This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Millikan, R. Articles by Logothetis, C. J. Search for Related Content PubMed PubMed Citation Articles by Millikan, R. Articles by Logothetis, C. J. Social Bookmarking                            What's this?

Randomized, Multicenter, Phase II Trial of Two Multicomponent Regimens in Androgen-Independent Prostate Cancer

Randall Millikan, Peter F. Thall, Sang-Joon Lee, Donnah Jones, M.W. Cannon, J. Phillip Kuebler, James Wade, III, Christopher J. Logothetis

From the Genitourinary Medical Oncology Department and Department of Biostatistics, University of Texas M.D. Anderson Cancer Center, Houston, TX; Cancer Center of Wichita/Clinical Community Oncology Program, Wichita, KS; Cancer Center of Columbus/Clinical Community Oncology Program, Columbus, OH; and Central Illinois Clinical Community Oncology Program, Decatur, IL.

Address reprint requests to Randall Millikan, PhD, MD, Department of Genitourinary Medical Oncology, University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Box 427, Houston, TX 77004-4009; email: rmillika{at}notes.mdacc.tmc.edu.

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Purpose: Several multicomponent regimens have been reported to be useful in advanced androgen-independent prostate cancer. We used a randomized phase II design to evaluate and compare two such regimens. Patients were accrued primarily in the community setting.

Patients and Methods: Patients with progressive, androgen-independent prostate cancer were randomly assigned to one of two treatments: either ketoconazole/doxorubicin alternating with vinblastine/estramustine (KA/VE) or paclitaxel, estramustine, and oral etoposide (TEE). Patients were prospectively stratified on the basis of disease volume. The primary end points were response and overall survival time.

Results: A total of 75 patients were registered; 71 are included in the analysis. By the criterion of an 80% prostate-specific antigen reduction maintained for at least 8 weeks, 11 (30%) of 37 patients in the TEE arm responded, whereas 11 (32%) of 34 assigned to KA/VE responded. Median survival was 16.9 months (95% confidence interval [CI], 10.5 to 21.2 months) in the TEE arm and 23.4 months (95% CI, 12.9 to 30.6 months) for patients treated with KA/VE. Many patients (24%) failed to complete at least 6 weeks of therapy, including five (8%) treatment-related early deaths.

Conclusion: Each of these regimens produced clinically significant responses, and the observed median survival (18.9 months for all 71 patients) compares favorably with previously published results, especially in the community setting. Nonetheless, it is apparent that these first-generation regimens must be applied judiciously, and thus we view efforts at better patient selection and the development of more tolerable therapies as higher priorities than carrying either of these regimens to phase III evaluation in the cooperative group setting.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
WIDELY METASTATIC androgen-independent prostate cancer (AIPC) is an aggressive disease, associated with high morbidity and a median survival of less than 1 year. Despite the long natural history from initial diagnosis, nearly all men who develop metastatic AIPC die as a result of inexorable disease progression. In the context of community use of mitoxantrone and prednisone (the only therapy so far approved by the United States Food and Drug Administration), median survival was reported to be 7.9 months.¹ Sadly, this represents the expected survival with purely symptomatic care. A great many single-agent phase II trials conducted at referral centers have consistently reported median survival times in the range of 8 to 11 months.² Combination chemotherapy regimens have recently become available³ that seem to alter the natural history of AIPC. However, the only available data come from referral centers, where considerable and hard-to-quantify selection biases exist. Thus we found substantial rationale to study some of the newer regimens in the community setting. Here we report a randomized phase II trial of two combination chemotherapy regimens. The goals of this trial were to obtain response and toxicity data and to screen for any obvious differences in toxicity or efficacy that could influence the choice of an experimental regimen to carry forward in more advanced trials. As far as we know, this is the first randomized trial between two multicomponent chemotherapy regimens to be reported in AIPC.

The regimen resulting from alternating weekly exposure to the doublet of ketoconazole and doxorubicin with the doublet of vinblastine and estramustine (KA/VE) was a benchmark development within the prostate cancer program at M.D. Anderson Cancer Center in Houston, TX.⁴ For the first time in our experience, palliative responses came to be expected in the majority of patients with far-advanced, metastatic AIPC. Moreover, the initially observed median survival time of 18 months has been amply confirmed by subsequent experience. It thus seems likely, although not yet rigorously demonstrated, that treatment with this regimen does indeed provide a modest improvement in survival. The toxicity of this regimen given at our center has been manageable; nonetheless, it was of interest to know how suitable this regimen would be in the community setting.

