General Information
Conflict of Interest
Request Clinical Trials
Associated Trial(s):- NCT00638690 - A Phase 3, Randomized, Double-Blind, Placebo-Controlled Study of Abiraterone Acetate (CB7630) Plus Prednisone in Patients With Metastatic Castration-Resistant Prostate Cancer Who Have Failed Docetaxel-Based Chemotherapy
- NCT01715285 - A Randomized, Double-blind, Comparative Study of Abiraterone Acetate Plus Low-Dose Prednisone Plus Androgen Deprivation Therapy (ADT) Versus ADT Alone in Newly Diagnosed Subjects With High-Risk, Metastatic Hormone-naive Prostate Cancer (mHNPC)
- NCT00887198 - A Phase 3, Randomized, Double-blind, Placebo-Controlled Study of Abiraterone Acetate (CB7630) Plus Prednisone in Asymptomatic or Mildly Symptomatic Patients With Metastatic Castration-Resistant Prostate Cancer
Request Clinical Trials
Data Request Status
Status: OngoingResearch Proposal
Project Title: Comparative Survival Analysis of Prior Prostatectomy vs Radiotherapy in Metastatic Prostate Cancer: Pooled Data from LATITUDE, COU-AA-301 and 302
Scientific Abstract:
Background: The comparative impact of prior definitive local therapy--specifically Radical Prostatectomy (RP) versus Radiotherapy (RT)--on the efficacy of subsequent Abiraterone Acetate (AA) in metastatic prostate cancer remains undefined. While RP offers maximal cytoreduction, RT may induce immunogenic effects. It is unknown if these distinct modalities differentially influence long-term survival, or if potential benefits diminish as the disease progresses from hormone-sensitive (mHSPC) to castration-resistant (mCRPC) states.
Objective: To conduct a head-to-head comparison of survival outcomes between patients with Prior-RP versus Prior-RT across the prostate cancer continuum.
Study Design: A retrospective pooled analysis of Individual Participant Data (IPD) from three landmark randomized Phase 3 trials: LATITUDE, COU-AA-302, and COU-AA-301.
Participants: The study population includes patients with high-risk mHSPC (LATITUDE cohort), chemotherapy-naïve mCRPC (COU-AA-302 cohort), and post-docetaxel mCRPC (COU-AA-301 cohort), providing a comprehensive view of the disease trajectory.
Primary and Secondary Outcome Measure: The Primary Outcome is overall survival (OS). The Secondary Outcome is radiographic progression-free survival (rPFS).
Statistical Analysis: Patients will be categorized by prior local therapy history (RP, RT, or None). To mitigate selection bias, Propensity Score Matching (PSM) will be employed to balance baseline covariates (including Age, ECOG status, Gleason score, and Visceral Metastases) between the RP and RT groups. Multivariable Cox proportional
Brief Project Background and Statement of Project Significance:
Background: Managing the primary tumor in metastatic prostate cancer remains controversial. Although systemic therapies like Abiraterone Acetate (AA) are standard for mHSPC and mCRPC, the long-term prognostic impact of the type of prior local therapy--Radical Prostatectomy (RP) versus Radiotherapy (RT)--is not well understood.
Biologically, these modalities differ significantly. RP offers maximal cytoreduction and removes the source of re-seeding metastases, whereas RT leaves the primary tumor in situ but may induce immunogenic effects. Most existing studies are retrospective and flawed by selection bias, often grouping all local therapies together. Consequently, we lack data on whether surgery or radiation offers better long-term priming for systemic therapy, and whether this benefit sustains as the disease progresses.
Statement of Project Significance: This study aims to conduct a direct comparison of Prior-RP versus Prior-RT using high-quality Individual Participant Data (IPD) from the LATITUDE, COU-AA-302, and COU-AA-301 trials. By analyzing data across the disease continuum, this project addresses the limitations of prior small-scale studies.
Our findings will have immediate clinical relevance:
Prognostic Modeling: Determining if prior modality is an independent predictor of AA efficacy will improve risk stratification.
Treatment Sequencing: Clarifying the distinct roles of RP and RT in the metastatic setting addresses a critical gap in urologic oncology.
Mechanistic Insight: The results will offer indirect evidence regarding the biological importance of cytoreduction versus immunomodulation in advanced disease control.
