Transition of a Clinical Trial into Translational Research: The Prostate Cancer Prevention Trial Experience

Prostate cancer is the most frequently diagnosed nonskin cancer and second leading cause of cancer death in U.S. men; 192,280 new diagnoses and 27,360 deaths are estimated to occur in 2009 despite advances in diagnosis and treatment (1). Furthermore, the absolute number of new cases of prostate cancer is expected to increase substantially in the years to come because of the aging male population. Research into improving control of this disease is progressing on many fronts including cancer prevention.

Although much has been learned in recent years about the molecular pathogenesis of prostate cancer, the only established risk factors are increased age, African American race, and family history of prostate cancer in a first-degree relative (2). Many questions remain about the etiologic roles of common genetic variants, inflammation, environmental carcinogens, and diet, obesity, and other lifestyle factors. Finasteride in the Prostate Cancer Prevention Trial (3–7) and dutasteride in the Reduction by Dutasteride of Prostate Cancer Events (8) trial reduced prostate cancer risk by 23% to 25%, but this reduction left many men in both studies who still developed prostate cancer while taking study medication. The development and implementation of strategies for more effective prevention, whether through more options or through larger risk reductions, requires advances in our understanding of prostate carcinogenesis and in prostate cancer risk assessment (9–12).

The Prostate Cancer Prevention Trial (PCPT) was a phase III randomized, double-blind, placebo-controlled clinical trial of a molecular targeted agent (3). Extensive clinical data, including clinical, lifestyle, behavioral, and anthropometric data, were collected at baseline and throughout the 7-year study, blood specimens were obtained annually for prostate-specific antigen (PSA) assessment, tissue from prostate biopsies and prostatectomies, and a special white blood cell (WBC) collection for the harvesting of DNA were stored in the PCPT biorepository. The clinical and biospecimen resources collected during the PCPT offered an opportunity to investigate hypotheses related both to the biology underlying the trial's findings and to the etiology of prostate cancer. This opportunity differs from that of a well-designed phase II trial, which would be more limited in the availability of data and biological samples and would not include a cancer endpoint for subsequent risk determinations. PCPT risk modeling also differs from population-based epidemiology, which is less controlled and more prone to the influence of confounding factors. With yearly blood samples, central pathology review, and definitive cancer endpoint data in a large prospective trial population that included clinically established controls (i.e., with no prostate cancer diagnosed), the PCPT is a unique resource for comprehensive risk modeling.

The overall goals of the program are to investigate hypotheses concerning a) the biological mechanisms underlying the results of the PCPT and b) the etiology of prostate cancer with an eye toward risk assessment and prevention strategies. The program encompasses 5 interrelated and interactive projects, biostatistical and pathology-genotyping teams, and 26 investigators. The strategy for accomplishing our research goals is to conduct interactive epidemiologic and molecular studies using the common PCPT database. Each projects ask 2 questions: (1) What are the associations of a purported risk factor with the risk of prostate cancer and high-grade prostate cancer, focusing on associations in the placebo arm? and (2) Do the effects of this purported risk factor on the risk of prostate cancer and high-grade prostate cancer differ for those taking finasteride? Interactions between projects are also investigated to determine the joint effects of genetic, metabolic, and environmental risk factors on the outcomes of the PCPT.

The project required extensive planning and coordination for the transition of a randomized clinical trial into a translational and epidemiologic scientific investigation. This article describes the comprehensive methodology and multidisciplinary collaborations, both national and international, which are essential components of this major risk-modeling research program. We provide a framework for doing collaborative research in an international setting structured around a common theme of a clinical trial.

The foundation of this risk-modeling program, the PCPT, was designed to establish whether finasteride, a 5α-reductase type II inhibitor, prevents prostate cancer. Between 1994 and 1997, a total of 18,882 men from 210 study sites located throughout the United States and Canada were randomized in the PCPT to either the finasteride or placebo arm and followed yearly with a digital rectal examination (DRE) and PSA determination. Men found to have abnormal DRE findings or elevated PSA levels were recommended for prostate biopsy. Following 7 years of study participation, men without a diagnosis of prostate cancer were recommended to undergo an end-of-study biopsy. The primary study endpoint was the 7-year period prevalence of prostate cancer. The results were published on July 17, 2003 (3).

