`
`GLUCOCORTICOIDS AND TREATMENT OF PROSTATE
`CANCER: A PRECLINICAL AND CLINICAL REVIEW
`
`MARWAN FAKIH, CANDACE S. JOHNSON, AND DONALD L. TRUMP
`
`Glucocorticoids have been used for a substan-
`tial period to treat patients with advanced an-
`drogen-independent prostate cancer (AIPC). Al-
`though known for their anti-inflammatory activity,
`glucocorticoids have antitumor activity in various
`hematologic malignancies. The role of glucocorti-
`coids as antineoplastic agents in epithelial tumors
`is less well defined. Their use in these tumors has
`been strictly palliative.
`Glucocorticoids have been widely used in the
`treatment of advanced prostate cancer and have
`served as the “standard” therapy arm in several
`randomized studies. Although multiple studies of
`glucocorticoid use in prostate cancer have been
`conducted, their therapeutic role remains unclear.
`We review the information regarding the mecha-
`nisms underlying glucocorticoid antitumor effects
`in prostate cancer and critically review the results
`of clinical trials using these agents.
`
`MECHANISMS OF ACTIVITY
`
`The cytotoxic effect of glucocorticoids in hema-
`tologic cells is well defined.1– 4 Glucocorticoids
`bind to a cytosolic receptor that localizes to the
`nucleus, leading to a variety of transcriptional
`modifications. This ultimately results in upregula-
`tion of multiple caspases, leading to apoptotic cell
`death.5 Glucocorticoids do not induce apoptosis in
`prostate cancer cells, yet growth inhibitory effects
`are well documented.6,7 We have demonstrated di-
`rect antiproliferative effects in the hormone-refrac-
`tory human PC-3 and rat Mat-Ly-Lu cell lines as
`
`Supported by grants from the Mary Hillman Jennings Foundation
`and CaPCURE. D. L. Trump and C. S. Johnson receive research
`funding from Bristol Myers Squibb, Aventis, and D-Novo.
`From the Departments of Medicine, Division of Hematology-
`Oncology and Pharmacology, University of Pittsburgh School of
`Medicine; and University of Pittsburgh Cancer Institute, Pitts-
`burgh, Pennsylvania
`Reprint requests: Donald L. Trump, M.D., Department of Med-
`icine, University of Pittsburgh Cancer Institute, Montefiore Uni-
`versity Hospital, N 723, 200 Lothrop Street, Pittsburgh, PA
`15213
`Submitted: January 30, 2002, accepted (with revisions): March
`28, 2002
`
`Amerigen Exhibit 1125
`© 2002, ELSEVIER SCIENCE INC.
`Amerigen v. Janssen
`IPR2016-00286
`
`ALL RIGHTS RESERVED
`
`measured by cell cycle arrest and modulation of the
`cdk inhibitors, p21 and p27. The mechanisms of
`this growth inhibitory effect are not clear. How-
`ever, several mechanisms have been postulated.
`
`SUPPRESSION OF ADRENAL ANDROGEN
`
`SECRETION
`
`Glucocorticoids may exert an antitumor effect
`on androgen-independent prostate cancer by sup-
`pression of adrenal androgens. Low-dose glu-
`cocorticoids produce negative feedback on the pi-
`tuitary gland,
`leading to a decrease in both
`testicular and adrenal androgens.8,9 Plowman et
`al.10 reported on 17 orchiectomized patients with
`progressive prostate cancer who were treated with
`hydrocortisone 30 mg/day. A significant decrease
`in testosterone, androstenedione, and dihydroan-
`drostenedione (DHEAS) levels was noted with
`therapy. Similarly, Tannock et al.11 noted a de-
`crease in testosterone, androstenedione, and
`DHEAS levels in association with low-dose pred-
`nisone therapy (7.5 to 10 mg daily) in surgically or
`medically castrated patients with advanced pros-
`tate cancer. Symptomatic relief was associated with
`a decrease in the adrenal androgen levels. Eight of
`13 patients who had a decrease in DHEAS levels by
`1 mol/L or greater had improvement in pain and
`only 1 of 8 patients with unchanged or increased
`DHEAS levels had symptomatic improvement.
`
`PARACRINE/AUTOCRINE FACTOR
`
`MODULATION
`
`Glucocorticoids can inhibit prostate cancer cell
`growth by modulating cellular growth factors such
`as lipocortin, tumor growth factor beta-1 (TGF-
`1), urokinase-type plasminogen activator (uPA),
`and interleukin-6 (IL-6).
