Bicalutamide for Advanced Prostate Cancer:
`The Natural Versus Treated History of Disease
`
`By Howard I. Scher, Christine Liebertz, W. Kevin Kelly, Madhu Mazumdar, Chris Brett, Lawrence Schwartz,
`Geert Kolvenbag, Lisa Shapiro, and Morton Schwartz
`
`Purpose: To determine the therapeutic effects of bica-
`lutamide 200 mg in patients with prostate cancers of
`different hormone sensitivities.
`Methods: Patients with progressive prostate cancer
`were treated with bicalutamide 200 mg daily. Before
`treatment, patients' tumors were classified on the basis
`of prior hormone exposure and by serum testosterone
`levels into androgen-dependent and androgen-indepen-
`dent groups. Prior exposure to flutamide and response
`to flutamide withdrawal was also considered. Outcomes
`were reported independently on the basis of posttherapy
`changes in prostate-specific antigen (PSA), measurable
`disease, and radionuclide bone scans.
`Results: Outcomes varied by prior hormone exposure
`as a higher proportion of patients with progression of
`androgen-dependent tumors showed posttherapy PSA
`decreases of more
`than 50% or more
`than 80%,
`measurable disease regression, and improvement on
`radionuclide bone scans than did patients with andro-
`gen-independent progression. Within the category of an-
`drogen-independent progression, clinical benefit was
`observed in patients who had previously progressed on
`flutomide, independent of the response to flutamide
`
`ORMONAL THERAPY has served as the mainstay
`of treatment for advanced prostate cancer for more
`than five decades. Controversies remain as to the optimal
`timing-initial versus delayed, the optimal type-com-
`bined androgen blockade versus monotherapy, and the
`role, if any, as second- and third-line treatment. When
`first described, hormone treatments were used primarily
`for patients with symptomatic metastatic disease. More
`recently, it is being used in the neoadjuvant setting before
`surgery or radiation therapy' in patients with increasing
`prostate-specific antigen (PSA) values after surgery or
`
`From the Genitourinarv Oncology Service. Division of Solid Tu-
`mor Oncology. Department of Medicine; Division of Biostatlstics,
`Department of Epidemiology and Biostatistics: and Departments of
`Radiology and Clinical Chemistry, Memorial Sloan-Kettering Can-
`cer Center, New York; Department of Medicine, Cornell University
`Medical College. New York, NY; and Zeneca Pharmaceuticals, Wil-
`mington, DE.
`Submitted February 14, 1997: accepted May 21, 1997.
`Supported by CA-05826, Zeneca Pharmaceuticals, Wilmington,
`DE, and the PepsiCo Foundation.
`Address reprint requests to Howard 1. Scher, MD, Memorial
`Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY
`10021; Email scherh@mskcc.org.
`© 1997 by American Society of Clinical Oncology.
`0732-183X/97/1508-0025$3.00/0
`
`withdrawal. Patients who had progressed on a gonado-
`tropin-releasing hormone (GnRH) analog alone had a
`low response proportion, whereas
`those who pro-
`gressed after two or more hormone therapies did not
`respond. Overall, the drug was well tolerated. After pro-
`gression on bicalutamide monotherapy, one third of pa-
`tients with androgen-dependent progression responded
`to medical castration with a GnRH analog.
`Conclusion: Classifying patient tumors on the basis of
`prior hormone exposure permits a more precise estimate
`of the potential benefit of a specific hormone therapy for
`the individual patient. The precision is further increased by
`reporting the effects of a drug on each parameter of disease
`independently. The difference in outcomes for patients with
`androgen-independent progression suggests that the spe-
`cific hormone therapy administered and the response to
`that therapy can influence the biology of the relapsing tu-
`mor and the sensitivity to subsequent therapies. The sensi-
`tivity to bicalutamide after progression on flutamide de-
`serves further study.
`J Clin Oncol 15:2928-2838.
`ciety of Clinical Oncology.
