The study by Scosyrev et al1 that estimated that stopping screening with prostate-specific antigen (PSA) testing in patients with early-stage prostate cancer would triple the number of men presenting with metastases seems unimportant and appears only to describe the desirable and expected consequence of reducing the current very high rates of overdiagnosis and the overtreatment of patients with early-stage prostate cancer caused by lead-time bias.
The authors estimated that in 2010, 217,000 men in the United States would be diagnosed and treated for early-stage prostate cancer (at a median age of 67 years) and 32,000 would die of the disease (at a median age of 80 years-81 years). The only purpose for diagnosing this vast number of men with early-stage prostate cancer is if early treatment would alter the natural history of the disease and prevent death from prostate cancer in some of those men. Robust data from 3 recent prospectively randomized studies performed in the era of PSA screening2-4 have demonstrated that the early diagnosis and treatment of early-stage prostate cancer only prevents an insignificant number of deaths from the disease, or possibly none at all, and the treatment causes significant and long-lasting morbidities in many otherwise healthy men. None of the 3 studies demonstrated any significant difference in overall survival rates. Of particular note is that at 12 years of minimum follow-up, the landmark Prostate Cancer Intervention Versus Observation Trial (PIVOT) study reported that, of 354 deaths overall, only 16 occurred in the patients treated with radical surgery versus 18 definite deaths from prostate cancer in the observation group.4
It is expected and hoped that ceasing PSA screening will achieve 2 outcomes for many of these 217,000 men who would otherwise be diagnosed with prostate cancer. First, it will save up to 180,000 or more men who are unnecessarily diagnosed with non–life-threatening early-stage prostate cancer each year from all the morbidities of unnecessary treatment. Second, it will delay the diagnosis of many of the 32,000 men who are destined to develop metastases but are currently undergoing radical treatment for early-stage prostate cancer. This delay will be many years and these patients too will saved from the toxicities of the futile treatment of early-stage prostate cancer.
Therefore, if we reduce the vast overdiagnosis of early-stage prostate cancer, would not the authors' projected number of 25,000 men who would subsequently be diagnosed with metastases in 2008 be well within the expected range and would not it occur regardless of whether they had been previously diagnosed using PSA and would not the overall length of survival and the mortality rate for these patients be unchanged?