PSA screening for prostate cancer has recently been the focus of scrutiny in the media, especially following a recent publication in the British Medical Journal. It is thus a good time to take a step back and examine what we know about PSA screening and its relationship to prostate cancer.
First, one might wonder why we screen healthy people for anything. Screening is very common in today’s society, as it helps us identify such problems as diabetes, scoliosis, colon cancer (colonoscopy) and breast cancer (mammography). To recommend screening a large population for a medical condition, several criteria need to be met. The condition needs to be common, the screening test must have minimal risks, and a useful treatment must exist so that detecting the condition earlier improves outcomes
.Does prostate cancer qualify as a condition suitable for screening?
Prostate cancer is definitely a significant health concern: it is the most common cancer in men in the U.S., diagnosed in about 200,000 men every year. It also is the second leading cause of cancer death, killing over 30,000 men each year. The PSA test is a simple blood test that only requires a few drops of blood, so the risks are essentially negligible. Surgery for prostate cancer has been clearly demonstrated to reduce deaths from prostate cancer, and it is likely that radiation does as well. However, prostate cancer can only be cured if caught at an early stage. Since PSA testing identifies prostate cancer far earlier than it would otherwise be found, it appears to be an ideal test for screening.
If prostate cancer warrants screening and PSA is a good test, what is all the controversy about?
The “problem” with PSA is that it is too good of a test. Prostate cancer is a very slow growing disease, often taking 10 to 15 years to cause significant health problems to occur. Since prostate cancer is more common in older men, many men who are diagnosed with prostate cancer (especially elderly men) may never have had any problems because of it. Such men do not need treatment, and the costs and side effects of such “overtreatment” are unnecessary. It is difficult to put a precise number on it, but it is widely agreed that a substantial proportion of men are overtreated for prostate cancer in the U.S.
Unfortunately, it is very difficult to know which men need treatment and which do not. Tests that we use to determine how advanced the cancer is (prostate biopsy results, PSA, rectal exam findings, and sometimes CT or bone scans) are often inaccurate. Looking at large populations, we can see that some men are being overtreated, but it is extremely difficult to predict which individual patients are being overtreated. We do know that the vast majority of men treated for prostate cancer, over 90 percent, are cured of the disease. Given the uncertainty in not treating and the near certainty of treating prostate cancer, most men opt for treatment.
What is the evidence to support or reject PSA screening?
As doctors, we base decisions on medical research – so called "evidenced-based medicine." Because of the slow growth of prostate cancer, it requires studying many patients for many years to see important differences. As indirect evidence, PSA screening became widespread in the U.S. in the early 1990s. Since that time, there has been a reduction in the rate of death from prostate cancer by about 4 percent per year, reversing what had been an increase in death rate. Most experts attribute this decrease to the use of PSA screening and early detection. More direct studies have been conducted, the results of three large ones were released last year, one from the U.S. and two from Europe. The results of these studies were different, causing a great degree of confusion. The two European studies showed that there was a benefit to screening, while the American trial did not show a benefit.
When research differs like this, we compare the details of the conflicting studies, looking more carefully at how they were designed. We look at such factors as the quality of the study and its size, as larger, better designed studies are more useful. Another thing we look at is the length of the study, as studies conducted over a longer time frame are generally more accurate.
What are the differences in the PSA screening articles?
Much of the difference comes from details within the composition of the studies. In the American trial, many men who were supposed to get PSA tests did not, and many in the “control” (no PSA) group did get one. These rates were better in the European trial and extremely good in the Goteborg (Sweden) trial, which should make the results more accurate. Length of follow-up was different, as well. The American and European trials each reported their results at 9 years, but the European trial was also analyzed at 12 years. The Swedish trial reported results at 14 years. As the length of follow-up increased, the benefit associated with screening increased. At 14 years, the Goteborg study showed that PSA screening cut the risk of dying from prostate cancer in half.
Which brings us to the recent article in BMJ. In this study, researchers began screening for prostate cancer in a randomly selected sample of men from a small town in Sweden in 1987. They then compared the rates of development of and death from prostate cancer in the men who were and were not screened. However, the men were only screened with rectal examination (which is much less accurate than PSA) until 1993, when they started including PSA in their screening. By this time, only 900 men were left in the screening group, who were followed for an average of less than 8 years. (Keep in mind that the European study screened 60,000 men and followed them for 12 years and the Goteborg study screened 7600 and followed for 14.) Of the men who were found to have prostate cancer, only half of them underwent treatment for it. Thus it is not surprising that there was minimal benefit to patients who underwent screening. However, when adjusting for age at the start of the trial, the trial did demonstrate a benefit for screening.
What does this mean?
Given the results of the better, larger trials, this new study does not change our approach to prostate cancer screening whatsoever. The American Urological Association and other organizations recommend screening with a PSA blood test and rectal examination starting at age 40, and others recommend stopping screening at age 75. PSA screening provides useful information to patients and can be followed over time to assess trends. An elevated PSA does not mean that a man has prostate cancer, and even having prostate cancer does not necessarily mean that a man needs surgery or radiation. As the large studies have longer follow-up, we will likely see more of a benefit for PSA screening.