A

s

c

(1w

ds

K

1

ptrzit

drt

1d

Original articlePhase II trial of thalidomide and daily oral dexamethasone

for treatment of hormone refractory prostate cancerprogressing after chemotherapy�

Silvia Romero, M.D.a, Gail Stanton, M.D.b, Joseph DeFelice, M.D.b, Fred Schreiber, M.D.c,Randy Rago, M.D.,d,1 Mayer Fishman, M.D., Ph.D.d,*

a Florida Cancer Specialists, Ft. Myers, FL, 33901-8108, USAb Morton Plant Hospital, Clearwater, FL, 33756-3809, USA

c Watson Clinic, Lakeland, FL, 33805-3019, USAd H. Lee Moffitt Cancer Center & Research Institute, Genitourinary Program, Department of Interdisciplinary Oncology, Tampa, FL 33612, USA

Received 17 May 2006; received in revised form 6 September 2006; accepted 7 September 2006

bstract

Background: Responses to monotherapy corticosteroid or thalidomide have been described in prostate cancer, in chemotherapy naïveubjects.

Methods: A total of 39 men with hormone refractory, metastatic prostate cancer who had progression during or after at least 1 conventionalytotoxic drug were treated on a single-arm Phase II trial with dexamethasone, 0.75 mg twice a day plus thalidomide, 100–400 mg/day.

Results: Best-observed responses included �50% prostate-specific antigen (PSA) reduction with no radiologic progression: 10 of 3926%; 95% confidence interval 13% to 42%). An additional 14 of 39 had decreased PSA but then with radiologic or other progression by2 weeks. Median progression-free survival was 84 days. Toxicity appeared treatable; there were 5 nonfatal thromboses. There was 1 subjectho had complete PSA and radiologic response; 4 responders tolerated treatment without progression for more than 1 year.Conclusions: PSA responses were frequent. Mostly, these were not durable, but some lasted more than a year. Further investigation on

eterminants of response durability for these or related compounds should be considered. The response rate of the present data does notupport Phase III testing of this regimen for this population. © 2007 Elsevier Inc. All rights reserved.

Urologic Oncology: Seminars and Original Investigations 25 (2007) 284–290

eywords: Thalidomide; Dexamethasone; Hormone refractory prostate cancer

adscbr

twfot

1

. Introduction

For patients with hormone refractory prostate cancerrogressing during or after chemotherapy, therapeutic op-ions are limited. There is an urgent need to identify newegimens with the potential to induce remission or stabili-ation. Considerations for developing such a treatment maynclude a limitation on the intensity and risk of side effectshat are acceptable, and the use of drugs with mechanisms of

� Celgene Corporation (Summit, NJ) provided financial support forata collection and, in many instances, thalidomide for subjects. M.F.eceived research funding from Celgene Corporation, the manufacturer ofhalidomide.

* Corresponding author. Tel.: �1-813-745-8343; fax: �1-813-745-8494.E-mail address: Mayer.Fishman@moffitt.org (M. Fishman).

c1 Deceased.

078-1439/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.oi:10.1016/j.urolonc.2006.09.017

ction different from conventional agents. In some cases,espite prostate-specific antigen (PSA), radiologic, orymptomatic response to a previous line of treatment, thelinical problems of fatigue or other toxicities, which maye cancer or treatment related, may be prohibitive for trial oretrial of intensive regimens.

In this single arm, single institution/affiliate networkrial, an all-oral regimen of dexamethasone and thalidomideas evaluated with the primary objectives of estimating the

requency of major (�50%) PSA responses, in the absencef other evidence of clinical or radiologic progression, andabulation of the toxicity experience.

