- Email: [email protected]
Translating clinical trials to clinical practice: outcomes of men with metastatic castration resistant prostate cancer treated with docetaxel and prednisone in and out of clinical trials*
original articles
Annals of Oncology Annals of Oncology 24: 2972–2977, 2013 doi:10.1093/annonc/mdt397 Published online 14 October 2013
Translating clinical trials to clinical practice: outcomes of men with metastatic castration resistant prostate cancer treated with docetaxel and prednisone in and out of clinical trials* A. J. Templeton1, F. E. Vera-Badillo1, L. Wang2, M. Attalla1, P. De Gouveia1, R. Leibowitz-Amit1, J. J. Knox1, M. Moore1, S. S. Sridhar1, A. M. Joshua1, G. R. Pond3, E. Amir1 & I. F. Tannock1* Divisions of 1Medical Oncology and Hematology, University of Toronto, Toronto; 2Biostatistics, Princess Margaret Cancer Center, Department of Medicine, University of Toronto, Toronto; 3Department of Oncology, McMaster University, Hamilton, Canada
Background: Multiple factors can influence outcomes of patients receiving identical interventions in clinical trials and in routine practice. Here, we compare outcomes of men with metastatic castrate-resistant prostate cancer (mCRPC) treated with docetaxel and prednisone in routine practice and in clinical trials. Patients and methods: We reviewed patients with mCRPC treated with docetaxel at Princess Margaret Cancer Centre. Primary outcomes were overall survival and PSA response rate. Secondary outcomes were reasons for discontinuation and febrile neutropenia. Outcomes were compared for men treated in routine practice and in clinical trials, and with data from the TAX 327 study. Results: From 2001 to 2011, 438 men were treated, of whom 357 received 3-weekly docetaxel as first-line chemotherapy: 314 in routine practice and 43 in clinical trials. Trial patients were younger and had better performance status. Median survival was 13.6 months [95% confidence interval (95% CI) 12.1–15.1 months] in routine practice and 20.4 months (95% CI 17.4–23.4 months, P = 0.007) within clinical trials, compared with 19.3 months (95% CI 17.6–21.3 months, P < 0.001) in the TAX 327 study. PSA response rates were 45%, 54%, and 53%, respectively (P = NS). Reasons for treatment discontinuation were similar although trial patients received more cycles (median: 6 versus 8 versus 9.5, P < 0.001). Rates of febrile neutropenia were 9.6, 0, and 3% (P < 0.001) while rates of death within 30 days of last dose were 4%, 0%, and 3%, respectively (P = NS). Conclusions: Survival of patients with mCRPC treated with docetaxel in routine practice is shorter than for men included in trials and is associated with more toxicity. Key words: docetaxel, routine practice, clinical trial, prostate cancer
introduction Data from clinical trials, although providing critical evidence of clinical activity, do not provide adequate information to judge the impact of a new treatments when used in the realworld setting [1]. The difference between results of clinical trials and outcomes in clinical practice may be explained by restrictive selection criteria for trials such as good prognosis and minimal comorbidities [2]. This is likely to be particularly important when older patients receive potentially toxic *Correspondence to: Dr Ian Tannock, Division of Medical Oncology and Hematology, Princess Margaret Cancer Centre, 610 University Avenue, Toronto, ON, M5G 2M9, Canada. Tel: +1-416-946-4501; Fax: +1-416-946-4563; E-mail: [email protected] *Results of this study were presented in part at the 2013 ASCO Genitourinary Cancers Symposium, 14–16 February 2013, Orlando, FL, Canada.
therapy, such as men with metastatic castration-resistant prostate cancer (mCRPC), since most will have pre-existing medical conditions. Despite an increasing role for secondary and tertiary hormonal-based therapies in men with mCRPC, standard initial treatment since 2004 remains chemotherapy with docetaxel and prednisone. This is based on the international TAX-327 study, which compared docetaxel in two schedules (3-weekly and weekly) with mitoxantrone (all with prednisone) [3]. The extent to which results from the pivotal trial translate to less selected patients treated in routine practice is unknown. Here we reviewed all men treated with docetaxel for mCRPC at our institution. We hypothesized that those treated outside of a clinical trial would have poorer survival and more toxicity than those recruited to clinical trials.
© The Author 2013. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
Received 1 July 2013; revised 2 August 2013; accepted 14 August 2013
original articles
Annals of Oncology
patients and methods study population All men who received at least one dose of docetaxel for mCRPC before January 2012 at Princess Margaret Cancer Centre (PMCC) in Toronto, Canada were identified through pharmacy records. Patients treated with weekly docetaxel were excluded as this was given to men who were frail or had a high risk of toxicity. Men who received docetaxel in the context of neoadjuvant or adjuvant trials, or as second-line chemotherapy for mCRPC were also excluded. Enrollment logs of prostate cancer trials were screened to identify patients treated with docetaxel in clinical trials. The institutional Research Ethics Board (REB) approved the study.
data collection
results patients and treatment
outcomes of interest The primary objective was to compare overall survival (OS) and PSA response rates in men with mCRPC treated with 3-weekly docetaxel in routine practice at PMCC with those receiving the same schedule at our institution in clinical trials evaluating docetaxel/prednisone alone (e.g. TAX327 [3]) or with GTI-2040 [4], GVAX [5], aflibercept [6], or OGX-011 [7, 8] (supplementary Table S1, available at Annals of Oncology online). We also compared these data with those for 332 men who received 3-weekly docetaxel in multiple institutions in the TAX-327 trial [3, 9, 10]. Secondary objectives were to compare reasons for treatment discontinuation, rates of febrile neutropenia (as described in the clinical notes), and death within 30 days of last administration of docetaxel.
statistical analysis Descriptive statistics for categorical variables were reported as percentages, and continuous variables were reported as medians and ranges. Categorical variables were compared using Fisher’s exact test and Kruskal–Wallis test for two and more variables, respectively. Continuous variables were compared by the Wilcoxon rank sum test. Dose intensity was calculated as the cumulative dose of docetaxel per body surface area divided by the time between first dose and end of the last cycle (and was 100% for 25 mg/m2/ week). If there was >12 weeks between two administrations of docetaxel, the patient was regarded as having two separate courses and only doses in the first course were included in calculation of dose intensity. Survival was calculated from the date of first docetaxel administration to death with censoring at date of last contact for patients alive at the cut-off date in August 2012. The Kaplan–Meier method was used to estimate survival, and survival curves were compared using the generalized Wilcoxon test; this test was preferred to log-rank since it places greater emphasis on the first part of the survival curves where the effect of treatment is more pronounced. The log-rank test was evaluated as a supportive analysis. Association of individual variables with survival was assessed using the Cox proportional hazard model. All variables with a P-value <0.1 in univariable analysis, variables with a P-value ≥0.1 but which changed the outcome of the variable ‘in trial’ by more than 10%, or variables where there was a significant interaction with the variable “in trial” (P < 0.05) were
Volume 24 | No. 12 | December 2013
Between February 2001 and December 2011, 438 men received treatment with docetaxel at PMCC, of whom 314 and 43 received first-line treatment every 3 weeks in routine practice and in a clinical trial, respectively (Figure 1). Of the 43 men in clinical trials, 7 received docetaxel as part of the TAX-327 study. Patients’ and tumor characteristics are presented in Table 1. Men receiving treatment in routine practice were older (median age 71 versus 68 years, P = 0.038) and more likely to have an ECOG performance status of 2 or higher (29% versus 3%, P < 0.001), when compared with those treated in a clinical trial.
