THE BENEFIT-HARM BALANCE OF PROSTATE CANCER SCREENING IN MEN WITH AVERAGE AND ELEVATED FAMILIAL RISK – PREDICTIONS OF THE ONCOTYROL PROSTATE CANCER OUTCOME AND POLICY MODEL

Purpose:

The benefit of prostate cancer (PCa) screening is still controversial due to potential harms by overdiagnosis and overtreatment. We applied the ONCOTYROL Prostate Cancer Outcome and Policy (PCOP) model to evaluate the benefit-harm balance of PCa screening followed by immediate treatment or active surveillance in men with average and elevated familial PCa risk, considering individual age and quality of life (QoL) weighting.

Method(s):

The PCOP model is a decision-analytic state-transition micro-simulation model simulating the natural history of PCa and the consequences of screening and treatment on duration and quality of life. Men with average and elevated familial PCa risk were simulated as separate cohorts, differing in familial risk parameters, which in the base-case analysis were assumed to affect both PCa onset and progression. Evaluated strategies included no screening, and various one-time and interval screening algorithms. Optimal screening strategies maximizing quality-adjusted life expectancy (QALE) were identified depending on age and individual QoL weighting (i.e. disutilities). Additionally, all screening strategies were evaluated in combination with biennial active surveillance biopsies delaying treatment of localized cancer until progression to Gleason score ≥ 7.

Result(s):

In men with average PCa risk, screening reduced QALE even under favorable assumptions. In men with elevated familial risk, screening gains QALE depending on age and disutilities. For men with familial risk aged 55 and 60 years annual screening to age 69 was the optimal strategy over most disutility ranges, whereas for 65 year-old men with average and above disutilities, no screening was the preferred option. Active surveillance strongly reduced overtreatment. However, gains by averted adverse events were opposed by losses due to delayed treatment and additional biopsies, and the reduction of overtreatment decreases with increasing speed of PCa progression. Compared to screening with immediate treatment, screening with active surveillance resulted in lower QALE losses in men with average PCa risk, and lower QALE gains in men with elevated PCa risk.

Conclusion(s):

Assumptions about PCa risk and prevalence strongly affect the benefit-harm balance of screening. Based on our assumptions, PCa screening is only beneficial for men with familial predisposition, in whom QALE gains depend on individual age and disutilities. Active surveillance should not delay treatment until Gleason score progression to 7. Alternative criteria for treatment initiation should be evaluated in further modeling studies.