`The
`
`of
`
` medicine
`
`original article
`
`Docetaxel and Estramustine Compared
`with Mitoxantrone and Prednisone
`for Advanced Refractory Prostate Cancer
`Daniel P. Petrylak, M.D., Catherine M. Tangen, Dr.P.H., Maha H.A. Hussain, M.D.,
`Primo N. Lara, Jr., M.D., Jeffrey A. Jones, M.D., Mary Ellen Taplin, M.D.,
`Patrick A. Burch, M.D., Donna Berry, Ph.D., R.N., Carol Moinpour, Ph.D.,
`Manish Kohli, M.D., Mitchell C. Benson, M.D., Eric J. Small, M.D.,
`Derek Raghavan, M.D., Ph.D., and E. David Crawford, M.D.
`
`abstract
`
`background
`Mitoxantrone-based chemotherapy palliates pain without extending survival in men
`with progressive androgen-independent prostate cancer. We compared docetaxel plus
`estramustine with mitoxantrone plus prednisone in men with metastatic, hormone-
`independent prostate cancer.
`methods
`We randomly assigned 770 men to one of two treatments, each given in 21-day cycles:
`280 mg of estramustine three times daily on days 1 through 5, 60 mg of docetaxel per
`square meter of body-surface area on day 2, and 60 mg of dexamethasone in three di-
`vided doses before docetaxel, or 12 mg of mitoxantrone per square meter on day 1 plus
`5 mg of prednisone twice daily. The primary end point was overall survival; secondary
`end points were progression-free survival, objective response rates, and post-treatment
`declines of at least 50 percent in serum prostate-specific antigen (PSA) levels.
`results
`Of 674 eligible patients, 338 were assigned to receive docetaxel and estramustine and
`336 to receive mitoxantrone and prednisone. In an intention-to-treat analysis, the medi-
`an overall survival was longer in the group given docetaxel and estramustine than in the
`group given mitoxantrone and prednisone (17.5 months vs. 15.6 months, P=0.02 by
`the log-rank test), and the corresponding hazard ratio for death was 0.80 (95 percent
`confidence interval, 0.67 to 0.97). The median time to progression was 6.3 months in
`the group given docetaxel and estramustine and 3.2 months in the group given mito-
`xantrone and prednisone (P<0.001 by the log-rank test). PSA declines of at least 50 per-
`cent occurred in 50 percent and 27 percent of patients, respectively (P<0.001), and ob-
`jective tumor responses were observed in 17 percent and 11 percent of patients with
`bidimensionally measurable disease, respectively (P=0.30). Grade 3 or 4 neutropenic
`fevers (P=0.01), nausea and vomiting (P<0.001), and cardiovascular events (P=0.001)
`were more common among patients receiving docetaxel and estramustine than among
`those receiving mitoxantrone and prednisone. Pain relief was similar in both groups.
`conclusions
`The improvement in median survival of nearly two months with docetaxel and estra-
`mustine, as compared with mitoxantrone and prednisone, provides support for this
`approach in men with metastatic, androgen-independent prostate cancer.
`
`From Columbia University, Herbert Irving
`Comprehensive Cancer Center, New York
`(D.P.P., M.C.B.); Southwest Oncology
`Group Statistical Center, Seattle (C.M.T.,
`C.M.); the University of Michigan Com-
`prehensive Cancer Center, Ann Arbor
`(M.H.A.H.); the University of California,
`Davis, Sacramento (P.N.L.); Baylor College
`of Medicine, Houston (J.A.J.); the Univer-
`sity of Massachusetts Medical Center, Wor-
`cester (M.E.T.); the Mayo Clinic, Rochester,
`Minn. (P.A.B.); Biobehavioral Nursing and
`Health Systems, University of Washington,
`Seattle (D.B.); the University of Arkansas
`for Medical Science, Little Rock (M.K.);
`the University of California, San Francisco,
`Cancer Center, San Francisco (E.J.S.); the
`Cleveland Clinic Foundation, Cleveland
`(D.R.); and the University of Colorado
`Health Science Center, Denver (E.D.C.).
`Address reprint requests to the Southwest
`Oncology Group (S9916) Operations Of-
`fice, 14980 Omicron Dr., San Antonio, TX
`78245-3217, or at pubs@swog.org.
`
`N Engl J Med 2004;351:1513-20.
`Copyright © 2004 Massachusetts Medical Society.