treatment received Comparisons of outcomes of patients treated on trial and offtrial, as well as those in the TAX-327 study, are presented in Table 2. Trial patients received more cycles of docetaxel (median: 8 versus 6, P = 0.001) whereas overall dose-intensity, dose reductions and delays by cycle 6 were similar. The median number of cycles and delayed infusions in TAX-327 were similar to those in PMCC patients whereas dose reductions occurred less often (P < 0.001).
overall survival and PSA response Median OS for all nontrial patients receiving 3-weekly docetaxel with prednisone was 13.6 months, compared with 20.4 months for PMCC trial patients (P = 0.007) and 19.3 months in TAX327 (P < 0.001; Figure 2). One-year and 2-year OS rates were higher for patients treated in clinical trials (Table 3). There was no significant difference in survival between trial patients in our institution and men treated on the TAX-327 study (with or without exclusion of the 7 PMCC patients who participated in TAX-327). In an exploratory analysis patients who would have met the selection criteria for the TAX-327 trial lived significantly longer compared with those not meeting the selection criteria (median survival 16.7 versus 9.0 months,
doi:10.1093/annonc/mdt397 |
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
Dates of treatment and doses were obtained from the PMCC pharmacy. Clinical data were extracted manually from the electronic patient record (EPR) by MA, PdG, and AT. Laboratory values were obtained electronically from EPR; baseline was defined as within 28 days before the first dose of docetaxel. Dates of primary diagnosis and death were obtained from the PMCC Registry, which includes 99% of patients treated in our institution. Data from all sources were linked using the patients’ Medical Record Number.
retained in the multivariable model. Variables included treatment in a clinical trial, age at start of treatment, time from diagnosis until start of docetaxel, Eastern Cooperative Oncology Group performance status (ECOG PS), year of first administration, presence of visceral metastases, Gleason score (<8 versus 8–10); serum levels of prostate-specific antigen (PSA), lactate dehydrogenase (LDH), alkaline phosphatase (ALP), hemoglobin (Hgb), and albumin. Laboratory values were treated as continuous variables. Variables with non-normal distribution (e.g. PSA, LDH, and ALP) were log transformed. PSA response was defined as ≥50% decline of PSA compared with baseline maintained for at least 3 weeks. Men with a rise in PSA within the first 12 weeks of treatment were considered as PSA responders if subsequent values of PSA satisfied the above criterion, as per Prostate Cancer Working group 2 (PCWG2) criteria [11]. All statistical tests were two sided, and statistical significance was defined as P < 0.05. Analyses were carried out using SAS version v9.2 (SAS Institute, Inc.) and SPSS version 20 (IBM Corp.); curves were plotted with R version 2.13.1 (www.r-project.org). Corrections for multiple significance testing were not applied.
original articles
Annals of Oncology
HR = 0.45, P < 0.001). Details are provided in the supplementary material, available at Annals of Oncology online. Compared with nontrial patients, patients receiving docetaxel in a clinical trial at PMCC had a reduced hazard of death in univariable analysis (HR = 0.74, Wilcoxon P = 0.007, Figure 2) but ‘trial treatment’ was not an independent prognostic factor in multivariable analysis (Table 4). Factors associated with longer survival were higher hemoglobin, lower LDH at baseline, longer time from initial diagnosis, and more recent treatment. There was no significant difference in PSA response rates between patients treated in daily practice, in a clinical trial at PMCC, or on the TAX-327 study (Table 3).
toxicity Overall, the rates for treatment discontinuation because of toxicity were the same, even when comparing discontinuation rates for toxicity by cycle 6 (Table 2). Febrile neutropenia occurred in 9.6% of nontrial patients (in 60% of the cases during cycle 1); in none of the PMCC trial patients (P = 0.035) and in 3% of men in TAX-327 (P < 0.001). Deaths within 30 days of the last administration of docetaxel occurred in 4% of men treated in routine practice whereas no such events were observed in PMCC trial patients (P = 0.24), and 3% in the TAX-327 study (P = 0.68).
discussion The purpose of the present study was to compare outcomes of men with mCRPC treated with docetaxel and prednisone in routine practice and in clinical trials. We therefore reviewed all patients treated with first-line 3-weekly docetaxel for mCRPC at our institution and compared outcomes of men treated in routine clinical practice with those in clinical trials, both within our institution and in the pivotal TAX-327 study. The results support our hypothesis that men treated in routine practice have
| Templeton et al.
