetastatic prostate cancer and rising PSA despite androgen deprivation therapy. Notable features of the study design include confirmation of a castrate testosterone level at study entry, documentation of PSA progression with three serial rises, and radiographic screening to exclude men with bone metastases at study entry. In addition, bone scans were obtained every 4 months in order to minimize PSA bias in defining the primary study outcome of time to first bone metastases.
Elevation in serum PSA levels after surgery or radiation therapy for early-stage prostate cancer typically predates clinically or radiographically detectable metastatic disease by several years, without additional treatment. In a retrospective series of 315 men with a rising serum PSA after radical prostatectomy for example, the median time to first bone metastasis was 8 years from the time of PSA elevation. 8 Three years following initial postoperative PSA elevation, only 27% of patients had radiographically detectable metastases. The results of our study suggest that rates of disease progression remain low for men with "PSA only" prostate cancer, even after androgen deprivation therapy.
These analyses extend other evidence linking PSA velocity to clinical outcomes. In men with rising PSA after surgery or radiation therapy for early-stage disease, PSA velocity predicts time to metastases 8- 12 and survival. 13- 15 PSA velocity also predicts survival in men with hormone-refractory metastatic prostate cancer. 16
The role of bisphosphonates for prevention of bone metastases in men with prostate cancer remains undefined. Mason et al recently reported the preliminary results of Medical Research Council Pr04, a randomized controlled trial of clodronate to prevent symptomatic bone metastases in men with nonmetastatic prostate cancer. 17 The study included 508 men receiving standard treatment for clinical stage T2 to T4 prostate cancer with no evidence of bone metastases. Men were assigned randomly to either oral clodronate (2,080 mg daily) or placebo for 5 years. Most of the subjects received either external beam radiation therapy, external beam radiation therapy with hormone therapy, or primary hormone therapy as standard treatment. The primary end point was time to development of symptomatic bone metastases or prostate cancer death. There were no scheduled radiographic studies to diagnose new asymptomatic metastases. At a median follow-up of 7 years, there were no significant differences between the groups in either time to symptomatic bone metastases or survival.
The early termination of our study bars conclusions about the efficacy of zoledronic to prevent bone metastases in men with castrate nonmetastatic prostate cancer. The results of this study, however, inform the design of future clinical trials to prevent bone metastases in men with prostate cancer. The relatively indolent natural history of castrate nonmetastatic prostate cancer suggests that future clinical trials in this setting should be very large or should select subjects at particularly high risk in order to have adequate statistical power to detect a clinically meaningful treatment effect. The independent predictive value of PSA velocity suggests that PSA kinetics may provide an effective strategy to identify men at high risk for bone metastases or death.
Our study has potential limitations. The study allowed treatment with secondary hormonal therapy and/or chemotherapy at physician discretion. This part of the study design increases the generalizability of the results. Because detailed information about secondary hormonal therapy and chemotherapy was not collected, however, we cannot assess the impact of these salvage therapies on clinical outcomes. Less than one third of subjects had a Gleason sum greater than 7 at initial prostate cancer diagnosis. The relatively low proportion of men with a Gleason sum greater than 7 suggests that our study may have selected subjects with lower Gleason grades. Alternatively, men with a Gleason sum greater than 7 may be more likely to develop bone metastases early (at diagnosis or first PSA failure) and may be underrepresented in studies of men with a rising PSA as the only indication of disease progression. Notably, the proportion of men with high-grade tumors in our study (29%) is similar to the proportion of men with Gleason sum more than 7 in a study of men with rising PSA after prostatectomy (32%). 8 Lastly, our analyses evaluated a limited number of variables, and additional studies are needed to identify additional predictors of clinical outcomes in this setting.
In summary, men with castrate nonmetastatic prostate cancer and rising PSA have an indolent natural history, with a median metastasis-free survival of approximately 30 months. Baseline PSA and PSA velocity independently predict time to first bone metastasis, overall survival, and metastasis-free survival.