`
`LIPOCORTIN
`Glucocorticoids mediate their anti-inflammatory
`effects in part by way of lipocortin, a member of the
`annexin family (calcium and phospholipid-bind-
`ing proteins).12,13 Lipocortin gene transcription is
`upregulated by glucocorticoids, resulting in in-
`
`UROLOGY 60: 553–561, 2002 • 0090-4295/02/$22.00
`PII S0090-4295(02)01741-7 553
`
`
`
`creased cellular levels. Lipocortin is subsequently
`secreted from the cells to mediate its anti-inflam-
`matory effects at the membrane level by inhibiting
`phospholipase A2.14,15 Lipocortin also mediates,
`at least in part, the antiproliferative effects of glu-
`cocorticoids. Carollo et al.6 showed that the inhib-
`itory effect of dexamethasone on androgen inde-
`pendent PC-3 cells is completely abolished if the
`cells are incubated with sheep antihuman lipocor-
`tin antibodies. The mechanism through which li-
`pocortin mediates its antiproliferative effect is not
`well defined.
`
`TRANSFORMING GROWTH FACTOR BETA-1
`TGF-1 is a member of a group of dimeric pep-
`tides that modulate multiple cellular functions, in-
`cluding extracellular matrix expression, differenti-
`ation, and cellular proliferation.16 TGF-1 has
`dual
`inhibitory and stimulatory effects on the
`growth of prostate cancer. Hsing et al.17 treated
`both tumorigenic (NRP-154) and nontumorigenic
`(NRP-152) rat prostate cell lines with TGF-1.
`TGF-1 induced cell death by apoptosis in both
`cell lines. The effects were enhanced by dexameth-
`asone and inhibited by insulin growth factor-1.
`Dexamethasone increased TGF-1 mRNA expres-
`sion in androgen-insensitive PA-III prostate cancer
`cell lines in association with growth inhibitory ef-
`fects.18 Co-treatment of PA-III with TGF-1 anti-
`body reversed the dexamethasone antiproliferative
`effects. This suggests that the inhibitory role of
`dexamethasone may be mediated by TGF-1.18
`Barrack19 and Morton and Barrack,20 on the other
`hand, showed that TGF-1 was implicated in tu-
`mor progression. They showed that although
`TGF-1 had no inhibitory effects on the growth of
`the hormone-independent, rat prostate Mat-Ly-Lu
`cancer cell line, it stimulated motility and fibronec-
`tin expression.20
`
`UROKINASE-TYPE PLASMINOGEN ACTIVATOR
`uPA, a serine protease, is implicated in the pro-
`gression of various malignancies, including pros-
`tate cancer. Serum uPA levels are higher in patients
`with metastatic prostate cancer than in those with
`localized disease.21 uPA increases the invasiveness
`and stimulates tumor migration and growth when
`expressed in hormone-responsive human prostate
`LNCaP cell lines.22,23 Furthermore, uPA has direct
`stimulatory effects on osteosarcoma cells and may
`play an important role in the pathophysiology of
`blastic lesions in prostate cancer.24 –26 Dexameth-
`asone downregulates uPA mRNA and its protein
`expression in both PC-3 and PA-III cell lines.24 By
`downregulating uPA, glucocorticoids may inhibit
`the growth and invasiveness of prostate cancer
`cells.
`
`INTERLEUKIN-6
`Recently, Nishimura et al.27 showed that dexa-
`methasone had growth inhibitory effects on the
`DU-145 AIPC human cell line. These inhibitory
`effects were associated with a dose-dependent up-
`regulation of inhibitor of kappa B (IB␣) a key
`inhibitor of nuclear factor-kappa B (NF-B). A
`subcellular localization evaluation of DU-145 cells
`pretreated with dexamethasone confirmed the loss
`of NF-B nuclear localization. NF-B is an impor-
`tant regulator of several cytokines, including IL-
`6.28 It is believed that loss of nuclear compartmen-
`talization of NF-B leads to inhibition of IL-6
`secretion, thus leading to a favorable growth inhib-
`itory effect. Consistent with this hypothesis is the
`inhibition of DU-145 cell growth by IL-6 antibod-
`ies, as well as the fourfold decrease in IL-6 levels in
`patients treated with low-dose dexamethasone.27
`
`SUPPRESSION OF ANDROGEN-DEPENDENT
`
`TRANSCRIPTION
`
`Androgen and glucocorticoid receptors share a
`significant degree of homology, especially in their
`DNA-binding domain (DBD). Both receptors bind
`through their DBD to a common DNA site termed
`the hormone receptor element. AR, GR, and MR
`(mineralocorticoid receptors) bind to hormone re-
`ceptor element as homodimers. Although these re-
`ceptors bind to the same DNA site, they result in
`different transcriptional activities. Factors other
`than DBD-hormone receptor element interaction
`determine the specificity of the transcriptional ac-
`tivity. Chen et al.29 has recently demonstrated that
`GR and AR can associate through their DBD to
`form a heterodimer. This heterodimer formation
`results in an inhibitory effect on androgen-specific
`transcription in vitro. This transcriptional inhibi-
`tion can lead to inhibition of the necessary down-
`stream events for the growth of prostate cancer
`cells. These results should be extrapolated with
`caution, because it has not been shown yet that GR
`and AR form heterodimers in vivo.