shorter survival and more toxicity than men treated within clinical trials. The difference in median survival between men treated in routine practice and in trials at PMCC (13.6 versus 20.4 months) was larger than reported in other series [12–14]. The difference between outcomes in clinical trials and clinical practice is known as the efficacy-effectiveness gap [1, 15, 16]. The outcome of men treated in clinical trials at our institution was similar to those treated in the TAX-327 study (median OS 19.3 months [9]), and we explored a potential ‘trial effect’ by performing multivariable analyses to assess whether treatment in a clinical trial was an independent prognostic factor. This might be expected because of strict rules for assessment or management of toxicity [13]. No such association was observed in multivariable analysis, suggesting that selection of patients with favorable prognosis accounts for the observed difference in OS between trial and nontrial patients. Patients treated in clinical trials were younger, had less comorbidity and better performance status. These findings are consistent with a meta-analysis that did not find evidence that enrollment in clinical trials is an independent prognostic factor [17]. Patients treated in routine practice in our institution had more toxicity than those treated in clinical trials: they received less cycles of docetaxel, but had a 9.6% rate of febrile neutropenia compared with 0 and 3% for PMCC trial patients and participants of the TAX-327 study, respectively. However, the rate of all-cause death within 30 days was not significantly different (4% in routine practice, 0% in trial patients, and 3% in TAX-327). The high rate of febrile neutropenia cautions the use of toxic treatment in men with mCRPC who would not have met criteria for trial selection. Registration trials use strict selection criteria to show maximum efficacy and minimal toxicity but most new therapies add toxicity [18], and these toxicities may only become apparent when treatment is applied in routine practice [19]. Ideally, registration trials should be
Volume 24 | No. 12 | December 2013
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
Figure 1. Men with metastatic castration resistant prostate cancer (mCRPC) treated with docetaxel at PMCC 2001–2011
original articles
Annals of Oncology Table 1. Patients’ characteristics Not in trial (N = 314)
In trial a (N = 43)
71 44–90
68 51–83
P
0.038
83 (30) 116 (41) 81 (29)
23 (62) 13 (35) 1 (3)
<0.001
196 (62) 118 (38)
32 (74) 11 (26)
0.13
27 17–42
27 17–45
68 42–92 193 (58)b 135 (41) 4 (1) NA
0.43
27 18–46 141 (57) 105 (43)
105 (43) 142 (58)
17 (45) 21 (55)
0.86
273 (87) 159 (51) 60 (19)
40 (93) 29 (67) 8 (19)
0.45 0.050 1.00
301 (91) 128 (39) 75 (23)
11 (4) 134 (43) 169 (54)
7 (16) 16 (37) 20 (47)
0.12
332 (100) 0 (0) 0 (0)
164 52–512
144 83–377
0.82
115 43–331
51 32–75
48 30–116
0.78
NA
265 217–381
250 179–337
0.056
NA
189 102–384
248 120–339
0.52
207 116–473
11.8 10.5–12.9
12.4 11.1–13.2
0.063
12.6 11.7–13.4
40 38–42
41 39–43
0.009
NA
Because of rounding, not all percentages total 100. Missing data are detailed in supplementary Table S2, available at Annals of Oncology online. a Includes seven PMCC patients in TAX 327. b PS was collected by Karnofsky: KPS 90–100 was assigned ECOG PS 0, KPS 70–80 ECOG PS 1, KPS ≤60 ECOG PS 2–3. ALP, alkaline phosphatase; BMI, body mass index; IQR, interquartile range; LDH, lactate dehydrogenase; NA, not available; PSA, prostate-specific antigen; PSA-dt, PSA doubling time.
designed such that the majority of patients seen in routine practice would be eligible, thereby minimizing the efficacyeffectiveness gap. Strengths of this study include its large sample size with a focus on unselected patients treated in routine practice, and the
Volume 24 | No. 12 | December 2013
Not in trial In trial a (N = 314) (N = 43)
TAX 327 (N = 332)
Cycles per patient, median (range) Median 6 Range 1–15 Overall dose intensity, n (%) ≥95% 159 (51) 85–94% 64 (20) 75–84% 49 (16) <75% 42 (13) Reason for stopping treatment, n (%) Toxicity 71 (23) Treatment completed 91 (29) Disease progression 111 (35) Other 28 (9) Death within 30 days of last 13 (4) dose ≥1 infusion delayed more 74 (24) than 1 week, n (%) By cycle 6 51 (16) Dose reduction to ≤60 mg/ 70 (22) m2, n (%) By cycle 6 65 (21) Discontinuation b/o 51 (16) toxicity by cycle 6, n (%) Febrile neutropenia, n (%) 30 (9.6)
P
TAX 327 (N = 332)
8 2–14
0.001
9.5 1–11
24 (56) 10 (23) 6 (14) 3 (7)
0.33
NA
8 (19) 16 (37) 19 (44) 0 0
0.44
36 (11) 154 (46) 126 (38) 16 (5) 11 (3)
15 (35)
0.13
79 (24)
11 (26) 5 (12)
0.14 0.16
NA 41 (12)
5 (12) 3 (7)
0.22 0.17
NA NA
0
0.035
9 (2.7)
a
Includes seven PMCC patients in TAX 327. Because of rounding, not all percentages total 100. NA, not available; NS, not significant.
Figure 2. Overall Survival for nontrial patients (black), PMCC trial patients (red), and TAX 327 patients (green). Hazard ratio for the comparison of nontrial patients and PMCC trial patients (including seven treated in TAX 327).
comparison with individual patients treated on clinical trials by the same physicians within the same institution, as well as with results from the pivotal TAX-327 study.
doi:10.1093/annonc/mdt397 |
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
Age, years Median Range ECOG PS, n (%) 0 1 2 or 3 Comorbidities, n (%) 0–1 >1 BMI, kg/m2 Median Range Gleason, n (%) ≤7 8–10 Extent of disease, n (%) Bones Lymph nodes Visceral Year started, n (%) 2001–2003 2004–2007 2008–2011 PSA, µg/l Median IQR PSA-dt, days Median IQR LDH, U/l Median IQR ALP, U/l Median IQR Hemoglobin, g/dl Median IQR Albumin, g/l Median IQR
Table 2. Treatment and outcomes
original articles
Annals of Oncology
Table 3. Overall survival and PSA response rates
Censored, n (%) Median OS (95% CI) in months 1-year OS % (95% CI) 2-year OS % (95% CI) PSA response rate (95% CI), %
Not in trial (N = 314)
In trial (N = 43)
Pa
TAX 327 (N = 332)
81 (26) 13.6 (12.1–15.1) 57 (51–63) 23 (18–29) 45 (39–51)
5 (12) 20.4 (17.4–23.4) 74 (58–85) 39 (24–53) 54 (38–69)
0.055 0.089 0.032 0.025 NS
50 (15) 19.3 (17.6–21.3) 74 (69–78) 36 (31–41) 53c (47–59)
Pb <0.001
0.061
a
for comparison in trial versus not in trial. for comparison not in trial versus TAX 327. c Note that this is an underestimate because in the TAX327 trial men who had increased PSA within the first 12 weeks with subsequent decrease to <50% of baseline were not included as PSA responders [9]. CI, confidence interval; NS, not significant; PSA, prostate-specific antigen; OS, overall survival. b
Table 4. Predictive factors for overall survival
95% CI
0.74 1.00 0.75 1.40 0.93 1.05 1.17 0.92 0.74 1.57 3.23 1.11
0.53 0.87 0.66 1.21 0.88 0.77 0.89 0.88 0.69 1.37 2.47 1.02
P 1.05 1.16 0.86 1.62 0.97 1.44 1.53 0.97 0.80 1.80 4.21 1.21
0.089 0.98 <0.001 <0.001 0.003 0.76 0.26 0.002 <0.001 <0.001 <0.001 0.021
Multivariable HR 95% CI
P
0.82
0.51
1.31
0.40
0.80 1.07 0.92
0.65 0.85 0.85
0.98 1.34 1.00
0.031 0.57 0.046
0.98 0.84 1.21 1.73 1.11
0.92 0.74 0.94 1.05 0.98
1.04 0.95 1.55 2.86 1.25
0.41 0.007 0.14 0.032 0.12
CI, confidence interval; ECOG PS, Eastern Cooperative Group Performance Status; HR, hazard ratio. Statistically significant variables are shown in bold typeface.