At the final analysis, prostate cancer status, either a diagnosis or negative end-of-study biopsy, was available in 59.6% of the participants in the finasteride arm and 63.0% of those in the placebo arm. Prostate cancer was detected in 803 of 4,368 (18.4%) in the finasteride arm and in 1,147 of 4,692 (24.4%) in the placebo arm; a 24.8% lower risk of prostate cancer for men in the finasteride group compared with men in the placebo group (P <.001). However, despite the reduction in risk of prostate cancer overall, there was an increased risk of a diagnosis of high-grade disease, defined as Gleason score 7–10. High-grade tumors were found in 280 of the evaluated men in the finasteride group (6.4%) versus 237 men in the placebo group (5.1%).

Subsequent analyses over the next 5 years helped to put the overall and high-grade disease findings into perspective. There is now convincing evidence that the initial observation of an excess risk of high-grade disease on the intervention arm of PCPT was due, at least in part, to detection bias: finasteride has been shown to significantly increase the sensitivity of both PSA (13) and prostate biopsy (5) for high-grade cancers, thereby enhancing their detection in men taking finasteride. In addition, statistical modeling studies, which adjusted for the fact that the majority of men with a diagnosis of prostate cancer did not undergo prostatectomy (the gold standard for determining prostate cancer grade), suggest that the risk of Gleason score 7–10 cancer is either unchanged or even reduced by finasteride (5–7).

The biological rationale for testing finasteride in the PCPT hinged largely on the nature of prostate cancer as an androgen-dependent disease (14, 15). Dihydrotestosterone (DHT) is the most biologically active androgen in the prostate and is synthesized through the irreversible reduction of testosterone (T) by the enzyme steroid 5α-reductase. The drug finasteride inhibits the type 2 isoenzyme of 5α-reductase, markedly decreasing the intraprostatic conversion of T to DHT (16). This rationale supported testing the preventive efficacy of finasteride in the PCPT. The initial results of the PCPT raised urgent questions regarding the use and activity of finasteride. A significant contribution of ongoing research based on PCPT will be to provide a better understanding of the precise role of hormone metabolism, with all its constitutional and somatic genetic variations, in prostate cancer risk and finasteride activity.

In the program concept phase, which began in 1998, an executive scientific review committee received and evaluated proposals solicited from a widely circulated requests for applications (RFA) from investigators across the United States and Canada for studies of genetic factors and other variables that affect an individual's risk of prostate cancer and the effect of the intervention with finasteride. Proposals were evaluated for their scientific merit, relevance of the project to the original PCPT study hypotheses, judicious use of biospecimens from this unique population, and the previous scientific accomplishment and experience of the investigators. The proposals need not have been related to the planned secondary objectives of the PCPT, and other than hypotheses related to the grade of carcinoma and the role of diet, they were not. Because of a numerous common themes across the accepted proposals, they were assembled into a program project (P01) application rather than released for submission as independent R01 applications. Proposals that were not accepted as part of the P01 still had the potential to move forward but would need to do so under a separate funding source (R01). Another year was spent refining the projects, identifying laboratories, performing pilot studies, and assembling the final P01 application that was submitted to the National Cancer Institute. Ultimately, the application received a favorable score and the notice of funding was received in December 2004.

The theme unifying the 5 key areas (or “studies”) within this project is the genetic, metabolic, and environmental factors associated with the risks of prostate cancer overall and high-grade prostate cancer specifically, and the effects of these factors on the efficacy of finasteride as a cancer preventive agent. The program also includes studies to better understand the mechanisms underlying these risk factor associations. Two major elements of this theme are (1) the study of genetic polymorphisms to identify molecular prostate cancer risk factors, from which we expect to determine pharmacogenetic profiles of men most likely to benefit from finasteride, and (2) the study of somatic mutations to discover the mechanisms underlying the increased risk of high-grade prostate cancer associated with finasteride use. Identifying factors or mutation that can distinguish aggressive or high-grade tumors in an era in which there is concern about the overdetection of prostate cancer will prove beneficial, and while the concern over the increase of high-grade disease in the finasteride arm has been tempered by subsequent research since the inception of this project, there is still interest in finding biologic results which support the detection bias finding.

