Loading...
Prostate-specific antigen (PSA) screening involves a tradeoff between potential mortality benefit and harms of overdiagnosis. In the European Randomized Study of Screening for Prostate Cancer (ERSPC), PSA screening lowered prostate cancer–related mortality at 11 years, but many men underwent surgery or radiation therapy for every death averted. Thus, researchers concluded previously that more analysis was needed “before general [screening] recommendations can be made” (JW Gen Med Mar 14 2012). Now, they present results of simulation modeling in which they integrated (1) ERSPC mortality outcomes, (2) probabilities of adverse diagnosis-related and treatment-related complications (e.g., incontinence, impotence), and (3) estimates of decrements in quality of life resulting from these adverse outcomes.
The model predicted that screening 1000 men (age range, 55–69) annually would extend the lives of 9 men for an average of 8 years each, adding about 73 life-years overall. However, when adverse effects on quality of life (experienced by all screened men who undergo biopsy and treatment) were factored into the model, the benefit of screening dropped from 73 nonadjusted life-years gained to 56 quality-adjusted life-years (QALYs) gained per 1000 men.
The authors then varied their estimates of adverse effects. Under the most favorable estimates, screening would add 97 QALYs per 1000 men; under the least favorable estimates, screening would result in 21 QALYs lost (meaning that screening would be harmful, on average).
Heijnsdijk EAM et al. Quality-of-life effects of prostate-specific antigen screening. N Engl J Med 2012 Aug 16; 367:595. (http://dx.doi.org/10.1056/NEJMoa1201637)
Sox HC. Quality of life and guidelines for PSA screening. N Engl J Med 2012 Aug 16; 367:669. (http://dx.doi.org/10.1056/NEJMe1207165)
Comment
This report reminds us that, at the population level, the value of PSA screening depends substantially on how men weigh the downstream benefits and harms of screening. One puzzling aspect of this theoretical analysis is its conclusion that screening would prevent 9 prostate cancer–related deaths per 1000 men; this benefit is ninefold larger than what actually has been demonstrated thus far in the ERSPC (1 death prevented per 1000 men screened, after mean follow-up of 11 years). Because this model assumes annual screening starting at age 55 and lifetime follow-up (whereas the ERSPC screened only every 4 years and started screening at various ages), the model's larger mortality benefit is plausible. However, the huge difference between the estimated benefit and actual ERSPC results is surprising. The authors indirectly acknowledge this issue in their cautious conclusion: “Longer follow-up data from both the ERSPC and quality-of-life analyses are essential before universal recommendations regarding screening can be made.”