`
`1997 by American So-
`
`radiation as the sole manifestation of progression, or as
`part of a formal intermittent (on and off) approach.2
`The role of hormones as second- and third-line treat-
`ment in patients who have progressed on first-line therapy
`is less clear because a range of response proportions have
`been reported. This is the result of patient- and tumor-
`related factors. 5 One pretreatment predictive factor
`rarely considered is the influence of the specific hormonal
`therapy(ies) used on the response to a subsequent hor-
`mone treatment. For example, a patient who has relapsed
`on an antiandrogen alone, with an elevated serum testos-
`terone level, might be expected to respond to a hormonal
`maneuver that lowers serum testosterone. In contrast, a
`lower response proportion would be anticipated to a hor-
`mone treatment in a patient who is showing progression of
`disease despite castrate levels of testosterone (androgen-
`independent progression). 6
`Bicalutamide (Casodex; Zeneca Pharmaceuticals, Wil-
`mington, DE) is a pure antiandrogen that blocks the bind-
`ing of dihydrotestosterone to the androgen receptor. It
`has a fourfold higher binding affinity to the androgen
`receptor than the active metabolite of flutamide, hydroxy-
`flutamide.7 Advantages over other medications in this
`class include an improved safety profile,8 a long-terminal
`half-life that permits once-daily administration, and a di-
`rect correlation between response proportions and drug
`
`2928
`
`Journal of Clinical Oncology, Vol 15, No 8 (August), 1997: pp 2928-2938
`
`Information downloaded from jco.ascopubs.org and provided by at UCLA on August 10, 2016 from 164.67.217.118
`Copyright © 1997 American Society of Clinical Oncology. All rights reserved.
`
`JANSSEN EXHIBIT 2055
`Mylan v. Janssen IPR2016-01332
`
`

`

`BICALUTAMIDE FOR PROSTATE CANCER
`
`exposure calculated from the area under the plasma-con-
`centration times time curve (AUC). 9 In one trial using
`sequentially higher doses of bicalutamide, 73% of pa-
`tients normalized their PSA at a dose of 50 mg daily,
`97% at 150 mg, and 98% at a dose of 200 mg daily.10
`The present study evaluated bicalutamide 200 mg in
`patients with progressive prostate cancer who were at
`different points in their illness with respect to disease
`extent and prior hormone exposure. It sought to determine
`the points in the natural and treated history of prostate
`cancer that an antiandrogen might be beneficial. Included
`were patients with androgen-dependent progression who
`had not received hormones, those who had relapsed after
`neoadjuvant therapy, and those with androgen-indepen-
`dent progression who had progressed after one or more
`hormonal agents. Associations with prior response to flu-
`tamide withdrawal were also considered. In recognition
`of the controversial nature of outcomes reporting in this
`disease, and the lack of uniform outcome criteria that
`have been validated using the end point of survival, the
`results were reported independently on the basis of post-
`therapy declines in PSA, changes in measurable disease,
`and changes in radionuclide bone scans.'
`
`PATIENTS AND METHODS
`
`Patient Demographics
`Patients with histologically confirmed progressive prostate cancer
`were eligible for the study. All had documented progression of dis-
`ease manifest by one or more of the following: (1) an increase of
`prostate-specific antigen of greater than 50% from an individual
`baseline value on three successive occasions; (2) new metastatic
`lesions on radionuclide bone scan; or (3) a greater than 25% increase
`in a bidimensionally measurable tumor mass. Patients with new
`symptoms of tumor growth as the sole manifestation of progression
`were not considered.
`
`Baseline Parameters
`All patients underwent a complete history and physical examina-
`tion with a determination of the Karnofsky performance status
`(KPS). Disease-related symptoms were recorded at each visit, at
`which time patients were asked to evaluate their libido as well as
`their ability to achieve an erection sufficient for penetration. Labora-
`tory studies included an automated blood and platelet count and an
`assay of serum alkaline phosphatase, lactate dehydrogenase, aspar-
`tate transglutaminase, blood urea nitrogen, creatinine, calcium, phos-
`phorus, uric acid, total protein, albumin, total bilirubin, and in the
`majority of cases, a baseline serum testosterone. Acid phosphatase
`was measured in the Department of Clinical Chemistry at Memorial
`Sloan-Kettering Cancer Center using an enzymatic assay with thy-
`mophthalein phosphate as substrate.' 2 Prostate-specific antigen was
`analyzed in the same laboratory with the Tandem-E prostate-specific
`antigen immunoenzymatic assay (Hybritech Corporation, San Diego,
`CA). Imaging studies included a radionuclide bone scan, a computed
`tomography scan or magnetic resonance imaging of the abdomen
`and pelvis, and a chest radiogram.
