Scandinavian Journal of Urology and Nephrology

ISSN: 0036-5599 (Print) 1651-2065 (Online) Journal homepage: https://www.tandfonline.com/loi/isju19

Parenteral Estrogen versus Combined Androgen Deprivation in the Treatment of Metastatic Prostatic Cancer - Scandinavian Prostatic Cancer Group (SPCG) Study No. 5

Per Olov Hedlund, Martti Ala-Opas, Einar Brekkan, Jan Erik Damber, Lena Damber, Inger Hagerman, Svein Haukaas, Peter Henriksson, Peter Iversen, Åke Pousette, Finn Rasmussen, Jaakko Salo, Sigmund Vaage & Eberhard Varenhorst

To cite this article: Per Olov Hedlund, Martti Ala-Opas, Einar Brekkan, Jan Erik Damber, Lena Damber, Inger Hagerman, Svein Haukaas, Peter Henriksson, Peter Iversen, Åke Pousette, Finn Rasmussen, Jaakko Salo, Sigmund Vaage & Eberhard Varenhorst (2002) Parenteral Estrogen versus Combined Androgen Deprivation in the Treatment of Metastatic Prostatic Cancer - Scandinavian Prostatic Cancer Group (SPCG) Study No. 5, Scandinavian Journal of Urology and Nephrology, 36:6, 405-413, DOI: 10.1080/003655902762467549 To link to this article: https://doi.org/10.1080/003655902762467549

Published online: 09 Jul 2009.

Submit your article to this journal

Article views: 72

Citing articles: 33 View citing articles

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=isju20 ORIGINAL ARTICLE

Parenteral Estrogen versus Combined Androgen Deprivation in the Treatment of Metastatic Prostatic Cancer ScandinavianProstatic Cancer Group(SPCG) Study No. 5

Per Olov Hedlund, 1 Martti Ala-Opas, 2 Einar Brekkan, 3 Jan Erik Damber, 4 Lena Damber, 5 Inger Hagerman, 6 Svein Haukaas, 7 Peter Henriksson, 8 Peter Iversen, 9 AÊ ke Pousette, 10 Finn Rasmussen, 11 Jaakko Salo, 12 Sigmund Vaage, 13 Eberhard Varenhorst 14 and the SPCG-5* Study Group

Fromthe Departments of Urology, 1KarolinskaHospital, Stockholm, Sweden, 2KuopioUniversity Hospital,Kuopio, Finland, 3UppsalaAcademic Hospital, Uppsala, Sweden and 4SahlgrensUniversity Hospital,Gothenburg, Sweden, 5OncologicCenter, UmeaÊ University, UmeaÊ , Sweden, 6Departmentof Cardiology, Huddinge Hospital, Huddinge, Sweden, 7Diakonissehjemmet Hospital,Bergen, Norway, 8Departmentof Medicine,Huddinge Hospital, Huddinge, Sweden, 9Departmentof Urology, Rigshospitalet,Copenhagen, Denmark, 10ResearchLaboratory for ReproductiveHealth, Karolinska Hospital, Stockholm, Swedenand the Departments of Urology, 11HerlevUniversity Hospital,Herlev, Denmark, 12Helsinki University Hospital,Helsinki, Finland, 13StavangerHospital, Stavanger, Norway and 14Norrko¨pingHospital, Norrko ¨ping,Sweden

(SubmittedFebruary 14, 2002. Accepted for publicationApril 24,2002)

Scand JUrol Nephrol 36: 405–413, 2002

Objective: Inthe mid-1980s, interest in parenteral estrogen therapy for prostate cancerwas renewed when it wasfound that it inuenced liver metabolism only marginally and had very fewcardiovascular side-effects.In this study high-dose polyestradiol phosphate (PEP; Estradurin 1)wascompared to combined androgen deprivation (CAD)for the treatmentof patients with metastaticprostate cancer.The aimof the study wasto compareanticancer efŽ cacy and adverseevents, especially cardiovascular side-effects. Material and Methods: Atotal of 917 patients with T0–4, NX,M1, G1-3 prostate cancerand an Eastern Cooperative Oncology Group performancestatus of 0–2 wererandomized to treatmentwith either PEP 240 mgi.m. twicea month for 2 months and thereafteronce amonth or utamide (Eulexin 1)250 mg t.i.d. per os in combination with either triptorelin (Decapeptyl 1)3.75 mgper month i.m. or, on an optional basis, bilateral orchidectomy. Atotal of 556 patients had died atthe timeof this analysis. Results: There wasno differencebetween the treatmentarms in termsof timeto biochemical or clinical progression and overall or disease-speciŽc survival. There wasno increasein cardiovascular mortality in the PEP arm.The PEP group had a higher prevalence of cardiovascular diseaseprior to the study and asigniŽcantly higher incidence of non-fatal ischemicheart events and heart decompensation during the study. Conclusions: PEP has an equal anticancerefŽ cacy to CADand does not increasecardiovascular mortality. Final evaluation of cardiovascular morbidity is awaiting further analysis and follow-up. PEP is considerably cheaperthan CAD. Key words: advanced disease,cardiovascular complications, combined androgen deprivation, hormone treatment,multicenter study, parenteral estrogen, prostate cancer.