References:
Fizazi K, et al. N Engl J Med. 2017;377(4):352-360.
Ryan CJ, et al. N Engl J Med. 2013;368(2):138-148.
de Bono JS, et al. N Engl J Med. 2011;364(21):1995-2005.
Specific Aims of the Project:
Hypothesis: We hypothesize that the modality of prior definitive local therapy differentially impacts long-term survival in metastatic prostate cancer patients treated with Abiraterone Acetate. Specifically, Prior-Radical Prostatectomy (RP) may confer a superior survival benefit compared to Prior-Radiotherapy (RT) due to maximal cytoreduction. We further hypothesize that this protective effect is stage-dependent, being most pronounced in the hormone-sensitive setting and attenuating in the late castration-resistant setting.
Specific Aims:
Primary Aim: To conduct a head-to-head comparison of Overall Survival (OS) and Radiographic Progression-Free Survival (rPFS) between patients with a history of Prior-RP versus Prior-RT. This analysis will focus on the mCRPC cohorts (COU-AA-301 and COU-AA-302), utilizing Propensity Score Matching (PSM) to balance baseline confounders such as age and visceral metastases.
Secondary Aim: To evaluate the prognostic value of Prior Local Therapy (Any) versus No Prior Local Therapy (De Novo metastatic presentation) within the high-risk mHSPC cohort (LATITUDE trial).
Exploratory Aim: To perform a pooled interaction analysis across all three datasets (mHSPC through late mCRPC) to determine if disease stage significantly modifies the association between prior local therapy modality and treatment efficacy.
Study Design: Meta-analysis (analysis of multiple trials together)
What is the purpose of the analysis being proposed? Please select all that apply.: New research question to examine treatment effectiveness on secondary endpoints and/or within subgroup populations Participant-level data meta-analysis Meta-analysis using only data from the YODA Project
Software Used: Python, R, RStudio, Open Office
Data Source and Inclusion/Exclusion Criteria to be used to define the patient sample for your study:
The study sample will be derived from pooled Individual Participant Data (IPD) from three Phase 3 randomized clinical trials requested via the YODA Project:
NCT01715285 (LATITUDE): High-risk mHSPC cohort.
NCT00887198 (COU-AA-302): Chemotherapy-naïve mCRPC cohort.
NCT00638690 (COU-AA-301): Post-docetaxel mCRPC cohort.
Inclusion Criteria:
Population: All randomized subjects (Intent-to-Treat [ITT] population) from the aforementioned trials.
Data Availability: Subjects must have accessible medical history or case report form (CRF) data sufficient to categorize their prior local therapy status (i.e., history of Radical Prostatectomy, Radiotherapy, or No Prior Local Therapy).
Exclusion Criteria:
Subjects with completely missing or unclassifiable data regarding prior local therapy modality.
Subjects with missing values for key baseline covariates required for Propensity Score Matching (e.g., Age, Visceral Metastases) that cannot be handled via multiple imputation.
Analysis Platform: We plan to pool the IPD from these three trials to create a combined dataset. All data harmonization and statistical analyses (including Cox regression and Propensity Score Matching) will be conducted strictly within the YODA Project Secure Data Environment (SDE) using the provided R software. We will not attempt to download raw IPD to a local machine.
Primary and Secondary Outcome Measure(s) and how they will be categorized/defined for your study:
Primary Outcome Measure: Overall Survival (OS): Defined as the time interval from the date of randomization to the date of death from any cause. For patients who are still alive at the time of analysis, data will be censored at the last date the patient was known to be alive.
Secondary Outcome Measure: Radiographic Progression-Free Survival (rPFS): Defined as the time from randomization to the first objective evidence of radiographic disease progression (assessed by CT/MRI for soft tissue or bone scan for bone lesions according to PCWG2/RECIST 1.1 criteria) or death from any cause, whichever occurs first.
Categorization and Definition for the Proposed Study: The primary independent variable of interest is the Type of Prior Local Therapy, which will be extracted from the "Medical History" or "Prior Procedures" datasets of each trial. This variable will be categorized into three mutually exclusive groups:
Prior Radical Prostatectomy (RP): Patients who underwent definitive surgery for the primary tumor.
Prior Radiotherapy (RT): Patients who received definitive External Beam Radiotherapy (EBRT) or Brachytherapy.