Each of the 5 projects constituting this collaborative risk-modeling program is outlined in Table 1, and fuller descriptions are provided in the following paragraphs.

This project directly examines the mechanisms underlying the primary hypothesis of the PCPT: whether inhibiting conversion of T to DHT, the primary steroid-driving cell growth in the prostate, reduces the risk of prostate cancer. The overall project 1 hypothesis is that genetic variations in the steroid metabolic genes and variations in hormone levels are associated with both prostate cancer risk and the efficacy of finasteride in preventing prostate cancer. Project 1 exemplifies the new field of “preventive pharmacogenetics,” or the study of genetic factors that account for interindividual variability in response to a preventive agent. Another hypothesis of project 1 is that long-term finasteride exposure, which results in a DHT-starved environment, selects for the growth of high-grade prostate cancer cells. This is being investigated by determining rates of somatic mutations in high-grade lesions to see whether they differ between the finasteride and placebo arms. Project 1 also examines whether rates of specific somatic mutations differ across normal and neoplastic prostate tissues. Finally, this project measures post–baseline finasteride concentrations in the intervention arm using LC/MS. These data will provide insight into the adherence to the intervention and will be used in secondary analyses.

This project improves the understanding of 1) the associations of diet and diet-related factors such as obesity with prostate cancer risk and 2) the role favorable dietary patterns or use of dietary supplements may play in enhancing the efficacy of finasteride as a prostate cancer preventive agent. This project investigates dietary factors that affect steroid hormone and insulin-like growth factor (IGF) metabolism, such as obesity, glycemic index, and dairy foods, as well as dietary factors that affect oxidative load through prooxidant, antioxidant, and anti-inflammatory effects, such as lycopene, ω-3 fatty acids, glucosinolates, and tocopherols.

This project provides information regarding the hypothesis that circulating levels of IGF-1, IGF-2, IGF-BP3 (total and intact), IGF-BP2, leptin, and C-peptide, considered jointly as well as individually, are related to prostate cancer risk. The metabolic factors tested in project 3 are IGFs, peptide growth factors known to play important roles in regulating cell proliferation, differentiation, and death. Suggestive evidence from human population studies show that IGF-1 and insulin resistance increase risk for prostate cancer whereas IGF-BP3 decreases risk. Project 3 is also assessing whether IGFs and IGF signaling in prostate tissue is related to measures of cell renewal dynamics. If IGF-insulin receptor signaling is found to be associated with aggressive prostate cancer, one model to account for the PCPT results regarding increased risk of high-grade lesions in the finasteride arm is that the reduction of androgen receptor signaling associated with finasteride use results in selective pressure for neoplasms that are, at least in part, stimulated by growth factors other than androgens.

This project tests the hypothesis that inflammation and focal atrophy, possible environmental contributors (serum antibodies against infectious agents), and genetics (polymorphisms in genes involved in the innate and adaptive immune response) influence prostate cancer risk. Given that inflammation is a clear target for intervention, we anticipate that this work will have important implications for prostate cancer prevention. This study is the first to systematically examine focal atrophy lesions in men with and without prostate cancer. Prostate tissue is not typically available for middle-aged men who do not have an indication for biopsy, such as an elevated serum PSA. For this reason, it is unknown how the presence and extent inflammation correlate with the occurrence of prostate cancer. This study is the first to evaluate this association in a large sample of men with a known negative prostate cancer status.