Inherent limitations of this study are its retrospective nature and manual extraction of clinical data. This may have led to under-detection of comorbidities since they were not systematically reported in clinical notes, inaccurate assessment of performance status (often not mentioned explicitly), and nonavailability of relevant data (e.g. Gleason scores). For some men, no single reason for treatment discontinuation could be identified, making assignment to a specific group difficult or arbitrary. Rates of febrile neutropenia were based on clinical notes, which did not always provide the degree of fever or the absolute neutrophil count and may have led to an overestimation. In contrast, some patients might have been admitted to peripheral hospitals with underdetection of febrile neutropenia. We evaluated year of first administration of docetaxel to account for possible earlier use of chemotherapy in more recent years, and increasing use of further treatments given after it (although only six patients included in this analysis received abiraterone acetate). Other limitations are that patients treated at a referral center are likely to represent a selected cohort, and our inability to adjust for unmeasured confounders. In conclusion, our results support the hypothesis that application of results from clinical trials establishing the efficacy of a new cancer treatment leads to poorer outcomes and greater
| Templeton et al.
toxicity when applied in routine practice. Specifically, routine treatment of men with mCRPC with docetaxel is associated with a higher rate of febrile neutropenia and shorter survival than for men who are recruited to clinical trials. These findings are probably due to less rigorous patient selection for treatment in routine practice.
acknowledgements The authors thank Roy Lee from the PMCC central pharmacy, Darlene Dale from the PMCC Registry, and Jonathan Ng from the electronic data warehouse for assistance in extracting data. They also thank all of their patients and their coworkers and patients who participated in the TAX 327 study.
funding Supported by a research grant from the Canadian Institutes of Health Research (grant number CIHR MOP 119297). A. Templeton’s work was supported by a grant from the Swiss Cancer Research Foundation.
Volume 24 | No. 12 | December 2013
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
In trial Age, per 10 years increase log (time from initial diagnosis) ECOG PS Year of first administration Visceral metastasis, yes versus no Gleason Score, <8 versus 8–10 Baseline albumin, g/l Hemoglobin, g/dl log (ALP, U/l) log (LDH, U/l) log (PSA, µg/l)
Univariable HR
Annals of Oncology
disclosure IT has received research funding from Sanofi. All other authors have declared no conflicts of interest.
references
Volume 24 | No. 12 | December 2013
9. Berthold DR, Pond GR, Soban F et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer: updated survival in the TAX 327 study. J Clin Oncol 2008; 26: 242–245. 10. Berthold DR, Pond GR, Roessner M et al. Treatment of hormone-refractory prostate cancer with docetaxel or mitoxantrone: relationships between prostatespecific antigen, pain, and quality of life response and survival in the TAX-327 study. Clin Cancer Res 2008; 14: 2763–2767. 11. Scher HI, Halabi S, Tannock I et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J Clin Oncol 2008; 26: 1148–1159. 12. Howard DN, Chambers C, Cusano F. Efficacy versus effectiveness–docetaxel and prednisone in hormone refractory prostate cancer. J Oncol Pharm Pract 2008; 14: 45–49. 13. Goyal J, Nuhn P, Huang P et al. The effect of clinical trial participation versus nonparticipation on overall survival in men receiving first-line docetaxel-containing chemotherapy for metastatic castration-resistant prostate cancer. BJU Int 2012; 110: E575–E582. 14. Chin SN, Wang L, Moore M et al. A review of the patterns of docetaxel use for hormone-resistant prostate cancer at the Princess Margaret Hospital. Curr Oncol 2010; 17: 24–29. 15. Sekine I, Takada M, Nokihara H et al. Knowledge of efficacy of treatments in lung cancer is not enough, their clinical effectiveness should also be known. J Thorac Oncol 2006; 1: 398–402. 16. Djulbegovic B, Paul A. From efficacy to effectiveness in the face of uncertainty: indication creep and prevention creep. JAMA 2011; 305: 2005–2006. 17. Peppercorn JM, Weeks JC, Cook EF et al. Comparison of outcomes in cancer patients treated within and outside clinical trials: conceptual framework and structured review. Lancet 2004; 363: 263–270. 18. Niraula S, Seruga B, Ocana A et al. The price we pay for progress: a meta-analysis of harms of newly approved anticancer drugs. J Clin Oncol 2012; 30: 3012–3019. 19. Seruga B, Sterling L, Wang L et al. Reporting of serious adverse drug reactions of targeted anticancer agents in pivotal phase III clinical trials. J Clin Oncol 2011; 29: 174–185.
doi:10.1093/annonc/mdt397 |
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
1. Sargent D. What constitutes reasonable evidence of efficacy and effectiveness to guide oncology treatment decisions? Oncologist 2010; 15(Suppl 1): 19–23. 2. George SL. Reducing patient eligibility criteria in cancer clinical trials. J Clin Oncol 1996; 14: 1364–1370. 3. Tannock IF, de Wit R, Berry WR et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004; 351: 1502–1512. 4. Sridhar SS, Canil CM, Chi KN et al. A phase II study of the antisense oligonucleotide GTI-2040 plus docetaxel and prednisone as first-line treatment in castration-resistant prostate cancer. Cancer Chemother Pharmacol 2011; 67: 927–933. 5. Small E, Demkow T, Gerritsen WR. A phase III trial of GVAX immunotherapy for prostate cancer in combination with docetaxel versus docetaxel plus prednisone in symptomatic, castration-resistant prostate cancer. Proc ASCO GU Cancer Symp 2009; http://meetinglibrary.asco.org/content/20295-64. 6. Tannock IF, Fizazi K, Ivanov S et al. Aflibercept versus placebo in combination with docetaxel and prednisone for treatment of men with metastatic castration-resistant prostate cancer (VENICE): a phase 3, double-blind randomised trial. Lancet Oncol 2013; 14: 760–768. 7. Chi KN, Hotte SJ, Yu EY et al. Randomized phase II study of docetaxel and prednisone with or without OGX-011 in patients with metastatic castrationresistant prostate cancer. J Clin Oncol 2010; 28: 4247–4254. 8. Chi K, De Bono JS, Higano C. SYNERGY: A randomized phase III study comparing first-ine docetaxel/prednisone to docetaxel/prednisone plus custirsen in metastatic castration-resistant prostate canceR (mCRPC). J Clin Oncol 2011; 29: suppl, abstr TPS180. http://meetinglibrary.asco.org/content/77749–102.
