PD03-06: Grade Migration of Prostate Cancer at Diagnosis Following the US Preventive Services Task Force Recommendations on Prostate Cancer Screening

Leonardo Borregales*, Gina DeMeo, Emily Cheng, Xiagnmei Gu, New York, NY, Jonathan Shoag, Cleveland, OH, Sigrid Carlsson, Andrew Vickers, Jim Hu, New York, NY

Introduction: Guidelines regarding prostate cancer (PCa) screening have changed rapidly over the last decade with the aim of reducing overdiagnosis and overtreatment. To examine these changes, we analyzed population-based tumor registry data to study temporal trends in PCa epidemiology.

Methods: Our primary objective was to examine temporal changes in PCa Gleason Grade Group (GGG) at diagnosis over the last decade using Surveillance, Epidemiology, and End Results (SEER) data comprising 438,432 men diagnosed during 2010 to 2018. Age-adjusted PCa incidence rates (per 100,000 men) were calculated and standardized to the 2000 US population. Temporal trends in incidence by GGG, Prostate-Specific Antigen (PSA) levels and PCa metastasis at diagnosis were examined, relative to annual incident cancers. The National Health Interview Survey was queried to evaluate trends of PCa screening. We then explored various hypotheses to explain the observed trends.

Results: During the study period, there was a decline in the incidence of GGG1 as a proportion of all PCa from 47% in 2010 to 32% in 2018(p < 0.001). There was also a decline in the proportion of GGG1 at radical prostatectomy surgical pathology from 31.5% in 2010 to 9.9% 2018(p < 0.001). Overall, the age-adjusted rate of GGG1 disease dropped from 52 to 26 cases per 100,000 over the same interim. Conversely, there was an increase in the incidence of GGG 2-5, particularly from 2014 to 2018. Notably, the incidence of GGG3 increased as a proportion of all PCa from 10.7% in 2014 to 13.5% in 2018, GGG4 from 9.6% in 2014 to 10.8% 2018, and GGG5 increased from 9.3% in 2010 to 11.0% in 2018. Additionally, the median PSA at diagnosis rose from 6.2 to 7.1 ng/ml(p <0.001). We confirmed a previously reported increase in the incidence of distant metastases at diagnosis, from 3.0% to 5.2%(p < 0.001).

Conclusions: We found a significant decline of GGG1 PCa at diagnosis with an upward grade migration. As there have been several approaches in recent years to enhance high-grade PCa detection rates and lower the detection of low-risk disease, there are several hypotheses that may explain the phenomenon highlighted here. Changes in screening practices are the best primary explanation for the grade shift observed, as there was relatively low uptake of pre-biopsy MRI during our study period and there was insufficient data to support a significant effect of biomarkers or increased incidence of risk factors of high-grade PCa such as obesity. Further research is needed to examine the downstream effects of these changes in PCa specific mortality.

Source of Funding: None