More recently, combinations of a taxane with estramustine have been widely investigated^5,6 as palliative therapy for AIPC. An early example of this paradigm coming from the University of Michigan program headed by Ken Pienta⁷ also produced palliative responses in our hands, including responses in some patients for whom front-line KA/VE failed to provide benefit. This first-generation regimen of paclitaxel, estramustine, and oral etopside (TEE) seemed to be promising enough to merit consideration for more advanced trials. On the basis of the observed activity and a favorable toxicity profile, this also seemed to be an appropriate regimen to study in the community setting. These considerations led us to conduct a randomized phase II trial of KA/VE and TEE, with patients accrued predominantly from clinical community oncology program (CCOP) affiliates of the University of Texas M.D. Anderson Cancer Center.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Eligibility
Eligible patients had metastatic prostate cancer that was progressing despite castrate levels of testosterone (defined for the purposes of this protocol as < 40 ng/dL) and withdrawal of antiandrogens. Patients on medical therapy (with luteinizing hormone–releasing hormone superagonists therapy) to suppress testicular function at registration were maintained continuously on that therapy. Eligible patients had serum prostate-specific antigen (PSA) of at least 4 ng/mL and had an increasing PSA level on three consecutive measurements before registration. All patients provided written informed consent to participate in this study, which was approved by the University of Texas M.D. Anderson Cancer Center institutional review board.

Treatment
The KA/VE regimen was administered as previously described.⁴ Briefly, ketoconazole was given orally at 400 mg tid for 7 days and doxorubicin was given at 20 mg/m² on the first day of the week. In week 2, patients received vinblastine at 4 mg/m² on the first day of the week and estramustine orally at 140 mg tid for 7 days. Treatment with ketoconazole/doxorubicin (KA) then continued to alternate with vinblastine/estramustine (VE) for a total of 6 weeks, followed by a 2-week break from therapy, to make an 8-week treatment cycle. Weeks 7 and 8 were designated rest periods, no matter how much therapy was actually delivered in weeks 1 to 6. Criteria for receiving therapy included an absolute neutrophil count of 1,000 cells/mL and platelet count of 75,000 cells/mL on the day intravenous treatment was due. If counts were still not adequate for treatment the next week, then both doxorubicin and vinblastine were reduced by 25%. The same dose reduction was taken for a second episode of inadequate counts lasting more than 1 week. If this happened a third time, or more than 3 weeks were required for count recovery, then patients were removed from study.

As described by Smith et al,⁷ patients assigned to TEE received paclitaxel at 135 mg/m² on day 2, following the usual premedication consisting of dexamethasone and combined histamine receptor blockade using both diphenhydramine and cimetidine. In addition, patients received estramustine orally at 280 mg tid for 14 days and etoposide orally at 50 mg bid for 14 days. Treatment was repeated on a 21-day cycle. If the neutrophil and platelet counts were not recovered to at least 1,000 and 75,000 cells/mL, respectively, then the therapy was delayed 1 week. If more than 28 days were required to restart, then etoposide was reduced to 100 mg alternating with 50 mg orally daily for 14 days (ie, a 25% reduction).

Response Criteria
The primary scientific objectives of this trial were to obtain unbiased estimates of the response rates and the median survival times associated with TEE and KA/VE. Response was defined in terms of PSA reduction for all patients and, where applicable, by the conventional criterion of a 50% reduction in the product of the greatest dimension and its perpendicular for patients with measurable nodal or visceral metastases. Because the magnitude of PSA response that is the best surrogate for clinical benefit continues to be controversial, we report marker responses according to both 50% and 80% PSA reduction from baseline, and always with the caveat that the PSA remain below these thresholds for at least 8 weeks. Time to progression was measured from protocol registration.