.1. Corticosteroid

Dexamethasone is a corticosteroid. Prior experience with

orticosteroid monotherapy for hormone refractory prostate

cmcasasaejP�tt

1

ititmctm

1

tratcttif

2

toM

iCqpovom

cwebscdrOoplnclBp(

2

0saamoda1ammwu

arTmiss

2

loppvtW

285S. Romero et al. / Urologic Oncology: Seminars and Original Investigations 25 (2007) 284–290

ancer includes the control arm of randomized trials testingitoxantrone [1–3]. Of responders on the control (hydro-

ortisone or prednisone only) arms, the responses weressociated with a superior toxicity, but inferior pain re-ponse profile compared to the mitoxantrone-containingrms, and shorter durations. Other single arm studies havehown PSA-control activity of corticosteroid monotherapyt higher frequencies (range 22% to 61%) [4–6]. Thesexperiences were with primarily chemotherapy naïve sub-ects and cannot address if there is a palliative effect of theSA response. Together, these independent reports of20% of patients having �50% PSA response rates suggest

hat some hormone refractory prostate cancer is responsiveo corticosteroid monotherapy.

.2. Thalidomide

The second part of the present regimen is thalidomide. Its a pleiotropic agent with several effects that could beheoretically useful for prostate cancer treatment, includingmmune modulation, anti-angiogenic effects, lowering ofumor necrosis factor-�, and alteration in cellular adhesionolecules [7,8]. Earlier single-agent testing in prostate can-

er explored 2 thalidomide dose levels [9] and in combina-ion with docetaxel [10,11]. Some responses on the 200-g/day dose were observed [9], but with limited durability.

.3. Combination design

Working from these experiences in corticosteroid andhalidomide monotherapy, we designed the present trialegimen to test the combination of daily oral dexamethasonend thalidomide in patients with hormone refractory pros-ate cancer who had previously been treated with cytotoxichemotherapy and subsequently had progression. Correla-ive laboratory testing of the pharmacodynamic impact ofhe combination was not pursued because it was unlikely todentify which of the many putative mechanisms postulatedor both drugs were operative.

. Materials and methods

Data management financial support was through a con-ract with Celgene Corporation (Summit, NJ). Coordinationf the offsite (Mease, Lakeland) centers was through theoffitt Cancer Center Affiliates program.Prospective subjects’ consent was obtained using local

nstitutional review board approved consent forms (Moffittancer Center, Watson Clinic, Morton Plant Hospital). Re-uired disease features included histologic diagnosis ofrostate cancer with metastatic disease present, with previ-us treatment for advanced disease having included a con-entional cytotoxic drug, and with subsequent progressionf the disease. Prior chronic steroids, except daily dexa-

ethasone, during which the current progression was oc- h

urring, were not exclusionary, but prior thalidomide useas exclusionary. Progression was defined as clinical wors-

ning attributed to the cancer, new lesions identified onone scan, new or enlarging adenopathy or other metastaticites on computed tomography (CT), or 2 consecutive in-reases of PSA (in the presence of documented metastaticisease and at a level of at least 5 ng/ml). Other featuresequired for patients’ inclusion were Eastern Cooperativencology Group (ECOG) performance status 0–2, absencef neuropathy worse than grade 1, neutrophil count �1,000/�l,latelet count �100,000/�l, creatinine less �1.5� upperimit of normal, and total bilirubin �1.2� upper limit oformal. Baseline blood tests included complete blood count,omprehensive metabolic panel, and PSA, which were col-ected (at least) once per 28-day treatment cycle thereafter.aseline imaging scans included at least chest x-ray, CTelvis, and bone scan, with subsequent testing at 3-cycle12-week) intervals for sites of identifiable disease.

.1. Study drugs

Dexamethasone was prescribed as oral dexamethasone.75 mg twice a day (b.i.d.) and obtained from commercialupply. Thalidomide, using 50, 100, or 200-mg capsules, asvailable, was prescribed on the following schedule: week 1t 100 mg/day; week 2 at 200 mg/day; week 3 at 300g/day; and week 4 at 400 mg/day. Patients were contacted

n a weekly basis during the first 28-day cycle to guide theosing. Dose adjustment for apparent thalidomide intoler-nce was to hold the dose until the toxicity resolved to gradeor lower, stop dose increase, or reduce the dose. Gener-

lly, during the first 4-week cycle, the individually titratedaximal dose between 50 and 400 mg/day could be deter-ined, and from the fifth week overall (counted as cycle 2eek 1), the same individually titrated dose was used,nless further reductions were required.