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Employment: Ronald Linnartz, Novartis Oncology; Ming Zheng, Novartis Oncology; Carsten Goessl, Novartis Oncology; Yong-Jiang Hei, Novartis Oncology. Stock Ownership: Ming Zheng, Novartis Oncology; Yong-Jiang Hei, Novartis Oncology. For a detailed description of these categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
Acknowledgment
We thank the following investigators who participated in the study: A. Avila (Santa Fe de Bogota, Colombia), A. Cajigas (Santa Fe de Bogota, Colombia), A. Torres (Mexico DF, Mexico), A. Malzyner (Sao Paulo, Brazil), B. Blank (Portland, OR), B. Aragao (Belo Horizonte, Brazil), C. Dzik (Sao Paulo, Brazil), C. Tosello (Sao Paulo, Brazil), C. McMurtry (Richmond, VA), D. Culkin (Oklahoma City, OK), D. Kim (Bay Shore, NY), D. Shevrin (Evanston, IL), D. Shevrin (Evanston, IL), D. Fiske (Stuart, FL), D. Bell (Halifax, Nova Scotia, Canada), D. Vaughn (Philadelphia, PA), D. Gleason (Tucson, AZ), D. Locke (Summerfield, FL), E.B. Alentorn (Vic, Spain), F. Boeminghaus (Neuss, Germany), F.H. de Mendoza (Lima, Peru), F. Desiderio (Rimini, Italy), F. Yunus (Memphis, TN), G. Comeri (Como, Italy), G. Greene (Little Rock, AK), H. Carloss (Paducah, KY), J. Lechner (Olympia, WA), J. McMurray (Hunstville, AL), J. Breza (Bratislava, Slovak Republic), J. Morales (San Juan, Puerto Rico), J. Metts III (Charleston, SC), J. Kaufman (Aurora, IL), J. Forrest (Tulsa, OK), J. Trachtenberg (Toronto, Ontario, Canada), J.C. Olmo (Granada, Spain), J. Camps (Tallahassee, FL), J. Chin (London, Ontario, Canada), K. Slawin (Houston, TX), L. Valansky (Kosice, Slovak Republic), L. Klotz (Toronto, Ontario, Canada), L. Kalman (Miami, FL), L. Reyno (Halifax, Nova Scotia, Canada), L. Balducci (Tampa, FL), L. Lacombe (Quebec, Montreal, Canada), L. Baez (San Juan, Puerto Rico), M. Colombel (Lyon, France), M. Ratner (Rockville, MD), M. Bondhus (Pinecrest, FL), M. Boyer (Concord, NH), M. McCleod (Ft Meyers, FL), M. Fodor (Santiago, Chile), M. Wiatrak (Milwaukee, WI), M. Spaulding (Buffalo, NY), N. Clarke (Manchester, UK), O. Bertetto (Torino, Italy), P. Donovan (Scottsdale, AZ), P. Miller (Lakewood, CO), P. Clingan (Wollongong, New South Wales, Australia), P. Nickers (Liege, Belgium), R. Bukowski (Cleveland, OH), A. Corica (Mendoza, Argentina), A. Dalul (Sante Fe, NM), A. Mandressi (Busto Arsizio, Italy), A. Muzio (Capital Federal, Argentina), A. Villaronga (Rem de Escalada, Argentina), A. Macan (Martinez, France), A. Comandone (Torino, Italy), A. Vasconcelos (Salvador, Brazil), A. DiStefano (Arlington, VA), B. Mellado (Barcelona, Spain), B. Paule (Creteil, France), B. Chern (Redcliff, Australia), B. Corser (Cincinnati, OH), B. Woodworth (Columbus, OH), B. Aragao (Belo Horizonte, Brazil), B. Hodge (Cumming, GA), C. Nolasco (Capital Federal, Argentina), C. Cordova (Hermosillo, Spain), C. Theodossiou (New Orleans, LA), D. Campos (San Isidro, Argentina), D. Rukstalis (Philadelphia, PA), D. Toledo (Santa Fe de Bogota, Colombia), D. Varcasia (Capital Federal, Argentina), D. Dearnaley (Surrey, UK), D. Gillatt (Bristol, UK), D. Joseph (Nedlands, Australia), D. Sahasharabude (Rochester, NY), D. Jocham (Lübeck, Germany), D. Leusmann (Kln, Germany), D. Riquet (Valenciennes, France), F. Boccardo (Genova, Italy), F.J. Marx (Kln, Germany), F. Hamdy (Sheffield, UK), G. Frenkel (Dallas, TX), G. Manikhas (St Petersburg, Russia), G. Porcile (Alba, Italy), H.U. Eickenberg (Bielefeld, Germany), H.J. Melchior (Kassel, Germany), I. Roussakov (Moscow, Russia), I. Klimberg (Ocala, FL), J.M. Wolff (Rostock, Germany), J. Kliment (Martin, Slovak Republic), J. Vinholes (Porto Alegre, Brazil), J. Gingrich (Memphis, TN), J. Picus (St Louis, MO), J. Grygiel (Darlinghurst, Australia), J. Hulbert (Minneapolis, MN), J. Ramsey (Indianapolis, IN), J. Esterellas (Rosario, Argentina), J.M. De Marco (Buenos Aires, Argentina), J. Premoli (Rosario, Argentina), J. Kish (Detroit, MI), K. Stockamp (Ludwigshafen, Germany), K.F. Klippel (Celle, Germany), Ken Pittman (Adelaide, SA), K. Clarke (Wodonga, Australia), K. Miller (Berlin, Germany), L. Weissbach (Berlin, Germany), L. Dogliotti (Orbassano, Italy), L. Montes de Oca (Capital Federal, Argentina), M. Siegsmund (Mannheim, Germany), M. Merlano (Cuneo, Italy), M. Gleave (Vancouver, British Columbia, Canada), M. Boyer (Camperdown, New South Wales, Australia), M. Lill (Los Angeles, CA), M. Carmel (Sherbrooke, Canada), M. Lopez (Capital Federal, Argentina), M. Moises (Tucuman, Argentina), N. Fredotovich (Capital Federal, Argentina), OA Dr Zimmermann (Greifswald, Germany), O. Damia (Capital Federal, Argentina), O.R. Tutor (Mendoza, Argentina), O. Aren (Uruguay), P. Benedetto (Miami, FL), P. Keane (Belfast, Ireland), P. McInerney (Plymouth, MA), P. Schelhammer (Norfolk, VA), P. Pizao (Campinas, Brazil), P. Venner (Edmonton, Alberta, Canada), P. Grise (Rouen, France), P.F. Conte (Pisa, Italy), P. Major (Hamilton, Ontario, Canada), P. Teillac (Paris, France), P. Lara (Sacramento, CA), R. Persad (Bristol, UK), R. Gerson (Mexico DF CP, Mexico), R. Anchelerguez (Mendoza, Argentina), R. Wainstein (Villa Sarmiento, Argentina), R. Costabile (Tacoma, WA), R. Kahnoski (Grand Rapids, MI), R. Mansour (Shreveport, LA), R. Middleton (Salt Lake City, UT), R. Feldman (Waterbury, CT), R. Smith (Columbia, SC), R. Quintana (Capital Federal, Argentina), R. Vega (Colonia La Raza, Mexico), R. Dmochowski (Rockville, MD), R. Carroll (Scarborough, UK), R. Lewis (Augusta, GA), R. Yanagihara (Gilroy, CA), R. Shah (Lake Bluff, IL), Roy MacKintosh (Reno, NV), R. Bengio (Cordoba, Argentina), S. Ernst (Calgary, Alberta, Canada), S. Metrebian (Cordoba, Argentina), S. Tchekmedyian (Long Beach, CA), S. Childs (Cheyenne, WY), S.C. Mueller (Bonn, Germany), S. Rosenberg (Des Moines, IA), T. Beck (Springdale, AR), T. Szlaby (Husum, Germany), T. Hlavinka (San Antonio, TX), T. Roberts (Newcastle, UK), U.W. Tunn (Offenbach, Germany), V. Benavente (Lima, Peru), V. Gorbunova (Moscow, Russia), W. Jacobsen (Kiel, Russia), W. Bohnert (Phoenix, AZ), W. Monnig (Cincinnati, OH), W. Moseley (San Diego, CA), W. Schubach (Seattle, WA), Y. Rahim (Toronto, Ontario, Canada), Z. Wajsman (Gainesville, FL). Dr. Smith is supported by the John and Claire Bertucci Center for Genitourinary Malignancies at Massachusetts General Hospital.