This project evaluates hypotheses related to oxidative stress and DNA repair and prostate carcinogenesis. This study investigates whether variations in genes encoding enzymes that generate reactive oxygen species (ROS), neutralize ROS, and repair DNA damage are associated with risk of prostate cancer and high-grade cancer. Serum levels of oxidized proteins, a biomarker of oxidative damage, are measured to determine whether baseline levels of oxidative stress are elevated in those who go on to develop cancer and, if so, whether the association is modified by finasteride.

Interactions among projects are motivated by the strategic incorporation of cross-project interactive specific aims within each project. Some examples of these aims are as follows:

• Project 1 (“Androgen Metabolism in the PCPT”) and project 2 (“Diet and Diet-Related Factors in the PCPT”): To investigate the joint associations of serum sex steroid hormones and obesity in relation to prostate cancer risk.

• Project 2 (“Diet and Diet-Related Factors in the PCPT”) and project 4 (“Genotypic and Phenotypic Studies of Inflammation in the PCPT”): To investigate the associations of dietary and serum phospholipid fatty acids with inflammation and focal atrophy in prostate biopsy tissue.

• Project 3 (“Insulin-like Growth Factor Axis and Insulin Resistance in the PCPT”) and project 5 (“Oxidative Damage and DNA Repair in the PCPT”): To assess how IGFs may influence the oxidative stress–prostate cancer risk association.

• Project 4 (“Genotypic and Phenotypic Studies of Inflammation in the PCPT”) and project 1 (“Androgen Metabolism in the PCPT”): To test whether the difference between prostate cancer cases and controls in the prevalence and extent of inflammation and atrophy varies by serum testosterone or estradiol.

• Project 5 (“Oxidative Damage and DNA Repair in the PCPT”) and project 2 (“Diet and Diet-Related Factors in the PCPT”): To test whether the associations of specific dietary factors with prostate cancer risk differ by polymorphisms in genes affecting oxidative load.

A final goal is to evaluate the joint effects of risk factors across the 5 projects, both in terms of predicting prostate cancer and predicting who may be the ideal candidates for the use of finasteride to prevent prostate cancer. This model spans all projects incorporating the key findings from each project and from the cross-project interactions and could include serum and tissue measures, pharmacokinetic assessment, and genotyping results. Finally, findings from this model could be used to enhance the prostate cancer risk model that was developed using data from the PCPT (2).

The biostatistics team supports the development of study design, manages data collection, and performs statistical analyses to test the research hypotheses. In the study design phase, biostatisticians assisted investigators in formulating studies that can feasibly address the questions of scientific interest, are amenable to analysis by statistical and data-mining methods, and will ultimately yield statistically valid and interpretable results.

An equally important role for the biostatistics team is the coordination of the clinical database and the accumulated results from analyses on serum, WBC, and tissue samples. Because the laboratory data are ultimately integrated into the overall PCPT database, the biostatistics team developed a quality control plan and worked with each laboratory to ensure quality sample processing and data handling. This includes the management of the overall data infrastructure for the entire program consisting of the maintenance of the clinical database, the secure electronic transfer of laboratory data, and ensuring the security and confidentiality of all data. Other key roles of the biostatistics team include the identification of the specific samples to be sent, verification that the appropriate material is sent to the correct laboratory at the correct time, the tracking of the laboratory analyses, and the overall progress of each project.

The pathology-genotyping team is responsible for 3 major areas with the overall goal of supporting and providing the foundation for the project. These include 1) archival biorepository management and distribution, 2) histopathology, laser capture microdissection, immunohistochemistry, and image analysis, and 3) specialized sample preparation (i.e., DNA) and genotyping. The 3 major areas are largely interrelated with the intent being to provide the highest quality samples prepared to meet the needs of the 5 research teams and coupled with high-quality standardized pathologic analysis and genotyping. The use of high-quality biological specimens that are handled, processed, and interpreted in a uniform manner and that can be directly linked to clinical information and outcome ensures that results across all projects are reliable and comparable.