PerOlov Hedlund, Skogsstigen 22, SE-131 42 Nacka, Sweden. E-mail: [email protected]

Oralestrogen treatment,once the commonest method changes in liver metabolism and grave deviations in forhormone manipulation of prostate cancerin the blood levels of coagulation factors seen afteroral Nordic countries, was largely abandoned in the 1970s therapy werereported to be absent or only marginally due its signiŽcant cardiovascular toxicity. In the mid- present afterparenteral therapy (1). Two Finnish 1980s, interest in estrogen therapy was renewed when studies showed that polyestradiol phosphate (PEP; it was found that estrogen administered parenterally Estradurin1;PharmaciaAB, Sweden) given atadose did not induce these side-effects. The signiŽcant of 160 mg i.m. per month did not lead to higher cardiovascular morbidity or mortality than bilateral *Acompletelist of themembers of theSPCG-5 studygroup is orchidectomy (2)or gonadotrophin-releasing hormone givenin theAppendix. analog therapy (3). This dose of PEPdid however have

Ó 2002Taylor & Francis. ISSN0036– 5599 ScandJ UrolNephrol 36 406 P.O.Hedlund etal. inferior anticancer efŽcacy, measured in termsof about the study and gave their oral consent to progression-free survival. This may be explained by participate. the factthat 160 mg of PEPper month does not Patients werestratiŽ ed according to country (with decrease serumtestosterone to castration levels (4). In Iceland being included in the Norwegian group), aSwedish pilot study it was found that PEPgiven ata ECOGperformance status 0–1 vs 2, alkaline phospha- dose of 240 mg per month decreased serumtestoster- tase level under or over 1.25 the upper limitof one to castration levels, and that this was most rapidly normal and whether or not they£ had aprevious or achieved by giving PEP240 mg every second week current history of cardiovascular disease. Randomiza- during the Žrst 2months (5). In an open study of 40 tion was done by the Oncologic Center of the prostate cancerpatients, no increase in cardiovascular Karolinska Hospital, Stockholm, atwhich the eligi- toxicity was noted during the Žrst 3years of this bility of the patients was initially checked. Patients therapy (6). wereallocated to treatmentaccording to their position The aimof the present study was to comparePEP on the stratiŽcation list. with combined androgen deprivation (CAD),with Polyestradiol phosphate (Estradurin) was given as overall survival as the primaryendpoint. Secondary i.m. injections of 240 mgtwicea month for2 months objectives wereto comparetime to biochemical and (total of Žve doses), and thereafteronce every month. clinical progression, cancer-speciŽc survival, cardio- CAD was given as utamide tablets (Eulexin 1; vascular toxicity, other adverse events and quality of Schering-Plough AB, Stockholm, Sweden), 250 mg life.Some data fromthe Žrst evaluation have pre- t.i.d., in combination with either triptorelin (Decapep- viously been published (7). The present paper gives a tyl1;Ferring AB,Malmo¨,Sweden), 3.75 mgi.m. per morecomprehensive presentation of the design and month, or, on an optional basis, bilateral orchidectomy. initial results. Flutamide treatmentwas started 1week before the Žrst triptorelin injection. Irradiation of the breasts prior to therapy was optional. MATERIALAND METHODS The patients werefollowed by visits to the trialist Between December1992 and June 1997, 917 menwith 1, 3and 6months afterthe start of the study and hormone-naive, T0–4, NX,M1, G1–3 prostate cancer thereafterevery 6months until clinical progression was and anEastern Cooperative Oncology Group (ECOG) clearly established. Between these visits, prostate- performancestatus of 0–2 wererandomized to receive speciŽc antigen (PSA)values weredetermined in treatmentwith either PEPor CAD. All patients had every patient, so that the PSA was measured every 3 skeletal metastases as evaluated frombone scans months. Ifrequired, patients could request more supplemented with X-rays when needed. The primary detailed examinations. At every visit patients were tumor was staged by means of digital rectalexamina- questioned and evaluated with regard tosymptoms or tion according tothe TNMclassiŽcation of 1987 and signs of disease progression and adverse events. Blood graded according to the World Health Organization pressure, body weight, performancestatus and pain and (WHO)system, either as aresult of Žne-needle analgesic scores (seeTable II)werenoted and levels of aspiration cytology (8)or histologically fromtrans- hemoglobin, creatinine, PSA and liver enzymes, urethral resection of the prostate specimens (9). The including alkaline phosphatase, weremeasured. Digital extent of bone disease was calculated frompretreat- rectalexamination, bone scan or X-rays werenot done ment bone scans according to amodiŽed Soloway regularly and only ifconsidered necessary forevalua- score as follows: score 1, the total areaof hot spots is tion of the disease status. The evaluation of clinical less than three bodies of alumbar vertebra; score 2, the progression was based on clinical signs and symptoms. total areaof hot spots is largerthan that of score 1, but When clinical progression was deŽnitively established, <75% of the total scan; and score 3, superscan (10). treatmentand follow-up of the patient wereat the Exclusion criteriawere as follows: patients who had discretion of the individual trialist. The timeand cause received previous systemic treatmentfor prostate of death wererecorded. canceror with amalignancy of any other kind, with Monitoring was done by researchnurses or monitor- the exception of basal cellcarcinoma of the skin; ing specialists by means of visits to the trialist atleast patients who had suffered myocardial or cerebral once ayear forthe Žrst 5years. Afterthe Žfth year, infarction 1month before the start of the study; monitoring was done only by means of telephone patients withµ previous or current liver disease with a conversations and correspondence. Some centers with bilirubin or alanine aminotransferase value above the many patients in the study weremonitored more upper limitof normal; and patients who it was felt closely. To check compliance with medication the would not be able to comply with the study protocol. patients had to bring their empty vials of Eulexin at The patients weregiven oral and written information their visits to the trialist and charts werealso kept on

