Eur Urol Suppl 2011;10(2):253 National Hospital Discharge Registry after PCa diagnosis (i.e. events leading to hospitalization) by treatment group (ET, curative treatment, or surveillance) and history of previous event(s). Results: We studied 76000 men diagnosed between 1997 and 2007, of whom 30642 men received primary ET, 26432 curative treatment, and 19526 surveillance. Mean follow up time was 4.1 years. 75% of all men did not have any event (fractures, stroke, CVD, or TED) during follow-up, 17% had one event and 9% had more than one event. The incidence of any event was 10/100 person- years. The proportion of men with an event was largest among men on ET and men on surveillance. Men who already had had an adverse event prior to PCa diagnosis were more likely to develop another event afterwards. Of men who had a fracture (n=3959) after PCa diagnosis, 64% had one fracture, 8% had several fractures, and 28% had one fracture combined with other event(s). Of men who had a stroke (n=4879) 63% had one stroke, 9% developed several strokes, and 28% had one stroke combined with other event(s). Of men who had a CVD (n=10589) 58% had one CVD (58%), 29% had several CVDs, and 13% had one CVD combined with other event(s). Of men who had a TED (n=1589) 64% had one TED, 5% had several TEDs, and 31% had one TED combined with other event(s). When comparing demographic and tumour characteristics (e.g. age, tumour stage, treatment, prior events) between those with 1, 2, or 3+ events after PCa diagnosis, there were no clear differences between patients. Multiple events occurred after all types of primary events, but more men had a HD prior to PCa diagnosis in the group with 2 or 3+ events than those with 1 event (36, 48, and 28%, respectively). Conclusions: This is the first study describing the combined risk patterns of side- effects following PCa treatment. About 75% of prostate cancer patients did not have any event of fractures, stroke, CVD, or TED; of the remaining 25% the majority of men had one event and about one third had a second event. Apart from a history of similar adverse events prior to PCa diagnosis, no particular characteristics were found for men who experienced multiple events. 800 PHASE III STUDY OF INTERMITTENT MAB VS. CONTINUOUS MAB Calais Da Silva Junior F.M. 1 , Bono A. 2 , Whelan P. 3 , Brausi M. 4 , Queimadelos A. 5 , Portillo J. 6 , Kirkali Z. 7 , Calais Da Silva Senior F. 1 SEUG 9401, Portugal 1 Central Hospital de Lisbon, Dept. of Urology, Lisbon, Portugal, 2 Hospital di Circolo e Foundation Macchi, Dept. of Urology, Varese, Italy, 3 Sant James Hospital, Dept. of Urology, Leeds, United Kingdom, 4 San Agostino Hospital, Dept. of Urology, Modena, Italy, 5 Policlinic la Rosaleda, Dept. of Urology, Santiago Compostela, Spain, 6 Hospital Valedecilla, Dept. of Urology, Santander, Spain, 7 Docuz Eylul University, Dept. of Urology, Izmir, Turkey Introduction & Objectives: Few randomised studies have compared intermittent hormonal therapy with continuous therapy for the treatment of advanced prostate cancer. The main publication in 2009 was based upon data with a median follow up of 51 months. We present updated results with extended follow up. The objective is to investigate if intermittent therapy is associated with a shorter survival time. Materials & Methods: 766 patients with locally advanced or metastatic prostate cancer received a three month induction treatment. 626 patients whose PSA decreased below 4 ng/ml or to 80% below the initial value, were randomised. Intervention: Patients received cyproterone acetate (CPA) 200 mg for two weeks and then monthly depot injections of a LHRH analogue plus 200 mg of CPA daily during induction. Patients randomised to the intermittent arm ceased treatment while those randomised to the continuous arm received 200 mg of CPA daily sa LHRH analogue. Results: A total of 474 patients are known to have died, 90 are lost to follow up, of which 37 withdrew mainly for patient refusal and change of therapy. There was no difference in survival, p = 0.61, with hazard ratio 0.96 (95% CI 0.80 to 1.14) and 239 deaths on the intermittent and 235 on the continuous arm. A slight excess of cancer deaths in the intermittent treatment arm (136 versus 109) is balanced by a slight excess of cardiovascular deaths in the continuous arm (68 versus 62), and deaths from other causes (58 versus 41). The hazard