original articles
Annals of Oncology Annals of Oncology 24: 2972–2977, 2013 doi:10.1093/annonc/mdt397 Published online 14 October 2013
Translating clinical trials to clinical practice: outcomes of men with metastatic castration resistant prostate cancer treated with docetaxel and prednisone in and out of clinical trials* A. J. Templeton1, F. E. Vera-Badillo1, L. Wang2, M. Attalla1, P. De Gouveia1, R. Leibowitz-Amit1, J. J. Knox1, M. Moore1, S. S. Sridhar1, A. M. Joshua1, G. R. Pond3, E. Amir1 & I. F. Tannock1* Divisions of 1Medical Oncology and Hematology, University of Toronto, Toronto; 2Biostatistics, Princess Margaret Cancer Center, Department of Medicine, University of Toronto, Toronto; 3Department of Oncology, McMaster University, Hamilton, Canada
Background: Multiple factors can influence outcomes of patients receiving identical interventions in clinical trials and in routine practice. Here, we compare outcomes of men with metastatic castrate-resistant prostate cancer (mCRPC) treated with docetaxel and prednisone in routine practice and in clinical trials. Patients and methods: We reviewed patients with mCRPC treated with docetaxel at Princess Margaret Cancer Centre. Primary outcomes were overall survival and PSA response rate. Secondary outcomes were reasons for discontinuation and febrile neutropenia. Outcomes were compared for men treated in routine practice and in clinical trials, and with data from the TAX 327 study. Results: From 2001 to 2011, 438 men were treated, of whom 357 received 3-weekly docetaxel as first-line chemotherapy: 314 in routine practice and 43 in clinical trials. Trial patients were younger and had better performance status. Median survival was 13.6 months [95% confidence interval (95% CI) 12.1–15.1 months] in routine practice and 20.4 months (95% CI 17.4–23.4 months, P = 0.007) within clinical trials, compared with 19.3 months (95% CI 17.6–21.3 months, P < 0.001) in the TAX 327 study. PSA response rates were 45%, 54%, and 53%, respectively (P = NS). Reasons for treatment discontinuation were similar although trial patients received more cycles (median: 6 versus 8 versus 9.5, P < 0.001). Rates of febrile neutropenia were 9.6, 0, and 3% (P < 0.001) while rates of death within 30 days of last dose were 4%, 0%, and 3%, respectively (P = NS). Conclusions: Survival of patients with mCRPC treated with docetaxel in routine practice is shorter than for men included in trials and is associated with more toxicity. Key words: docetaxel, routine practice, clinical trial, prostate cancer
introduction Data from clinical trials, although providing critical evidence of clinical activity, do not provide adequate information to judge the impact of a new treatments when used in the realworld setting [1]. The difference between results of clinical trials and outcomes in clinical practice may be explained by restrictive selection criteria for trials such as good prognosis and minimal comorbidities [2]. This is likely to be particularly important when older patients receive potentially toxic *Correspondence to: Dr Ian Tannock, Division of Medical Oncology and Hematology, Princess Margaret Cancer Centre, 610 University Avenue, Toronto, ON, M5G 2M9, Canada. Tel: +1-416-946-4501; Fax: +1-416-946-4563; E-mail: [email protected] *Results of this study were presented in part at the 2013 ASCO Genitourinary Cancers Symposium, 14–16 February 2013, Orlando, FL, Canada.
therapy, such as men with metastatic castration-resistant prostate cancer (mCRPC), since most will have pre-existing medical conditions. Despite an increasing role for secondary and tertiary hormonal-based therapies in men with mCRPC, standard initial treatment since 2004 remains chemotherapy with docetaxel and prednisone. This is based on the international TAX-327 study, which compared docetaxel in two schedules (3-weekly and weekly) with mitoxantrone (all with prednisone) [3]. The extent to which results from the pivotal trial translate to less selected patients treated in routine practice is unknown. Here we reviewed all men treated with docetaxel for mCRPC at our institution. We hypothesized that those treated outside of a clinical trial would have poorer survival and more toxicity than those recruited to clinical trials.
© The Author 2013. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
Received 1 July 2013; revised 2 August 2013; accepted 14 August 2013
original articles
Annals of Oncology
patients and methods study population All men who received at least one dose of docetaxel for mCRPC before January 2012 at Princess Margaret Cancer Centre (PMCC) in Toronto, Canada were identified through pharmacy records. Patients treated with weekly docetaxel were excluded as this was given to men who were frail or had a high risk of toxicity. Men who received docetaxel in the context of neoadjuvant or adjuvant trials, or as second-line chemotherapy for mCRPC were also excluded. Enrollment logs of prostate cancer trials were screened to identify patients treated with docetaxel in clinical trials. The institutional Research Ethics Board (REB) approved the study.
data collection
results patients and treatment
outcomes of interest The primary objective was to compare overall survival (OS) and PSA response rates in men with mCRPC treated with 3-weekly docetaxel in routine practice at PMCC with those receiving the same schedule at our institution in clinical trials evaluating docetaxel/prednisone alone (e.g. TAX327 [3]) or with GTI-2040 [4], GVAX [5], aflibercept [6], or OGX-011 [7, 8] (supplementary Table S1, available at Annals of Oncology online). We also compared these data with those for 332 men who received 3-weekly docetaxel in multiple institutions in the TAX-327 trial [3, 9, 10]. Secondary objectives were to compare reasons for treatment discontinuation, rates of febrile neutropenia (as described in the clinical notes), and death within 30 days of last administration of docetaxel.
statistical analysis Descriptive statistics for categorical variables were reported as percentages, and continuous variables were reported as medians and ranges. Categorical variables were compared using Fisher’s exact test and Kruskal–Wallis test for two and more variables, respectively. Continuous variables were compared by the Wilcoxon rank sum test. Dose intensity was calculated as the cumulative dose of docetaxel per body surface area divided by the time between first dose and end of the last cycle (and was 100% for 25 mg/m2/ week). If there was >12 weeks between two administrations of docetaxel, the patient was regarded as having two separate courses and only doses in the first course were included in calculation of dose intensity. Survival was calculated from the date of first docetaxel administration to death with censoring at date of last contact for patients alive at the cut-off date in August 2012. The Kaplan–Meier method was used to estimate survival, and survival curves were compared using the generalized Wilcoxon test; this test was preferred to log-rank since it places greater emphasis on the first part of the survival curves where the effect of treatment is more pronounced. The log-rank test was evaluated as a supportive analysis. Association of individual variables with survival was assessed using the Cox proportional hazard model. All variables with a P-value <0.1 in univariable analysis, variables with a P-value ≥0.1 but which changed the outcome of the variable ‘in trial’ by more than 10%, or variables where there was a significant interaction with the variable “in trial” (P < 0.05) were
Volume 24 | No. 12 | December 2013
Between February 2001 and December 2011, 438 men received treatment with docetaxel at PMCC, of whom 314 and 43 received first-line treatment every 3 weeks in routine practice and in a clinical trial, respectively (Figure 1). Of the 43 men in clinical trials, 7 received docetaxel as part of the TAX-327 study. Patients’ and tumor characteristics are presented in Table 1. Men receiving treatment in routine practice were older (median age 71 versus 68 years, P = 0.038) and more likely to have an ECOG performance status of 2 or higher (29% versus 3%, P < 0.001), when compared with those treated in a clinical trial.