`
`n engl j med
`
`351;15
`
`www.nejm.org october
`
`7, 2004
`
`1513
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at SIDLEY AUSTIN LLP on July 15, 2016. For personal use only. No other uses without permission.
`
` Copyright © 2004 Massachusetts Medical Society. All rights reserved.
`
`JANSSEN EXHIBIT 2059
`Amerigen v. Janssen IPR2016-00286
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`m
`
` new england journal
`The
`
` medicine
`of
`
`en with newly diagnosed
`meta-
`
`static prostate cancer have a rapid re-
`sponse to surgical or medical castration,
`with improvement in bone pain, regression of soft-
`tissue metastases, and a decline in serum prostate-
`1
`specific antigen (PSA) levels.
` Nevertheless, in vir-
`tually all patients the tumor ultimately becomes
`androgen-independent a median of 18 to 24 months
`1,2
`after castration.
` During this terminal phase in the
`natural history of prostate cancer, approximately
`29,900 affected men in the United States will die
`3
`
`Patients with metastatic
`of the disease in 2004.
`androgen-independent prostate cancer have a pro-
`gressive and morbid disease with a median survival
`of 10 to 12 months; currently, no treatment offers
`a survival advantage. Chemotherapy for androgen-
`4
`independent prostate cancer is ineffective
`: mito-
`xantrone plus prednisone or hydrocortisone, the
`current standard of care, palliates bone pain in ap-
`proximately 30 percent of patients but does not im-
`5,6
`prove survival.
`Immunohistochemical studies have demonstrat-
`ed that the antiapoptotic protein Bcl-2 is increased
`in metastatic cells from androgen-independent
`7
`prostate tissue.
` Docetaxel, a taxane used to treat
`a variety of solid tumors, phosphorylates Bcl-2 in
`vitro, leading to its inactivation and to eventual cell
`8
`9
`death by apoptosis.
` Estramustine,
` which disrupts
`microtubule-associated proteins in vitro, has syner-
`gistic activity with docetaxel against human pros-
`10,11
` Phase 1 and 2 studies of
`tate-cancer cell lines.
`docetaxel plus estramustine in men with androgen-
`independent prostate cancer demonstrated a de-
`cline in serum PSA levels of at least 50 percent in 68
`to 84 percent of patients, a measurable disease re-
`sponse in 28 to 55 percent, and a median survival of
`11-14
`up to 23 months.
` These data provided the foun-
`dation for this prospective, randomized, phase 3
`trial (Southwest Oncology Group [SWOG] Inter-
`group protocol 99-16), which we conducted to de-
`termine whether docetaxel plus estramustine im-
`proves survival over that afforded by mitoxantrone
`plus prednisone in men with androgen-indepen-
`dent prostate cancer.
`
`methods
`
`patients
`Patients were enrolled by institutions affiliated with
`SWOG, Cancer and Leukemia Group B, the North
`Central Cancer Treatment Group, the Clinical Trials
`
`Support Unit, and the extended participation pro-
`ject program through the National Cancer Institute.
`Eligibility required pathologically confirmed ade-
`nocarcinoma of the prostate and progressive meta-
`static disease (stage D1 or D2) despite androgen-
`ablative therapy and cessation of antiandrogen
`treatment. Criteria for progressive disease were pro-
`gression of a bidimensionally measurable lesion,
`as assessed within 28 days before study registration;
`progression of disease that could be evaluated but
`not measured (e.g., by bone scanning), as assessed
`within 42 days before registration; or an increase in
`the serum PSA level over the baseline level in at least
`two consecutive samples obtained at least 7 days
`15
`
`apart.
`Antiandrogen therapy was discontinued
`before registration, at least six weeks before in the
`case of nilutamide or bicalutamide and four weeks
`before in the case of flutamide or other secondary
`hormonal therapy. To ensure continued androgen
`ablation, patients continued taking luteinizing-
`hormone–releasing hormone agonists throughout
`study treatment. Patients were required to discon-
`tinue bisphosphonates at least 28 days before reg-
`istration. Prior radiotherapy (to less than 30 percent
`of the bone marrow only) or one prior systemic
`therapy (except with estramustine, taxanes, anthra-
`cyclines, or mitoxantrone) was permitted if at least
`four weeks had elapsed since the completion of
`that therapy. Adequate renal, hepatic, and cardiac
`function and a SWOG performance-status score
`of 0 to 2 (a performance status of 3 was allowed if
`the score was due to bone pain) were also required.