Statistical Methods
A randomized phase II design was used. As is well known, observed differences between average outcomes from two single-arm trials of two treatments conducted sequentially or at different sites are in fact the confounded effects of the treatments and unobserved, latent variables. We randomly assigned patients to the arms to avoid such confounding and obtain unbiased estimates of the differences between the average effects of TEE and KA/VE on patient outcome. A Simon⁸ optimal two-stage phase II design was used within each of the two treatment arms. For this design, response was defined as a 50% or greater reduction in PSA maintained for 8 weeks. Within each treatment arm, the Simon design was based on a test of a null response probability of 0.40 versus an alternative of 0.60, with the type I (false-positive) and type II (false-negative) error rates each controlled to be 10%. In this case, the Simon optimal design within each arm is to first treat 18 patients in stage 1. If there are seven or fewer responses, then the arm is terminated early and the treatment is rejected for further study; if there are eight or more responses, then an additional 28 patients are treated in stage 2, with the treatment rejected if there are 22 or fewer responses among the total of 18 + 28 = 46 patients and accepted for future study if there are 23 or more responses. This design has a 56% chance of terminating a treatment arm early if the true response rate is the null value 0.40. Considering both arms, with respect to the goal of selecting the better of the two treatments, the probability that the treatment with a truly superior response rate would be selected is 80% if the true difference between the two rates was at least 0.10. The maximum accrual goal of 46 + 46 = 92 patients for the two treatment arms was considered realistic based on our experience in a previous trial of identical design,⁹ which accrued 90 patients in just under 12 months. In addition, this provided a 95% confidence interval (CI) for the response probability within each treatment arm having width less than 0.30.

Patients were prospectively stratified into one of three groups: (1) low-volume disease, defined as disease limited to the area of the prostate, confined to lymph nodes, or with two or fewer lesions on bone scan; (2) intermediate-volume disease, defined as more than two bone lesions but confined to the pelvis, spine, ribs, or calvarium; or (3) high-volume disease, defined as bone lesions outside the central skeleton or with visceral metastases. Random assigment was performed within each of these three prospectively identified strata to improve prognostic balance in this small trial. For the multivariate Cox model analysis, however, patients were stratified differently, based on the following four categories, listed in order of expected prognostic impact: (1) disease confined to nodes; (2) bone involvement at two or fewer sites; (3) bone involvement at three or more sites; and finally, (4) visceral involvement (regardless of the number of bone lesions).

Unadjusted probabilities of survival were estimated using the method of Kaplan and Meier.¹⁰ Unadjusted between-group comparisons of survival were made using the log-rank test.¹¹ The Cox proportional hazards regression model^12,13 was used to assess the ability of patient characteristics or treatments to predict survival, with goodness-of-fit assessed by the Grambsch-Therneau test¹⁴ and martingale residual plots.¹³ All scatter plots were smoothed using the lowest method of Cleveland,¹⁵ with predictive variables transformed as appropriate based on these plots. Comparison of the response probabilities in the two treatment groups was made using Fisher’s exact test. All computations were carried out in S-plus¹⁶ (Statistical Sciences, Seattle, WA) using standard S-plus functions and the S-plus survival analysis package of Therneau.¹⁷

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Patients
Between January 1998 and May 2000, 75 patients were registered. In view of the sluggish accrual and higher than anticipated morbidity, it was decided to close the trial short of the original accrual goal of 92, especially because this was felt to be an adequate number to estimate response rates with reasonable reliability. The Data Monitoring Committee was not involved in the decision to stop accrual, because the differences between the two arms were not sufficient to threaten equipoise.

Three patients withdrew consent before receiving any therapy (two were assigned to TEE and one was assigned to KA/VE) and were not included in the analysis. In addition, it was discovered in retrospect that one patient was registered despite never having had a trial of sustained androgen ablation, a major eligibility violation. This patient was also excluded from the analysis. The remaining 71 patients are all included in the analysis of response and toxicity, regardless of how much therapy was received. Baseline characteristics of these 71 patients are shown in Table 1. Significantly, 69% of the patients were symptomatic at baseline, and most (77%) had evidence of skeletal metastases on bone scan. A total of 32 patients (45%) had conventionally measurable disease, 15 in the KA/VE arm and 17 in the TEE arm. Note that 46 patients (65%) were enrolled by the CCOPs, and 25 (35%) were enrolled by the department of genitourinary medical oncology at the University of Texas M.D. Anderson Cancer Center. Approximately half (18 of 37 in the TEE arm and 17 of 34 in the KA/VE arm) of the patients had no prior local therapy to the primary site.

View larger version (19K): [in this window] [in a new window]   Fig 1. Overall survival of 71 assessable patients by assigned treatment.

Toxicity
Adverse events were graded according to the National Cancer Institute common toxicity criteria version 2.0 and are summarized in Table 2. As expected with estramustine-containing regimens, thromboembolic events were the most often encountered serious adverse event. Remarkably, 19 (27%) of 71 patients had either deep venous thrombosis requiring anticoagulant therapy or pulmonary embolism.