No Prior Local Therapy: Patients with no history of definitive local treatment (including those with de novo metastatic disease or those managed with active surveillance/watchful waiting).
Analysis Plan: The primary comparison of interest is the head-to-head comparison of Prior-RP vs. Prior-RT regarding OS and rPFS. Hazard Ratios (HR) will be estimated for this specific contrast, adjusting for baseline covariates.
Main Predictor/Independent Variable and how it will be categorized/defined for your study:
Variable Name: Type of Prior Local Therapy
Definition & Source: This variable represents the modality of definitive treatment received for the primary tumor prior to study enrollment. It will be derived from the "Medical History," "Prior Procedures," or "Surgeries" domains of the IPD.
Categorization: Patients will be classified into three mutually exclusive categories:
Prior Radical Prostatectomy (Prior-RP): Patients documented to have undergone radical prostatectomy.
Prior Radiotherapy (Prior-RT): Patients documented to have received definitive external beam radiotherapy (EBRT) or brachytherapy to the prostate.
No Prior Local Therapy: Patients with no documented history of definitive local treatment (representing de novo metastatic disease or initial management with active surveillance/watchful waiting).
Primary Comparison: The analysis will focus on the head-to-head comparison of Prior-RP versus Prior-RT to evaluate differential survival outcomes.
Other Variables of Interest that will be used in your analysis and how they will be categorized/defined for your study:
Demographics:Age at Randomization: Continuous (years) and Categorical (<75 vs. >=75 years) to address geriatric heterogeneity.
Race/Ethnicity: As reported in the CRF.
Clinical Characteristics:
Visceral Metastasis Status: Presence of metastases in visceral organs (Liver and/or Lung) at baseline (Yes vs. No).
ECOG Performance Status: Baseline functional status (Categorized as 0-1 vs. >=2).
Gleason Score: Histological grade at initial diagnosis (Categorized as <7 vs. >=8).
Time from Diagnosis: Time interval from initial histological diagnosis to randomization (Continuous) to distinguish between rapid progressors and those with indolent disease history.
Biomarkers:
Baseline PSA: Serum Prostate-Specific Antigen level (Continuous, log-transformed).
Laboratory Values: Baseline Hemoglobin, Lactate Dehydrogenase (LDH), and Alkaline Phosphatase (ALP) levels as markers of tumor burden.
Statistical Analysis Plan:
Software and Computing Environment: All data management and statistical analyses will be performed using R (version 4.0.0 or later, accessed via RStudio) and Python (version 3.8 or later) within the secure YODA Project Data Analysis Environment.
1. Data Harmonization and Preparation: Individual Participant Data (IPD) from the three trials (LATITUDE, COU-AA-302, COU-AA-301) will be harmonized into a single master dataset. Python (pandas) and R (tidyverse) will be utilized to map variables to a common data model, ensuring consistent definitions across protocols.
2. Descriptive Statistics: Baseline characteristics will be summarized. Continuous variables will be reported as medians (IQR) and categorical variables as frequencies (%).
3. Addressing Selection Bias (Dual-Method Verification Strategy): To ensure robust causal inference and minimize selection bias, we will employ two complementary propensity score methods to mutually verify our findings:
Method A: Propensity Score Matching (PSM):
Propensity Score Estimation: Scores will be estimated using logistic regression with key covariates (Age, ECOG, Visceral Metastases, Gleason, PSA).
Matching: We will perform 1:1 nearest-neighbor matching (caliper = 0.2 SD) using R (MatchIt). This method creates intuitive, balanced cohorts but may sacrifice sample size.
Balance Check: Standardized Mean Differences (SMD) < 0.1.
Method B: Inverse Probability of Treatment Weighting (IPTW):
Weighting: We will calculate stabilized inverse probability weights based on the same propensity score model. This allows us to utilize the full study population without discarding unmatched patients.
Balance Check: Weighted SMD will be assessed to ensure the pseudo-population is balanced.
4. Outcome Analysis (Survival Analysis): The association between Prior Local Therapy and survival (OS, rPFS) will be evaluated in parallel:
Primary Analysis (PSM Cohort): Stratified Cox proportional hazards models applied to the matched population.
Confirmatory Analysis (IPTW Cohort): Weighted Cox proportional hazards models applied to the full weighted population.
Consistency: Results will be considered robust only if the direction and magnitude of the Hazard Ratios (HR) are consistent between the PSM and IPTW analyses.