`
`2929
`
`Hormonal Sensitivity
`Patients' tumors were categorized on the basis of prior hormone
`exposure as previously described." A tumor was considered to be
`androgen dependent if the serum testosterone level was in the non-
`castrate (> 35 ng/mL) range. Operationally, this included patients
`who had not received hormones, those who were being treated with
`an intermittent approach, and those who had been treated with neoad-
`juvant therapy before surgery or radiation and who subsequently
`relapsed off hormonal therapy. Androgen-independent tumors were
`those proliferating despite castrate levels (< 35 ng/mL) of testoster-
`one. This category was subdivided further into relapse-1 (relapse
`after orchiectomy or gonadotropin-releasing hormone analog mono-
`therapy); those who had progressed on combined androgen blockade
`with a gonadotropin-releasing hormone (GnRH) analog and flutam-
`ide who did or did not respond to flutamide withdrawal; and those
`who progressed on two or more hormone treatments exclusive of
`antiandrogen withdrawal (Table 1).
`
`Extent of Disease
`
`Disease extent was recorded on the basis of PSA values, the
`presence or absence of metastatic disease on bone scan, and the
`presence or absence of measurable disease by physical examination
`or imaging studies. Sites of measurable disease included lymph
`nodes, pulmonary nodules, or visceral sites. All imaging studies were
`reviewed independently (L.S.) without knowledge of the patient's
`clinical status.
`
`Treatment and Follow-Up
`Patients who met the eligibility criteria for the study were asked
`to sign a statement of informed consent that was reviewed by the
`Memorial Hospital Institutional Review Board. Bicalutamide was
`then administered at a fixed dose of 200 mg: four 50-mg tablets
`given once daily. Pill counts were performed at each outpatient visit
`to ensure compliance. Different follow-up schedules were used for
`patients with androgen-dependent versus those with androgen-inde-
`pendent tumors based on the anticipated response to the study agent.
`Those with androgen-dependent tumors were monitored monthly for
`the first 3 months and at 3-month intervals thereafter. At each visit,
`symptoms and adverse events were recorded, and biochemical stud-
`ies including a screening profile and PSA were obtained. Imaging
`studies that were abnormal at baseline were repeated at 3 months,
`6 months, and at 6-month intervals thereafter until disease progres-
`sion was documented. Imaging studies that were negative at baseline
`were only repeated if clinically indicated. After progression on bica-
`lutamide monotherapy, a hormonal intervention that lowered serum
`testosterone level (eg, a gonadotropin-releasing hormone analog or
`orchiectomy) was added and the patient was evaluated for a second-
`ary response while bicalutamide was continued. Patients were then
`followed-up until a second progression or nonresponse was docu-
`mented. At that point, bicalutamide was discontinued while GnRH
`analogs were continued, and the patient was evaluated for a blcalu-
`tamide withdrawal response.
`tumors were monitored
`Patients with androgen-independent
`monthly with serial examinations. symptom and adverse event as-
`sessments, and biochemical studies. Abnormal imaging studies were
`repeated at months 3, 6, and every 6 months thereafter, at which time
`overall response assessments were made. Patients were continued on
`treatment until progression of disease was documented.
`
`Information downloaded from jco.ascopubs.org and provided by at UCLA on August 10, 2016 from 164.67.217.118
`Copyright © 1997 American Society of Clinical Oncology. All rights reserved.
`
`

`

`SCHER ET AL
`
`at a minimum of monthly intervals. The duration of the decline was
`measured from the start of therapy to the date of the first sequential
`increase that was confirmed by a subsequent value.
`Measurable disease. Standard phase II response criteria were
`used. A partial response required a more than 50% reduction in the
`sum of the products of the largest perpendicular diameters of all
`measured lesions with no growth of other lesions or appearance of
`new lesions. Progression required a more than 25% increase or
`appearance of new lesions. All other responses were classified as
`stable.
`Radionuclide bone scan. Posttherapy bone scans were classified
`as showing improvement, stability, or progression relative to the
`baseline scan on the basis of a visual inspection 3 by a blinded
`independent review. Progression required the appearance of new
`lesions. An increase in the intensity of a preexisting lesion was not
`considered to be progression.
`
`Statistical Analysis
`The time to progression was calculated from the start of therapy
`to the time of the first confirmed increase in PSA, increase in measur-
`able disease, new lesions on bone scan, or increase in disease-related
`symptoms, whichever was shorter. For patients with androgen-de-
`pendent tumors, the time to treatment failure was calculated from
`the start of treatment to the time of progression after the addition of
`a GnRH analog or surgical castration to bicalutamide monotherapy.
`Progression-free survival distributions were estimated by the method
`of Kaplan and Meier.l4
`
`RESULTS
`One hundred five patients were entered, 104 of whom
`were assessable. One patient was enrolled but opted not
`to receive treatment after signing informed consent. The
`demographics including prior therapy and baseline bio-
`chemical parameters are listed in Table 1. As illustrated,
`the median age was 70 years, and median KPS was 90%.
