Search

Persons experiencing homelessness (PEH) face barriers to cancer screening, leading to cancer diagnosis at more advanced stages. Data on prostate specific antigen (PSA) screening in this population is scarce. The aim of this study was to 1) assess PSA testing rates in PEH and identify factors associated with screening and 2) compare PSA screening rates in PEH with persons not experiencing homelessness (non-PEH). This is an IRB approved study of PEH and non-PEH cared for at a large metropolitan hospital system from 2014 to 2021. We identified PEH from a prospective institutional registry which started screening for homelessness in January 2014. Homelessness was defined by either the presence of the Z-code for homelessness (Z59) within the medical record, the listed address was matched to homeless shelters or other transitional housing, or the patient identified as homeless. Institutional CPT and HCPCS billing codes for PSA testing were available beginning 01/01/2017 and cross-referenced to identify patients who underwent PSA testing at least once between 01/01/2017 and 12/31/2021. We extracted PSA testing information for a matched cohort of non-PEH based on sex, age, county of residence, and date of last encounter for comparison. We included men ages 50-69 that lived in our institution's county or an immediately adjacent county. The primary outcome was to assess the rate of PSA testing in the past 5 years in accordance with national screening guidelines (USPSTF, AUA, ACS) and to identify factors associated with PSA testing in PEH. A total of 9,249 PEH were identified, with 1,605 eligible for PSA screening (men between ages 50-69) within the study timeframe; 3,421 non-PEH were eligible for screening comparison. The median age of PEH was 58.0 years old and the majority of PEH were Black (50%) or White (43%). 67% of PEH were current tobacco users, 52% current alcohol users, and 34% current illicit drug users. Medicaid was the most common insurance (51%) followed by Medicare (19%) and private insurance (4%). PEH were less likely to have a PCP (58% vs 81%, p<0.001) and had a significantly lower PSA testing rate (13% vs 34%, p<0.001) compared to non-PEH. On univariate analysis, PEH were more likely to be Black (p<0.001), single (p<0.001), unemployed (p<0.001) and have Medicaid/no insurance (p<0.001). On multivariate analysis, each of these associations remained significant; most notably, lack of a PCP (OR=1.78; 95% CI 1.35-2.30, p<0.001) or PSA testing (OR=1.63; 95% CI, 1.25-2.14, p<0.001) were associated with increased odds of being homeless. Within the homeless population, multivariate analysis showed that having a PCP (OR=2.58; 95% CI, 1.64-4.05) significantly increased the odds of getting PSA testing. PEH experience low rates of prostate cancer screening. Interventions to increase screening in this population, including increased PCP access, are warranted. [ABSTRACT FROM AUTHOR]

Morphologic features such as cribriform and intraductal carcinoma (CF/IDC) are associated with unfavorable oncologic outcomes in prostate cancer (PCa) patients. Few studies have investigated whether commonly utilized genomic classifiers such as Decipher are associated with CF and/or IDC. The objective of our study was to identify any correlations between Decipher genomic risk score and CF/IDC status while also assessing PCa transcriptomic features of these pathologic features. A retrospective review of PCa patients who had Decipher testing at a single high-volume center between 2009-2020 was performed. A genitourinary pathologist reviewed specimens for the presence of CF and IDC features and annotated three groups: absent CF/IDC (CF-/IDC-), CF positive only (CF+/IDC-), and CF/IDC positive (CF+/IDC+). Categorical clinical, genomic, and pathologic variables were assessed using the Pearson Chi-Square test, quantitative with the Kruskal-Wallis test. Molecular subtyping using the PAM50 classifier was performed. Gene set enrichment analysis (GSEA) was performed to identify differentially expressed pathways between patients that were CF-/IDC- compared to CF+/IDC+. Four hundred and sixty-three patients were included. Patients that were CF+/IDC+ had the highest mean Decipher scores (CF+/IDC+: 0.77 vs. CF+/IDC-: 0.71 vs. CF-/IDC-: 0.61, p<0.001). Patients who were CF+/IDC also had the highest rates of Grade Group ≥ 3 disease on final pathology (CF+/IDC+: 79% vs. CF+/IDC-: 52% vs. CF-/IDC-: 52%, p<0.001). Gene network level transcriptomic analysis revealed upregulation of MYC pathway genes (Normalized Enrichment Score (NES): 1.65, p=0.04) and E2F targets (NES: 1.69, p=0.03) respectively. Molecular subtyping revealed that CF+/IDC+ tumors were associated with luminal proliferating type B classification (CF+/IDC+: 53% vs. CF+/IDC-: 48% vs. CF-/IDC-: 29%, p<0.001). We report an association between Decipher score and Cribriform/Intraductal morphology in PCa. CF+/IDC+ tumors had higher average Decipher scores and upregulated adverse gene programs (e.g., MYC, pRB1). CF+/IDC+ was associated with the luminal proliferating B subtype, which has been shown to have the worst oncologic outcomes but most sensitivity to ADT. [ABSTRACT FROM AUTHOR]