ScandJ UrolNephrol 36 Estradurin versus combined androgen deprivation 407 which the nurses noted all injections of Estradurin or and ventilatory scintigraphies or angiography. Anon- Decapeptyl. Astatement fromthe trialist regarding fatalevent was only counted once inapatient even if compliance was included in the case report form(CRF) the samecategory of event happened several times. If forclinical progression. such an event was ultimately the cause of death it was counted again in the cause-of-death table. DeŽnition and analysis of endpoints Timeto biochemical progression was calculated from Evaluation of cardiovascular toxicity: the blind randomization to the timewhen the Žrst risein PSA observer was observed, followed by acontinuous risein In order to optimize the evaluation of cardiovascular consecutive measurements. Timeto clinical progres- events, all such events wereevaluated by ablind sion was deŽned asthe timefrom randomization to the observer, acardiologist with aspecial interest in Žrst suspicion of clinical deterioration of the disease, cardiovascular side-effectsof estrogen treatment. which could be supported by further clinical deteriora- When acardiovascular event was suspected, the tion in consecutive evaluations. Clinical deterioration patient’s Žle, blinded as to the treatmentthat the was thus based on symptoms, most commonly pain or patient had received forprostate cancer, was sent to the decreased performancestatus, but also in afewcases blind observer, who decided whether the event on micturition problems, lymph edemaor uremia.The qualiŽed as acardiovascular event or not, according evaluation of biochemical and clinical progression was to the criteriaof the trial.This was also done forthe done according to the retrospective sequential method categorization of the cause of death ifdeath was not (11). Overall survival was the timefrom randomization obviously caused by the malignant disease. The blind to death fromany cause. Analysis of cancer-speciŽc observer was also to be contacted ifthere was any survival included deaths fromprostate canceror deaths uncertainty atstratiŽ cation as to whether apatient’s fromanother disease with signiŽcant contribution to disease prior toenrollment in the trialcould be counted the prostatic malignancy, i.e. in these patients clinical as acardiovascular disease according tothe criteriaof progression had occurred before death (11). the trial.

Evaluation of adverse events Evaluation of other toxicity Strictcriteria for different cardiovascular events were Gastrointestinal and liver toxicity, as wellas allergic or established as follows. cutaneous manifestations, weregraded according to the WHO scale (12). Hot ushes wereevaluated in terms Heart decompensation. Signs of pulmonary crepita- of frequency and annoyance according to Fro¨din etal. tions or rales, stasis of pulmonary vessels on X-rayor (13). The extent of gynecomastia was recorded as athird heart sound should be registered. Use of follows: slight, only involving the mammarygland and diuretics, angiotensin-converting enzyme inhibitors or nipple; moderate, involving the mammarygland and digitalis should be needed. including lipomastia; or severe, areaof the breast involved largerthan the patient’s Žst. Ischemic heart disease. Myocardial infarction accord- Datawere continuously collected in adata bank. ing to the WHO criteriaor unstable angina with Prior to locking the data afterthe death of the patient, a progression to stable angina with frequent attacks or Žnal check of all CRFs was done by aDataQuality along-lasting attack (>15 min), with newly devel- Committee, consisting of 10 experienced urologists, oped ST-change or T-inversion. two each fromDenmark, Finland and Norway and four fromSweden. Cerebral ischemic event. Cerebralinfarction seen on The trialwas analyzed on an intent-to-treat basis. computerized tomography or transient ischemic at- The study was performedin accordance with the tacks with clearneurological symptoms fromregions recommendations of the Helsinki Declaration (World of the internal carotid or vertebral arteries. Medical Assembly, Helsinki, 1964; amended in Tokyo, 1975, Venice, 1983 and Hong Kong, 1989). The study Intermittent claudication. Only severe intermittent was approved by the Ethical Committeeof the claudication ata maximumwalking distance of Karolinska Institute, Stockholm and by the local ethical 200 mto be counted. committeesof the participating centers.

Venous thromboembolism. Thromboses to be diag- Statistical methods nosed using phlebography, pletysmography, 125I Ž bri- The aimof the study, according to the protocol, was to nogen scanning or thermography. Pulmonary embol- show that PEPtreatmentwas equally effectiveas CAD ismto be veriŽed by means of combined perfusion in termsof overall median survival by testing the

ScandJ UrolNephrol 36 408 P.O.Hedlund etal.

Table I. Parameters of stratiŽcation Table II. Other demographic characteristics at entry