`Patients were ineligible if they had received prior
`radioisotope or anticoagulant therapy (excluding
`aspirin), had active thrombophlebitis or hypercoag-
`ulability, had a history of pulmonary embolus, or
`pleural effusions or ascites.
`
`stratification
`Patients were classified at registration according
`to the following factors: type of progression (i.e.,
`progression of disease that could be measured or
`evaluated vs. increasing PSA level alone), grade of
`bone pain according to the Common Terminology
`Criteria of the National Cancer Institute (grade 1
`[mild, not interfering with function] vs. grade 2
`[moderate pain interfering with function but not
`interfering with the activities of daily life], grade 3
`[severe pain, severely interfering with the activities
`of daily living], or grade 4 [disabling pain]), and
`SWOG performance-status score (0 or 1 vs. 2 or
`
`1514
`
`n engl j med
`
`351;15
`
`www.nejm.org october
`
`,
`
`7
`
`2004
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at SIDLEY AUSTIN LLP on July 15, 2016. For personal use only. No other uses without permission.
`
` Copyright © 2004 Massachusetts Medical Society. All rights reserved.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`chemotherapy for metastatic prostate cancer
`
`16
`3).
` All patients provided written informed con-
`sent, and the study was approved by the institution-
`al review board of each participating institution.
`
`treatment
`Patients were randomly assigned to one of two
`treatments, each given in 21-day cycles: 280 mg of
`estramustine (Emcyt, Pfizer) three times daily one
`hour before or two hours after meals on days 1
`through 5 plus 60 mg of docetaxel (Taxotere, Aven-
`tis) per square meter of body-surface area intrave-
`nously on day 2, preceded by 60 mg of dexametha-
`sone orally in three divided doses, starting the night
`before docetaxel, or 12 mg of mitoxantrone (No-
`
`intravenously on
`vantrone, OSI) per square meter
`day 1 plus 5 mg of prednisone twice daily. Doses
`of docetaxel and mitoxantrone were increased to
`70 mg per square meter and 14 mg per square meter,
`respectively, if no grade 3 or 4 adverse events were
`observed during the first cycle. A report that pro-
`phylactic anticoagulation decreased estramustine-
`associated vascular effects prompted an amend-
`ment of the protocol on January 15, 2001, to include
`daily warfarin (2 mg) plus aspirin (325 mg) in the
`16
`group assigned to receive estramustine.
` Treat-
`ment continued until disease progression or unac-
`ceptable adverse effects occurred or until a maxi-
`mum of 12 cycles of docetaxel and estramustine or
`144 mg of mitoxantrone per square meter had
`been administered.
`
`evaluation
`The pretreatment evaluation included a history tak-
`ing, a physical examination in which weight and
`performance status were recorded, computed to-
`mography (CT) of the abdomen and pelvis, bone
`scanning, nuclear ventriculography (multiple gated
`acquisition [MUGA] scanning), a complete blood
`count, and measurement of serum PSA, serum cre-
`atinine, and serum testosterone. MUGA scans were
`repeated every four cycles among patients in the
`group given mitoxantrone and prednisone. At every
`cycle, the pretreatment evaluation was repeated (ex-
`cluding MUGA scanning, measurement of serum
`testosterone, and baseline imaging studies). Ad-
`verse events were evaluated by means of the Com-
`mon Toxicity Criteria of the National Cancer Insti-
`tute, version 2.0. Imaging studies were repeated
`every six cycles; if positive, they were repeated every
`three cycles.
`Objective responses were defined on the basis
`
`of the sum of bidimensional measurements of
`metastatic lesions. Confirmed objective responses
`required a follow-up scan (a minimum of four
`weeks later) that demonstrated a continued re-
`sponse. Progression was defined by one of the
`following: a 50 percent increase or an increase of
`2
`10 cm
`, whichever was smaller, in the sum of mea-
`surements of metastatic lesions over the sum at
`baseline; a clear worsening of nonmeasurable dis-
`ease; reappearance of any lesion that had disap-
`peared; appearance of any new lesion; or death.
`A confirmed partial response of nonmeasurable
`disease was defined as a reduction by more than 50
`percent over baseline in two or more PSA measure-
`ments obtained at least four weeks apart, with no
`evidence of disease progression on imaging. Pro-
`gressive disease was defined as a 25 percent in-
`crease in the serum PSA level — to at least 5 ng per
`milliliter — over the last preregistration measure-
`ment, with confirmation of the increase at least four
`weeks later. For patients with a decrease in serum
`PSA levels during the trial, progressive disease was
`defined as a confirmed increase of 25 percent, to at
`14
`least 5 ng per milliliter over the nadir.