`
`GLUCOCORTICOID-DRIVEN CELLULAR
`
`PROLIFERATION
`
`Although the data discussed above point to an
`inhibitory effect of glucocorticoids on prostate
`cancer cell lines, a stimulatory effect has been also
`documented. Zhao et al.30 described two cell lines
`(MDA PCa 2a and 2b) derived from bony metasta-
`ses of a patient with progressing prostate cancer
`despite hormonal therapy. These cells carry a mu-
`tated AR (L701H and T877A) that has a high affin-
`ity to cortisol and cortisone. Both cortisol and cor-
`tisone resulted in enhanced cellular proliferation
`and prostate-specific antigen (PSA) secretion in
`these two cell lines.30 Chang et al.31 studied the
`
`554
`
`UROLOGY 60 (4), 2002
`
`
`
`effects of various endogenous and synthetic glu-
`cocorticoids on LNCaP prostate cancer cell line
`(AR T877A mutant) growth and PSA secretion.
`Both 11-deoxycorticosterone (DOC) and dexa-
`methasone stimulated the AR-dependent PSA se-
`cretion. Furthermore, DOC was able to stimulate
`cellular growth at physiologically achievable lev-
`els, and dexamethasone seemed to have only a
`modest agonistic effect. These effects were medi-
`ated through the mutated T877A AR.
`
`CLINICAL UTILITY OF GLUCOCORTICOIDS
`
`IN AIPC
`
`Despite the limited preclinical data regarding the
`mechanism of their inhibitory effects on prostate
`cancer cell growth, glucocorticoids are used exten-
`sively in the treatment of AIPC. Miller and Hin-
`man32 were the first to report on the clinical activ-
`ity of glucocorticoids in AIPC. Ten patients with
`advanced prostate cancer who had progressed de-
`spite prior orchiectomy and estrogen therapy were
`treated with cortisone 50 to 200 mg/day. Subjective
`symptomatic improvements were noted in 8 of 10
`patients.32 Since this initial publication, multiple
`studies have evaluated single-agent glucocorti-
`coids in the treatment of AIPC. These studies have
`ranged from retrospective analyses to Phase III tri-
`als.
`
`RETROSPECTIVE AND PHASE II TRIALS
`Phase II and retrospective studies before 1995
`did not use PSA as a marker of disease activity
`and relied on subjective response criteria
`(Table I). 11,33,34 Tannock et al.11 evaluated retro-
`spectively in 28 patients and prospectively in 37
`patients, the palliative effects of daily prednisone
`7.5 to 10 mg. All enrolled patients had symptom-
`atic bony metastases with evidence of disease pro-
`gression despite orchiectomy and/or estrogen ther-
`apy. Of 28 patients analyzed retrospectively, 7 had
`improvement in pain and reduced requirements
`for analgesics lasting a median of 5 months. In 37
`patients evaluated prospectively, 14 patients
`(38%) had improvements in pain and a leveling or
`a decrease in their analgesic requirements for at
`least 1 month. The reduction in pain was associ-
`ated with improvements in quality of life (QOL).
`The clinical benefits lasted 3 to 30 months in 7
`patients. Alkaline phosphatase and acid phospha-
`tase were evaluated in this study but did not show
`any evidence of correlation with clinical response.
`Patel et al.34 treated patients with prostate cancer
`and measurable disease that had progressed de-
`spite orchiectomy or estrogen therapy with low-
`dose dexamethasone. Fifty-eight patients were
`randomized to receive either megestrol acetate or
`dexamethasone 0.75 mg daily. The study had com-
`
`plex response criteria that involved evaluations of
`multiple variables (bone scan, computed tomogra-
`phy, chest radiography, pain evaluation, acid phos-
`phatase, weight, performance status, alkaline
`phosphatase, and hemoglobin). Two of 29 patients
`on the dexamethasone arm had evidence of disease
`regression each lasting 359 and 512 days. Twenty-
`one patients had stable disease at a median of 86
`days. The median survival on the dexamethasone
`arm was 246 days.