PEP CAD Characteristic PEP CAD Parameter (n = 455) (n = 455) Age (years)a 71.9 (71.2–72.6) 72.2 (71.5–72.9) Country Body weight (kg) a 75.2 (74.4–76.0) 76.2 (75.4–77.0) Denmark 106 107 Pain scoreb Finland 46 44 0 192 185 Norway and Iceland 51 51 1 137 145 Sweden 254 255 2 94 98 Performance status 3 29 24 ECOG 0–1 375 384 4 3 3 ECOG 2 80 71 T stage Alkaline phosphatase level T0 1 4 Low 232 233 T1 14 19 High 225 222 T2 68 78 Previous cardiovascular disease 78 66 T3 244 249 Ischemic heart disease 46 40 T4 110 98 Heart decompensation 13 12 Grade of malignancy Ischemic cerebral disease 6 4 1 67 69 Venous thromboembolism 11 8 2 211 203 Intermittent claudication 2 2 3 163 177 Soloway category of bone metastases 1 152 167 2 250 233 3 49 49 hypothesis using two one-sided level alpha-tests (14). Pretreatment irradiation of 234 67 Adifferencein timeto death of <20% was not the breasts B-hemoglobin(g/ l) a 92 (86–98) 91 (86–99) considered clinically signiŽcant. To demonstrate an a S-creatinine ml/l 98 (94–102) 102 (97–106) equivalence in median survival with apower of 80% S-testosterone nmol/l a 14 (13–14) 14 (13–15) a and atan alpha level of 5%, 371 deaths wereneeded per S-PSA mg/l 823 (632–1014) 719 (597–841) group. Atotal of 900 randomized patients were a Mean and 95% conŽdence interval. recommended in order to obtain the necessary number b 0=no pain; 1=slight pain, non-opioid analgesics occasionally; of deaths. 2=moderate pain, non-opioid analgesics regularly; 3=severe pain, Atotal of 917 patients wereincluded in the study. opioids occasionally; 4=intolerable pain, opioids regularly. The analyses of overall and cancer-speciŽc survival and of the timeto progression werebased on the events in the 556 deceased patients forwhom data were tion obtained fromthe total available materialfrom all examined by the DataQuality Committee. Survival randomized patients. Fisher’s exact test was used in was measured fromthe date of randomization to the comparisons between the two treatmentgroups. The date of death by any cause foroverall survival or to the softwareused forall analyses was SPSS forWindows date of death by prostate cancerfor cancer-speciŽ c (SPSS Inc., Cary, NC). survival or to the most recentfollow-up available in the database. The deaths of the 556 deceased patients occurred between March1993 and December1999. RESULTS The most recentdate of follow-up forthe remaining patients ranged between January 1995 and May 2000. In the present evaluation the median follow-up was Timeto progression, clinically or biologically, was 27.1 months forthe PEPgroup and 27.4 months forthe determined fromthe date of randomization to the date CAD patients. Atotal of 556 patients (61%) had died. of progression, death or most recentfollow-up. Threepatients werefound to be non-eligible as they ACox proportional hazard regression model was had aECOGperformancestatus of 3atrandomization, used to estimatethe relative effectsof treatmentswith leaving 914 patients available forevaluation. Only 901 regard to overall survival. The hypothesis stated in the of these had atleast one follow-up documentation. In protocol (assuming aconstant hazard) corresponds to a the CAD group, 159 patients (35%) underwent B relative hazard expressed as e = ¶PEP(t)/¶CAD(t) < orchidectomy and 298 (65%) weretreated with 1.25, where Bwas estimated fromCox regression. triptorelin. The parametersof stratiŽcation arelisted Kaplan–Meier plots wereperformed for overall and in Table I.The different formsof previous cardiovas- cancer-speciŽc survival and timeto progression. cular disease areincluded although stratiŽcation was Differencesbetween the two treatmentgroups in terms performedonly on the basis of whether or not the of timeto progression weretested using the log rank patient had ahistory of cardiovascular disease. As is test. A p-value of 0.05 was considered signiŽcant. apparent fromTable I,there weresome differences Analyses foradverse events werebased on informa- between the groups with regard to cardiovascular

ScandJ UrolNephrol 36 Estradurin versus combined androgen deprivation 409

Fig. 1. Kaplan–Meier estimates of biochemical progression; Fig. 3. Kaplan–Meier estimates of overall survival. p =0.58 (log rank test). Median time to biochemical progression: PEP10.2 (9.4–11.0), CAD10.1 (9.3–10.9) months.

uremiawere the reasons forassessing the condition as clinical progression. morbidity. As shown in Table II,other baseline Overall and disease-speciŽc survival areshown in characteristics werewell balanced. Figs 3and 4. The calculated hazard ratio foroverall Timeto biochemical and clinical progression forthe survival was 0.96, with a95% conŽdence interval of 556 patients that had died is shown in Figs 1and 2. 0.82–1.12. The corresponding Žgures forcancer- Therewere no signiŽcant differences between the two speciŽc survival were0.91 and 0.77–1.08. The PEP treatments: p =0.58 and 0.87, respectively. Patients and CAD treatmentswere equally effectivein termsof who died of another disease without having progressed median overall survival: p = 0.001. (n =50), patients who progressed rapidly and died of Cause of death is shown in Table III.There was no prostate cancerwithout having been registered fora differencebetween the two groups. Patients who were period of progression in the 6-month interval between found dead athome and in whom no post-mortem the routine follow-up visits ( n =20), patients who died investigation was done wereconsidered to have died fromunknown causes ( n =13) or those forwhom PSA fromunknown causes. Causes of death other than values weremissing ( n =14) werecensored. Clinical prostatic malignancy, with and without contribution of progression was Žrst seen as recurrenceof pain and prostatic cancer, areshown inTable IV.Therewas no decline in performancestatus in most patients but in signiŽcant differencebetween the two treatmentarms. 43/556 patients lymph edema, micturition problems or Non-fatal cardiovascular events areshown in Table V.Therewas signiŽcantly moreischemic heart disease and heart decompensation in the PEPgroup, with other

Fig. 2. Kaplan–Meier estimates of clinical progression; p = 0.87 (log rank test). Median time to clinical progression: PEP13.7 (12.5–14.9), CAD13.5 (12.4–14.6) months. Fig. 4. Kaplan–Meier estimates of cancer-speciŽc survival.