`
`statistical analysis
`The primary objective of the study was to compare
`overall survival in the two groups. Assuming an ex-
`ponential distribution of survival times, 3.5 years
`for accrual, an additional year of follow-up, and a
`sample size of 310 patients per group, this study
`had a statistical power of 0.80 to detect an improve-
`ment of 33 percent in median survival in the group
`given docetaxel and estramustine, as compared with
`the group given mitoxantrone and prednisone, with
`the use of a one-sided log-rank test at a P value of
`0.025. Interim analyses were to be conducted when
`half the patients had been enrolled and again when
`enrollment was complete. The null and alternative
`hypotheses were to be tested at a one-sided P level
`of 0.0025 at each analysis. The significance level
`for the final analysis, performed one year after study
`closure, was specified as a one-sided P value of 0.02.
`Jour-
`However, in accordance with the policy of the
`nal,
` only two-sided P values are reported. Secondary
`end points included progression-free survival, the
`objective-response rate, the rate of PSA response
`(defined as a decline in the serum PSA level of at
`least 50 percent), and adverse events. The data set
`was locked and analyzed on March 9, 2004.
`Kaplan–Meier curves were used to estimate rates
`
`n engl j med
`
`351;15
`
`www.nejm.org october
`
`7, 2004
`
`1515
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at SIDLEY AUSTIN LLP on July 15, 2016. For personal use only. No other uses without permission.
`
` Copyright © 2004 Massachusetts Medical Society. All rights reserved.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
` new england journal
`The
`
` medicine
`of
`
`of overall survival and progression-free survival.
`Survival was defined from the date of randomiza-
`tion to the date of death from any cause or censored
`at the date of last contact. Progression-free survival
`was defined as the time from randomization to the
`first occurrence of objective or PSA progression or
`death from any cause. The general chi-square test
`was used to compare rates of response (objective
`and PSA) and adverse events between the two treat-
`ment groups. All analyses were performed with the
`use of SAS software, version 9.0.
`The study was designed by the Genitourinary
`
`Table 1. Baseline Characteristics of the Patients.
`
`Characteristic
`
`No. randomized
`No. eligible
`Age (yr)
`Median
`Range
`Race or ethnic group (%)*
`White
`Black
`Hispanic
`Asian
`Unknown
`SWOG performance-status score (%)
`0 or 1
`2 or 3
`Type of progression (%)
`Measurable or able to be evaluated
`Increased PSA only
`PSA (ng/ml)
`Median
`Range
`Sites of disease (%) *
`Bone
`Soft tissue
`Lymph node
`Liver
`Lung
`Bone pain (%)
`Grade <2
`Grade ≥2
`
`Docetaxel
`and Estramustine
`
`Mitoxantrone
`and Prednisone
`
`386
`338
`
`70
`47–88
`
`86
`12
`7
`1
`1
`
`90
`10
`
`81
`19
`
`384
`336
`
`70
`43–87
`
`82
`15
`6
`1
`1
`
`88
`12
`
`82
`18
`
`84
`0.1–10,820
`
`90
`0.1–8378
`
`84
`
`24
`8
`10
`
`64
`36
`
`88
`
`26
`9
`10
`
`64
`36
`
`* Patients could be included in more than one category. Race or ethnic group
`was self-reported.
`
`Committee of SWOG and was approved by the
`Cancer Treatment and Evaluation Program of the
`National Cancer Institute. The SWOG Statistical
`Center received funding from Aventis Pharmaceu-
`ticals for the additional cost of collecting data on
`the quality of life. Aventis was allowed to review the
`protocol and make comments before enrollment
`began. Aventis had no access to the data but re-
`ceived a semiannual summary of enrollment and
`adverse events.
`
`results
`
`characteristics of the patients
`A total of 770 patients were enrolled between Octo-
`ber 1999 and January 2003. Ninety-six patients (12
`percent) were found to be ineligible: 30 owing to
`the lack of adequate withdrawal of antiandrogen or
`other hormonal therapy, 11 because of missing
`documentation, 31 because of inadequate baseline
`laboratory studies, 17 because of rising PSA levels
`without evidence of metastatic disease, and 7 for
`miscellaneous reasons. The baseline characteristics
`of the 674 eligible patients in both treatment groups
`were similar (Table 1). The sole evidence of disease
`progression was a rising PSA level in 18 percent of
`patients.