treatment received Comparisons of outcomes of patients treated on trial and offtrial, as well as those in the TAX-327 study, are presented in Table 2. Trial patients received more cycles of docetaxel (median: 8 versus 6, P = 0.001) whereas overall dose-intensity, dose reductions and delays by cycle 6 were similar. The median number of cycles and delayed infusions in TAX-327 were similar to those in PMCC patients whereas dose reductions occurred less often (P < 0.001).
overall survival and PSA response Median OS for all nontrial patients receiving 3-weekly docetaxel with prednisone was 13.6 months, compared with 20.4 months for PMCC trial patients (P = 0.007) and 19.3 months in TAX327 (P < 0.001; Figure 2). One-year and 2-year OS rates were higher for patients treated in clinical trials (Table 3). There was no significant difference in survival between trial patients in our institution and men treated on the TAX-327 study (with or without exclusion of the 7 PMCC patients who participated in TAX-327). In an exploratory analysis patients who would have met the selection criteria for the TAX-327 trial lived significantly longer compared with those not meeting the selection criteria (median survival 16.7 versus 9.0 months,
doi:10.1093/annonc/mdt397 |
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
Dates of treatment and doses were obtained from the PMCC pharmacy. Clinical data were extracted manually from the electronic patient record (EPR) by MA, PdG, and AT. Laboratory values were obtained electronically from EPR; baseline was defined as within 28 days before the first dose of docetaxel. Dates of primary diagnosis and death were obtained from the PMCC Registry, which includes 99% of patients treated in our institution. Data from all sources were linked using the patients’ Medical Record Number.
retained in the multivariable model. Variables included treatment in a clinical trial, age at start of treatment, time from diagnosis until start of docetaxel, Eastern Cooperative Oncology Group performance status (ECOG PS), year of first administration, presence of visceral metastases, Gleason score (<8 versus 8–10); serum levels of prostate-specific antigen (PSA), lactate dehydrogenase (LDH), alkaline phosphatase (ALP), hemoglobin (Hgb), and albumin. Laboratory values were treated as continuous variables. Variables with non-normal distribution (e.g. PSA, LDH, and ALP) were log transformed. PSA response was defined as ≥50% decline of PSA compared with baseline maintained for at least 3 weeks. Men with a rise in PSA within the first 12 weeks of treatment were considered as PSA responders if subsequent values of PSA satisfied the above criterion, as per Prostate Cancer Working group 2 (PCWG2) criteria [11]. All statistical tests were two sided, and statistical significance was defined as P < 0.05. Analyses were carried out using SAS version v9.2 (SAS Institute, Inc.) and SPSS version 20 (IBM Corp.); curves were plotted with R version 2.13.1 (www.r-project.org). Corrections for multiple significance testing were not applied.
original articles
Annals of Oncology
HR = 0.45, P < 0.001). Details are provided in the supplementary material, available at Annals of Oncology online. Compared with nontrial patients, patients receiving docetaxel in a clinical trial at PMCC had a reduced hazard of death in univariable analysis (HR = 0.74, Wilcoxon P = 0.007, Figure 2) but ‘trial treatment’ was not an independent prognostic factor in multivariable analysis (Table 4). Factors associated with longer survival were higher hemoglobin, lower LDH at baseline, longer time from initial diagnosis, and more recent treatment. There was no significant difference in PSA response rates between patients treated in daily practice, in a clinical trial at PMCC, or on the TAX-327 study (Table 3).
toxicity Overall, the rates for treatment discontinuation because of toxicity were the same, even when comparing discontinuation rates for toxicity by cycle 6 (Table 2). Febrile neutropenia occurred in 9.6% of nontrial patients (in 60% of the cases during cycle 1); in none of the PMCC trial patients (P = 0.035) and in 3% of men in TAX-327 (P < 0.001). Deaths within 30 days of the last administration of docetaxel occurred in 4% of men treated in routine practice whereas no such events were observed in PMCC trial patients (P = 0.24), and 3% in the TAX-327 study (P = 0.68).
discussion The purpose of the present study was to compare outcomes of men with mCRPC treated with docetaxel and prednisone in routine practice and in clinical trials. We therefore reviewed all patients treated with first-line 3-weekly docetaxel for mCRPC at our institution and compared outcomes of men treated in routine clinical practice with those in clinical trials, both within our institution and in the pivotal TAX-327 study. The results support our hypothesis that men treated in routine practice have
| Templeton et al.