ScandJ UrolNephrol 36 410 P.O.Hedlund etal.

Table III. Causes of death Table VII. Temporary or persistent dose reduction/termination of therapy due to adverse events PEP CAD p PEPFlutamide Triptorelin Prostate cancer 223 223 Other disease with contribution of prostate Nausea/diarrhea 13/20 cancer 16 29 NS Increase in liver enzymes 7/11 Other disease without contribution of Cardiovascular events 1/10 prostate cancer 29 21 NS Cutaneous (allergy) 0/2 0/2 Unknown 9 6 Pain of injections 0/3 Flush 0/2 NS=not signiŽcant ( p > 0.05). Impotence 0/2 Gynecomastia 0/1 Fatigue, depression 0/2 0/2 a Misunderstanding 2/1 Table IV. Causes of death other than prostate cancer Miscellaneous 0/4 With Without contribution of contribution of prostate cancer prostate cancer PEPCADPEPCAD cardiovascular events being equally distributed. The Ischemic heart disease 2 3 5 3 principal trialinvestigator advised the trialist to stop Heart decompensation 2 3 7 2 PEPtreatmentin 4patients who suffered amyocardial Ischemic cerebral disease 4 3 2 5 infarction during treatment,as this was considered to Venous thromboembolism 0 0 1 1 Cerebral hemorrhage 0 1 0 2 be in the interests of the patients. Other non-fatal events Other malignancy 2 2 9 3 areshown in Table VI.Temporary or persistent Septicemia 1 1 2 1 reduction of the dosage or cessation of therapy Suicide 1 0 0 1 Trauma 0 1 0 1 occurred in 85 patients, the reasons forwhich are Other non-malignant disease b 4 1 3 2 shown in Table VII.

a The groups are too small for calculation of any signiŽcant differences. DISCUSSION b Heterogeneous group. This study shows that the PEPregimen, which is Table V. Non-fatal cardiovascular events considerably cheaper than CAD, has an anticancer efŽcacy equal to that of CADwhen evaluated in terms PEP CAD of timeto biochemical and clinical progression and (n = 455) (n = 455) p overall and disease-speciŽc survival. The patient Ischemic heart disease 17 5 0.009 materialrepresents many patients with alarge tumor Heart decompensation 20 9 0.035 Ischemic cerebral disease 9 10 NS burden, as can be seen fromthe pretreatmentPSA Venous thromboembolism 9 9 NS values and Soloway scores. The periods of disease-free Intermittent claudication 2 3 NS survival wererelatively short in the present study, indicating that many patients progressed and died NS=not signiŽcant ( p > 0.05). rapidly.

Table VI. Other non-fatal events stratiŽed by severity and presented as percentages of the groups as follows: PEP( n = 446)/CAD (n = 451)

Severity 0 1 2 3 4 Nausea 80.5/74.9 14.8/17.5 3.8/5.3 0.9/2.0 0.0/0.2 Diarrheaa 90.4/74.3 7.6/10.9 1.6/7.1 0.4/7.3 0.0/0.4 Cutaneous b 96.4/94.9 2.2/2.4 0.9/2.4 0.2/0.2 0.2/0.0 Flush frequency c 69.5/25.7 27.6/40.4 2.9/25.5 0.0/8.4 0.0/0.0 Flush bother d 74.0/33.5 20.6/29.7 4.7/29.0 0.7/7.8 0.0/0.0 Gynecomastia(%) Prophylactic irradiation 40.7/52.3 47.8/44.6 9.6/3.0 1.7/0.0 No irradiation 15.5/63.9 52.6/30.9 27.6/4.8 4.2/0.2 Pain after drug injections e 93.5/100.0 2.8/0.0 1.0/0.0

a 0=none; 1=slight; 2=moderate; 3=severe, requiring therapy; 4=intolerable. b 0=none; 1=erythema; 2=dry desquamation/ pruritus; 3=ulceration; 4=exfoliative dermatitis. c 0=none; 1=1–3/ day; 2=4–10/ day; 3=>10/day d 0=not; 1=slightly; 2=moderately; 3=greatly. e Only for the 556 patients who died. For other deŽnitions, see Material and Methods.