Standard supportive care instructions included the use ofcathartic (senna or similar, 2 per day). Subjects were

equired to comply with the manufacturer’s System forhalidomide Education and Prescribing Safety program. Inany instances, subjects received their medication supply

n 28-day increments, directly from Celgene. In other in-tances, the drug was obtained through the commercialupply manufactured and distributed by Celgene.

.2. Response assessment

Responses were defined as downward PSA changes of ateast 50%, or PSA nonprogression in conjunction with otherbjective regression meeting partial response. Accrual waslanned in a 2-stage design. In the first stage, at least 12atients in either the “no previous corticosteroids” or “pre-ious corticosteroids” strata were analyzed for responses. Ifhere was at least 1 response, then accrual was to continue.

ith this early stopping plan, if the combination treatment

ad an underlying response rate of 5%, there would be a

5w

sufimohrsu

2

ccwoelawcdlo

3

pTw

scr

ssa

TP

C

AYPPPP

P

S

TW

A

C

C

G

G

H

L

N

S

T

V

286 S. Romero et al. / Urologic Oncology: Seminars and Original Investigations 25 (2007) 284–290

4% chance of stopping; if the underlying response rateere 15%, there would be a 14% chance of stopping early.The total number of ever-treated patients planned was 40

o that an estimate of the chance of response would have anncertainty (standard error of the mean) of about �9%. Thenal conclusion on whether to use this combination treat-ent experience as a basis for further study was to be based

n a threshold of at least 3 of 30 subjects, in either stratum,aving a partial or complete response. This result would beeached with a probability of 16% if the underlying re-ponse rate was 5% and with a probability of 77% if thenderlying response rate was 15%.

.3. Adverse event tabulation

Adverse events were tabulated using CTC version 2 (http://tep.cancer.gov/reporting/ctc__archive.html) when events oc-urred, or study staff evaluations at baseline, 2 weeks, 4eeks, and approximately 4-week intervals thereafter. Ance every 4 weeks office visit with a history and physicalxamination and blood tests as stated previously, and (ateast) once per 12 weeks radiologic imaging encompassingll known sties of measurable disease by CT and bone scanere used to assess disease response. In the absence of

linical or radiologic progression, “PSA progression” wasefined as 2 PSA test results at least 2 weeks apart, both ateast 50% higher than the nadir level reached since the startf study testing.

. Results

Between October 2000 and June 2003, 45 men withrostate cancer signed forms consenting to participate.here were 6 men who never started treatment because theyithdrew consent, were ineligible for absence of progres-

able 1atient characteristics

haracteristics Median Range

ge (yrs) 73 42–97rs since diagnosis 7 1–19rior docetaxel 35rior mitoxantrone 16rior ketoconazole 12SA level at baseline (ng/ml)Range 2.4–3,386Average 421Median 123

erformance statusECOG 0 16ECOG 1 21ECOG 2 3Median ECOG performance status 1

ite of diseaseBone by CT or bone scan 35

mPathologic lymph node identifiable by CT 9

ion, or ineligible for worsened performance status. Theharacteristics of the 39 ever-treated subjects are summa-ized in Table 1.