`
`treatment
`There were 11 major protocol deviations. Six pa-
`tients in the group given docetaxel and estramus-
`tine and four patients in the group given mitoxan-
`trone and prednisone did not receive the assigned
`treatment and were not included in the evaluation
`of adverse events. One patient in the latter group
`who received intermittent radiotherapy while re-
`ceiving the assigned treatment, a major protocol de-
`viation, was included in the evaluation of adverse
`events. Six patients who discontinued treatment
`within one week after starting mitoxantrone and
`prednisone (four men) or docetaxel and estramus-
`tine (two men) were not included in the evaluation
`of adverse events; however, in the case of all these
`men, the reported results and statistical analyses
`are based on the treatment group to which the pa-
`tients were assigned.
`
`response and survival
`During a median follow-up of 32 months, 217 of
`the 338 patients in the group given docetaxel and
`estramustine died (64 percent), as did 235 of the
`336 patients in the group given mitoxantrone and
`
`1516
`
`n engl j med
`
`351;15
`
`www.nejm.org october
`
`,
`
`7
`
`2004
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at SIDLEY AUSTIN LLP on July 15, 2016. For personal use only. No other uses without permission.
`
` Copyright © 2004 Massachusetts Medical Society. All rights reserved.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`chemotherapy for metastatic prostate cancer
`
`under review, a fourth was due to gastrointestinal
`bleeding thought to be due to aspirin, a fifth was
`caused by sepsis arising from necrotic prostate
`tissue, a sixth (due to liver and renal failure, atrial
`fibrillation, and pulmonary edema) occurred with-
`in a week after treatment was started, a seventh was
`associated with granulocytopenia and neutropenia,
`and the eighth was caused by a respiratory tract in-
`
`P=0.02
`
`Docetaxel+estramustine
`(217 deaths; median, 17.5 mo)
`
`Mitoxantrone+prednisone
`(235 deaths; median, 15.6 mo)
`
`12
`
`24
`Months after Enrollment
`
`36
`
`48
`
`100
`
`80
`
`60
`
`40
`
`20
`
`Overall Survival (%)
`
`0
`
`0
`
`No. at Risk
`Docetaxel+
`estramustine
`Mitoxantrone+
`prednisone
`
`338
`
`336
`
`218
`
`185
`
`60
`
`50
`
`13
`
`10
`
`Figure 1. Kaplan–Meier Estimates of Overall Survival among Men with Andro-
`gen-Independent Prostate Cancer Treated with Mitoxantrone and Prednisone
`or Docetaxel and Estramustine.
`
`Docetaxel+estramustine
`(311 events; median, 6.3 mo)
`
`P<0.001
`
`Mitoxantrone+prednisone
`(312 events; median, 3.2 mo)
`
`12
`
`24
`Months after Enrollment
`
`36
`
`48
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`0
`
`Progression-free Survival (%)
`
`No. at Risk
`Docetaxel+
`estramustine
`Mitoxantrone+
`prednisone
`
`338
`
`176
`
`336
`
`116
`
`75
`
`56
`
`20
`
`15
`
`6
`
`6
`
`Figure 2. Kaplan–Meier Estimates of Progression-free Survival among Men
`with Androgen-Independent Prostate Cancer Treated with Mitoxantrone
`and Prednisone or Docetaxel and Estramustine.
`
`prednisone (70 percent). According to the inten-
`tion-to-treat analysis, the median survival was 17.5
`months among the patients assigned to docetaxel
`and estramustine and 15.6 months among the pa-
`tients assigned to mitoxantrone and prednisone
`(P=0.02) (Fig. 1); the corresponding hazard ratio
`for death was 0.80 (95 percent confidence interval,
`0.67 to 0.97). The median time to progression was
`6.3 months in the group given docetaxel and estra-
`mustine and 3.2 months in the group given mito-
`xantrone and prednisone (P<0.001) (Fig. 2).