shorter survival and more toxicity than men treated within clinical trials. The difference in median survival between men treated in routine practice and in trials at PMCC (13.6 versus 20.4 months) was larger than reported in other series [12–14]. The difference between outcomes in clinical trials and clinical practice is known as the efficacy-effectiveness gap [1, 15, 16]. The outcome of men treated in clinical trials at our institution was similar to those treated in the TAX-327 study (median OS 19.3 months [9]), and we explored a potential ‘trial effect’ by performing multivariable analyses to assess whether treatment in a clinical trial was an independent prognostic factor. This might be expected because of strict rules for assessment or management of toxicity [13]. No such association was observed in multivariable analysis, suggesting that selection of patients with favorable prognosis accounts for the observed difference in OS between trial and nontrial patients. Patients treated in clinical trials were younger, had less comorbidity and better performance status. These findings are consistent with a meta-analysis that did not find evidence that enrollment in clinical trials is an independent prognostic factor [17]. Patients treated in routine practice in our institution had more toxicity than those treated in clinical trials: they received less cycles of docetaxel, but had a 9.6% rate of febrile neutropenia compared with 0 and 3% for PMCC trial patients and participants of the TAX-327 study, respectively. However, the rate of all-cause death within 30 days was not significantly different (4% in routine practice, 0% in trial patients, and 3% in TAX-327). The high rate of febrile neutropenia cautions the use of toxic treatment in men with mCRPC who would not have met criteria for trial selection. Registration trials use strict selection criteria to show maximum efficacy and minimal toxicity but most new therapies add toxicity [18], and these toxicities may only become apparent when treatment is applied in routine practice [19]. Ideally, registration trials should be
Volume 24 | No. 12 | December 2013
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
Figure 1. Men with metastatic castration resistant prostate cancer (mCRPC) treated with docetaxel at PMCC 2001–2011
original articles
Annals of Oncology Table 1. Patients’ characteristics Not in trial (N = 314)
In trial a (N = 43)
71 44–90
68 51–83
P
0.038
83 (30) 116 (41) 81 (29)
23 (62) 13 (35) 1 (3)
<0.001
196 (62) 118 (38)
32 (74) 11 (26)
0.13
27 17–42
27 17–45
68 42–92 193 (58)b 135 (41) 4 (1) NA
0.43
27 18–46 141 (57) 105 (43)
105 (43) 142 (58)
17 (45) 21 (55)
0.86
273 (87) 159 (51) 60 (19)
40 (93) 29 (67) 8 (19)
0.45 0.050 1.00
301 (91) 128 (39) 75 (23)
11 (4) 134 (43) 169 (54)
7 (16) 16 (37) 20 (47)
0.12
332 (100) 0 (0) 0 (0)
164 52–512
144 83–377
0.82
115 43–331
51 32–75
48 30–116
0.78
NA
265 217–381
250 179–337
0.056
NA
189 102–384
248 120–339
0.52
207 116–473
11.8 10.5–12.9
12.4 11.1–13.2
0.063
12.6 11.7–13.4
40 38–42
41 39–43
0.009
NA
Because of rounding, not all percentages total 100. Missing data are detailed in supplementary Table S2, available at Annals of Oncology online. a Includes seven PMCC patients in TAX 327. b PS was collected by Karnofsky: KPS 90–100 was assigned ECOG PS 0, KPS 70–80 ECOG PS 1, KPS ≤60 ECOG PS 2–3. ALP, alkaline phosphatase; BMI, body mass index; IQR, interquartile range; LDH, lactate dehydrogenase; NA, not available; PSA, prostate-specific antigen; PSA-dt, PSA doubling time.
designed such that the majority of patients seen in routine practice would be eligible, thereby minimizing the efficacyeffectiveness gap. Strengths of this study include its large sample size with a focus on unselected patients treated in routine practice, and the
Volume 24 | No. 12 | December 2013
Not in trial In trial a (N = 314) (N = 43)
TAX 327 (N = 332)
Cycles per patient, median (range) Median 6 Range 1–15 Overall dose intensity, n (%) ≥95% 159 (51) 85–94% 64 (20) 75–84% 49 (16) <75% 42 (13) Reason for stopping treatment, n (%) Toxicity 71 (23) Treatment completed 91 (29) Disease progression 111 (35) Other 28 (9) Death within 30 days of last 13 (4) dose ≥1 infusion delayed more 74 (24) than 1 week, n (%) By cycle 6 51 (16) Dose reduction to ≤60 mg/ 70 (22) m2, n (%) By cycle 6 65 (21) Discontinuation b/o 51 (16) toxicity by cycle 6, n (%) Febrile neutropenia, n (%) 30 (9.6)
P
TAX 327 (N = 332)
8 2–14
0.001
9.5 1–11
24 (56) 10 (23) 6 (14) 3 (7)
0.33
NA
8 (19) 16 (37) 19 (44) 0 0
0.44
36 (11) 154 (46) 126 (38) 16 (5) 11 (3)
15 (35)
0.13
79 (24)
11 (26) 5 (12)
0.14 0.16
NA 41 (12)
5 (12) 3 (7)
0.22 0.17
NA NA
0
0.035
9 (2.7)
a
Includes seven PMCC patients in TAX 327. Because of rounding, not all percentages total 100. NA, not available; NS, not significant.
Figure 2. Overall Survival for nontrial patients (black), PMCC trial patients (red), and TAX 327 patients (green). Hazard ratio for the comparison of nontrial patients and PMCC trial patients (including seven treated in TAX 327).
comparison with individual patients treated on clinical trials by the same physicians within the same institution, as well as with results from the pivotal TAX-327 study.
doi:10.1093/annonc/mdt397 |
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
Age, years Median Range ECOG PS, n (%) 0 1 2 or 3 Comorbidities, n (%) 0–1 >1 BMI, kg/m2 Median Range Gleason, n (%) ≤7 8–10 Extent of disease, n (%) Bones Lymph nodes Visceral Year started, n (%) 2001–2003 2004–2007 2008–2011 PSA, µg/l Median IQR PSA-dt, days Median IQR LDH, U/l Median IQR ALP, U/l Median IQR Hemoglobin, g/dl Median IQR Albumin, g/l Median IQR
Table 2. Treatment and outcomes
original articles
Annals of Oncology
Table 3. Overall survival and PSA response rates
Censored, n (%) Median OS (95% CI) in months 1-year OS % (95% CI) 2-year OS % (95% CI) PSA response rate (95% CI), %
Not in trial (N = 314)
In trial (N = 43)
Pa
TAX 327 (N = 332)
81 (26) 13.6 (12.1–15.1) 57 (51–63) 23 (18–29) 45 (39–51)
5 (12) 20.4 (17.4–23.4) 74 (58–85) 39 (24–53) 54 (38–69)
0.055 0.089 0.032 0.025 NS
50 (15) 19.3 (17.6–21.3) 74 (69–78) 36 (31–41) 53c (47–59)
Pb <0.001
0.061
a
for comparison in trial versus not in trial. for comparison not in trial versus TAX 327. c Note that this is an underestimate because in the TAX327 trial men who had increased PSA within the first 12 weeks with subsequent decrease to <50% of baseline were not included as PSA responders [9]. CI, confidence interval; NS, not significant; PSA, prostate-specific antigen; OS, overall survival. b
Table 4. Predictive factors for overall survival
95% CI
0.74 1.00 0.75 1.40 0.93 1.05 1.17 0.92 0.74 1.57 3.23 1.11
0.53 0.87 0.66 1.21 0.88 0.77 0.89 0.88 0.69 1.37 2.47 1.02
P 1.05 1.16 0.86 1.62 0.97 1.44 1.53 0.97 0.80 1.80 4.21 1.21
0.089 0.98 <0.001 <0.001 0.003 0.76 0.26 0.002 <0.001 <0.001 <0.001 0.021
Multivariable HR 95% CI
P
0.82
0.51
1.31
0.40
0.80 1.07 0.92
0.65 0.85 0.85
0.98 1.34 1.00
0.031 0.57 0.046
0.98 0.84 1.21 1.73 1.11
0.92 0.74 0.94 1.05 0.98
1.04 0.95 1.55 2.86 1.25
0.41 0.007 0.14 0.032 0.12
CI, confidence interval; ECOG PS, Eastern Cooperative Group Performance Status; HR, hazard ratio. Statistically significant variables are shown in bold typeface.