ScandJ UrolNephrol 36 Estradurin versus combined androgen deprivation 411

No signiŽcant increase in cardiovascular mortality metastaticpain on closer evaluation and with knowl- with PEPwas observed. Therewas, however, a edge of the further development of the disease in the signiŽcant increase in non-fatal ischemic heart disease patient. Ithas been shown that ischemic complications and heart decompensation. These results arein of oral estrogen treatmentcan be reduced by prophy- accordance with those of the Finnprost study no. 6, laxis with aspirin (19). In aFinnish study this wheretreatment with the samedose of PEPwas prophylaxis was given to patients treated with compared to orchidectomy in 440 patients (15). The 160 mg of PEPper month (2). No cardiovascular Žnal implication of our Žnding must, however, await complications occurred in the PEPgroup, but the further follow-up and analysis owing to the discre- anticancer effectwas inferior tothat of orchidectomy. pancy between the mortality and morbidity data and the No data areavailable concerning the efŽcacy of aspirin higher prevalence of cardiovascular disease prior to the prophylaxis or any other anticoagulant therapy during study inthe PEPgroup. treatmentwith 240 mg of PEPper month. Adetailed Itis well known that oral estrogen treatmentinduces analysis of the patients who werereceiving anti- grave deviations in liver metabolism leading to an coagulant therapy atthe startof this trialwill be done increase in coagulation factors VIIand X,adecrease in separately. The incidence of cardiovascular events antithrombin IIIand changes inthe Žbrinolytic system, encountered in the PEPgroup was very much lower all of which aresupposed to contribute to cardiovas- than that registered in earlierstudies in which oral cular toxicity. Ithas been postulated that the increase in estrogen treatmentwas administered (18) and can be factorVII is strongly related to myocardial ischemia as considered amodest problem. Despite stratiŽcation, the increase is directly proportional to the depression of the PEPgroup contained morepatients with previous the ST-segment on the electrocardiogram (16). How- cardiovascular disease and amoredetailed evaluation ever, afterparenteral estrogen therapy, factorVII was of cardiovascular adverse events and pre-trialcardio- only slightly and insigniŽcantly elevated and anti- vascular disease will be done separately when amore thrombin IIIdecreased slightly (1). Ithas also been complete follow-up is available. Itis important that the found that elderly menwith advanced prostatic urologist prescribing PEPis awareof the cardiovas- malignancy arealready in ahypercoagulable state cular risk involved, monitors the situation closely and before therapy, with increased levels of coagulation can interfereif symptoms and signs occur. factorV, Žbrinogen, Žbrinogen degradation products The treatmentarms showed well-known differences and plasminogen inhibitors (17). with regard to other side-effects(Table VI). Itmay When the present study was designed much interest seemsurprising that so many PEPpatients experienced was focused on the evaluation of cardiovascular side- slight or moderate nausea and diarrhea. Itwas also effects.Only very fewpatients wereexcluded from unexpected that so many PEPpatients reported slight randomization due to previous cardiovascular morbid- hot ushes, as estrogen is often used to alleviate ushes ity, i.e. those who had suffered amyocardial or cerebral in castrated patients. This phenomenon, which has also ischemic event within 1month of the date of been reported in other trials (3, 20, 21), is dealt with in randomization. Atotal of 142 patients with previous aseparate publication fromthis study (22). Irradiation cardiovascular disease according tothe strictly deŽned of the breasts seemed to have amorepronounced effect protocol criteriawere included inthe study. In addition, in PEPpatients than in the CADgroup. Dose reduction patients with non-protocol-deŽned cardiovascular con- or termination of therapy was most frequent in ditions, such as hypertension and atrialŽ brillation, utamide-treated patients. In most patients with in- werealso included. Itis wellknown that previous creased liver enzymes the increase was very modest cardiovascular disease is astrong risk factorfor and no criticalcases wereseen during the trial. Itis cardiovascular complications during oral estrogen likely that registration of side-effectswas relatively therapy (18). complete in this study as, instead of registering side- The blind observer was consulted on 255 occasions effectsonly afterquestioning the patient regarding and her evaluation of cardiovascular event reports whether there wereany problems overall, the CRFs proved to be of great importance. Itcan easily happen contained separate boxes that could be ticked for that apatient with leg edemaor apatient who reports nausea, diarrhea, cutaneous, anaphylactic, gynecomas- some dyspnea on uphill walking can be incorrectly tia, ush frequency, ush bother, ischemic heart, registered with heart decompensation, especially ifthe ischemic cerebral,venous thromboembolism, intermit- CRFs contain questions and boxes wherethis can be tent claudication, heart decompensation and others. It ticked and especially ifthe patient is on estrogen is probable that the pain associated with PEPinjections treatment.Also, pain in the legs or thorax maybe is morefrequent than was reported. Frompersonal misinterpreted as intermittent claudication or myocar- experience, slight pain during the day is rathercommon dial ischemia, respectively, which may turn out to be even ifthe injections aregiven with care.

ScandJ UrolNephrol 36 412 P.O.Hedlund etal.

