A new study from Sweden revealed that screening men between 50 and 65 years old with the PSA test for prostate cancer reduced deaths by nearly a half over 14 years and concluded that prostate cancer screening compares favorably with screening for other cancers, but also cautioned that the risk of overdiagnosis is substantial, causing an expert to point out that such findings should not be taken as a reason to introduce universal PSA testing.
You can read about the study, which was funded by the Swedish Cancer Society, the Swedish Research Council, and the National Cancer Institute, and led by Dr Jonas Hugosson from the University of Gothenburg, in the 1 July Early Online issue of The Lancet Oncology.
Hugosson and colleagues wrote in their background information that one way to decrease the risk of death from prostate cancer, a leading cause of death from malignant disease among men in the developed world, is to screen for PSA (prostate-specific antigen, a biomarker that can indicate the presence of cancer), but there is a lot of controversy about whether the benefits of screening outweigh the harms.
For the study, which began in December 1994, the researchers randomized 20,000 men aged from 50 to 65 to either a screening group or to a control group.
The men had been randomly selected from a population register, and those in the screening group were invited to have a PSA test every two years, until they reached an upper age limit of between 67 to 71 years (median 69), while those in the control group were not.
Men in the screening group with high PSA could opt to take further tests, including rectal exams and biopsies.
The researchers are still monitoring the groups, with men who have not reached the upper age limit still being offered a PSA test.
The results reported in The Lancet Oncology paper include cumulative prostate-cancer incidence and mortality data up to the end of December 2008, and show that:
- 76 per cent of the screening group had at least one PSA test.
- Over a median follow up of 14 years, the cumulative incidence of prostate cancer was 12.7 per cent in the screening group (1,138 men diagnosed), and 8.2 per cent in the control group (718 diagnosed).
- This resulted in a hazard ratio (HR) of 1.66, with 95 per cent Confidence Interval (CI) ranging from 1.50 to 1.80, with a high statistical significance value of p
The authors wrote that the study showed deaths were nearly halved over 14 years with screening, but that the “risk of over-diagnosis is substantial and the number needed to treat is at least as high as in breast-cancer screening programmes”.
They concluded that:
“The benefit of prostate-cancer screening compares favourably to other cancer screening programs.”
In a “reflection and reaction” article in the same issue of the journal, Dr David Neal, a professor in the Department of Oncology at University of Cambridge in England, comments on the findings by Hugosson and colleagues in the light of the overdiagnosis and overtreatment controversy surrounding PSA testing.
The PSA test remains controversial “because it detects some cancers that will never present during the patient’s lifetime”, and “results in unnecessary treatments that can damage men’s quality of life”, writes Neal.
He notes that although these points are valid, the study shows that in certain circumstances, PSA testing reduces deaths from prostate cancer, but added that it does not imply “PSA screening programs should now be introduced internationally”.
“Mortality results from the Göteborg randomised population-based prostate-cancer screening trial.”
Jonas Hugosson, Sigrid Carlsson, Gunnar Aus, Svante Bergdahl, Ali Khatami, Pär Lodding, Carl-Gustaf Pihl, Johan Stranne, Erik Holmberg, Hans Lilja.
DOI:10.1016/S1470-2045(10)70146-7
Written by: Catharine Paddock, PhD