ratio of a cancer death is 1.27 (95% CI 0.98, 1.64) , p = 0.06 in the intermittent arm compared to the continuous, For cardiovascular deaths the hazard ratios are 1.05 (95% CI 0.75, 1.49), p =0.77, continuous compared to intermittent, and for other deaths the hazard ratio for continuous compared to intermittent is 1.38 (95% CI 0.93, 2.06), p=0.11. The extended follow up has accumulated a further 135 deaths since the last analysis which used data up to 2005 and exceeds the number of events specified in the original power analysis. The extra 5 years of follow up now means that the study has accumulated almost 3000 person years at risk among the 626 randomised patients and the median follow up is now 57 months compared to 51 months in the publication. Conclusions: Intermittent therapy should be considered for use in routine practice since it is associated with no reduction in survival, no clinically meaningful impairment in quality of life, better sexual activity, and considerable economic benefit to individual and community. 801 TIME TO PROGRESSION SHOULD NOT BE USED TO INFER THE ULTIMATE BENEFIT OF IMMEDIATE VERSUS DEFERRED/SALVAGE ANDROGEN DEPRIVATION THERAPY FOR PROSTATE CANCER Studer U.E. 1 , Whelan P. 2 , Wimpissinger F. 3 , Casselman J. 4 , De Reijke Th.M. 5 , Knönagel H. 7 , Madersbacher S. 8 , Isorna S. 9 , Sundaram S.K. 10 , Collette L. 11 EORTC Genito-Urinary Tract Cancer Group, Belgium 1 Inselspital, Dept. of Urology, Berne, Switzerland, 2 Sint James Hospital, Dept. of Urology, Leeds, United Kingdom, 3 Rudolfstiftung Hospital, Dept. of Urology, Vienna, Austria, 4 Damiaan Hospital, Dept. of Urology, Oostende, Belgium, 5 Academic Medical Centre, Dept. of Urology, Amsterdam, The Netherlands, 6 University Hospital of Zürich, Dept. of Urology, Zürich, Switzerland, 7 Hospital Barmherzige Schwestern, Dept. of Urology, Linz, Austria, 8 Hospital Nuestra Senora Del Pino, Dept. of Urology, Las Palmas, Spain, 9 Mid Yorkshire NHS Trust, Pinderfields Hospital, Dept. of Urology, Wakefield, United Kingdom, 10 EORTC Headquarters, Dept. of Statistics, Brussels, Belgium Introduction & Objectives: Recent trials of androgen deprivation therapy (ADT) in prostate cancer often assess the benefit of immediate ADT on the basis of differences in time to progression. We argue that this endpoint is not appropriate because it does not capture the mechanism of androgen refractoriness and the impact of the initial treatment on the effectiveness of further salvage therapies. Materials & Methods: EORTC trial 30891 compared immediate ADT (n=492 pts) with orchiectomy or LH-RH analogue to deferred ADT (n=493 pts) initiated at the time of symptomatic disease progression or life-threatening complications in patients with T0-4 N0-2 M0 prostate cancer. Overall survival was primary endpoint, prostate cancer mortality, time to first objective progression (metastases documented on imaging or by significant rise in alkaline phosphatase or by doubling serum PSA value above 20 ng/ml within 1 year) and time to ADT refractory progression were secondary endpoints. We assess the treatment effect on these endpoints and contrast them with a view on surrogacy. Results: Overall, 362 of 985 patients had objective disease progression on study, 291 had ADT refractory progression, 769 patients died (78.1%), of which 269 (27.3%) of prostate cancer. Deferred treatment was significantly worse than immediate treatment in terms of the time to first disease progression (P<0.0001) with 10-year progression rates of 40.5% vs 27.2%, resp. Time to ADT refractory progression after immediate or deferred ADT did not differ significantly (P=0.64,fig.1). Prostate cancer mortality was also not improved with immediate ADT (fig.2). Conclusions: Time to progression should not be used to assess the eventual benefit of immediate ADT compared to an initial watchful waiting approach. A survival endpoint should be used instead. 802 NEOADJUVANT LHRH ANALOG PLUS ESTRAMUSTINE PHOSPHATE COMBINED WITH THREE-DIMENSIONAL CONFORMAL RADIOTHERAPY FOR INTERMEDIATE- TO HIGH-RISK PROSTATE CANCER: A RANDOMIZED STUDY Hirano D., Nagane Y., Okada Y., Satoh K., Yamanaka Y., Yamaguchi K., Takahashi S. Nihon University School of Medicine, Dept. of Urology, Tokyo, Japan Introduction & Objectives: Estramustine phosphate (EMP) has been widely used for treatment of advanced carcinomas, including hormone-refractory prostate