During the first cycle, thalidomide was prescribed astated previously. The doses identified as tolerable at thetart of the second cycle ranged from 100 to 400 mg, withmedian of 250 mg/day thalidomide. No early dose adjust-

able 2orst grade of AE per category, per subject, over entire period on-study

Grade1

Grade2

Grade3

Grade4

lkaline phosphatase and LFTsAlkaline phosphatase 2 1 4 1Bilirubin, AST, ALT 2 1

ardiac and edemaArhythmia, cardiac (bradycardia) 4 3Hypotension 1Edema 10 3 1

onstitutional & moodAnxiety, irribability, insomnia

(1 each)3

Fatigue 4 8 3Hot flashes 1 2

astrointestinal & nutritionConstipation 6 22 1Weight gain 3Weight loss or anorexiaDry mouth 2Other GI 15 2

enitourinaryUreteral obstruction 1 1Other urologic 4 1 1

ematologicThrombocytopenia 2 1 1Anemia 4 3Leukopenia or neutropenia 2 3 1 1

ungsPulmonary/respiratory 2 5 1

europathy & musculoskeletalSensory neuropathy 8 9 1Other neuropathy 8 5 1Dizzinies 2 1 2Ataxia 2 3Pain 4 8 4 1Leg cramps 2

kinMisc Dermatologic 12 3 1 0Skin cancers (Second

malignancy, unrelated)2

Cyst on arm (1)Thumb infection (1)

hrombosis & bleedingDVT-thrombosis grade 3

(w/o PE)2

Pulmonary embolus: Thrombosisgrade 4 (5 subjects continuedtreatment while on coumadin)

3

isionMisc. ocular 3 1One case diplopia from brain

metastasis

ents of dexamethasone were required.

3

tmmuwpi

sfitTenpbopfco

twmv

ibbgd

3

afpht

vdt(w

dlw2wr

Fb3

287S. Romero et al. / Urologic Oncology: Seminars and Original Investigations 25 (2007) 284–290

.1. Toxicities

Toxicity events were mostly grade 1–2. Because cathar-ics were systematically prescribed with the initial thalido-ide prescription, “constipation requiring medical treat-ent” (grade 2) was identified unless the subject stopped

sing cathartics. Some adverse events appeared consistentith attribution to thalidomide (e.g., neuropathy or consti-ation) and others to dexamethasone (e.g., easy bruising orrritable mood).

Fatigue, neuropathy, and constipation were the principleymptoms used to adjust the thalidomide dose during therst 28-day cycle. Observed events, regardless of attribution

o drug, prostate cancer, or other cause, are tabulated inable 2. There were 5 subjects who had serious thromboticvents, 2 deep venous thromboses, and 3 additional pulmo-ary emboli. Attribution of the thrombotic events to thereexisting prostate cancer, or to thalidomide or the com-ination treatment remains unknown in this data set; thalid-mide may increase the risk of thrombosis in patients withrostate cancer [12]. No cases of pulmonary embolus wereatal or required prolonged hospitalization; some subjectsontinued on-study on warfarin after having a thrombosisn-study or preceding study treatment.

Frequent low-grade edema was observed, and generallyreated with diuretics and evaluation for thrombosis. Thereere 3 subjects who had bradycardia (33–48 beats perinute) but were asymptomatic. Neutropenia has been pre-

ig. 1. Absolute and relative values of PSA. (A) Absolute value of PSA byy cycle number; scale shows 0–200 ng/ml. (C) Relative value of PSA bycycles, for the 4 subjects who were on treatment for more than 1 year.

iously observed with thalidomide therapy, but attribution i

n the single observed case is unknown. The grade 4 throm-ocytopenia on bone marrow biopsy appeared to be causedy marrow involvement by prostate cancer. There were norade 5 toxicities attributable to study drugs or occurringuring treatment.

.2. Disease response

The PSA response to treatment is shown using absolutend relative scales in Figs. 1A–C. As illustrated, there wasrequently some decrease in the PSA at some point; 14 of 39atients had at least a 50% decrease of PSA, and 10 of 39ad at least a 50% decrease of PSA and continued onreatment through the fourth cycle.