All projects with serum-based measurements use a single, shared, nested case–control sample of participants to evaluate associations of risk factors with the risk or prostate cancer so that each subject will provide a more complete biomarker and genetic profile for the evaluation of joint effects of these putative factors. In addition, several projects are using prostate tissue from the biopsy cores from both cases and controls and the prostatectomy tissue from cases. These studies use a variety of designs and laboratory methods to address questions about somatic mutation, inflammation, focal atrophy, proliferation, and growth factor signaling. In the following, we highlight features of the case–control study design, including the definition of cases and controls.

Data from the PCPT, a randomized clinical trial, can be thought of as a prospective cohort, with data collected on exposures before diagnosis of cancer being used to predict prostate cancer risk. However, even though this is a cohort study, a time-to-event is inappropriate. Because the PCPT had interim events and the detection of the interim events was biased between the arms due to the enhanced sensitivity of PSA, DRE and needle biopsy in the finasteride group, the only method to determine whether a participant was prostate cancer free or not was an end-of-study biopsy that assessed the endpoint in an unbiased fashion.

As a result of this detection method, our endpoints are not time-to-event analysis; rather, they are simply the presence or absence of prostate cancer by the end of 7 years of follow-up, as it was unknown when many of the cancers would have been detected without the end-of-study biopsy. We therefore define the outcome in the cohort study as the 7-year period prevalence of prostate cancer found at either a for-cause procedure or an end-of-study biopsy. Because the tissue, serum, and genetic assays are expensive, the prospective nested case–control study design was selected to efficiently examine a sample of men without prostate cancer.

The primary case definition for this program is biopsy-proven presence of prostate cancer. From an initial pool of 2,401 potential cases, cases for this project were excluded because the detection of the prostate cancer was after the trial was unblinded (n = 173) or was outside the established time frame of the 7-year end-of-study biopsy (n = 91) or an adequate baseline serum sample was not available (n = 328). This resulted in a total of 1,809 cases of prostate cancer with available baseline serum included in this project. Approximately 40% of our cases are interim and 60% are cases detected at the end-of-study biopsy. All analyses examining prostate cancer cases are stratified by Gleason score, defining high-grade cancer as Gleason score 7–10 (n = 498; 218 in placebo arm, 280 in finasteride arm) and low-grade cancer, defined as Gleason score 2–6 (n = 1,233; 782 in placebo arm, 451 in finasteride arm). Although the subsequent analyses, described earlier (4–7), have tempered concerns regarding the excess of high-grade cancer diagnosed in men receiving finasteride, all projects continue to examine high-grade cancers, given the clinical relevance of high-grade disease and the urgent need to develop markers of aggressive disease that go beyond Gleason score. In addition, given the relatively small numbers of Gleason score 8–10 cancers, it was not possible draw definitive conclusions regarding the potential of finasteride to affect this important subgroup.

Two endpoints often used in epidemiologic studies of prostate cancer, stage and prostate cancer–specific death, cannot be examined in the project research proposed because nearly all of the men with a diagnosis of prostate cancer in the PCPT had early-stage disease (i.e., 98% were T1 or T2). This is most likely due to the requirement for serum PSA levels of 3 ng/mL or less and normal DRE at randomization and annual PSA and DRE testing.

Controls were randomly selected in a 1:1 ratio from men who completed the end-of-study prostate biopsy and had no evidence of prostate cancer. Controls include all eligible non-White men (primarily African Americans and Hispanics) to better support exploratory analyses in these subgroups. Remaining controls were frequency matched to cases by distributions of age at randomization, in 5-year intervals, first-degree family history of prostate cancer (established risk factors for prostate cancer), and intervention arm. By oversampling the non-White men and matching on the factors listed earlier, the estimated race-specific odds ratios for prostate cancer are not interpretable.

Because of the impossibility of sampling the entire prostate using needle biopsies, it is possible that some men with prostate cancer will be misclassified as controls. However, the controls in this program will yield a much lower misclassification of prostate cancer disease status than that typically seen in epidemiologic studies using cumulative incidence sampling of controls. This reduced misclassification is especially important when the expected associations for some of the research questions being addressed are in the small to moderate range [e.g., single-nucleotide polymorphisms (SNP) and prostate cancer].

