Prostate Cancer in High-Risk Population

Abstract

In 2018, the United States Preventive Services Task Force (USPSTF) recommended that African American men and those with a family history of prostate cancer should discuss the pros and cons of PSA testing with their physician and engage in shared decision making. Identifying risk factors of prostate cancer, calculating individualized prediction of the risk of prostate cancer, and ensuring that the patients are well informed with knowledge to understand these risks so as to easily make decisions, are extremely important in shared decision making. The overarching goal of this dissertation is to assist in these three steps and facilitate shared decision making in a high-risk population. Aim 1 assessed the association of demographic characteristics, clinical markers and genitourinary symptoms with the diagnosis of prostate cancer (Aim 1a), as well as with the diagnosis of significant prostate cancer (Aim 1b), in a high-risk population. A nested case-control study for Aim 1a and a case-control study for Aim 1b was conducted using the Prostate Risk Assessment Program (PRAP) data, which enrolls African American men and those with a family history of prostate cancer. Multivariate conditional logistic regression was used to assess the association between the risk factors for any prostate cancer, while multivariate logistic regression was used for clinically significant prostate cancer. The risk of any prostate cancer increased with increasing age, presence of family history and increasing PSA levels, while the risk of significant prostate cancer was associated with increasing PSA level. This suggests that PSA level, as a continuum, is extremely important while predicting prostate cancer, especially significant prostate cancer within a high-risk population. Using Aim 2, we compared the performance of two prostate cancer risk calculators, Prostate Cancer Prevention Trial – Risk Calculator 2.0 (PCPT-RC 2.0) and the European Randomized Study of Screening for Prostate Cancer -Risk Calculator 3/5 (ERSPC-RC 3/5) to predict any prostate cancer and clinically significant prostate cancer in a high-risk American population. All men who underwent prostate biopsy with the PRAP data registry since 1996 were included in the study. The probability of being diagnosed with any prostate cancer and significant prostate cancer (Gleason score > 6) was calculated using the online versions of PCPT-RC 2.0 and ERSPC-RC 3/5. The performance of these calculators was compared using calibration (calibration plot and calibration-in-the-large), discrimination (comparing AUC curves using DeLong’s method) and decision curve analysis (to assess clinical utility). The calibration suggested that both risk calculators under-predicted the probability of any prostate cancer while PCPT-RC 2.0 over-predicted the probability of significant prostate cancer. Analysis of the AUC curves suggested that the PCPT-RC 2.0 (AUC: 0.59, 95% CI 0.52 to 0.66) showed a trend towards better discrimination for any prostate cancer as compared to ERSPC-RC 3/5 (AUC: 0.55, 95% CI 0.48-0.63, p= 0.3819). Similarly, PCPT-RC 2.0 (AUC: 0.71, 95% CI 0.61-0.82) showed a trend towards better discrimination for significant prostate cancer as compared to ERSPC-RC 3/5 (AUC: 0.63, 95% CI 0.51 to 0.75, p= 0.2335). PCPT-RC 2.0 proved to be clinically beneficial to predict significant prostate cancer in the range of lower prediction thresholds. These results suggest that the PCPT-RC 2.0 is superior to ERSPC-RC 3/5 in a high-risk American population. Aim 3 utilized a systematic review of the decision aids used to improve prostate cancer knowledge, improve risk perception, reduce confusion, involve in shared decision making or utilize PSA tests in men at high-risk of prostate cancer, defined as those with African descent or those with a family history of prostate cancer. Data was extracted by searching MEDLINE, CINAHL, EMBASE, and PsycINFO via Ovid and EBSCOhost. After screening titles and abstracts, the resulting full-text articles were assessed for inclusion and exclusion criteria. A data extraction table was created, and the methodological quality of the studies was assessed based on three criteria – randomization, double blinding and intention-to-treat analysis. Due to the clinical heterogeneity of the studies, a descriptive analysis of all the studies was conducted and tabulated. A total of 2605 articles were retrieved after literature search, of which 8 articles met the inclusion criteria and were included in the qualitative analysis. Of these 8 articles, 6 were targeted at those who were African American or those with an African descent and 2 articles included interventions targeted at those with a family history of prostate cancer. Majority of the studies targeted at African American men demonstrated an improvement in knowledge and reduction in decisional conflict in the intervention group compared to the control group. The two studies that included men with a family history of prostate cancer did not show any change in knowledge or decisional conflict in the intervention group compared to the comparison group. All studies were of low quality, except one which was medium quality. Thus, this review unveiled that tailored decision aids would be helpful in improving knowledge and reducing decisional conflict in African American men while decision aids designed for men with family history of prostate cancer would not significantly change prostate cancer knowledge or decisional conflict compared to the standard decision aid. Thus, one of the tailored decision aids can be used to help African American men improve their knowledge of prostate cancer and reduce decisional conflict, while the standard decision aid can be used in men with a family history of prostate cancer. These conclusions can be assimilated into the USPSTF recommended shared decision-making sessions between the patients and the physicians.