`Studies conducted since 1995 used PSA as a
`marker for response (Table I). PSA is not a perfect
`response endpoint, but it is agreed as an indicator
`of biologic activity.35,36 The Prostate-Specific An-
`tigen Working Group has developed guidelines for
`using PSA as a measurement of outcome. Only a
`PSA decrease of more than 50% with a sustained
`decrease for 4 weeks is considered a PSA re-
`sponse.36 Such a decrease has been associated with
`an extension in overall survival in several stud-
`ies.37,38
`Storlie et al.39 retrospectively evaluated 38 pa-
`tients who had progressive disease despite orchiec-
`tomy and were treated with 0.75 mg dexametha-
`sone two or three times a day. They reported a 63%
`subjective symptomatic improvement and a 79%
`PSA decrease of more than 50%. The median time
`to biochemical progression in responders (PSA in-
`creases of greater than 50%) was 245 days (range
`99 to 660). Thirty-five percent of PSA responders
`had some evidence of bony disease regression and
`none of the nonresponders did. The study included
`only patients who had adequate follow-up. Thus,
`the study has potentially excluded nonresponders
`with early dropout because of disease progres-
`sion.39
`Kelly et al.40 conducted a Phase II trial in which
`30 patients with AIPC were treated with hydrocor-
`tisone 40 mg/day. Suramin was added to the treat-
`ment regimen at the time of PSA progression. All
`patients must have had evidence of disease pro-
`gression and documented castrate testosterone lev-
`els before entry on study. Patients who were taking
`flutamide had to have the drug discontinued; they
`were entered on trial only if they showed disease
`progression after flutamide withdrawal. Six pa-
`tients (20%) had a decrease in PSA of more than
`50% and the median duration of the decline was 16
`weeks. One patient had a PSA decline of 99% last-
`ing more than 52 weeks and 1 patient had PSA and
`bone scan stabilization lasting more than 44 weeks.
`Sartor37 retrospectively reviewed the results of
`29 consecutive patients with AIPC treated with
`prednisone 10 mg twice daily. All patients had to
`have a rising PSA despite adequate androgen abla-
`tion to be entered in the trial. Flutamide with-
`drawal was ruled out as a possible confounding
`variable in all assessable patients. Ten (34%) of 29
`
`UROLOGY 60 (4), 2002
`
`555
`
`
`
`TABLE I. Phase II and retrospective studies evaluating glucocorticoids in the treatment of progressive hormone-resistant prostate cancer
`
`Study
`
`Tannock et al.,11
`1989, prospective
`study
`
`Patients (n) and
`Treatment
`
`Entry Criteria
`
`Objective Responses and
`Post-Therapy PSA Decline
`
`MST and TTP
`(mo)
`
`Quality of Life and
`Subjective Responses
`
`37; prednisone 7.5–10
`mg
`
`Failed orchiectomy and/or
`estrogen therapy
`
`NA
`
`—
`
`38% had improvement
`in pain score at 1 mo;
`19% had sustained
`improvement in pain
`for 3–30 mo
`—
`
`—
`
`—
`
`Patel et al.,34 1990,
`prospective study
`
`29; dexamethasone
`0.75 mg BID
`
`Nishiyama,41 1998,
`prospective study
`
`7; low-dose
`dexamethasone 1.5–
`0.5 mg/day
`
`Sartor,37 1998,
`retrospective study
`Storlie et al.,39 1995,
`retrospective study
`
`Kelly et al.,40 1995,
`prospective study
`
`Weitzman et al.,42
`2000, prospective
`study
`Nishimura et al.,43
`2000, prospective
`study
`
`29; prednisone 10 mg
`BID
`38; low-dose
`dexamethasone 0.75
`BID/TID
`30; hydrocortisone 25
`mg QAM and 15 mg
`QPM
`12; dexamethasone 20
`mg PO Q6 ⫻ 3 Q3
`wk
`37; dexamethasone
`0.5–2 mg QD
`
`Failed orchiectomy or
`diethyl-stilbestrol
`therapy
`Patients with orchiectomy
`or LHRH therapy with
`progression after
`secondary hormonal
`therapy withdrawal
`Failed orchiectomy or
`LHRH agonist
`Failed orchiectomy;
`progressive disease
`
`Progressive AIPC
`
`Progressive AIPC
`
`Progressive AIPC
`
`2/29 objective responses
`(decrease in measurable
`disease)
`4/7 (PSA decrease of ⬎90%)
`after 3 mo of therapy
`
`MST 6.7
`
`TTP 3–11
`
`34% (PSA decrease of
`⬎50%)
`PSA response: 79% had a
`PSA decrease ⬎50% (not
`confirmed at 4 wk)
`20% (PSA decrease of
`⬎50%); 1 Pt had ⬎99%
`PSA decrease
`0% PSA response
`
`MST 12.8; median
`TTP 2
`
`—
`
`—
`
`—
`
`63% had symptomatic
`improvement
`
`—
`
`—
`
`62% PSA response (⬎50%
`decrease confirmed in 4
`wk)
`
`MTP 9
`
`11/18 had improvement
`in pain
`
`KEY: PSA ⫽ prostate-specific antigen; MST ⫽ mean survival time; TTP ⫽ time to progression; NA ⫽ not available; BID ⫽ twice daily; LHRH ⫽ luteinizing hormone-releasing hormone; TID ⫽ three times daily; Q ⫽ every; Pt ⫽ patient; PO ⫽ orally;
`AIPC ⫽ androgen-independent prostate cancer; QD ⫽ every day; MTP ⫽ median time to progression.