The baseline demographics and potential prostate cancer risk factors [diabetes, body mass index (BMI), physical activity, and smoking status] for the cases and controls, as well as for those with noncancers in the PCPT who were eligible to be in the controls but were not chosen, are presented in Table 2. Also presented are the characteristics of the men in the PCPT who had an endpoint evaluated (i.e., either had an interim prostate cancer or had an end-of-study biopsy within the required time frame) and those who were not evaluated. As shown, the men who were evaluated for an endpoint were younger, more likely to have had a family history of prostate cancer, and tended to be Caucasian. Redman et al. reported that by accounting for the nonrandom missing biopsy results, the true rate of cancer would have been slightly less in both intervention arms, but the relative risk was minimally changed (4).

Because the population to be studied in this project is frequency-matched case–control with specific oversampling of non-White race and positive family history, men in the project were older and have a higher baseline PSA than those men not in the project.

The approach for the analysis of the primary aims for each of the projects is to apply the same methods across all projects when possible. This includes using the same set of cases and controls and definitions of exposure variables and endpoints. The preliminary analyses produce standard tables and refinements are made to address issues specific to individual projects and/or analyses. For the serum analyses, this has been relatively straightforward to implement once agreement among the investigators was achieved. Because the tissue analyses are done in differing subsets, the analyses are more project-specific.

Each biological analyte is assessed separately in each of the intervention (placebo and finasteride) groups. If there is no evidence of an interaction between the exposure analyte and intervention, the 2 arms are pooled. This increases the sample size and greatly increases the power to identify associations.

In the design phase, each investigator consulted with the statisticians to determine the achievable power for each of the primary and cross-project aims given a sample size of 1,809 cases and 1,809 controls and their prespecified, clinically meaningful differences. Power was calculated for the main effects, intervention interactions, and effect modification by genetic variation. Tables 3–5 provide some examples of the minimally detectable odds ratios with 80% power and a 2-sided α of 0.05.

Most cross-project interactive aims take the analytic form of tests of interactions: between genes, between genes and metabolic and behavioral measures, and between metabolic and behavioral measures. These cross-project collaborations are feasible because of both the large number of cases and the coordination of study designs (including sample selection) across projects. There is also reasonable power to detect statistically significant tests of cross-project aims. In general, if odds ratios comparing high to low quartiles of an exposure differ by a ratio of about 2 between 2 groups (e.g., by polymorphism), there is 80% power to detect such a difference.

Another focus of our analyses is to maximize the ability to address questions related to race/ethnicity and prostate cancer risk. African American men have an elevated risk of prostate cancer, and there is some evidence that this may be linked to genetic characteristics (e.g., short AR CAG repeat length, CYP3A4 polymorphisms) and higher serum androgen levels (17). Extensive efforts were made to recruit minorities into the PCPT; however, only 8% of participants were non-White and only 4% were African American. We have oversampled controls to include all non-White men, and race/ethnicity is used as a stratification variable in exploratory analyses. Although limited in power to address hypotheses stratified by race/ethnicity, this program includes the largest African American cohort (216 controls and 90 cases) we are aware of to have been included in comprehensive molecular epidemiologic studies of prostate cancer.

For the genetic aims, the program investigators are primarily using a candidate gene and SNP approach. Several criteria were used to select the specific genes and polymorphisms to be assayed in the different pathways of interest. Data on functional significance or data demonstrating a significant role in prostate or any other cancer were the primary criteria used for selection. Other criteria include SNPs that result in amino acid changes in the protein (i.e., nonsynonymous SNPs) and changes in promoter regions and splice sites. In general, variants with a minor allele frequency of 5% or higher were selected to have good statistical power. The major exception to this is SNPs in project 1, in which a number of variants at very low frequency will be studied in the SRD5A2 gene because of their known functional significance and relevance to the finasteride intervention. Additional genotyping will be done on tagging SNPs for key genes. Each project prioritized which tagging SNPs would be genotyped on the basis of their knowledge of the gene variants and the number that would be needed for coverage.