Inherent limitations of this study are its retrospective nature and manual extraction of clinical data. This may have led to under-detection of comorbidities since they were not systematically reported in clinical notes, inaccurate assessment of performance status (often not mentioned explicitly), and nonavailability of relevant data (e.g. Gleason scores). For some men, no single reason for treatment discontinuation could be identified, making assignment to a specific group difficult or arbitrary. Rates of febrile neutropenia were based on clinical notes, which did not always provide the degree of fever or the absolute neutrophil count and may have led to an overestimation. In contrast, some patients might have been admitted to peripheral hospitals with underdetection of febrile neutropenia. We evaluated year of first administration of docetaxel to account for possible earlier use of chemotherapy in more recent years, and increasing use of further treatments given after it (although only six patients included in this analysis received abiraterone acetate). Other limitations are that patients treated at a referral center are likely to represent a selected cohort, and our inability to adjust for unmeasured confounders. In conclusion, our results support the hypothesis that application of results from clinical trials establishing the efficacy of a new cancer treatment leads to poorer outcomes and greater
| Templeton et al.
toxicity when applied in routine practice. Specifically, routine treatment of men with mCRPC with docetaxel is associated with a higher rate of febrile neutropenia and shorter survival than for men who are recruited to clinical trials. These findings are probably due to less rigorous patient selection for treatment in routine practice.
acknowledgements The authors thank Roy Lee from the PMCC central pharmacy, Darlene Dale from the PMCC Registry, and Jonathan Ng from the electronic data warehouse for assistance in extracting data. They also thank all of their patients and their coworkers and patients who participated in the TAX 327 study.
funding Supported by a research grant from the Canadian Institutes of Health Research (grant number CIHR MOP 119297). A. Templeton’s work was supported by a grant from the Swiss Cancer Research Foundation.
Volume 24 | No. 12 | December 2013
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
In trial Age, per 10 years increase log (time from initial diagnosis) ECOG PS Year of first administration Visceral metastasis, yes versus no Gleason Score, <8 versus 8–10 Baseline albumin, g/l Hemoglobin, g/dl log (ALP, U/l) log (LDH, U/l) log (PSA, µg/l)
Univariable HR
Annals of Oncology
disclosure IT has received research funding from Sanofi. All other authors have declared no conflicts of interest.
references
Volume 24 | No. 12 | December 2013
9. Berthold DR, Pond GR, Soban F et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer: updated survival in the TAX 327 study. J Clin Oncol 2008; 26: 242–245. 10. Berthold DR, Pond GR, Roessner M et al. Treatment of hormone-refractory prostate cancer with docetaxel or mitoxantrone: relationships between prostatespecific antigen, pain, and quality of life response and survival in the TAX-327 study. Clin Cancer Res 2008; 14: 2763–2767. 11. Scher HI, Halabi S, Tannock I et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J Clin Oncol 2008; 26: 1148–1159. 12. Howard DN, Chambers C, Cusano F. Efficacy versus effectiveness–docetaxel and prednisone in hormone refractory prostate cancer. J Oncol Pharm Pract 2008; 14: 45–49. 13. Goyal J, Nuhn P, Huang P et al. The effect of clinical trial participation versus nonparticipation on overall survival in men receiving first-line docetaxel-containing chemotherapy for metastatic castration-resistant prostate cancer. BJU Int 2012; 110: E575–E582. 14. Chin SN, Wang L, Moore M et al. A review of the patterns of docetaxel use for hormone-resistant prostate cancer at the Princess Margaret Hospital. Curr Oncol 2010; 17: 24–29. 15. Sekine I, Takada M, Nokihara H et al. Knowledge of efficacy of treatments in lung cancer is not enough, their clinical effectiveness should also be known. J Thorac Oncol 2006; 1: 398–402. 16. Djulbegovic B, Paul A. From efficacy to effectiveness in the face of uncertainty: indication creep and prevention creep. JAMA 2011; 305: 2005–2006. 17. Peppercorn JM, Weeks JC, Cook EF et al. Comparison of outcomes in cancer patients treated within and outside clinical trials: conceptual framework and structured review. Lancet 2004; 363: 263–270. 18. Niraula S, Seruga B, Ocana A et al. The price we pay for progress: a meta-analysis of harms of newly approved anticancer drugs. J Clin Oncol 2012; 30: 3012–3019. 19. Seruga B, Sterling L, Wang L et al. Reporting of serious adverse drug reactions of targeted anticancer agents in pivotal phase III clinical trials. J Clin Oncol 2011; 29: 174–185.
doi:10.1093/annonc/mdt397 |
Downloaded from http://annonc.oxfordjournals.org/ at Kainan University on May 10, 2015
1. Sargent D. What constitutes reasonable evidence of efficacy and effectiveness to guide oncology treatment decisions? Oncologist 2010; 15(Suppl 1): 19–23. 2. George SL. Reducing patient eligibility criteria in cancer clinical trials. J Clin Oncol 1996; 14: 1364–1370. 3. Tannock IF, de Wit R, Berry WR et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004; 351: 1502–1512. 4. Sridhar SS, Canil CM, Chi KN et al. A phase II study of the antisense oligonucleotide GTI-2040 plus docetaxel and prednisone as first-line treatment in castration-resistant prostate cancer. Cancer Chemother Pharmacol 2011; 67: 927–933. 5. Small E, Demkow T, Gerritsen WR. A phase III trial of GVAX immunotherapy for prostate cancer in combination with docetaxel versus docetaxel plus prednisone in symptomatic, castration-resistant prostate cancer. Proc ASCO GU Cancer Symp 2009; http://meetinglibrary.asco.org/content/20295-64. 6. Tannock IF, Fizazi K, Ivanov S et al. Aflibercept versus placebo in combination with docetaxel and prednisone for treatment of men with metastatic castration-resistant prostate cancer (VENICE): a phase 3, double-blind randomised trial. Lancet Oncol 2013; 14: 760–768. 7. Chi KN, Hotte SJ, Yu EY et al. Randomized phase II study of docetaxel and prednisone with or without OGX-011 in patients with metastatic castrationresistant prostate cancer. J Clin Oncol 2010; 28: 4247–4254. 8. Chi K, De Bono JS, Higano C. SYNERGY: A randomized phase III study comparing first-ine docetaxel/prednisone to docetaxel/prednisone plus custirsen in metastatic castration-resistant prostate canceR (mCRPC). J Clin Oncol 2011; 29: suppl, abstr TPS180. http://meetinglibrary.asco.org/content/77749–102.
original articles