`The predominant local treatment was external beam radia-
`tion therapy, whereas 34% of the patients had undergone
`a radical prostatectomy. The high median hemoglobin
`reflects the good performance status of the patient popula-
`tion. Approximately half of the patients had androgen-
`dependent tumors (hormone naive, relapse after prior
`neoadjuvant or intermittent treatment) and half androgen-
`independent progression (relapse after orchiectomy or
`GnRH analog alone, 12 patients; in one case, diethylstil-
`bestrol). combined androgen blockade with flutamide (26
`patients), or multiple (two or more) hormonal therapies
`(12 cases).
`Table 2 shows the outcomes based on PSA decreases,
`changes in radionuclide bone scans, and changes in mea-
`surable disease based on the hormone sensitivity of the
`tumor. Overall, patients with androgen-dependent tumors
`had more favorable outcomes relative to those with andro-
`gen-independent tumors. For these two groups, 88% (47
`of 53; 95% confidence interval, 77% to 96%) versus 10%
`(five of 51; 95% confidence interval, 8% to 31%) showed
`
`2930
`
`Table 1. Patient Characteristics
`
`104
`70
`48-82
`90
`70-90
`
`22
`12
`3
`
`55
`
`3 9
`
`11
`
`14 1
`9 1-165
`
`19
`11-49
`
`1670
`106-351
`
`1 1
`6-2 4
`
`60
`1-72
`
`22.9
`10-1,300
`
`No. of patients
`Median age, years
`Range
`Median KPS
`Range
`Prior therapy
`Surgery
`Prostatectomy
`Prostatectomy and EBRT
`Cryosurgery
`Radiation
`External beam
`1-125 implantation
`External beam/palliative
`Palliative
`Baseline biochemical parameters
`Hemolgobin
`Median
`Range
`AST
`Median
`Range
`LDH
`Median
`Range
`Creatinine
`Median
`Range
`Acid phosphatase
`Median
`Range
`PSA
`Median
`Range
`
`No of
`Cases
`
`Testosterone
`Median
`
`Hormonal Status
`Androgen dependent
`Naive
`Intermittent/sensitive
`Androgen independent
`Relapse-1
`Flutamide withdrawal responders
`Flutamide withdrawal nonresponders
`2 2 Hormone therapies
`
`53
`37
`16
`51
`13
`14
`12
`12
`
`384
`500
`
`146-778
`299-987
`
`12
`12
`6
`3
`
`0-34
`0-34
`0-23
`0-18
`
`Abbreviations: EBRT, external-beam radiation therapy, LDH, lactate de-
`hydrogenase
`
`Outcomes
`Posttherapy changes in PSA, and if present at baseline, measurable
`disease and/or radionuclide bone scanning, were evaluated indepen-
`dently in the following manner.
`Posttherapy PSA changes. The percentage decline in PSA from
`baseline was calculated from the lowest posttherapy nadir relative
`to the pretreatment baseline for groups of patients on the basis of
`hormonal status. To meet the criteria for a given degree (> 50% or
`> 80%) of decline required a minimum of three measurements taken
`
`Information downloaded from jco.ascopubs.org and provided by at UCLA on August 10, 2016 from 164.67.217.118
`Copyright © 1997 American Society of Clinical Oncology. All rights reserved.
`
`

`

`BICALUTAMIDE FOR PROSTATE CANCER
`
`Disease Manifestation
`
`Increasing PSA
`
`Bone
`
`Soft tissue
`
`Table 2. Outcomes Based on Overall Hormone Sensitivity
`Androgen Dependent
`No
`%
`
`No
`
`%
`
`Outcome
`
`80% decline
`50% decline
`Stable
`Progression
`
`Improved
`Stable
`Progression
`Nonassessable
`
`> 50% decrease In area
`Stable
`Progression
`Nonassessable
`
`(n = 104)
`
`(n = 53)
`
`52
`9
`11
`32
`
`11
`15
`21
`1
`
`7
`10
`14
`3
`
`(n = 48)
`
`(n = 34)
`
`50
`9
`11
`32
`
`23
`31
`44
`2
`
`21
`29
`41
`9
`
`47
`2
`2
`2
`
`7
`6
`
`2
`
`7
`8
`3
`1
`
`(n = 15)
`
`0
`
`(n = 19)
`
`88
`4
`4
`4
`
`48
`40
`
`12
`
`37
`42
`16
`5
`
`2931
`
`Androgen Independent
`No
`%
`
`(n = 51)
`
`5
`7
`9
`30
`
`4
`9
`19
`1
`
`3
`11
`2
`
`(n = 33)
`
`(n = 16)
`0
`
`10
`14
`17
`59
`
`12
`27
`58
`3
`
`19
`69
`12
`
`NOTE. One patient was entered onto the study but withdrew before receiving treatment. Nonassessable indicates that the specific site of disease was
`not reassessed
`
`a more than 80% decrease in PSA, 48% (seven of 15;
`95% confidence interval, 21% to 73%) versus 12% (four
`of 33; 95% confidence interval, 3% to 28%) an improve-
`ment in bone scan, whereas 37% (seven of 19; 95% con-
`
`fidence interval, 16% to 62%) versus 0% (zero of 16)
`showed regression of soft tissue disease. These differ-
`ences in outcomes are reflected in Fig 1, which shows
`the time to progression based on PSA elevations for these
`
`.o
`
`o.9
`
`0.8
`
`a
`
`0.?