With extensive serum measurements and genotyping from several different projects, the program provides the opportunity to study the complexity of potential joint effects within and between projects. A number of methods will be explored including logistic regression adjusting for “upstream” effects and potential interactions, classification and regression trees (CART), or logic regression to identify prognostic groups based on combinations of SNPs or other biomarkers. Bayesian model averaging, which draws inferences about particular effects, taking into account the uncertainty about the correct model form, will also be used. A pharmacokinetic approach that attempts to use physiologically based representations of the biochemical pathways to statistically model the dependencies between factors and pathway analysis providing estimates of hypothesized causal connections between variables are other planned analytic methods. All of these methodologies rely on interdisciplinary collaboration. These models are useful not only to identify the complex joint effects on prostate cancer risk but also to identify groups of men who are either more or less likely to benefit from the chemopreventive effects of finasteride.

At the time of the development of this project, it was clear that the statistical analysis for genetic polymorphisms was a rapidly evolving field. Over the course of the project, improved and less expensive methods have become available, which have permitted SNP analyses to be conducted at lower cost while providing flexibility to select new candidate SNPs and haplotype tagging SNPs.

This program is unique in that all cases in the study sample are identified and the control group contains only men who are negative for biopsy-detected cancer. Case–control studies to date have been based on control groups without definitive knowledge of cancer status. The PCPT demonstrated that a moderate prevalence of prostate cancer, including high-grade cancer, occurs even among men with low PSA levels (19). This prevalence is especially problematical for studies completed in the PSA era because they included early-stage, low-grade cancers in the case group and cannot rely on PSA screening to eliminate such cancers from the control group. This misclassification weakens studies attempting to understand the risk factors for early-stage cancer and may explain both inconsistent findings across studies and the apparent lack of associations of many risk factors with early-stage disease. This issue may be particularly important for studies in which the magnitude of effects may be small but population-attributable risk is potentially very high.

The cases and controls, and the statistical methods to evaluate associations of risk factors with disease, are shared across all projects. The selection of prostate tissues and serum specimens is carefully integrated across projects to maximize the number of cases and controls for which genetic, metabolic, and histopathologic data are available, which will maximize the ability to complete cross-project interactive aims.

The number of cases in this program is based on a larger number of cases than most prospective studies, and this program is one of the largest prospective studies with access to prostate tissue and multiple serum samples collected throughout the follow-up period. The comprehensive set of data available on cases and controls allows this program to address many important questions in prostate cancer etiology and prevention.

This program uses a central pathology core to histopathologically classify cancer and other abnormalities. While many studies have used clinical stage to identify aggressive disease, no large prevention studies have had access to uniform pathologic grading based on a central pathology facility using rigorously standardized protocols. This is especially important for studies of high-versus low-grade disease in which it is difficult to standardize results across different pathologists and laboratories.

An overarching aim of this program is to examine joint associations among prostate cancer risk factors such as gene–gene and gene–environment interactions (diet, hormones, growth factors) that will be integrated into comprehensive risk models.

The cases identified during the PCPT were mostly early, clinically localized prostate cancers. Although the PCPT has been criticized to prevent clinically insignificant cancers, the cancers detected reflect the characteristics of the current cases that are being detected and treated in the PSA screening era. Subsequent analysis has determined that many of the criticized end-of-study cancers are “clinically meaningful,” using the most accepted definitions for classification based on biopsy pathology (18). The criteria for prostate biopsy have changed since the implementation of this trial, from a PSA cutoff of greater than 4.0 ng/mL to greater than 2.5 ng/mL. As a result, many of the cancers found on end-of-study biopsy would have been found during the course of the trial if this lower cutoff had been the threshold for biopsy.