`Declines in serum PSA levels of at least 50 per-
`cent occurred more frequently after treatment with
`docetaxel and estramustine (155 of 309 patients, or
`50 percent) than after treatment with mitoxantrone
`and prednisone (82 of 303 patients, or 27 percent;
`P<0.001). A partial response in measurable disease
`occurred in 17 percent of patients in the group given
`docetaxel and estramustine (17 of 103, 4 uncon-
`firmed) and 11 percent of patients in the group given
`mitoxantrone and prednisone (10 of 93, 4 un-
`confirmed). This difference was not significant
`(P=0.30). Patients with an inadequate assessment
`were assumed to have had no response. There was
`no significant difference in pain relief, as reported
`by the patients, between the two groups (data not
`shown).
`
`adverse events
`As of December 2003, all surviving patients had
`stopped the protocol treatment. Adverse events led
`to the withdrawal of 54 patients in the group as-
`signed to docetaxel and estramustine (16 percent)
`and 32 patients in the group assigned to mitoxan-
`trone and prednisone (10 percent). The rates of
`severe or life-threatening (grade 3 or 4) and fatal
`(grade 5) adverse events are summarized in Table 2.
`The rate of grade 3, 4, or 5 neutropenia in the group
`given mitoxantrone and prednisone did not differ
`significantly from that in the group given docetaxel
`and estramustine (12.5 percent vs. 16.1 percent,
`P=0.22). As compared with the group given mito-
`xantrone and prednisone, the group given docetaxel
`and estramustine had significantly higher rates of
`grade 3 or 4 neutropenic fevers (5 percent vs. 2 per-
`cent, P=0.01), cardiovascular events (15 percent vs.
`7 percent, P=0.001), nausea and vomiting (20 per-
`cent vs. 5 percent, P<0.001), metabolic disturbances
`(6 percent vs. 1 percent, P<0.001), and neurologic
`events (7 percent vs. 2 percent, P=0.001). There
`were eight treatment-related deaths in the group
`given docetaxel and estramustine: three are still
`
`n engl j med
`
`351;15
`
`www.nejm.org october
`
`7, 2004
`
`1517
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at SIDLEY AUSTIN LLP on July 15, 2016. For personal use only. No other uses without permission.
`
` Copyright © 2004 Massachusetts Medical Society. All rights reserved.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
` new england journal
`The
`
` medicine
`of
`
`fection. Four patients had grade 5 adverse events
`attributed to mitoxantrone and prednisone. Three
`died within 30 days of receiving protocol treatment,
`and another died from a respiratory tract infection
`and grade 4 anorexia. Vascular complications and
`their relationship to prophylactic warfarin treatment
`in the group given docetaxel and estramustine are
`shown in Table 3; these findings are observational
`and were not part of the planned primary analysis.
`
`discussion
`
`This randomized trial demonstrated that the treat-
`ment of androgen-independent metastatic prostate
`cancer with estramustine and docetaxel results in
`a longer median survival than treatment with mito-
`xantrone and prednisone (17.5 months vs. 15.6
`months, P=0.02). The hazard ratios for death
`
`(0.80 and 0.76, respectively) and median survival
`rates (17.5 months and 18.9 months, respectively)
`were similar in our docetaxel group and the group
`given docetaxel every three weeks in the study re-
`17
` elsewhere in this issue of
`ported by Tannock et al.
`Journal.
`the
` Although we did not meet our primary
`aim of detecting a 33 percent improvement in me-
`dian survival with estramustine and docetaxel, this
`trial had reasonable power to detect smaller differ-
`ences in survival. Relative to mitoxantrone and pred-
`nisone, docetaxel and estramustine reduced the
`mortality rate by 20 percent (95 percent confidence
`interval, 3 to 33 percent). The rates of reduced PSA
`levels and progression-free survival were signifi-
`cantly higher in the group given docetaxel and es-
`tramustine than in the group given mitoxantrone
`and prednisone. The survival estimate of the group
`given estramustine and docetaxel in our trial fell
`
`Table 2. Adverse Events.