`
`0.8
`
`Z 0.5
`
`0 I
`
`L0 n
`
` 0.3
`
`0.2
`
`0.1
`
`0.0
`
`0
`
`6
`
`12
`
`a8
`MONTHS
`
`000 Androgen-D epndent (53 Painths.
`ndrogen-lndependent
`(51 Patients.
`000
`
`S8 Ceneored)
`4 Cnseord)
`
`Tick mark(t) IndleaLe lst follow-up
`
`24
`
`30
`
`38
`
`Fig 1. Time to progression based on PSA elevations for patients with androgen-dependent and androgen-independent progression.
`
`Information downloaded from jco.ascopubs.org and provided by at UCLA on August 10, 2016 from 164.67.217.118
`Copyright © 1997 American Society of Clinical Oncology. All rights reserved.
`
`

`

`2932
`
`Disease Maonifestation
`
`Increasing PSA
`
`Bone
`
`Soft tissue
`
`Table 3. Outcomes for Androgen-Dependent Disease
`
`SCHER ET AL
`
`No Prior Hormones
`
`Prior Neoadjuvant or
`Intermittent
`
`Outcome
`
`80% decline
`> 50% to < 80% decline
`Stable
`Progression
`
`Improved
`Stable
`Progression
`Nonossessable
`
`> 50% decrease In area
`Stable
`Progression
`Nonassessable
`
`No
`
`47
`2
`2
`2
`
`7
`6
`2
`
`7
`8
`3
`1
`
`%
`
`(n = 53)
`
`(n =15)
`
`0
`in = 19)
`
`88
`4
`4
`4
`
`48
`40
`13
`
`37
`42
`16
`5
`
`No
`
`34
`1
`1
`1
`
`7
`3
`
`7
`6
`3
`
`(n = 37)
`
`(n = 10)
`
`0
`0
`In = 16)
`
`0
`
`%
`
`92
`2
`2
`2
`
`70
`30
`
`44
`37
`19
`
`%
`
`81
`7
`7
`7
`
`60
`40
`
`67
`
`No
`
`13
`1
`1
`1
`
`3
`2
`
`2
`
`(n = 16)
`
`(n = 5)
`0
`
`0
`(n = 3)
`0
`
`0
`1
`
`NOTE. One patient was entered onto the study but withdrew before receiving treatment. Nonassessable indicates that the specific site of disease was
`not reassessed.
`
`two groups of patients, the most stringent parameter for
`progression. As illustrated, the medians were 14 months
`and 3 months, respectively.
`Table 3 subdivides patients with androgen-dependent
`progression into those who had no prior hormone expo-
`sure and those who were treated with neoadjuvant or
`intermittent therapy and later progressed. For those with
`no prior hormonal exposure, 92% (34 of 37; 95% confi-
`dence interval, 78% to 98%) had a greater than 80%
`decrease in PSA, 70% (seven of 10; 95% confidence
`interval, 35% to 93%) showed bone scan improvement,
`whereas 44% (seven of 16; 95% confidence interval, 20%
`to 70%) showed regression of a measurable lesion. A
`smaller proportion of patients who had been treated with
`neoadjuvant therapy or by a policy of intermittent treat-
`ment showed benefit. However, based on the number of
`patients evaluated, this difference was not significant.
`Only three patients had measurable tumor sites, two of
`whom remained stable. None of the five abnormal bone
`scans improved. The median time to progression for these
`two groups was 14 months (range, 3 to 33+ months) and
`not reached (range, 3 to 23+ months), respectively, with
`62% of patients in each group having progressed based
`on PSA elevations.