Prognostic Covariates
Many clinical parameters that reflect disease burden and/or physiologic reserve have been shown to be prognostically important in AIPC. Among those known to adversely affect outcome are compromised performance status, elevated PSA, elevated alkaline phosphatase, anemia, and extent of disease. In addition, the time from initiation of hormone therapy to clinically progressive AIPC may be a useful indicator of biologic aggressiveness and, therefore, prognostically informative. A fitted Cox model including assigned treatment and each of these covariates is summarized in Table 3. PSA was transformed to log(PSA). Alkaline phosphatase was dichotomized at the median value of 128 U/L. Performance status was dichotomized as 0 to 1 versus 2. Time to hormone failure (in years), log(PSA), and hemoglobin (in g/dL) were considered as continuous variables. We illustrate the remarkable effect of pretreatment alkaline phosphatase, ignoring the other covariates, on overall survival in Fig 2.

View larger version (19K): [in this window] [in a new window]   Fig 2. Overall survival of all 71 patients by pretreatment alkaline phosphatase, dichotomized at the 120 U/L (ie, the upper level of normal).

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
To date, there is no definitive demonstration of improved survival by virtue of any therapy for patients with AIPC. However, several investigators have reported median survival times in the range of 18 to 24 months, and this may well represent an improvement from the long-established benchmark of 8 to 11 months that has been observed in multiple phase II studies reported over the last two decades. In this context, we felt that it was important to carefully consider which regimen(s) should be taken on to more advanced trials to formally test the hypothesis that treatment with these regimens does improve survival. Furthermore, we thought it was of special interest to investigate the performance of some of the newer multicomponent regimens in the context of community oncology practice.

We had previously conducted a randomized phase II trial of ketoconazole with or without doxorubicin in this same group of CCOP affiliates.⁹ Our experience in that trial, in which 90 patients were accrued in less than 12 months, was that community oncologists were keen to offer patients with AIPC some form of therapy beyond analgesics. We were thus somewhat surprised that the currently reported trial did not accrue well, with only 46 patients registered by the CCOP affiliates in 28 months. (It is worth noting that we anticipated that many oncologists would see mitoxantrone/prednisone as a standard of care, and thus patients previously treated with one prior regimen were eligible for this trial.) The reasons for the sluggish accrual are unknown but could perhaps reflect the judgment that these regimens are too toxic for many patients with AIPC, too logistically complex, or too expensive. As of December 2001, the average wholesale price for just the oral medications (ie, those typically not covered by Medicare) are $320/mo for KA/VE and $2,440/mo for TEE.

The observed median survival of 18.9 months for all assessable patients in this trial compares favorably with the recent report¹ of mitoxantrone/prednisone (7.9 months) in a similar cohort also studied in the community setting. Indeed, the results of this trial were quite comparable to the performance of these regimens in the referral centers. For TEE, Smith et al⁷ reported 16 (40%) of 40 patients with more than 75% PSA reduction and median survival of 12.8 months. For KA/VE, Ellerhorst et al⁴ reported 27 (52%) of 46 patients with 80% PSA reduction and median survival of 18 months. Importantly, the present study confirms a high correlation between PSA response and objective response and confirms the powerful prognostic impact of PSA response status. Thus we continue to urge that an 80% PSA reduction maintained for at least 8 weeks be reported as a readily applied, objective standard of response that is a meaningful index of patient benefit in the conduct of clinical trials in AIPC. When this study was conducted we did not measure PSA every 4 weeks, and thus we cannot report our results according to the consensus criterion of the PSA working group for response to be a 50% reduction maintained for 4 weeks.¹⁸

Although intuitive and familiar for clinicians, it is nonetheless striking that a short-lived response to androgen ablation was so highly predictive of biologically aggressive cancer and short survival, despite treatment with the chemotherapy regimens we studied. It would seem that this feature should be accounted for when assessing the prognostic features of patients with AIPC.