Fig. 2A shows the time course of progression-free sur-ival. Median progression free survival was reached at 84ays. Fig. 2B illustrates the progression-free survival amonghose 13 patients with at least 1 of the first 2 PSA values4 and 8 weeks) reduced by at least 50% compared those 26ithout such a reduction.Overall, 10 of 39 subjects had at least a 50% PSA

ecrease, on 2 consecutive measurements, with no radio-ogic or clinical progression at the point that the second PSAas identified. This result corresponds to a proportion of6% (95% exact binomial confidence interval 13% to 42%),hich supports the original hypothesis that the underlying

esponse frequency would exceed 15% and, using the orig-

number; logarithmic scale showing all points. (B) Absolute value of PSAumber versus first available (set at 100%). (D) Relative value of PSA, first

cyclecycle n

nal trial design, would be a basis to consider the further

dp

mt2rotsa

1swd1ral

naPp

ds[

4

eo“tgpotataidvang

4

tTdpprs(tbtaGhhoirarh

Fefa1

288 S. Romero et al. / Urologic Oncology: Seminars and Original Investigations 25 (2007) 284–290

evelopment of this combination treatment for this group ofatients.

Time on-treatment ranged from 6 to 515 days, with aedian of 84 days. The median corresponds to the evalua-

ion point at the completion of the third 28-day cycle (3 �8 � 84). Considering the clinical importance of durableesponses as compared to frequent, nondurable responses orf PSA-responses in the face of other signs of progression,here were no instances of discordance (i.e., PSA better butcan worse or PSA worse but scan improving) between PSAnd radiologic testing.

There were 4 subjects who had protracted responses overyear. Retrospective examination of the early PSA re-

ponse indicates that it was prompt and less than 50%ithin the first 2 months (Fig. 1D). These subjects’ toleratedoses of thalidomide during cycle 2 were 400, 200, 200, and00 mg, respectively, although all had reduction or inter-uption of doses because of apparent cumulative neuropathyfter about 1 year. In 1 subject who had a single bone scan

0 56 112 168 224 280 336 392 448 504 5600.00

0.25

0.50

0.75

1.00

Progression Free Survival

Days

frac

tion

with

out

prog

ress

ion

0 100 200 300 400 500 6000

0.25

0.5

0.75

1

A

B

ig. 2. Progression-free survival in days. (A) There were 39 subjectsver-treated, with a median of 84 days; none are ongoing. (B). Progression-ree survival in days, with the cohort retrospectively divided into those witht least 1 PSA �50% below baseline in the first 2 evaluations (solid, n �3), versus not (dashed, n � 26).

esion at the position of the sacroiliac joint, there was t

ormalization of the bone scan, durable for several months,fter the PSA had become undetectable. Subsequently, theSA began to increase, and the bone scan again becameositive at that location.

The median overall survival was 265 days. This survivaloes not appear superior to that observed in other largereries that describe therapy on chemotherapy naïve subjects1–3,13,14].

. Discussion

The population of patients with prostate cancer has het-rogeneous age, nonprostate cancer comorbidities, andther disease characteristics. Varying definitions of the termhormone refractory” can include PSA progression despiteestosterone suppression plus anti-androgen, or anti-andro-en-withdrawal. The use of cytotoxic drugs in advancedrostate cancer is typically considered after that point, butther secondary hormone-based maneuvers may be insti-uted as well. These maneuvers can include the use ofdrenal suppression or estrogenic substances. Diverse cyto-oxic drug classes that have shown clinical activity includelkylating drugs and platinum compounds, topoisomerasenhibitors such as mitoxantrone, and microtubule bindingrugs, including docetaxel, paclitaxel, estramustine, andinblastine. This two-part combination of dexamethasonend thalidomide would theoretically present new mecha-isms of anticancer activity compared to the cytotoxicroup.