`
`Response to Chemical Castration After Progression on
`Bicalutamide
`A total of 15 patients treated with bicalutamide with
`androgen-dependent disease was assessable for a response
`to the addition of a GnRH analog after sequential eleva-
`tions in PSA were observed. Overall, five (33%; 95%
`
`confidence interval, 12% to 63%) showed a greater than
`80% decrease in PSA for a median of 6 months (range,
`6 to 11). Of these five patients, four had parallel improve-
`ments in an abnormal bone scan, but none had measurable
`disease at the start of bicalutamide therapy that could
`be assessed for response. Ten patients did not respond
`biochemically to subsequent chemical castration.
`
`Response to Bicalutamide Withdrawal
`Currently, eight of 15 patients who progressed after a
`GnRH analog was added to bicalutamide monotherapy
`are assessable for a bicalutamide withdrawal response.
`Of these, three showed significant (> 80%) decreases in
`PSA for 3, 5, and 5+ months. Of interest was that one
`of the patients who responded to bicalutamide withdrawal
`initially responded to bicalutamide but did not respond
`to the addition of a GnRH analog after progression on
`bicalutamide monotherapy. Figure 2 illustrates the se-
`quential PSA values for a patient who had previously
`received neoadjuvant hormonal therapy and later re-
`lapsed. After a 12-month response to bicalutamide, a
`GnRH analog was added with a subsequent secondary
`response. At month 18, bicalutamide was discontinued
`while the GnRH analog was continued. As shown, 8
`weeks after discontinuation of bicalutamide, the PSA
`level decreased for 5 months, after which progression of
`disease was observed.
`Table 4 considers patients with androgen-independent
`progression. As shown, patients who relapsed after an
`orchiectomy or GnRH analog alone (relapse-l) showed
`modest benefit, as none of the 13 patients showed a more
`
`Information downloaded from jco.ascopubs.org and provided by at UCLA on August 10, 2016 from 164.67.217.118
`Copyright © 1997 American Society of Clinical Oncology. All rights reserved.
`
`

`

`BICALUTAMIDE FOR PROSTATE CANCER
`
`2933
`
`250
`
`200
`
`150
`
`100
`
`50
`
`0
`
`-_j
`
`E0
`
`)
`c<C
`U)
`EL
`
`Fig 2. Sequential PSA values
`for a patient who responded to
`bicalutamide for 12 months. A
`GnRH analog was added at sec-
`ond progression and no
`re-
`sponse was observed. Later, bi-
`calutamide was discontinued
`and after a delay of 8 weeks, a
`decline in PSA was observed that
`was sustained
`for 5 months.
`is manufactured by
`*Casodex
`Zeneca Pharmaceuticals, Wil-
`mington, DE. tLupron is manu-
`factured by TAP Pharmaceuti-
`cals.
`
`Months
`
`than 80% decline in PSA, whereas two (15%; 95% confi-
`dence interval, 1% to 45%) showed a more than 50%
`decline with parallel improvements in bone scan. In con-
`trast, patients who had received two or more hormone
`treatments did not respond by any criteria. Of interest
`was the observation that patients who had previously been
`exposed to flutamide showed benefit, as 38% (10 of 26;
`95% confidence interval. 20% to 59%) had a more than
`50% decrease in PSA. An equal proportion of patients
`who responded or who did not respond to flutamide with-
`drawal had PSA decreases, five of 12 (41%; 95% confi-
`dence interval, 12% to 72%) versus five of 14 (35%; 95%
`confidence interval, 11% to 61%) showed a more than
`
`50% decline in PSA. Improvements in bone scans were
`also observed in these patients as additional evidence
`of clinical benefit, whereas patients also had soft-tissue
`lesions to assess response. For these two groups, the me-
`dian time to PSA increase was 4 months (range, 1 to 17 +
`months) and 3 months (range, 1 to 7 months), respec-
`tively. The differences could not be explained on the basis
`of serum testosterone levels, which were similar between
`the groups. Figure 3 shows the sequential PSA values for
`a patient who progressed on combined androgen blockade
`after 17 months, followed by a 12-month response to
`flutamide withdrawal. At that point (month 31), bicalu-
`tamide was added and the patient responded for an addi-
`
`Table 4. Outcomes for Patients with Androgen-Independent Progression
`
`Disease
`Manlfestahton
`
`Outcome
`
`PSA
`
`Bone
`
`80% decline
`> 50% to < 80% decline
`Stable
`Progression
`
`Improved
`Stable
`Progression
`Nonassessable
`
`Soft tissue
`
`> 50% decrease in area
`Stable
`Progression
`Nonassessable
`
`Total
`
`No
`
`(n = 51)
`10
`5
`14
`7
`17
`9
`59
`30
`(n =33)
`4
`12
`9
`27
`19
`58
`1
`3
`(n= 16)
`0
`
`3
`11
`2
`
`19
`69
`2
`
`Relapse-l (No
`Prior Flutamide)
`
`Flutamide
`Withdrawal
`Responders
`
`Flutoamide
`Withdrawal
`Nonresponders
`
`Relapse After
`Two or More
`Hormones
`
`No.