For both of these regimens, toxicity was striking in the community setting. In our original report of the KA/VE regimen,⁴ we observed no toxic deaths and only 11 episodes of neutropenia (three complicated by fever) in 648 courses of therapy delivered. It was also noted in the original report that almost uniformly, the observed neutropenia occurred between days 11 and 14 of the first cycle. This information concerning predictable neutropenia in this interval was emphasized in the current protocol. In the original disclosure of Smith et al,⁷ there were two toxic deaths in 40 patients (5%) treated with 223 cycles; there were 10 episodes of grade 3/4 neutropenia and six episodes of neutropenic fever. It should be noted that 65% of the patients in that study had prior chemotherapy, whereas the patients reported here were almost entirely chemotherapy-naive. We have no explanation for the increased morbidity and mortality observed in the community setting, but we suspect that it may in part reflect the selection bias of receiving treatment at referral centers that effectively precludes patients with a significantly compromised performance status.

The high rate of thromboembolic events we observed (27%) undoubtedly reflects the well-known thrombogenic effect of estramustine. The role of prophylactic administration of warfarin and aspirin in the face of estrogenic therapy for prostate cancer remains controversial. A small Canadian study¹⁹ of diethylstilbestrol did not suggest a protective effect of low-dose warfarin. More recently, Kelly et al⁶ have disclosed their results with paclitaxel, estramustine (at 280 mg tid for 5 days each week), and carboplatin. In this study, neither low-dose warfarin nor low-dose aspirin seemed to provide significant amelioration of the substantial risk of thromboembolic morbidity. Full anticoagulation and combinations of warfarin and aspirin remain to be investigated. It is also possible that a lower estramustine dose would have a more favorable therapeutic ratio, and this is also under investigation.

The reported median survival of patients with advanced AIPC is improving. Undoubtedly, this reflects many factors such as stage migration, waning therapeutic nihilism, the availability of new bone consolidation strategies, and so on. Also of potential importance is the availability of multiple active regimens. In this trial, many patients had second-line therapy, often with further response. For example, one patient with liver metastases that grew in the face of front-line therapy with KA/VE had striking success with a subsequent taxane-based regimen. He not only responded but lived another 34 months after progression on front-line therapy. Observations like this in the setting of serial courses of treatment have motivated us to develop novel statistical methods to simultaneously evaluate several therapies given in a multicourse structure.²⁰ This new methodology may allow for more powerful selection in the context of randomized phase II evaluation and streamline the process of ranking contenders for more advanced clinical evaluation. Such a trial comparing four regimens in patients with AIPC is currently underway at M.D. Anderson Cancer Center.

In conclusion, both KA/VE and TEE produced palliative responses in patients with AIPC, with approximately one third of treated patients achieving a threshold of response that we have come to recognize as clinically significant. Response and survival in the community setting compared remarkably well with results previously reported from tertiary care centers. Despite these encouraging observations, we believe that further refinements aimed at easing the morbidity of these therapies are required. Even though a palliative benefit from available therapy is now beyond question, and life-extending activity is probable, these particular first-generation combination regimens do not seem to be well suited for widespread use without better patient selection or some refinement of the regimens themselves. That being said, the survival results reported here are by far the best yet observed in a community-based trial of any therapy for patients with AIPC and, in our view, provide ample rationale for moving forward with phase III trials of promising, better-tolerated combinations. Furthermore, these results suggest that the taxane-based regimens are not necessarily better than anthracycline-based regimens, and indeed, it seems that these approaches may be more complementary than mutually exclusive.

	NOTES

 
Supported by grant no. CA45809 from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Bethesda, MD.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Dowling AJ, Czaykowski PM, Krahn MD, et al: Prostate specific antigen response to mitoxantrone and prednisone in patients refractory prostate cancer: Prognostic factors and generalizability of a multicenter trial to clinical practice. J Urol 163:1481–1485, 2000[CrossRef][Medline]

2. Yagoda A, Petrylak D: Cytotoxic Chemotherapy for advanced hormone-resistant prostate cancer. Cancer 71:1098–1109, 1993[CrossRef][Medline]

3. Kelly WK, Slovin SF: Chemotherapy for androgen-independent prostate cancer: Myth or reality. Curr Oncol Rep 2:394–401, 2000[Medline]

4. Ellerhorst JA, Tu SM, Amato RJ, et al: Phase II trial of alternating weekly chemohormonal therapy for patients with androgen-independent prostate cancer. Clin Cancer Res 3:2371–2376, 1997[Abstract/Free Full Text]

5. Petrylak DP, Macarthur R, O’Connor J, et al: Phase I/II studies of docetaxel (Taxotere) combined with estramustine in men with hormone-refractory prostate cancer. Semin Oncol 26:28–33, 1999 (suppl 17)[Medline]