The selection of the projects and investigators for this program resulted from an open process of evaluating responses to widely distributed RFAs to use the PCPT biorepository and data set. This open and impartial review process continues and will continue into the future. The Southwest Oncology Group Executive Committee Review, independent from either PCPT or this program's leadership, is in place to evaluate new proposals to use the PCPT biospecimens. This review process provides no funding but only permission to include the formal approval to use PCPT specimens/data in applications for funding from other sources. The committee's review includes determining whether proposed future research will have potential conflicts (with respect to duplication, feasibility etc.) with the ongoing projects. In general, approval based on priority is given only to projects that address prostate cancer prevention and at least 1 of the following: 1) Questions related to understanding the PCPT clinical outcomes; 2) Questions that require the PCPT biorepository and cannot be addressed otherwise; and 3) Questions that expand and contribute to the ongoing risk-modeling program. Studies that do not address prostate cancer prevention but meet one of the other criteria will be considered on the basis of scientific merit and amounts and types of materials requested. The application can be accessed online at http://www.cancer.gov/pcpt.

The project described here is unparalleled in the combined biology and etiology of cancer in its access to the resources, in particular annual blood samples, from a large, placebo-controlled, randomized prevention trial with a definitive cancer primary endpoint. The program benefits from parallel study designs and integrated selection of tissue and serum specimens across projects; the selection of cases and controls and the statistical methods to evaluate associations of risk factors with disease are shared across all projects. Furthermore, the selection of prostate tissue and serum specimens is integrated across projects to maximize the number of cases and controls for which genetic, metabolic, and histopathologic data are available.

This project represents a unique opportunity to test biological hypotheses related to a clinical trial in a coordinated manner. The strengths of the program are multifold and include the wealth of biological samples, both in total quantity and in the multiple sources (blood, tissue, sputum) that represent and enable the biological basis for the project. Another key strength comes from the interdisciplinary approach that spans the 5 individual projects and the common set of methods used across studies, including the central repositories, sample distribution, and centralized genotyping that minimize variation across studies and permits the analysis of main effects in individual projects and joint effects across studies. Furthermore, the unified approach to the analysis, which incorporates a common set of definitions, cross-project aims and a cohesive analysis plan that addresses each of the individual projects, and ultimately an analysis that will span all of the projects, gives this program a strong and unique backbone.

The approach taken by this project does not come without challenges. The diverse locations of the 5 projects and biostatistics and pathology-genotyping centers can be cumbersome for coordination and communication and requires a commitment to open and effective communication. This is achieved through monthly teleconferences, twice yearly in-person meetings, and is augmented by focused discussion by phone and e-mail discussions that are topic-specific or related to specific analysis or procedural issues. Another challenge is in managing multiple laboratories with differing sample and data-handling capabilities. This challenge is managed by developing specific procedures within some laboratories, training, and close regulation of all steps of the laboratory process. Finally, the program was conceived prior to publication of the results of the PCPT. The results of the trial and the evolving interpretation based on subsequent analyses of the data have resulted in some of the hypotheses being modified and additional hypotheses developed.

The program entered its most productive period in mid-2009, with completion of the vast majority of laboratory analyses of serum and tissue. The initial round of genotyping has been completed and a second expanded round is in process. A number of papers are under review by the authors, submitted for publication, accepted, or already have been published. The topics of these reports cover some of the pilot work that was done to prepare for the program, methodologic issues, and results from specific projects.

This program represents a model for team science and the use of the biospecimens generated from a large clinical trial. One example is an ongoing SNP-based risk-modeling program nested within a randomized, placebo-controlled trial to prevent head and neck cancer–related second primary tumors (20). All large, definitive prevention trials could be designed with an eye toward the development of future programs that utilize their biorepositories. For example, the definitive Selenium and Vitamin E Cancer Prevention Trial (21) in 35,533 men was carefully designed with data and biological specimen collection methods and objectives that would facilitate follow-on risk-modeling and other molecular biological research. This approach assembles experts in a number of cancer research areas into a single collaborative effort. Although benefiting from the expertise, the approach has a cost in the learning curve to understand the study design, science, and procedures of the parent randomized, controlled trial, as illustrated by this program founded on the PCPT.

No potential conflicts of interest were disclosed.

Grant Support

This work was funded by the National Cancer Institute, National Institutes of Health (CA37429 and P01CA108964). The content of this work is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.