.1. Corticosteroid experiences

There were 3 randomized trials using steroid-only con-rol arms that showed activity for steroid-only treatments.annock et al. [1] in 1996 reported on a 161 patient ran-omized trial, men with pain who had hormone refractoryrostate cancer who were randomized to treatment withredinsone 10 mg/day, with or without mitoxantrone. Painesponse was the primary endpoint; PSA response was aecondary endpoint. In the prednisone group, 12 of 5422%) men had �50% PSA reduction, and in the combina-ion group, 19 of 57 (33%) had similar reduction fromaseline PSA values. Studying men with hormone refrac-ory prostate cancer not previously treated with chemother-py, Kantoff et al. [2] reported on the Cancer and Leukemiaroup B 9182 trial. This study involved 242 men withormone refractory prostate cancer, randomized to dailyydrocortisone with or without mitoxantrone. For the sec-ndary endpoint of �50% PSA response, results were sim-lar to the previous report [1], with the combination armesponses at 42 of 112 (38%) and the hydrocortisone alonerm showing 25 of 116 (22%) [2]. In 2002, Berry et al. [3]eported a 120 patient trial in which asymptomatic men withormone refractory prostate cancer were randomized be-

ween predinsone 5 mg given orally (PO) b.i.d., with or

whc

twSpaddsth9

cpItp

swtsshr

bdcpsFlr

4

c[t(Pgtm

wlfsp

sccTptt

4

ttmrhsmptds

ibdpitacoohrcctb

britcrm

5

taI

289S. Romero et al. / Urologic Oncology: Seminars and Original Investigations 25 (2007) 284–290

ithout mitoxantrone every 3 weeks. The combination armad 48% of patients with �50% PSA reduction and theontrol arm 24%.

In a nonrandomized study of men with prostate cancerhat had progressed after orchiectomy (i.e., most not treatedith antiandrogens and not treated with chemotherapy),torlie et al. [4] in 1995 reported 23 of 38 (61%) men withrostate cancer who had �50% PSA response to dexameth-sone 0.75 mg PO b.i.d. Saika et al. [5] showed in 2001 thatexamethasone 0.75 mg PO b.i.d. therapy resulted in �50%ecreases of PSA in 7 of 19 (22%) men with bone meta-tatic prostate cancer. Akakura et al. [6] in 2003 reportedhat 11 of 25 (44%) patients with prostate cancer withormone refractory disease had �50% PSA reductions, andof 11 (82%) had improved pain responses.Another application of chronic daily steroids for prostate

ancer therapy is in combination with medical adrenal sup-ression therapy using ketoconazole or aminoglutethimide.n this case, it is the latter agents that are considered the keyherapeutic component, by eliminating adrenal androgenroduction.

These trials do not specifically address whether cortico-teroid has palliative activity compared to placebo, orhether the frequency of PSA decreases imply benefit in

erms of delayed overall disease progression or onset ofymptoms, or which corticosteroid could have a better re-ponse rate. However, they are a basis to consider that someormone refractory prostate cancer responds to corticoste-oids.

Two pathways within the prostate cancer cell that coulde affected by the choice of glucocorticoid, in this caseexamethasone, include the androgen receptor and glu-ocorticoid receptor pathways. Mutated androgen receptorrotein, harbored by some prostate cancer cell lines, may beusceptible to ligation by non-androgen steroids [15–17].rom this perspective, the choice of glucocorticoid based on

ess chance of ligation of endogenous mutated androgeneceptor could be a determinant of response probability.