Aquality-of-life analysis using the European Or- icaleffects, of aparenteral estrogen regimen. Prostate ganisation forResearch and Treatmentof Cancer 1999; 40: 76–82. Quality-of-LifeQuestionnaire C-30, with the addition 6. Henriksson P,Eriksson A,Stege R,Collste L,Pousette A,von Schoultz B,etal. Cardiovascular follow-up of of special questions forprostate cancer, will be patients with prostatic cancertreated with single-drug published later.It is of interest to see whether the polyestradiol phosphate. Prostate 1988; 13: 257–61. two formsof castration studied, one involving the 7. Hedlund PO, Henriksson P.Parenteralestrogen versus addition of femalesteroid hormones and the other total androgen ablation in the treatmentof advanced involving emptying the body of malesteroids, will prostate carcinoma:effects on overall survival and have differenteffects on malepsychology and quality cardiovascular mortality. The Scandinavian Prostatic CancerGroup (SPCG)-5 Trial Study. Urology 2000; 55: of life. 328–33. In summary, the PEPregimen used in this study had 8. Esposti P.Cytologic malignancy grading of prostatic an equal anticancer efŽcacy to that of CAD. Further- carcinomaby transrectalaspiration biopsy. Scand JUrol more,PEP treatmentwas not associated with any Nephrol 1971; 5: 199–209. increase in cardiovascular mortality. SigniŽcantly 9. MostoŽFK, Sesterhenn IA, Sobin LH.In: Histologic morenon-fatal ischemic heart disease and heart typing of prostate tumors. Geneva, Switzerland: edn World Health Organization, 1980: 17–21. decompensation events occurred in the PEPgroup 10. Soloway MS. The importance of prognostic factorsin which unfortunately had ahigher prevalence of advanced prostate cancer.Cancer 1990; 66: 1017–21. cardiovascular disease prior to the study despite the 11. Andersson L.Design of clinical trials on prostate cancer. stratiŽed randomization. Until amorecomprehensive Urology 1997; 49 (Suppl 4A): 39–45. analysis of the cardiovascular data obtained in this 12. Miller AB,Hoogstraten B,Staquet M,Winkler A. study isdone weare not able to identify any risk factors Reporting results of cancertreatment. Cancer1981; 47: 207–14. forcardiovascular complications attributable to this 13. Fro¨din T,A c lund G,Varenhorst E.Measurementof skin PEPregimen. blood-ow and waterevaporation asa meansof objectively assessinghot ushes afterorchidectomy in patients with prostatic cancer.Prostate 1985; 7: 203–8. ACKNOWLEDGEMENTS 14. Schuirmann DJ.Acomparison of the two one-sided tests procedure and the power approach for assessingthe We acknowledge statisticians Robert Johansson and O ¨ rjan Nordle, equivalence of averagebioavailability. JPharmacokinet monitors Marie Botilsrud, Maria Concha, Hildeburg Eriksson, Biopharm 1987; 15: 657–80. Charlotte de Molade, Thea Nieminen, Kerstin Paulsson, Paula 15. Mikkola AK,Ruutu ML, Aro JL,Rannikko SA, Salo JO. Saari, Britt Tryde and Lena Wiman, data technician Gunilla Andersson and secretaries Anneli Lahtinen and Hilkka Lindberg. Parenteralpolyestradiol phosphate vs orchidectomy in This study was Žnancially supported by Ferring AB,Malmo¨, the treatmentof advanced prostatic cancer.EfŽ cacy and Sweden, Ferring Laegemidler A/S,Copenhagen, Denmark, Phar- cardiovascular complications: a2-year follow-up report macia AB,Sweden and Schering-Plough AB,Stockholm, Sweden. of anational, prospective prostatic cancerstudy. Finnprostate Group. Br JUrol 1998; 82: 63–8. 16. Henriksson P,Blomback M,Bratt G,Edhag O,Eriksson A,Vesterquist O.Effectsof oestrogen therapy and REFERENCES orchidectomy on coagulation and prostanoid synthesis in patients with prostatic cancer.Med Oncol Tumor 1. Henriksson P,Blomba¨ck M,Eriksson A,Stege R, Pharmacother 1989; 3: 219–25. Carlstro¨mK.Effectof parenteral oestrogen on the 17. Blomback M,Hedlund PO, SaweU. Changes in blood coagulation systemin patients with prostatic carcinoma. coagulation and Žbrinolysis in patients on different Br JUrol 1990; 65: 282–5. treatmentregimens for prostatic cancer.Predictors for 2. Aro JL,Haapiainen RK,Rannikko SA, Alfthan OS. High cardiovascular complications? Thromb Res1988; 49: dose polyestradiol phosphate with and without aceto- salicylic acid versus orchiectomy in the treatmentof 111–21. prostatic cancer.Finnprostate Group. Br JUrol 1989; 63: 18. Hedlund PO, Gustafson H,Sjo¨gren S.Cardiovascular 512–4. complications to treatmentof prostate cancerwith 3. Lukkarinen O,Kontturi M,and the Finnish Zoladex estramustinephosphate (Estracyt 1)or conventional Multicentre Study Group. Comparison of along-acting estrogen. Scand JUrol Nephrol Suppl 1980; 55: 103. LHRHagonist and polyestradiol phosphate in the 19. Eisen M,Napp HE,Vock R.Inhibition of platelet treatmentof advanced prostatic carcinoma.Scand J aggregation causedby estrogen treatmentin patients Urol Nephrol 1994; 28: 171–8. with carcinomaof the prostate. JUrol 1975; 114: 93–7. 4. Stege R,Carlstrom K,Collste L,Eriksson A,Henriksson 20. The Leuprolide Study Group. Leuprolide versus diethyl- P,Pousette A,etal. Single-drug parenteral estrogen stilbestrol for metastaticprostate cancer.N Engl JMed treatmentin prostatic cancer:a study of two main- 1984; 311: 1281. tenance-dose regimens. Prostate 1989; 14: 183–8. 21. Cox RL,Crawford D.Estrogens in the treatmentof 5. Henriksson P,Carlstrom K,Pousette A,Gunnarsson PO, prostate cancer.J Urol 1995; 154: 1991–8. Johansson CJ,Eriksson B,etal. Timefor revival of 22. Spetz AC, HammarM, Lindberg B,Spacngberg S, estrogens in the treatmentof advanced prostatic Varenhorst E.Prospective evaluation of hot ushes carcinoma?Pharmacokinetics, and endocrine and clin- during treatmentwith parenteral estrogen or complete

