Study methods influence estimates of lead time and overdiagnosis in prostate cancer

Mar 11 2009

In prostate cancer, estimates of lead time (how long screening advances diagnosis of cancer) and overdiagnosis (the fraction of cancers detected by screening that would not have been diagnosed during the patient's lifetime without screening) vary widely, depending on the definition of lead time used, the population, and the estimation methodology, according to a report in the March 10 online issue of the Journal of the National Cancer Institute.

Previous studies have aimed to calculate the mean lead time due to PSA screening, but the results have varied greatly from 3 to 12 years. Similarly, past estimates of the proportion of PSA-diagnosed patients who are overdiagnosed have varied substantially from 25 percent to 80 percent.

In the current study, Gerrit Draisma, Ph.D., of Erasmus MC in Rotterdam, The Netherlands, and colleagues used three different models to more accurately estimate lead time and overdiagnosis of prostate cancer among U.S. men aged 50 to 84 years in 1985 to 2000. They also tested three different definitions of lead time to determine whether the definition itself made a difference.

Among the three models, the estimated mean lead time ranged from 5.4 to 6.9 years, and the proportion of patients overdiagnosed ranged between 23 percent and 42 percent of PSA-detected cancers.

The researchers found that the estimates of lead time were relatively consistent between the three models but varied depending on which definition was used. Additionally, the context or population's clinical practice (for example what PSA score triggers a biopsy) affected the estimates of overdiagnosis derived from a given model.

"This article is the first, to our knowl¬edge, to closely examine the reasons for discrepancies across stud¬ies. Our results clearly show that the context or population used to derive the estimates, the definition of lead time used, and the esti¬mation methodology all have important roles," the authors write. "We feel strongly that for future studies to be correctly interpreted, analysts should specify the definition used in their publications."

In an accompanying editorial, Michael Barry, M.D., and Albert Mulley, Jr., M.D., of Massachusetts General Hospital in Boston, note that the level of overdiagnosis and lead time for prostate cancer screening are higher than for other types of cancer screening, including breast and colon cancers. In fact, they argue, PSA screening may not be beneficial at the population level when the costs of overdiagnosis, such as the added burden of so many more men having to face a prostate cancer diagnosis and the side effects of unnecessary treatment, are weighed against the number of prostate cancer deaths that might be prevented.

Additionally the long lead times estimated in the current study and others will make it challenging to evaluate the risk-benefit equation for PSA screening even when more clinical trial data become available, according to the editorialists. "In all likelihood, at the individual level, a shared decision-making approach to prostate cancer screening, including discussion of the trade-offs between benefits and harms, may still be the optimal strategy," they conclude.