`
`
`
`patients achieved a PSA decline of more than 50%.
`Patients who had more than a 50% decline in their
`PSA level had a median survival time (MST) of 17.4
`months compared with 12.8 months in the overall
`patient population. The median progression-free
`survival was 2 months; however, 14% had a pro-
`gression-free survival of more than 6 months. Al-
`though the PSA response rate of 34% was higher
`than what has been reported with a daily 10-mg
`prednisone regimen, the retrospective nature of
`the study and the small number of patients limited
`result interpretation. The results of the study, how-
`ever, suggest an association between a decline of
`more than 50% in PSA and prolonged survival.
`Nishiyama41 evaluated in a small prospective
`study 16 patients with bony metastases in whom
`hormonal therapy had failed. All patients had been
`treated with orchiectomy or a luteinizing hor-
`mone-releasing hormone analogue in combination
`with chlormadinone acetate, estramustine, or flut-
`amide. Patients underwent hormonal withdrawal
`before study entry. Seven patients progressed de-
`spite hormonal withdrawal and were treated with
`dexamethasone 1.5 mg daily. Three of these pa-
`tients had a PSA response of more than 50%, and
`all responses lasted more than 6 months.
`Weitzman et al.42 evaluated a high-dose inter-
`mittent schedule of dexamethasone in patients
`with progressive prostate cancer despite androgen
`ablation. Twelve patients were treated with 20 mg
`dexamethasone every 6 hours for three doses re-
`peated every 3 weeks. None of the patients had a
`decrease in PSA of more than 50%. Although the
`number of patients was small, these data suggest a
`lack of efficacy of an intermittent schedule using
`glucocorticoids.
`Most recently, Nishimura et al.43 evaluated, in a
`prospective trial, the use of low-dose dexametha-
`sone in AIPC. Thirty-seven patients with a rising
`PSA and castrate testosterone levels were treated
`with daily dexamethasone 0.5 to 2 mg. Forty-nine
`of these patients had symptomatic metastases on
`entry. The median pretreatment PSA was 38 ng/mL
`(range 2.4 to 3570). Antiandrogens were discon-
`tinued at least 4 weeks before initiation of dexa-
`methasone. Twenty-three patients (62%) had a
`PSA decline of more than 50% that was sustained
`for more than 4 weeks. Four of the responders had
`decreasing PSA after antiandrogen withdrawal and
`before the start of dexamethasone. The median
`time to PSA progression in the responders was 9
`months. The MST in the PSA responders was 22
`months versus 8 months in the nonresponders.
`The favorable results seen in this trial were, at least
`partially, influenced by the responses to antiandro-
`gen withdrawal and perhaps the relatively low vol-
`ume of systemic disease.
`
`PHASE III TRIALS
`Glucocorticoids have served as the control arm
`in several Phase III trials of cytotoxic or hormonal
`therapies (Table II). Tannock et al.44 randomized
`161 symptomatic patients with AIPC to mitox-
`antrone plus prednisone or prednisone alone at 10
`mg/day and evaluated QOL as an endpoint. They
`described a 12% palliative response in patients re-
`ceiving single-agent prednisone. The PSA level de-
`creased by more than 50% in 22% of patients, but
`these responses were not reconfirmed in 4 weeks.
`Osoba et al.45 analyzed this same patient popula-
`tion for health-related QOL.45 Patients receiving
`prednisone had a significant
`improvement
`in
`health-related QOL scores at 6 weeks. This statis-
`tical significance was lost at 12 weeks of therapy.
