ADVANCED PROSTATE CANCER

PRESENTATION TO ODAC Towards a Consensus in Measuring Outcomes in New Agents for Prostate Cancer Derek Raghavan MD PhD Cleveland Clinic Taussig Cancer Center Cleveland, OH. Important Specific Issues for Prostate Cancer   Potentially long natural history – 10+ years Elderly patients   intercurrent disease deaths from competing risks Advanced “conventional” disease – clinical metastases Hormone treated – relapsed, resistant, refractory New imaging techniques used more actively  earlier stage PSA-only disease after treatment Changes in imaging PSA and other tumor markers Quality of life measurement – new indices  Variable clinical manifestations – the “states” model      Stage migration     Changing surrogate measures of outcome CLINICAL STATES: A FRAMEWORK Points of Intervention Initial Prostate Evaluation: No Cancer Diagnosis Clinically Localized Disease Rising PSA Clinical Metastases: Noncastrate Clinical Metastases: Castrate 1. Assess and reassess for the presence of disease, or probability of a clinically significant event in a given time frame. 2. Treat to eliminate or, depending on probability, to prevent the occurrence of the event(s). 3. Defer treatment if probability is low. Important Specific Issues for Prostate Cancer   Potentially long natural history – 10+ years Elderly patients   intercurrent disease deaths from competing risks   Variable clinical manifestations – the “states” model    Advanced “conventional” disease – clinical metastases Hormone treated – relapsed, resistant, refractory New imaging techniques used more actively PSA-only disease after treatment  Stage migration    Changes in imaging PSA and other tumor markers Quality of life measurement – new indices  Changing surrogate measures of outcome Brief Historical Perspective: 1  1900-1960’s  Animal models    Survival Acid phosphatase Huggins & Hodges  Nobel Prize: CAP & Castration  Human studies     Usually characterized by VERY advanced disease Some imprecise endpoints – tried to assess response Acid phosphatase Survival Brief Historical Perspective: 2  1970’s-1990’s:  Refinement of assessment: – “response” and “stable disease”  Earlier and more precise diagnosis  Evolving measures of quality of life  Introduction of concept of “PSA RESPONSE”  NPCP Development of PSA (TM Chu et al)  Stage migration  Education of the Public regarding treatment  Brief History of FDA Approval of Agents for Prostate Cancer AGENT Docetaxel  Zolidronic Acid  Mitoxantrone   YEAR 2004 2003 1996 ENDPT. survival QOL QOL Estramustine 1981 old rules Current Status   Impact of Community Pressure & Patient Advocacy Stage Migration    New imaging – PET scans, tomographic scans Super-sensitive assays of PSA, etc. Refinement of “PSA response” Refined measurement of quality of life Absence of progression – the “static” agents PSA and other antigens – time dependent fluxes  New Endpoints presented:      “Response” – PSA, symptoms, objective Is survival still THE standard? KEY QUESTION FOR APPROVAL: Does New Therapy ALTER True Outcomes?  Is survival the “ultimate” test?   Confounded by death from other causes Confounded by a series of “salvage” therapies    What about quality of life & toxicity of treatment? Do surrogates reflect real changes in outcome? Do measures of outcome change with the clinical states of the disease? Goals of Treatment vs. States Clinical State No Cancer Aim Prevent Cancer Outcome No cancer Localized Disease Delay recurrence Maximize cure Minimize toxicity Prevent clinical metastases PSA level - after surgery - after radiotherapy Absolute PSA level? Altered kinetics? Rising PSA OBJECTIVES BY CLINICAL STATE INITIAL LOCALIZED EVALUATION: DISEASE NO