.2. Thalidomide experiences

Single-agent testing in hormone refractory prostate can-er explored thalidomide dose levels at 200 and 800 mg/day9]. Observed responses, including some patients previouslyreated with chemotherapy, were reported showing 9 of 5018%) of the patients with 200 mg/day having at least a 50%SA decrease, and further 20 of 50 (40%) in the low-doseroup and 9 of 13 (69%) in the high-dose group with 10%o 49.9% reductions. There were 4 patients on treatment forore than 8 months.Later, Phase II testing of thalidomide in combination

ith docetaxel, a cytotoxic drug now associated with pro-ongation of median survival in patients with hormone re-ractory disease [13,14], on a weekly schedule (not theuperior schedule [14]) in patients with hormone refractory

rostate cancer not previously treated with chemotherapy m

howed a higher frequency of �50% PSA response in theombination (53% vs. 37%) and improved survival in theombination arm, although not statistically significant [10].his result suggests that the limited frequency of anti-rostate cancer activity of thalidomide is independent ofhat attributed to docetaxel. In the present study, most pa-ients had previously been treated with docetaxel.

.3. Combination experience

The median progression-free survival of this combina-ion of thalidomide and dexamethasone indicates clearlyhat the majority of patients do not have durable, clinicallyeaningful disease suppression with this regimen. The du-

ability of the low frequency, longer responses over 1 year,owever, appears to exceed the experience in hydrocorti-one or prednisone monotherapy series or thalidomideonotherapy, cited previously. Considering that these are

atients with different prior therapy characteristics and thathese are independent series, a direct comparison of theual-agent to single-agent response frequency is outside thecope of the present experience.

Another concept that the present data are consistent withs the idea that the (unknown relevant) mechanism of actiony thalidomide, dexamethasone, or the combination is in-ependent of the mechanism of response to docetaxel. In theresent case, the patients with prior docetaxel treatment,ncluding all 4 longer responding subjects, had responses tohe thalidomide-dexamethasone combination. This resultppears not dissimilar in magnitude to the incremental in-reased activity of the thalidomide-docetaxel combinationver thalidomide observed in the randomized Phase II studyf Dahut et al. [10]. One may hypothesize that among theeterogeneous population of prostate cancer with hormoneefractory disease, even after the point that progression afterhemotherapy has occurred, a subset not defined by prior oroncurrent docetaxel exposure may have a useful responseo agents with mechanisms comparable to the present com-ination.

The toxicity experience, including incidence of throm-oses and neuropathy, especially among the longer termesponders, remain issues that have to be managed carefullyn an individualized risk-benefit therapeutic decision. Allhe subjects in this series had been previously treated withytotoxic drugs, including docetaxel. The application of aegimen such as this could be considered after failure onore conventional agents.

. Conclusions

The low frequency of response identified argues againsthe routine use of this combination in the post-chemother-py setting or commitment of resources for Phase III testing.f used, the early pattern of PSA response may be a useful

arker to gauge the chance for longer duration response,

attpcgt

adkupsstpm

A

osaiMse

R

[

[

[

[

[

[

[

[

290 S. Romero et al. / Urologic Oncology: Seminars and Original Investigations 25 (2007) 284–290

lthough this retrospective, descriptive observation was nothe primary intent of the present report. It is not surprisinghat for the present data, a key determinant for protractedrogression-free survival was consistent with other prostateancer therapeutic experiences. A prompt PSA responseenerally portends better overall survival, no matter whathe intervention being used.

On the other hand, identification of the mechanism ofction for the long-term responding patients, whether byirect antitumor effect or more subtle alteration of cyto-ines, is of interest. It is possible that in the future, molec-lar features of prostate cancer, or proteomic or cytokinerofiling, could identify a subset of patients likely to re-pond well to this regimen, even in the post-chemotherapyetting. Alternatively, amplification of this mechanismhrough judicious choice of synthetic glucocorticoid com-ounds and with newer thalidomide-related compoundsay merit clinical testing.

cknowledgments

Randy Rago, M.D., deceased, was the author of theriginal protocol document. Critical reading of the manu-cript by Catriona Byrne (Celgene Corp.) is particularlyppreciated. Data coordinators at Moffitt Cancer Centerncluded Kathy McCollister, Diana Ewbank, and Brenda

oore-Shiggs. Dr. Alan Cantor provided assistance withtatistical issues in the study design. Anita Bruce providedditorial assistance.

eferences

[1] Tannock IF, Osoba D, Stockler MR, et al. Chemotherapy with mi-toxantrone plus prednisone or prednisone alone for symptomatichormone-resistant prostate cancer: A Canadian randomized trial withpalliative end points. J Clin Oncol 1996;14:1756–64.