`Kantoff et al.38 randomized 242 patients with
`AIPC to mitoxantrone plus hydrocortisone or hy-
`drocortisone 45 mg daily and evaluated the sur-
`vival and response rates as endpoints. The MST in
`the hydrocortisone arm was 12.6 months, and the
`median time to progression (MTP) was 2.3
`months. Twenty-two percent of patients had a
`maximal PSA decrease of more than 50%. Patients
`who had a maximal decrease in PSA responses of
`more than 50% had a significantly higher MST
`(20.5 versus 10.3 months). This study did not
`show a statistically significant difference between
`the two arms in survival or quality-of-life (QOL)
`measures.
`Gregurich46 randomized 120 asymptomatic pa-
`tients with progressive hormone-resistant prostate
`cancer to mitoxantrone plus prednisone or pred-
`nisone 5 mg twice daily. A PSA decrease of more
`than 50% occurred in 24% of patients on the pred-
`nisone arm. The MTP in the prednisone arm was
`3.8 months, significantly lower than in the combi-
`nation arm. The overall survival was equivalent in
`both arms.
`Small et al.47 randomized 460 patients with
`symptomatic, metastatic AIPC to low-dose
`suramin plus hydrocortisone or hydrocortisone 40
`mg daily. The primary endpoint of the trial was the
`evaluation of pain and analgesic use as primary
`indicators of response. Twenty-eight percent of pa-
`tients receiving single-agent hydrocortisone had a
`pain response and 8% had both a pain response and
`a decrease in opioid analgesic use. The PSA level
`decreased by more than 50% in 16% of patients
`receiving single-agent prednisone, and the MST
`was 9.2 months.47
`Fossa et al.48 randomized 201 patients with
`symptomatic AIPC to receive prednisone 5 mg four
`times a day or flutamide 250 mg 3 times a day. The
`subjective response was assessed on the basis of the
`performance status, reduction in analgesic use, and
`reduction of the pain score (World Health Organi-
`zation criteria). At 6 weeks of therapy, the subjec-
`
`UROLOGY 60 (4), 2002
`
`557
`
`
`
`TABLE II. Phase III trials with glucocorticoids as control arm
`
`Phase III Trials
`
`Tannock et al.,44 1996
`Osobay et al.,45
`1999
`
`Patients in GC arm
`(n) and Treatment
`
`81; prednisone 5 mg
`BID
`
`Entry Criteria
`
`Symptomatic, metastatic
`disease, progressing
`despite standard
`hormonal therapy
`
`Kantoff et al.,38 1999
`(hydrocortisone and
`mitoxantrone ⫹
`hydrocortisone)
`
`123; hydrocortisone
`30 mg QAM, 15 mg
`QPM
`
`Progressive, metastatic
`disease despite
`orchiectomy or LHRH
`analogue therapy
`
`Gregurich,46 2000
`(prednisone and
`mitoxantrone ⫹
`prednisone)
`
`Small et al.,47 2000
`(hydrocortisone and
`suramin ⫹
`hydrocortisone)
`Fossa et al.,48 2001
`(prednisone and
`flutamide)
`
`60; prednisone 5 mg
`BID
`
`231; hydrocortisone
`40 mg daily
`
`101; prednisone 5
`mg QID
`
`Asymptomatic
`metastatic HRPC with
`progressive disease
`despite standard
`hormonal therapy
`Symptomatic, metastatic
`disease, progressing
`despite standard
`hormonal therapy
`Symptomatic metastatic
`disease with
`progression despite
`orchiectomy or LHRH
`analogues
`
`Objective Responses
`and Post-Therapy
`PSA Decline
`
`22% (PSA decrease of
`⬎50%)—not
`reconfirmed in 4 wk
`
`21.5% (PSA decrease
`of ⬎50%); PR 4%
`(PR based on PSA
`necessitates a
`decrease in PSA of
`⬎80% lasting 6 wk
`24% (PSA decrease
`⬎50%)
`
`MST and Median
`TTP (mo)
`
`Quality of Life and
`Subjective Responses
`
`MST 11
`
`MST 12.6; median TTP
`2.3
`
`12.3% had a palliative
`response; HQL improved
`at 6 wk of therapy
`compared with initial time
`of recruitment; statistical
`significance lost at 12 wk
`—
`
`Median TTP 3.8
`
`—
`
`16% (PSA decrease
`⬎50% lasting 4 wk)
`
`MST 9.2
`
`21% PSA decrease
`⬎50%
`
`Median TTP 3.4; OS 10.6
`
`28% pain response based on
`decrease in pain level or
`analgesic use
`
`56% had a subjective
`response (physician
`assessed: decrease in pain
`and improved general
`condition); QOL scores
`favored prednisone arm
`
`KEY: GC ⫽ glucocorticosteroids; HQL ⫽ health-related quality of life; PR ⫽ partial response; HRPC ⫽ hormone-refractory prostate cancer; OS ⫽ overall survival; QOL ⫽ quality of life; other abbreviations as in Table I.