CANCER DIAGNOSIS RISING PSA CLINICAL CLINICAL METASTASES: METASTASES: NON-CASTRATE CASTRATE PREVENTION MINIMIZE PREVENT MORBIDITY/ METASTASES MAXIMIZE CURE ELIMINATE / PREVENT SYMPTOMS DEATH OF DISEASE Presentation of Advanced Prostate Cancer: Syndromes Bone pain  Constitutional symptoms – the great mimic  Urinary obstruction   Slow stream, nocturia, frequency, hematuria  Acute/chronic renal failure Bone marrow failure  Unusual sites – liver, lungs, nodes, skin  PSA only disease  “Imaging only” disease  Hormone Refractory Prostate Cancer: Median Time from Progression to Death PARAMETER PSA increase Bone scan change Alkaline phos increase Pain increase Performance status decline Hemoglobin decline Weight loss Liver metastases Newling, et al. Cancer. 1993;72:3793-3798. Weeks 52 41 35 32 24 22 12 10 Changing Endpoints in Prostate Cancer Therapy  Impact of stage migration  PSA only disease    Rising after radiotherapy or surgery Asymptomatic disease with known metastases New PSA targets – e.g. rate of rise or fall  Earlier intervention for +ve bone scans and refinement in measurement of changes in scans    Measurement of quality of life Measurement of time to progression Adjuvant trials Overall Survival 100% 80% 60% 40% 20% 0% 0 12 Months 24 36 (SWOG 9916) D+E M+P # at Risk 338 336 # of Median Deaths in Months 217 18 235 16 HR: 0.80 (95% CI 0.67, 0.97), p = 0.01 48 PSA Response Rate 50% % of patients with a > 50% decrease in PSA p < 0.0001 40% 30% 50% 27% 20% 10% 0% Docetaxel/estramustine n = 303 Mitoxantrone/prednisone n = 303 Criterion 1b: Survival by Surrogate (50% Decrease in PSA during first 3 months) 100% 80% 60% 40% 20% 0% no 50% dec 50% dec Median At Risk Deaths in Months 291 214 14 238 130 21 P < .0001 0 12 24 Months After Registration 36 48 Criterion 1c: Survival by Treatment and Surrogate 100% 80% 60% 40% 20% 0% Median At Risk Deaths in Months D + E, no 50% dec 99 71 15 D + E, 50% dec 162 91 21 M + P, no 50% dec 192 143 14 M + P, 50% dec 76 39 21 P < .0001 0 12 24 Months After Registration 36 48 Changing Endpoints in Prostate Cancer Therapy  Impact of stage migration  PSA only disease    Rising after radiotherapy or surgery Asymptomatic disease with known metastases New PSA targets – e.g. rate of rise or fall  Earlier intervention for +ve bone scans and refinement in measurement of changes in scans    Measurement of quality of life Measurement of time to progression Adjuvant trials Measures of Quality of Life    Difficulty of assessing response within “stable” disease category Use of patient reported symptom response widens the goalposts Contrast:    symptoms of age symptoms of cancer side effects of therapy   Dichotomy between objective vs. subjective vs. PSA response Optimal technology not defined Patient Reporting Domains  Non-specific     Tumor-related     Fevers, sweats Pruritis Malaise      Well-being Mood Activities Quality of life Sexuality Pain Weight Performance status Metastases Hormone effects RT effects Surgical effects Chemotherapy impact  Treatment-related     Examples of Patient Report Instruments o McGill Melzack – Present Pain Intensity o o o What is optimal measure of reduction? Significance of 2 point reduction? Impact of baseline severity? o o o EORTC QLQ 30 EORTC Prostate Cancer Specific Module PROSQOLI Methodological Problems Impact of baseline variables – e.g. pain  How to score  Dealing with missing data  Statistical analysis  under curve  Kaplan Meier vs. Wilcoxson  ROC curves   Area Confounding variables Patient knowledge of PSA fluxes – impact?  