Adequacy of prostate cancer prevention and screening recommendations provided by an artificial intelligence-powered large language model.

PURPOSE: We aimed to assess the appropriateness of ChatGPT in providing answers related to prostate cancer (PCa) screening, comparing GPT-3.5 and GPT-4. METHODS: A committee of five reviewers designed 30 questions related to PCa screening, categorized into three difficulty levels. The questions were formulated identically for both GPTs three times, varying the prompts. Each reviewer assigned a score for accuracy, clarity, and conciseness. The readability was assessed by the Flesch Kincaid Grade (FKG) and Flesch Reading Ease (FRE). The mean scores were extracted and compared using the Wilcoxon test. We compared the readability across the three different prompts by ANOVA. RESULTS: In GPT-3.5 the mean score (SD) for accuracy, clarity, and conciseness was 1.5 (0.59), 1.7 (0.45), 1.7 (0.49), respectively for easy questions; 1.3 (0.67), 1.6 (0.69), 1.3 (0.65) for medium; 1.3 (0.62), 1.6 (0.56), 1.4 (0.56) for hard. In GPT-4 was 2.0 (0), 2.0 (0), 2.0 (0.14), respectively for easy questions; 1.7 (0.66), 1.8 (0.61), 1.7 (0.64) for medium; 2.0 (0.24), 1.8 (0.37), 1.9 (0.27) for hard. GPT-4 performed better for all three qualities and difficulty levels than GPT-3.5. The FKG mean for GPT-3.5 and GPT-4 answers were 12.8 (1.75) and 10.8 (1.72), respectively; the FRE for GPT-3.5 and GPT-4 was 37.3 (9.65) and 47.6 (9.88), respectively. The 2nd prompt has achieved better results in terms of clarity (all p < 0.05). CONCLUSIONS: GPT-4 displayed superior accuracy, clarity, conciseness, and readability than GPT-3.5. Though prompts influenced the quality response in both GPTs, their impact was significant only for clarity.

A Ten Year Experience of Men's Health Events in a Socioeconomically Diverse City in the United States - Lessons Learned.

Community-based health events provide an opportunity to increase knowledge, awareness, and screening for acute and chronic diseases among individuals living in a socioeconomically diverse community. Because there are limited reports of such events, here we describe our ten-year experience of annual men's health fairs. This retrospective study of the Michigan Institute of Urology Foundation evaluated Men's Health Events held in Detroit, Michigan, from 2012 to 2021. Over 10 years, 11,129 men were screened and > 100,000 screenings were performed. The majority of the attendees were African-American men (61%), had a college degree (67%) or a high school diploma (26%), and had an annual income of <$35K (47%) or $35-60 K (30%). From 2012 to 2021, participants who saw a doctor in the past year rose from 62 to 70%; the median age of men rose from 52 to 58; their median testosterone levels increased from 353 ng/dL to 412 ng/dL, and men with concerning prostate-specific antigen values (≥ 4 ng/mL) doubled from 5% to 10%. Among participants, 59% had cholesterol levels of < 200 mg/dL, 28% of 200-240 mg/dL, and 13% of > 240 mg/dL; 7% had glucose levels of < 70 mg/dL, 68% of 70-105 mg/dL, and 25% of > 105 mg/dL ; 24% had ≥ 140 mmHg systolic and 18% had ≥ 90 mmHg diastolic blood pressure. Our findings suggest that community health events are successful at attracting and screening diverse community members. Such events should emphasize screening of high-risk individuals for acute and chronic diseases and promote other health-related behaviors.

Comparison of patient background between a real-world North American cohort and the Göteborg-2 trial.

OBJECTIVES: To analyze the generalizability of the Göteborg-2 findings to a North American cohort. METHODS: We replicated the Göteborg-2 inclusion criteria in our Henry Ford Health (HFH) cohort, by identifying all patients 50-60 years old who had a PSA test from 2013 to 2018. The first PSA within the study period was considered PSA at entry, and included in the analysis. Chi-square test was used to compare categorical variables between the Göteborg-2 and HFH cohort, with a particular focus on Black men, who were also analyzed separately. RESULTS: The HFH patients included in the cohort were 49 456, of which 8562 were Black. In patients within the entire HFH cohort, HFH Black cohort, Göteborg Reference cohort, and Göteborg Experimental cohort, the rate of PSA ≥3 ng/mL was, respectively, 6.8%, 10.2%, 6.8%, and 6.6%. The rate of biopsy performed was, respectively, 1.8%, 4.1%, 5.8%, and 2.5%. PCa was found in, respectively, 1.4%, 3.0%, 2.3%, and 1.5%; Gleason score 3 + 3 in, respectively, 0.5%, 0.8%, 1.2%, and 0.6%; Gleason score > 3 + 3 in, respectively, 0.9%, 2.2%, 1.1%, and 0.9%. CONCLUSIONS: Our cohort had a lower biopsy rate and a lower incidence of non-csPCa diagnosis than both Göteborg cohorts, while still maintaining the same incidence of csPCa. This implies that the benefits of reducing non-csPCa diagnosis, as observed in the Experimental Göteborg cohort, are not necessarily replicable in U.S. "real-world practice" patients. Also noteworthy, we had a significantly higher percentage of Black men, who showed more aggressive disease.

