Maryland Prostate Cancer Case-Control Study

Division Of Basic Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

Our study was implemented in two phases. The first phase, which started in April of 2005, constituted a pilot study to evaluate recruitment procedures. This phase was successful, and the full study was initiated with minor changes to the protocol in April of 2006. The full study was completed in 2015 [976 cases (489 African American and 487 European American) and 1034 population-based controls (486 African American and 548 European American)]. We collected blood and urine from all individuals and paraffin-embedded and fresh-frozen tissue specimens form 135 prostatectomy surgeries. Cases are from two Baltimore hospitals, the Veterans Affairs Medical Center and the University of Maryland Medical Center. Cases have pathologically confirmed prostate cancer. They had a disease diagnosis within the last two years prior to recruitment and presented with prostate cancer at all stages of the disease. One-hundred and thirty-eight cases had advanced stage disease (n = 79 with T3 and n = 59 with T4 disease). Cases were also assigned to National Comprehensive Cancer Network risk groups, based on the patients' TNM stage, Gleason score, Gleason pattern, and PSA level at diagnosis according to the guidelines, for prostate cancer, version 2.2021. They were classified into low, intermediate, high, and very high risk based on the likelihood of their disease to progress to lethal prostate cancer. Lastly, we have secured current information on overall and disease-specific survival from National Death Index records for the cases. A most recent update brought the number to 247 deaths from all causes with 66 of them being reported as a prostate cancer-specific mortality. The median survival follow-up time for the cohort is >10 years with an NDI follow up through December 31, 2022. We will continue to collect this information for both cases and controls. More recently, we were able to determine PSA values at recruitment for 888 men from the control group. Only 16 of them had PSA > 4 at time of recruitment (< 2%). Furthermore, we defined 823 patients as incident cases (422 African American, 401 European American) when they were recruited into the study within one year after the disease diagnosis, having an average interval between diagnosis and enrollment into our study of 4.8 months (4.4 months for African American and 5.2 months for European American men). The population-based controls were identified through the Maryland Department of Motor Vehicles database and were frequency-matched by age and race to cases. The study involved the administration of a survey and collection of blood and urine from all study subjects. Tumor specimens were obtained from cancer patients if they were available after prostatectomy. Our survey evaluates tobacco use, medication use, occupational history, diet, medical and sexual history, familial cancer history, and socioeconomic status. Current activities in this study include the collection of additional data from pathology and medical records to have clinicopathology for all cases and information on disease recurrence as available. Participants were evaluated for their West African, European, and Native American ancestry using 105 ancestry informative markers. The genotyping data showed that self-identified African American participants have an average West African ancestry of 75.5% among cases and 72.1% among controls whereas the European ancestry in the self-identified European American participants ranged from an average 85.8% among cases to 89.9% among the controls. For a subset (83%) of the cases and controls, we obtained additional West African ancestry estimates using the Infinium HumanOmni5-Quad BeadChip array (genotyping data for 706 cases and 744 controls were submitted to dbGaP). West African ancestry estimates using the two approaches were highly similar (r=0.98). In 2021, we completed a multi-year effort to link neighborhood measures of poverty to participants in our studies. We have completed analyses using neighborhood deprivation and gentrification as exposures. A neighborhood deprivation index was generated following the guidelines by Messer at al. Our index contains the following variables: percent (%) households in poverty, % female headed households with dependent children, % households on public assistance, % households earning under $30,000/year, % households with no car, and % males and females unemployed (see PMID: 36662526). This Baltimore, Maryland, based study is aimed at identifying differences in risk factor exposure and tumor biology between African American and European American men. Molecular work will be used to examine race/ethnic differences in tumor biology and circulating biomarkers. Our research is also aimed at identifying environmental and inherited factors (e.g., infections and immune response, smoking exposure, ancestry germline genetic factors, low penetrance susceptibility loci) that promote the development of an aggressive disease and specifically contribute to the survival disparity between African American and European American men. A major research focus is the role of tumor and systemic inflammation in disease progression because of our previous observation that tumors of African American patients contain a prominent immune-inflammation gene signature (PMID: 18245496). We are also investigating the relationships of neighborhood factors and African genetic ancestry with prostate cancer. For example, we used this resource to evaluate West African ancestry as a risk factor for prostate cancer. We found it is - but in interaction with the neighborhood environment. Importantly, our data showed that West African genetic ancestry was a greater risk factor for men residing in neighborhoods with high levels of deprivation than for men living in more affluent neighborhoods (JAMA Network Open; PMID: 39283637). As ongoing research, we are currently developing a measure for allostatic load that is a surrogate for stress exposure in prostate cancer patients in our cohort. We are analyzing urine-based surrogate markers for stress exposure together with traditional allostatic load markers derived from medical records and evaluate their interrelationship as well as their relationship with prostate cancer outcomes. Research question: Are the allostatic load variable and the surrogate markers for stress exposure similarly associated with an increased disease mortality?