6. Kelly WK, Curley T, Slovin S, et al: Paclitaxel, estramustine phosphate, and carboplatin in patients with advanced prostate cancer. J Clin Oncol 19:44–53, 2001[Abstract/Free Full Text]

7. Smith DC, Esper P, Strawderman M, et al: Phase II trial of oral estramustine, oral etoposide, and intravenous paclitaxel in hormone-refractory prostate cancer. J Clin Oncol 17:1664–1671, 1999[Abstract/Free Full Text]

8. Simon R: Optimal two-stage designs for phase II clinical trials. Control Clin Trials 10:1–10, 1989[Medline]

9. Millikan RE, Baez L, Banerjee T, et al: Randomized phase 2 trial of ketoconazole and ketoconazole/doxorubicin in androgen-independent prostate cancer. Urol Oncol 6:111–115, 2001[CrossRef][Medline]

10. Kaplan EL, Meier P: Nonparametric estimator from incomplete observations. J Am Stat Assoc 53:457–481, 1958[CrossRef]

11. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 60:163–170, 1966

12. Cox DR: Regression models and life tables (with discussion). J R Stat Soc B 34:187–220, 1972

13. Therneau TM, Grambsch PM: Modeling Survival Data. New York, NY, Springer, 2000

14. Grambsch PM, Therneau TM: Proportional hazards tests and diagnostics based on weighted residuals. Biometrika 81:515–526, 1994[Abstract/Free Full Text]

15. Cleveland WS: Robust locally weighted regression and smoothing scatterplots. J Am Stat Assoc 74:829–836, 1979[CrossRef]

16. Venables WN, Ripley BD: Modern Applied Statistics With S-Plus (ed 3). New York, NY, Springer, 1999

17. Therneau TM: A package for survival analysis in S. Rochester, MN, Mayo Clinic Foundation, 1997

18. Bubley GJ, Carducci M, Dahut W, et al: Eligibility and response guidelines for phase II clinical trials in androgen-independent prostate cancer: Recommendations from the Prostate-Specific Antigen Working Group. J Clin Oncol 17:3461–3467, 1999[Abstract/Free Full Text]

19. Klotz L, McNeill I, Fleshner N: A phase 1–2 trial of diethylstilbestrol plus low dose warfarin in advanced prostate carcinoma. J Urol 161:169–172, 1999[CrossRef][Medline]

20. Thall PF, Millikan RE, Sung H-G: Evaluating multiple treatment courses in clinical trials. Stat Med 19:1011–1028, 2000[CrossRef][Medline]

Submitted April 5, 2002; accepted November 20, 2002.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				R. E. Millikan, S. Wen, L. C. Pagliaro, M. A. Brown, B. Moomey, K.-A. Do, and C. J. Logothetis Phase III Trial of Androgen Ablation With or Without Three Cycles of Systemic Chemotherapy for Advanced Prostate Cancer J. Clin. Oncol., December 20, 2008; 26(36): 5936 - 5942. [Abstract] [Full Text] [PDF]

				P. F. Thall, C. Logothetis, L. C. Pagliaro, S. Wen, M. A. Brown, D. Williams, and R. E. Millikan Adaptive Therapy for Androgen-Independent Prostate Cancer: A Randomized Selection Trial of Four Regimens J Natl Cancer Inst, November 7, 2007; 99(21): 1613 - 1622. [Abstract] [Full Text] [PDF]

				A. Casati, V. S. Zimmermann, F. Benigni, M. T. S. Bertilaccio, M. Bellone, and A. Mondino The Immunogenicity of Dendritic Cell-Based Vaccines Is Not Hampered by Doxorubicin and Melphalan Administration J. Immunol., March 15, 2005; 174(6): 3317 - 3325. [Abstract] [Full Text] [PDF]

				A. Stangelberger, A. V. Schally, J. L. Varga, M. Zarandi, K. Szepeshazi, P. Armatis, and G. Halmos Inhibitory Effect of Antagonists of Bombesin and Growth Hormone-Releasing Hormone on Orthotopic and Intraosseous Growth and Invasiveness of PC-3 Human Prostate Cancer in Nude Mice Clin. Cancer Res., January 1, 2005; 11(1): 49 - 57. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Millikan, R. Articles by Logothetis, C. J. Search for Related Content PubMed PubMed Citation Articles by Millikan, R. Articles by Logothetis, C. J. Social Bookmarking                            What's this?