`
`
`
`tive response rate was 49% in the prednisone
`group. Subjective responses lasted for a median of
`4.8 months. Twenty-one percent of patients in the
`prednisone arm had a decrease in PSA of more than
`50%. In 9% of patients receiving prednisone, a PSA
`decline of more than 50% lasted more than 4
`weeks. A PSA decline was associated with a subjec-
`tive response. A PSA decline of more than 50% was
`associated with a favorable survival outcome. The
`MTP and MST for the prednisone arm was 3.4 and
`10.6 months, respectively.
`
`COMMENT
`
`We conclude that glucocorticoids have an inhib-
`itory effect on AIPC in vitro and in vivo. Cell
`growth inhibition is linked to several potential
`mechanisms: downregulation of adrenal andro-
`gens, modulation of cellular growth factors, and
`downregulation of AR-dependent transcription.
`Other, nongenomic pathways of glucocorticoids
`have recently been described, especially in the ner-
`vous system. The role of such pathways in prostate
`cancer antitumor activity has not yet been delin-
`eated. Thus, a full understanding of the mecha-
`nism of activity or resistance to glucocorticoids is
`far from complete.
`In total, the laboratory data show a heteroge-
`neous effect of glucocorticoids on prostate cancer.
`This heterogeneous effect reflects the interplay of
`two main variables: glucocorticoids and tumor bi-
`ology. Glucocorticoid activity varies depending on
`the chemical agent used, as well as on its dosage.
`The stimulatory effects of DOC compared with the
`neutral effect of dexamethasone on LNCaP exem-
`plify class-dependent activity.31 A dose-dependent
`response is manifested in the inhibitory effects of
`dexamethasone on the DU 145 cell line in which
`increasing activity is seen at concentrations be-
`tween 10⫺9M and 10⫺7M.27 Glucocorticoid activ-
`ity is also dependent on tumor biology such as GR
`expression or the presence of AR mutations. Dexa-
`methasone inhibitory activity is lacking in GR-neg-
`ative LNCaP cells, but it is clearly present in GR-
`expressing cells such as PC-3 and DU-145.27 AR
`mutations such as the combined mutation L701H/
`T877A or T877A have been associated with a par-
`adoxical stimulatory response to glucocorticoids,
`and the wild-type AR has not been shown to pro-
`duce such effects. Limited data are available on GR
`expression in human prostate cancer tissue. One
`study of 20 patients with localized prostate adeno-
`carcinoma reported minimal GR staining in neo-
`plastic epithelial cells.49 More recently, Nishimura
`et al.27 reported intense nuclear GR staining in 8 of
`16 patients with localized prostate cancer. To our
`knowledge, GR expression in AIPC samples has
`not been evaluated.
`
`The clinical data clearly indicate that glucocorti-
`coids have an overall palliative effect in the treat-
`ment of hormone-refractory prostate cancer when
`used on a daily basis. On the basis of data collected
`in Phase III trials, a PSA decrease of more than 50%
`in 16% to 20% of patients, a MTP of 2.3 to 3.4
`months, and a MST ranging between 9.2 and 12.6
`months are to be expected. The reported greater
`than 50% decrease in most of the studies listed
`does not constitute a partial response by the Pros-
`tate-Specific Antigen Working Group criteria be-
`cause of the lack of a 4-week confirmation of re-
`sponse. However, patients with a PSA decrease of
`greater than 50% had a longer lasting palliative
`effect and survival advantage than did non-PSA re-
`sponders. The QOL is improved, but the improve-
`ment is often limited to the first 6 weeks of therapy.
`Phase II and retrospective studies using low-dose
`dexamethasone have reported higher response
`rates.
`None of the randomized studies so far has shown
`a survival advantage of any combination therapy
`over glucocorticoids; however, the combination of
`mitoxantrone and prednisone or hydrocortisone
`seems to provide better palliation and an improved
`MTP and QOL. When randomized against the an-
`tiandrogen flutamide, prednisone resulted in a
`similar median survival but was associated with
`improved subjective responses and QOL. This fur-
`ther supports the use of glucocorticoids in the pal-
`liative setting or on failure of primary chemother-
`apy in AIPC.
`Although the clinical activity of glucocorticoids
`has been consistently reproduced, the benefits
`have been usually limited to a small subgroup of
`the target patient population. Careful study to de-
`termine the factors associated with clinical benefit
`in AIPC should be done. Do these responders have
`specific patterns of adrenal androgen production?