`
`Type or Site of Adverse Event
`
`Docetaxel and Estramustine
`(N=330)
`
`Mitoxantrone and Prednisone
`(N=328)
`
`P Value*
`
`Drug reaction
`Cardiovascular
`Clotting
`Dermatologic
`Endocrine
`Influenza-like symptoms
`Nausea and vomiting
`Hematologic
`Hemorrhage
`Immunologic
`Infection
`Liver
`Lung
`Metabolic
`Musculoskeletal
`Neurologic
`Pain
`Renal or bladder
`Maximal grade of any adverse event†
`
`0
`37
`2
`1
`0
`29
`61
`17
`11
`3
`36
`9
`12
`14
`8
`21
`34
`8
`114
`
`0
`10
` 0
`0
`0
`3
`5
`47
`2
`0
`7
`1
`2
`6
`0
`2
`1
`0
`62
`
`Grade 3 Grade 4 Grade 5
`Grade 3 Grade 4 Grade 5
`number of patients
`3
`0
`1
`16
`0
` 0
`0
`1
`0
`1
`0
`20
`0
`16
`1
`18
`1
`6
`0
`0
`2
`20
`1
` 11
`1
`8
`0
`2
`0
`1
`0
`5
`0
`18
`1
`3
`8
`63
`
`0
`6
`0
`0
`0
`2
`1
`33
`0
`0
`2
`1
`1
`0
`2
`0
`5
`0
`46
`
`3
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`1
`0
`0
`0
`0
`0
`4
`
`1.00
` 0.001
`0.50
`1.00
`0.50
`0.20
`<0.001
`0.18
`0.11
`0.25
`0.004
`0.84
`0.42
`<0.001
`0.22
`0.001
`0.13
`0.14
`<0.001
`
`* P values were calculated by means of a two-sided Fisher’s exact test and are for the comparison of the percentage of pa-
`tients in each treatment group with grade 3 or 4 adverse events with the percentage of patients with grade 1 or 2 adverse
`events; there was no adjustment for multiple comparisons.
`† The maximal grade of adverse event was 0, 1, or 2 in 146 patients in the group given docetaxel and estramustine and 215
`patients in the group given mitoxantrone and prednisone.
`
`1518
`
`n engl j med
`
`351;15
`
`www.nejm.org october
`
`,
`
`7
`
`2004
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at SIDLEY AUSTIN LLP on July 15, 2016. For personal use only. No other uses without permission.
`
` Copyright © 2004 Massachusetts Medical Society. All rights reserved.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`chemotherapy for metastatic prostate cancer
`
`within the confidence intervals of smaller phase 1
`11-14
`and 2 studies of this combination.
`The median survival of 15.6 months among pa-
`tients treated with mitoxantrone and prednisone
`is longer than that reported by Tannock et al. (12
`5
`6
`
`months),
` Kantoff et al. (12.3 months),
`or Ernst
`18
`et al. (11.5 months).
` The median survival was sim-
`ilar to that reported in the current study by Tannock
`et al. (16.5 months). This difference may be due in
`part to the use of different eligibility criteria, in par-
`ticular the requirement for symptomatic disease in
`the studies by Tannock et al. and Ernst et al. In con-
`trast, in a randomized trial of 101 asymptomatic
`men with a rising serum PSA level, Berry et al. found
`a nonsignificant 4-month difference in median sur-
`vival: 23 months among men treated with mitoxan-
`trone and prednisone, as compared with 19 months
`19
` In our
`among men treated with prednisone alone.
`trial, 18 percent of patients had an increase in PSA
`but were asymptomatic with metastatic disease.
`The median PSA level at entry (87 ng per milliliter)
`was somewhat lower than in the studies by Kantoff
`6
`et al. (150 ng per milliliter),
` Tannock et al. (158 ng
`5
` and Ernst et al. (150 ng per millili-
`per milliliter),
`18
`ter),
` but pretreatment PSA levels do not predict
`survival in this patient population, although they
`probably reflect tumor burden.
`Crossover treatment may also partially account
`for the small difference in survival between the two
`
`treatment groups. Of all the patients we treated,
`about half received at least one other antineoplastic
`regimen after having had no response to the as-
`signed treatment. It is difficult to judge the effect of
`these additional treatments on overall survival, be-
`cause multiple variables influence response and
`survival after crossover treatment.
`Continuous corticosteroid treatment can reduce
`serum PSA levels by at least 50 percent in 20 to 74
`percent of men with hormone-refractory prostate
`20
`cancer,
` but the regimen of premedication with
`dexamethasone (60 mg every three weeks) that we
`used is unlikely to have affected the results. In a
`phase 2 study, the same dose and schedule of dexa-
`methasone that we used were employed until the
`serum PSA level rose at least 25 percent over base-
`line levels or clinical progression occurred, at
`which point docetaxel and estramustine were giv-
`en. None of the patients treated with dexametha-
`sone had a decline in PSA of at least 50 percent (the
`level actually increased by a median of 47 percent).
`After progression during dexamethasone therapy,
`92 percent of patients treated with docetaxel and
`estramustine had a decline in serum PSA levels of
`21
`at least 50 percent.