5. Heterogeneity and Interaction Testing: We will test for stage-dependent efficacy by including an interaction term between Prior_Therapy_Type and Cohort_Stage in the Cox models.
6. Sensitivity Analyses: Multivariable adjustment on the crude (unmatched/unweighted) population will be performed as a final consistency check.
Narrative Summary:
Patients with advanced prostate cancer are often treated with Abiraterone. Many previously underwent either surgery (removing the prostate) or radiation (zapping the tumor) years earlier. However, doctors do not know if one of these past treatments leads to better long-term survival than the other when the cancer eventually spreads.
This study will combine data from three major clinical trials (LATITUDE, COU-AA-301, COU-AA-302) involving over 3,000 patients across different disease stages. We will use statistical matching techniques to fairly compare the survival outcomes of patients who had prior surgery versus those who had prior radiation. The findings will help determine if aggressive removal of the primary tumor offers a lasting "protective" benefit, helping doctors personalize treatment plans.
Project Timeline:
Estimated Project Start Date: Upon approval of data access request (Month 0).
Phase 1: Data Setup & Harmonization (Months 1-3)
Access YODA secure environment.
Clean and merge IPD from LATITUDE, COU-AA-301, and COU-AA-302.
Map variables to common definitions using Python/pandas and R/tidyverse.
Define cohorts based on prior local therapy history.
Phase 2: Statistical Analysis (Months 4-6)
Perform descriptive statistics and missing data handling.
Execute Propensity Score Matching (PSM) and Inverse Probability of Treatment Weighting (IPTW) in R.
Conduct primary survival analyses (Cox regression) and interaction testing (Stage x Therapy).
Perform sensitivity analyses and validate consistency between PSM and IPTW results.
Phase 3: Manuscript Drafting (Months 7-9)
Interpret findings and generate high-quality figures/tables.
Draft the full manuscript focusing on the "Surgery vs. Radiotherapy" comparison.
Internal review and revision.
Phase 4: Submission & Reporting (Months 10-12)
Month 10: Submit manuscript to a high-impact peer-reviewed journal.
Month 11: Submit abstract to a major international oncology congress (e.g., ASCO, ESMO, or EAU).
Month 12: Report final results/publication status back to the YODA Project and close the project or request extension if revision is ongoing.
Dissemination Plan:
Anticipated Products: The primary deliverable of this project will be a full-length original research manuscript titled (provisional): "Comparative Impact of Prior Radical Prostatectomy versus Radiotherapy on Survival Outcomes Across the Metastatic Prostate Cancer Continuum: A Pooled Analysis of LATITUDE, COU-AA-302, and COU-AA-301."
Target Audience: The findings will be highly relevant to a multidisciplinary audience, including Urologists, Radiation Oncologists, and Medical Oncologists. The results will directly inform decision-making regarding the potential long-term benefits of primary tumor control strategies.
Target Journals: Given the large sample size (>3,000 pts), the high quality of the randomized trial data, and the clinical significance of the "Surgery vs. Radiation" debate, we intend to submit this work to top-tier journals such as:
European Urology (Impact Factor ~24)
Journal of Clinical Oncology (JCO)
JAMA Oncology
The Lancet Oncology
Conference Presentations: We also plan to disseminate preliminary findings via oral or poster presentations at major international meetings, such as the American Society of Clinical Oncology (ASCO) Annual Meeting, the European Society for Medical Oncology (ESMO) Congress, or the European Association of Urology (EAU) Congress.
Bibliography:
Fizazi K, Tran N, Fein L, et al; LATITUDE Investigators. Abiraterone plus Prednisone in Metastatic, Castration-Sensitive Prostate Cancer. N Engl J Med. 2017;377(4):352-360.
Ryan CJ, Smith MR, de Bono JS, et al; COU-AA-302 Investigators. Abiraterone in metastatic prostate cancer without previous chemotherapy. N Engl J Med. 2013;368(2):138-148.
de Bono JS, Logothetis CJ, Molina A, et al; COU-AA-301 Investigators. Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med. 2011;364(21):1995-2005.
Scher HI, Halabi S, Tannock I, et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J Clin Oncol. 2008;26(7):1148-1159.
Haukoos JS, Lewis RJ. The Propensity Score. JAMA. 2015;314(15):1637-1638.