`
`%
`
`No
`
`%
`
`No
`
`%
`
`No
`
`%
`
`(n = 13)
`0
`
`15
`2
`8
`1
`77
`10
`(n = 12)
`17
`2
`
`0
`
`10
`
`83
`
`0
`(n = 6)
`0
`
`33
`67
`
`2
`4
`
`0
`
`(n = 12)
`2
`3
`4
`3
`(n = 10)
`2
`5
`3
`
`17
`25
`33
`33
`
`20
`50
`30
`
`0
`(n = 2)
`0
`0
`
`2
`
`100
`
`0
`
`(n = 14)
`21
`3
`14
`2
`21
`3
`43
`6
`(n = 5)
`0
`
`40
`40
`20
`
`2
`2
`1
`(n = 3)
`0
`0
`
`2
`1
`
`66
`33
`
`(n = 12)
`0
`0
`
`1
`11
`
`(n = 6)
`0
`
`2
`4
`
`0
`(n = 5)
`0
`
`1
`3
`1
`
`8
`92
`
`33
`67
`
`20
`60
`20
`
`Information downloaded from jco.ascopubs.org and provided by at UCLA on August 10, 2016 from 164.67.217.118
`Copyright © 1997 American Society of Clinical Oncology. All rights reserved.
`
`

`

`SCHER ET AL
`
`Fig 3. A patient treated with
`combined androgen blockade
`who later progressed and subse-
`quently responded to flutamide
`withdrawal. After the addition of
`bicalutamide, a second response
`was observed. *Eulexin is manu-
`factured by Schering-Plough,
`Kenilworth, NJ.
`
`2934
`
`200
`
`150
`
`-_j
`
`E c
`
` 100
`
`50
`
`0
`
`0
`
`10
`
`20
`
`30
`Months
`
`40
`
`50
`
`60
`
`tional 10 months. Bicalutamide was then withdrawn and
`the patient was evaluated for a bicalutamide withdrawal
`response; none was observed.
`
`Adverse Experiences
`
`Table 5 shows effects of bicalutamide on libido, on
`the ability to have an erection, and on whether a patient
`experienced hot flashes, breast swelling, or breast tender-
`ness. These outcomes were assessed by the patient re-
`sponse to a direct question at each office visit. Only pa-
`tients who reported a normal libido and erectile capability,
`or who did not have hot flashes, breast swelling, or breast
`tenderness at the start of bicalutamide therapy were con-
`sidered. As shown, only 6% of the 32 patients who re-
`ported normal libido before treatment reported normal
`libido after treatment. whereas 17% reported normal erec-
`tions. No attempt was made to correlate the reported ad-
`verse experiences with the definitive treatment (surgery or
`
`radiation therapy with or without neoadjuvant hormonal
`therapy) of the primary tumor. Hot flashes developed in
`40% of cases, breast tenderness in 59%, and breast swell-
`ing in 72% of cases who did not report these symptoms
`at the start of treatment. However, in no case were these
`symptoms dose limiting, although two patients underwent
`breast reduction surgery for gynecomastia.
`Other adverse experiences were graded with the Na-
`tional Cancer Institute common toxicity criteria and are
`listed in Table 6. No distinction was made between a
`disease- versus a treatment-related symptom. Thus, al-
`though 14% of patients reported pain and 35% urinary
`frequency, in no case could this experience be attributable
`to the bicalutamide per se. The most frequent and signifi-
`cant adverse event was fatigue and malaise. As shown,
`the overall type and frequency of adverse events is similar
`5 although several
`to that reported
`in other studies.8Y'
`unique experiences were noted. One patient developed
`
`Table 5. Effects on Libido and Sexual Function
`
`Symptom
`
`Libido
`Erections
`
`Hot flashes
`Breast swelling
`Breast tenderness
`
`Boseline
`Status
`
`Normal
`Normal
`
`Absent
`Absent
`Absent
`
`No of
`Poatents
`
`32
`23
`
`67
`68
`91
`
`Normal
`
`Impaired
`
`Absent
`
`No
`
`18
`16
`
`%
`
`56
`70
`
`Absent
`
`No
`
`2
`4
`
`No
`
`40
`19
`37
`
`%
`
`6
`17
`
`%
`
`60
`28
`41
`
`Present
`
`No
`
`12
`3
`
`No
`
`27
`49
`54
`
`%
`
`38
`13
`
`%
`
`40
`72
`59
`
`Information downloaded from jco.ascopubs.org and provided by at UCLA on August 10, 2016 from 164.67.217.118
`Copyright © 1997 American Society of Clinical Oncology. All rights reserved.