NOTES
Supported by Novartis Oncology. M.R.S. is supported by the John and Claire Bertucci Center for Genitourinary Malignancies at Massachusetts General Hospital.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Jemal A, Tiwari RC, Murray T, et al: Cancer statistics, 2004. CA Cancer J Clin 54:8-29, 2004
Anonymous: Immediate versus deferred treatment for advanced prostatic cancer: Initial results of the Medical Research Council Trial—The Medical Research Council Prostate Cancer Working Party Investigators Group. Br J Urol 79: 235-246, 1997
Bolla M, Gonzalez D, Warde P, et al: Improved survival in patients with locally advanced prostate cancer treated with radiotherapy and goserelin. N Engl J Med 337:295-300, 1997
Messing EM, Manola J, Sarosdy M, et al: Immediate hormonal therapy compared with observation after radical prostatectomy and pelvic lymphadenectomy in men with node-positive prostate cancer. N Engl J Med 341:1781-1788, 1999
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 351:1513-1520, 2004
Tannock IF, de Wit R, Berry WR, et al: Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 351:1502-1512, 2004
Kalbfleisch J, Prentice R: The Statistical Analysis of Failure Time Data (ed 2). Indianapolis, IN, John Wiley and Sons, 2002
Pound CR, Partin AW, Eisenberger MA, et al: Natural history of progression after PSA elevation following radical prostatectomy. JAMA 281:1591-1597, 1999
Lee WR, Hanks GE, Hanlon A: Increasing prostate-specific antigen profile following definitive radiation therapy for localized prostate cancer: Clinical observations. J Clin Oncol 15:230-238, 1997
Patel A, Dorey F, Franklin J, et al: Recurrence patterns after radical retropubic prostatectomy: Clinical usefulness of prostate specific antigen doubling times and log slope prostate specific antigen. J Urol 158:1441-1445, 1997
Roberts SG, Blute ML, Bergstralh EJ, et al: PSA doubling time as a predictor of clinical progression after biochemical failure following radical prostatectomy for prostate cancer. Mayo Clin Proc 76:576-581, 2001
Sartor CI, Strawderman MH, Lin XH, et al: Rate of PSA rise predicts metastatic versus local recurrence after definitive radiotherapy. Int J Radiat Oncol Biol Phys 38:941-947, 1997
Sandler HM, Dunn RL, McLaughlin PW, et al: Overall survival after prostate-specific-antigen-detected recurrence following conformal radiation therapy. Int J Radiat Oncol Biol Phys 48:629-633, 2000
D'Amico AV, Cote K, Loffredo M, et al: Determinants of prostate cancer-specific survival after radiation therapy for patients with clinically localized prostate cancer. J Clin Oncol 20:4567-4573, 2002
D'Amico AV, Moul JW, Carroll PR, et al: Surrogate end point for prostate cancer-specific mortality after radical prostatectomy or radiation therapy. J Natl Cancer Inst 95:1376-1383, 2003
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 83:1989-1994, 1998
Mason MD, Collaborators MP: Development of bone metastases from prostate cancer: First results of the MRC PR04 trial (ISCRTN 61384873). Proc Am Soc Clin Oncol 22: 384S, 2004 (abstr 4511)(Matthew R. Smith, Fairooz)