Personalized Risks of Over Diagnosis for Screen Detected Prostate Cancer Incorporating Patient Comorbidities: Estimation and Communication.

Abstract

Shared patient-physician decision making regarding the treatment of prostate cancer detected by prostate specific antigen screening involves a complex calculus weighing cancer risk and patient life expectancy. We sought to quantify these competing risks using the probability that the cancer was over diagnosed, ie would not have been clinically diagnosed (diagnosed without screening) during the remaining lifetime of the patient. Using an established model of prostate cancer screening and clinical diagnosis we simulated screen detected cases and determined whether a modeled clinical diagnosis would occur before noncancer death. Time of noncancer death was based on comorbidity adjusted population lifetables. Logistic regression models were fitted to the simulated data and used to estimate over diagnosis probabilities given patient age, prostate specific antigen level, Gleason sum and comorbidity category. An online calculator was developed to communicate over diagnosis estimates. Face validity and ease of use were assessed by surveying 32 clinical experts. Estimated probabilities of over diagnosis ranged from 4% to 78% across clinicopathological variables and comorbidity status. When ignoring comorbidity, the estimated probability of over diagnosis in a 70-year-old man with prostate specific antigen 9.4 ng/ml and Gleason 6 was 34%. With severe comorbidities the estimate increased to 51%. Such a personalization may help inform the choice between active surveillance and definitive treatment. Based on responses from 20 of 32 experts we modified the explanation of over diagnosis for the online calculator and the input method for comorbid conditions. The probability of over diagnosis is strongly influenced by comorbidity status in addition to age. Personalized estimates incorporating comorbidity may contribute to shared decision making between patients and providers regarding personalized treatment selection.