e.g. Race and socio-demographic factors  e.g. Chemotherapy for Prostate Cancer Impact of Patient-Reported Outcomes (usually added to hormone effects) Mitoxantrone Phase III Canadian Trial: Study Design Mitoxantrone R A N D O M I Z E + Prednisone Primary Endpoint: Palliation Prednisone* N=81 N=80 Symptomatic HRPC *Crossover on progression (N=50) Tannock, et al. J Clin Oncol. 1996;14:1756-1764. Mitoxantrone for Advanced Prostate Cancer: Overall Survival Tannock et al, J. Clin. Oncol., 1996 Mitoxantrone for Advanced Prostate Cancer: Quality of LIfe Tannock et al, J. Clin. Oncol., 1996 PSA, Palliative Response & Survival (Dowling et al, Ann. Oncol., 2001, 12: 773-8)          Retrospective analysis of MP vs P trial Cox proportional hazards model & landmark analysis at 9 weeks Absence of PSA data = “non-responders” 34% of M+P  PSA response 11% of P  PSA response 53% with PSA response  palliative response 29% without PSA response  palliative response Multivariate factors: PS, high Hb, PSA response Palliative response did NOT predict survival Summary of Estramustine-Based Chemotherapy: Nonrandomized Trials Study Amato Seidman Hudes Hudes Petrylak Phase II II II I+II I Agents Combined with EMP vinblastine vinblastine vinblastine paclitaxel docetaxel N 22 25 36 38 32 Number PR/ Number Measured 3/7 2/5 1/7 6/12 5/16 N (%)  50% PSA  11/22 (50) 13/24 (54) 11/36 (31) 19/35 (54) 20/32 (63) Imp. Pain N (%) — 6/9 (66) 12/28 (43) — 8/15 (53) Median Survival (Months) — — 11.5 17 — Amato, et al. Proc Am Assoc Cancer Res. 1999. Seidman, et al. J Urol. 1992;147:931-934. Hudes, et al. J Clin Oncol. 1992;10:1754-1761. Hudes, et al. J Clin Oncol. 1997;15:3156-3163. Petrylak, et al. Proc Am Soc Clin Oncol. 1997. (Courtesy of Maha Hussain, MD) TAX 327: Secondary Objectives Response Rates Docetaxel 3 wkly Pain Response Rate* n, evaluable Response rate (%) P-value (vs. mitoxantrone) PSA Response Rate* n, evaluable PSA response rate (%) P-value (vs. mitoxantrone) Tumor Response Rate* n, evaluable Response rate (%) P-value (vs. mitoxantrone) 153 35 0.01 291 45 0.0005 141 12 0.1 Docetaxel wkly 154 31 0.07 282 48 <0.0001 134 8 0.5 Mitoxantrone 157 22 300 32 137 7 - * Determined only for patients with pain or PSA 20 or measurable disease at baseline, respectively Taxotere-Calcitriol (Beer et al, J Clin Oncol, 2004, 100: 758-763)  Indices: point reduction on PPI (or 0 if PPI=1)  50% reduction in analgesic use  Other measures of QOL 2 Analgesic response in 48%  BUT worse QOL on QLQ-C30QOL      Physical and role functioning Fatigue Appetite Global health status Placebo Effects in Oncology (Chvetzoff & Tannock, JNCI, 2003, 95: 19-29)        Reviewed 37 placebo-controlled trials & 10 with best supportive care Studies not all double blinded Improved pain with placebo in 2/6 trials  (0%  21% of individual patients) (8%  27% of individual patients) Improved appetite in 1/7 trials  No improvement in weight with placebo No improvement in QOL in 10 trials (assessed by pt’s) No improvement in ECOG performance status in 9 trials (via MD’s) Patient Reports of Symptoms      Assessment of symptomatic response leads to stage/response migration – c.f. “objective” response Measures of Quality of Life and Symptom Response still being developed and validated PSA response vs. Symptom Response vs. Toxicity of Treatment  “disconnect” Discordant QOL results – which should “dominate”? Confounding symptoms:   Toxicity of treatment Age-related disorders – BPH, arthritis, anemia  This area should be regarded as “work in progress” by FDA SUMMARY   Survival has been the standard Surrogates under evaluation     Quality of Life and Patient Reporting PSA response PSA time dependent kinetics Markers of bone turnover Need for new parameters? Are they really useful?  New agents that cause cytostatic effects    Definition of a “new era” – insufficient data in 2005