`Do specific AR mutations correlate with a pro-
`longed response to glucocorticoids? Is the re-
`sponse dependent on GR expression or overex-
`pression? Such efforts should identify the most
`active class and dose to be used, and most impor-
`tantly, in which patients. Until then, the optimal
`dose and schedule of glucocorticoids in AIPC will
`remain a subject of debate. Future progress in the
`use of glucocorticoids in AIPC can only happen by
`achieving a better understanding of the molecular
`pathology of this disease.
`
`REFERENCES
`1. Riccardi C, Zollo O, Nocentini G, et al: Glucocorticoid
`hormones in the regulation of cell death. Therapie 55: 165–
`169, 2000.
`2. Smets LA, Salomons G, and van den Berg J: Glucocor-
`ticoid induced apoptosis in leukemia. Adv Exp Med Biol 457:
`607–614, 1999.
`
`UROLOGY 60 (4), 2002
`
`559
`
`
`
`3. Miyoshi H, Ohki M, Nakagawa T, et al: Glucocorticoids
`induce apoptosis in acute myeloid leukemia cell lines with a
`t(8;21) chromosome translocation. Leuk Res 21: 45–50, 1997.
`4. Arai Y, Nakamura Y, Inoue F, et al: Glucocorticoid-
`induced apoptotic pathways in eosinophils: comparison with
`glucocorticoid-sensitive leukemia cells. Int J Hematol 7: 340 –
`349, 2000.
`5. Mann C, Hughes F, and Cidlowski J: Delineation of
`signaling pathways involved in glucocorticoid induced and
`spontaneous apoptosis of rat thymocytes. Endocrinology 141:
`528 –538, 2000.
`6. Carollo M, Parente L, and D’Alessandro N: Dexameth-
`asone induced cytotoxic activity and drug resistance effects in
`androgen-independent prostate tumor PC-3 cells are medi-
`ated by lipocortin 1. Oncol Res 10: 245–254, 1998.
`7. Smith RG, Syms AJ, and Norris JS: Differential effects of
`androgens and glucocorticoids on regulation of androgen re-
`ceptor concentrations and cell growth. J Steroid Biochem 20:
`277–281, 1984.
`8. MacAdams MR, Kamisischke A, Kemper DE, et al: Tes-
`tosterone levels in men with chronic obstructive pulmonary
`disease with or without glucocorticoid therapy. Eur Respir J
`11: 41–45, 1998.
`9. White RH, and Chipps BE: Reduction of serum testos-
`terone levels during chronic glucocorticoid therapy. Ann In-
`tern Med 104: 648 –651, 1986.
`10. Plowman PN, Perry LA, and Chard T: Androgen sup-
`pression by hydrocortisone without aminoglutethimide in or-
`chiectomised men with prostate cancer. Br J Urol 59: 255–257,
`1987.
`11. Tannock I, Gospodarowicz M, Meakin W, et al: Treat-
`ment of metastatic prostatic cancer with low dose prednisone:
`evaluation of pain and quality of life as pragmatic indices of
`response. J Clin Oncol 7: 590 –597, 1989.
`12. Perretti M: Lipocortin-derived peptides. Biochem Phar-
`macol 47: 931–938, 1994.
`13. Flower RJ, and Rothwell NJ: Lipocortin 1: cellular
`mechanisms and clinical relevance. Trends Pharmacol Sci 15:
`71–76, 1994.
`14. Choudhury Q, Newman S, et al: Lipocortin and the
`control of cPLA2 activity in A549 cells: glucocorticoids block
`EGF stimulation of cPLA2 phosphorylation. Biochem Phar-
`macol 52: 351–356, 1996.
`15. Croxtall JD, Choudhury Q, Tokumoto H, et al: Lipocor-
`tin and the control of arachidonic acid release in cell signaling:
`glucocorticoids inhibit G-protein dependent activation of
`cPLA2 Activity. Biochem Pharmacol 50: 465–474, 1995.
`16. Roberts AB, and Sporn MB. The transforming growth
`factor, in: Handbook of Experimental Pharmacology: Peptide
`Growth Factors and Their Receptors. Berlin, Germany, Spring-
`er-Verlag, 1990, pp 419-472.
`17. Hsing A, Kadomatsu K, Bonham MJ, et al: Regulation of
`apoptosis induced by transforming growth factor-1 in non-
`tumorigenic and tumorigenic rat prostatic epithelial cell lines.
`Cancer Res 56: 5146 –5149, 1996.
`18. Reyes-Moreno C, and Koutsiliers M: Glucocorticoid re-
`ceptor function possibly modulates cell-cell interactions in
`osteoblastic metastases on rat skeleton. Clin Exp Met 15: 205–
`217, 1997.
`19. Barrack E: TGF in prostate cancer: a growth inhibitor
`that can enha