`
`

`

`BICALUTAMIDE FOR PROSTATE CANCER
`
`Table 6. Other Adverse Experiences
`
`Grade
`
`Event
`
`Fatigue/malaise
`Constipation
`Diarrhea
`Dizziness
`Edema
`Elevated bilirubin
`Fever
`Gastrointestinal upset
`Nausea
`Pain
`Pulmonary
`Rash
`Stomatitis
`Urgency/frequency
`
`1 +
`
`56%
`15%
`5%
`2%
`3%
`0
`3%
`2%
`7%
`11%
`5%
`12%
`2%
`34%
`
`2+
`
`2%
`2%
`1%
`1%
`0
`1%
`0
`0
`0
`2%
`0
`0
`0
`1%
`
`3+
`
`0
`0
`0
`0
`2%
`0
`0
`0
`0
`1%
`0
`0
`0
`0
`
`4+
`
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`1%
`0
`0
`0
`
`of the following were also noted.
`NOTE. Single events (maximal grade
`dysuria, headache, cough, increase in AST, weight gain, weight loss, dry
`eyes, change in mood, change in vision, myalgia, and rectal discomfort.
`
`biopsy proven grade IV pneumonitis that required hospi-
`talization and treatment with corticosteroids. This re-
`solved completely after 12 weeks, but the patient was
`not rechallenged. There was one death on study from a
`presumed arrhythmia in a patient with a history of cardiac
`disease. No autopsy was obtained. Not related to bicalu-
`tamide was the finding that two patients developed a sec-
`ond malignancy on treatment: one a superficial transi-
`tional cell tumor controlled by transurethral resection
`(TUR), and a second underwent a colon resection for a
`Dukes C lesion. Two patients underwent a cholecystec-
`tomy while on treatment, and both resumed bicalutamide
`therapy after recovery from surgery.
`
`DISCUSSION
`The present study shows the influence of prior hormone
`exposure on response to bicalutamide. Androgen-depen-
`dent tumors were more sensitive than androgen-indepen-
`dent tumors by whatever criteria used. Clinically, this
`included patients who had not received hormones, those
`who had progressed after neoadjuvant therapy, or those
`who were being treated with an intermittent hormone
`approaches. The overall results in patients with androgen-
`dependent disease were similar to those reported by other
`investigators in large comparative trials. 8' 161 8 Although
`only small numbers of patients were treated, the response
`proportions and time to progression were similar in pa-
`tients who were hormone naive and those who had pre-
`viously received neoadjuvant hormonal therapy and later
`progressed off treatment.
`An important question is whether patients with tumors
`
`2935
`
`that are progressing on bicalutamide monotherapy with
`a normal or elevated serum testosterone level are still
`androgen dependent, ie, will they respond to a hormonal
`treatment that lowers serum testosterone levels? In this
`study, five of 15 (33%; 95% confidence interval, 12% to
`62%) patients showed a more than 80% decrease in PSA
`when a GnRH analog (which lowers serum testosterone
`levels) was added at the time of progression. These out-
`comes are similar to those of Fabozzi et al.19 who noted
`a more than 50% decline in PSA in 50% of cases (14 of
`28; 95% confidence interval, 31% to 69%) who were
`treated with an orchiectomy (22 cases) or Zoladex (Zen-
`eca Pharmaceuticals, Wilmington, DE; six cases) after
`bicalutamide 50 mg. Sogani et al2u studied the effect of
`diethylstilbestrol (DES) or orchiectomy after a response
`followed by progression flutamide monotherapy. Thirty-
`nine percent of patients (13 of 32; 95% confidence inter-
`val, 24% to 59%) showed pain relief, but no other objec-
`tive outcomes were reported. In contrast, Kasimis et a117
`studied 54 patients who initially received bicalutamide 50
`mg as monotherapy followed by a testosterone-lowering
`treatment at the time of progression. Overall, 37 (69%:
`95% confidence interval, 54% to 81%) showed a response
`defined as complete or partial remission, or stabilization
`of disease. Further analyses suggested that patients who
`treatment