[2] Kantoff PW, Halabi S, Conaway M, et al. Hydrocortisone with orwithout mitoxantrone in men with hormone-refractory prostate can-cer: Results of the cancer and leukemia group B 9182 study. J Clin

Oncol 1999;17:2506–13.

[3] Berry W, Dakhil S, Modiano M, et al. Phase III study of mitoxantroneplus low dose prednisone versus low dose prednisone alone in pa-tients with asymptomatic hormone refractory prostate cancer. J Urol2002;168:2439–43.

[4] Storlie JA, Buckner JC, Wiseman GA, et al. Prostate specific antigenlevels and clinical response to low dose dexamethasone for hormone-refractory metastatic prostate carcinoma. Cancer 1995;76:96–100.

[5] Saika T, Kusaka N, Tsushima T, et al. Treatment of androgen-independent prostate cancer with dexamethasone: A prospectivestudy in stage D2 patients. Int J Urol 2001;8:290–4.

[6] Akakura K, Suzuki H, Ueda T, et al. Possible mechanism of dexa-methasone therapy for prostate cancer: Suppression of circulatinglevel of interleukin-6. Prostate 2003;56:106–9.

[7] Kumar S, Witzig TE, Rajkumar SV. Thalidomide: Current role in thetreatment of non-plasma cell malignancies. J Clin Oncol 2004;22:2477–88.

[8] Macpherson GR, Franks M, Tomoaia-Cotisel A, et al. Current statusof thalidomide and its role in the treatment of metastatic prostatecancer. Crit Rev Oncol Hematol 2003;(Suppl 6):S49–57.

[9] Figg WD, Dahut W, Duray P, et al. A randomized phase II trial ofthalidomide, an angiogenesis inhibitor, in patients with androgen-independent prostate cancer. Clin Cancer Res 2001;7:1888–93.

10] Dahut WL, Gulley JL, Arlen PM, et al. Randomized phase II trial ofdocetaxel plus thalidomide in androgen-independent prostate cancer.J Clin Oncol 2004;22:2532–9.

11] Cox MC, Dahut WL, Figg WD. The use of thalidomide in androgen-independent prostate cancer. Urol Oncol 2006;24:246–9.

12] Horne MK III, Figg WD, Arlen P, et al. Increased frequency ofvenous thromboembolism with the combination of docetaxel andthalidomide in patients with metastatic androgen-independent pros-tate cancer. Pharmacotherapy 2003;23:315–8.

13] Petrylak DP, Tangen CM, Hussain MH, et al. Docetaxel and estra-mustine compared with mitoxantrone and prednisone for advancedrefractory prostate cancer. N Engl J Med 2004;351:1513–20.

14] Tannock IF, de Wit R, Berry WR, et al. Docetaxel plus prednisone ormitoxantrone plus prednisone for advanced prostate cancer. N EnglJ Med 2004;351:1502–12.

15] Matias PM, Carrondo MA, Coelho R, et al. Structural basis for theglucocorticoid response in a mutant human androgen receptor(AR(ccr)) derived from an androgen-independent prostate cancer.J Med Chem 2002;45:1439–46.

16] Krishnan AV, Zhao XY, Swami S, et al. A glucocorticoid-responsivemutant androgen receptor exhibits unique ligand specificity: Thera-peutic implications for androgen-independent prostate cancer. Endo-crinology 2002;143:1889–900.

17] Zhao XY, Malloy PJ, Krishnan AV, et al. Glucocorticoids can pro-mote androgen-independent growth of prostate cancer cells through a

mutated androgen receptor. Nat Med 2000;6:703–6.