Impact of lymphovascular invasion on survival in surgically treated upper tract urothelial carcinoma: a nationwide analysis.

OBJECTIVES: To assess the prognostic ability of lymphovascular invasion (LVI) in upper tract urothelial carcinoma (UTUC) as a predictor of overall survival (OS) using a large North American cohort. PATIENTS AND METHODS: Our cohort included 5940 patients with clinical M0 UTUC who underwent a radical nephroureterectomy (RNU), between 2010 and 2016, within the National Cancer Database. The main variable of interest was LVI status, and its interaction with pathological nodal (pN) status. Kaplan-Meier curves were used to depict the OS also stratifying patients on LVI status. Cox regression analysis tested the impact of LVI status on OS after accounting for the available covariates. RESULTS: The median (interquartile range [IQR]) age at diagnosis was 71 (63-78) years and most patients had pathological T1 stage disease (48.6%). Nodal status was pN0, pN1 and pNx in 45.8%, 6.3% and 47.9%, respectively. Overall, 22.1% had LVI. The median (IQR) follow-up time was 32.6 (16.0-53.3) months. At the 5-year postoperative follow-up, the estimated OS rate was 28% in patients with LVI vs 66% in those without LVI (P < 0.001). When patients were stratified based on nodal status those rates were 32% vs 68% in pN0 patients (P < 0.001), 23% vs 30% in pN1 patients (P = 0.8), and 28% vs 65% in pNx patients (P < 0.001). On multivariable analysis, the presence of LVI was associated with less favourable OS (hazard ratio 1.79, 95% confidence interval 1.60-1.99; P < 0.001). CONCLUSION: Our study assessed the impact of LVI on OS in patients with UTUC in a large North American nationwide cohort. Our series, as the largest to date, indicate that LVI is associated with less favourable survival outcomes in patients with UTUC after RNU, and this variable could be used in counselling patients about their prognosis and might be a useful tool for future trials to risk-stratify patients.

Assessing the impact of lymphovascular invasion on overall survival in surgically treated renal cell carcinoma patients: A nationwide cohort analysis.

INTRODUCTION: Lymph-vascular invasion (LVI) is recognized as an adverse pathological feature in patients with renal cell carcinoma (RCC). However, its impact on overall survival (OS) is not clear and scarcely addressed in the literature. We aimed to assess the prognostic ability of LVI as a predictor of OS in RCC patients using a large, North American cohort. METHODS: We included 95,783 cM0 RCC patients, diagnosed between 2010 and 2015, who underwent partial or radical nephrectomy within the National Cancer Database. Kaplan-Meier curves and log-rank tests were used to depict and compare survival curves. Cox regression analysis tested the impact of LVI on OS, after adjusting for all available confounders. RESULTS: Mean age (SD) was 59 (12), and most patients had pT1 stage (72.2%). Nodal status was pN0, pN1, and pNx, in 14.5%, 2.3%, and 83.3%, respectively. Overall, 9.0% of patients had LVI. The mean (SD) follow-up of the cohort was 39 months (24). At 5 years, OS was 65% in patients with LVI vs. 86% in patients without LVI (p<.0001). When patients were stratified based on nodal stage, these rates were 64% vs. 78% in pN0 patients, 31% vs. 41% in pN1 patients, and 69% vs. 87% in pNx patients (all P < 0.001). On multivariable analysis, and in comparison to patients without LVI, those with LVI had 1.37- (P < 0.001), 1.18- (P = 0.068), and 1.53-fold (P < 0.001) greater risk of death, when also harboring pN0, pN1, and pNx disease, respectively. CONCLUSIONS: Our findings are the first, to our best knowledge, to illustrate the clear detrimental impact of LVI on OS in surgically treated RCC patients. These findings might be useful in postoperative patient counseling and need to be accounted for when designing future clinical trials.