`In phase 2 studies, estramustine was associated
`with an increased risk of nausea, thromboembolic
`11-14,22
`events, and cardiovascular events.
` In our trial,
`cardiovascular and gastrointestinal events were
`
`Table 3. Adverse Events among Patients Receiving Docetaxel and Estramustine, According to Whether They Were
`Receiving Prophylactic Anticoagulation.
`
`Adverse Event
`
`No Anticoagulation
`(N=111)
`
`Prophylactic Anticoagulation
`(N=218)
`
`Abnormal troponin T level
`Cardiac ischemia or myocardial infarction
`Cardiovascular accident
`Thrombosis or embolism
`Epistaxis
`Melena or gastrointestinal bleeding
`Hematochezia
`Hematemesis
`Hemoptysis
`Hematuria
`Increase in partial-thromboplastin time
`Increase in prothrombin time
`
`0
`3 (3)
`1 (1)
`3 (3)
`0
`4 (4)
`0
`0
`0
`2 (2)
`0
`0
`
`1 (1)
`2 (2)
`0
`3 (3)
`0
`1 (1)
`0
`1 (1)
`0
`0
`0
`0
`
`Grade 3 Grade 4 Grade 5
`Grade 3 Grade 4 Grade 5
`number of patients (percent)
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`1 (1)
`0
`0
`0
`0
`0
`0
`0
`0
`3 (1)
`0
`2 (1)
`0
`1 (1)
`
`0
`1 (1)
`2 (1)
`8 (4)
`0
`1 (1)
`1 (1)
`0
`0
`0
`0
`0
`
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`
`n engl j med
`
`351;15
`
`www.nejm.org october
`
`7, 2004
`
`1519
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at SIDLEY AUSTIN LLP on July 15, 2016. For personal use only. No other uses without permission.
`
` Copyright © 2004 Massachusetts Medical Society. All rights reserved.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`chemotherapy for metastatic prostate cancer
`
`more frequent among patients given docetaxel and
`estramustine than among those given mitoxantrone
`and prednisone. However, this difference was not
`associated with an increased rate of treatment-
`related deaths or discontinuation of treatment in
`the former group.
`In conclusion, treatment with estramustine and
`docetaxel moderately increases survival at the cost
`of an increased rate of adverse events. These factors
`must be balanced when one is considering the use
`of docetaxel and estramustine as first-line therapy
`for men with metastatic androgen-independent
`prostate cancer.
`Supported in part by grants (CA38926, CA32102, CA37135,
`CA25224, CA46441, CA37981, CA45808, CA27057, CA12644,
`
`CA68183, CA22433, CA35261, CA58861, CA20319, CA46113,
`CA58882, CA76447, CA04919, CA16385, CA35090, CA03096,
`CA67663, CA45450, CA35431, CA45807, CA58416, CA14028,
`CA45377, CA63845, CA42777, CA46136, CA11083, CA35119,
`CA58658, CA46282, CA76129, CA46368, CA35176, CA86780,
`CA46462, CA35192, CA35178, CA67575, CA63844, CA12213,
`CA74647, CA35128, CA35996, CA58686, CA13612, CA45461,
`CA58723, CA63848, CA35281, CA63850, CA76132, and CA74811)
`from the National Cancer Institute, Department of Health and Hu-
`man Services, and by Aventis.
`Drs. Petrylak and Taplin report having received grant support,
`lecture fees, and consulting fees from Aventis; Dr. Hussain consult-
`ing fees and lecture fees from and having equity in Aventis; Dr. Lara
`lecture fees from Aventis and AstraZeneca; Dr. Benson lecture fees
`from Aventis; Drs. Small and Raghavan consulting fees from Aven-
`tis; Dr. Crawford consulting fees and lecture fees from Aventis; and
`Dr. Moinpour consulting fees from AstraZeneca.
`We are indebted to Dr. Robert Fine and Dr. Karen Antman for
`helpful comments during the preparation of this manuscript.
`
`references
`
`Crawford ED, Eisenberger MA, McLeod
`1.
`DG, et al. A controlled trial of leuprolide
`with and without flutamide in prostatic car-
`cinoma. N Engl J Med 1989;321:419-24.
`[Erratum, N Engl J Med 1989;321:1420.]
`Eisenberger MA, Blumenstein BA,
`2.
`Crawford ED, et al. Bilateral orchie