ScandJ UrolNephrol 36 Estradurin versus combined androgen deprivation 413

androgen ablation for metastaticcarcinoma of the Ha¨ggarth, Per Olov Hedlund; Kristinehamn Hospital: Bruno prostate. JUrol 2001; 166: 517–20. Larsson; Kunga¨lv Hospital: Bengt Lindberg, Sture Carlsson; Linko¨ping Hospital: Inge Ho¨jgaard, Anders Spacngberg; Lund Hospital: PeterElfving, Rolf Lundgren; Mora Hospital: Bengt Hahne; Norrta¨ljeHospital: Rickard Idestro¨m;Sahl- APPENDIX grenska Hospital, Go¨teborg: HansHedelin, Jan-Erik Dam- ber; SalaHospital: Bengt Andersson; Sandviken Hospital: PARTICIPANTS INTHESPCG-5 TRIAL Torsten Sandin; Sa¨fe Hospital: Mauritz Wallden; Trelle- Denmark borg Hospital: Ebbe Telhammar,Ronnie Olsson; Umeac Aalborg Hospital: Torben Krarup; Bispebjerg Hospital: Hospital: Torvald Granfors, Per Stattin; Uppsala Hospital: ValdemarHvidt, PeterMogensen; Frederiksberg Hospital: Einar Brekkan, Bo Johan Norlen; Varberg Hospital: Jan Michael Luke; Frederikshavn Hospital: OleSo ¨rensen; Hammarsten;Va ¨rnamo Hospital: Ib Mikkelsen, Anders Gentofte Hospital: CaiFrimodt-Mo ¨ller, Henrik Willumsen; Ramsing; Va¨stervik Hospital: Inger Wall; Va¨xjo¨Hospital: LarsAdell, Agneta Ullman; Ystad Hospital: Curt-Eric Glostrup Hospital: PeterKlarskov, Svend Mortensen; Herlev ¨ ¨ Hospital: Finn Rasmussen,Jesper Rye Andersen; Holbaek Nelson; Angelholm Hospital: Susanne Ljungeryd; O rebro Hospital: Palle Rosenkilde; Hvidovre Hospital: HansGeorg Hospital: Sven-Olof Andersson, Jan Erik Johansson. Jensen; NaestvedHospital: Erik Larsen; Nykoebing Hospi- tal: Paul Skaarup; Odense Hospital: John Farber, Hans Melchior-Nissen; Randers Hospital: So¨ren Mommsen; Rigs- hospitalet: PeterIversen, Jo¨rgen Kvist Kristensen; Skejby MEMBERS OFTHE SCIENTIFIC COMMITTEE Hospital: HansWolf, Finn Lundbeck; Viborg Hospital: Arne Ho¨jsgaard. Kjell Carlstro¨m,Department of Endocrinology, Huddinge University Hospital, Huddinge Sweden; Jan Erik Damber, Finland Departmentof Urology, Sahlgrenska University Hospital, Diakonissa Hospital: Antero Halme;Helsinki University Gothenburg, Sweden; Lena Damber,Oncology Centre, Hospital: Jaakko Salo, Mirja Ruutu; Hyvinka Hospital: Eero Umeac University, Umeac,Sweden; Per Olov Hedlund, Kasinen; Jorvi Hospital: HarriJuusela; Kuopio Hospital: Departmentof Urology, Karolinska University Hospital, Martti Ala-Opas;Lohja Hospital: Risto Salminen; Mikkeli Stockholm, Sweden; PeterHenriksson, Departmentof Hospital: Tapani Liukkonen; Joensu Hospital: Jouko Viita- Internal Medicine, Danderyds University Hospital, Stock- nen; Pa¨ijat Ha¨meenHospital: Martti Talja; Lappenranta holm, Sweden; Per-AageHo ¨isaeter,Department of Urology, Hospital: Jaakko Permi, Veli-Matti Puolakka; Turku Hospi- Bergen University Hospital, Bergen, Norway (Chairman); tal: Martti Nurmi; VaasaHospital: Erkki Hansson. PeterIversen, Departmentof Urology, Rigshospitalet, Copenhagen University, Copenhagen, Denmark(Secretary); ¨ Iceland Jan Erik Johansson, Departmentof Urology, O rebro Uni- versity Hospital, O ¨ rebro, Sweden; Teuvo Tammela,Depart- Borgarspitalinn Hospital: Gudmundur Geirsson. ment of Urology, TampereUniversity Hospital, Tampere, Finland; OleSteen Nielsen, Departmentof Oncology, A c rhus Norway Community Hospital, A c rhus, Denmark; Bo Johan Norlen, Bergen Hospital: Per-AageHo ¨isaeter,Svein Haukaas; Departmentof Urology, Uppsala University Hospital, Levanger Hospital: Fredrik Hesselberg, KarstenVada; Uppsala, Sweden; A c ke Pousette, Laboratory for Reproduc- Stavanger Hospital: Sigmund Vaage,Ilker Tasdemir;Trond- tive Health, Karolinska Hospital, Stockholm, Sweden; Sakari heim Hospital: Per Lundmo; UllevaclHospital: Per O ¨ greid, Rannikko, Departmentof Urology, Helsinki University Carl-Johan Ribbegren. Hospital, Helsinki, Finland; Finn Rasmussen,Herlev Uni- versity Hospital, Copenhagen, Denmark; Reinhard Stege, Sweden Departmentof Urology, Huddinge University Hospital, Boden Hospital: Andrzej Owczarski;Eskilstuna Hospital: Huddinge, Sweden (Co-chairman); Eberhard Varenhorst, Torsten Lindeborg, Leif Borck; Hudiksvall Hospital: Stig Departmentof Urology, Linko¨ping University Hospital, Susskind; KarlshamnHospital: Bo Hedberg; KarlstadHos- Linko¨ping, Sweden; HansWolf, Departmentof Urology, pital: Bjo¨rn Kihl Jan-Olof Olsson; Karolinska Hospital: Lars Skejby University Hospital, A c rhus, Denmark.

ScandJ UrolNephrol 36