Final Keynote Address Highlights the Uncertainties Surrounding Prostate Cancer Screening and Treatment
In this "PSA era" that is, a time when PSA (prostate-specific antigen) blood tests are detecting prostate cancer at a much earlier stage than in the past there is a great deal of uncertainty about what the best treatments are and whether finding and treating the cancer early makes a difference in survival. Dr. Michael Barry, the first primary care physician ever asked to speak at ASTRO's annual meeting, reviewed what evidence we do have during his Wednesday morning keynote address. As Chief of the General Medicine Unit at Massachusetts General Hospital in Boston, Barry is actively involved in prostate cancer treatment and research.
His hour-long address, "Beliefs and Evidence in the Outcomes of Treatment and Screening in Prostate Cancer," led to the all-too-common conclusion that more evidence from clinical trials is needed before we can make definitive statements about prostate cancer screening and treatment. In the meantime, though, it may be useful to note how radiation oncologists and urologists come at the problem from different points of view. After devoting the first part of his talk to this topic, Dr. Barry moved on to what he called the "64,000 dollar question:" "Does it all work? Are population-based data for mortality confirming that early detection and treatment make a difference?" So far, the answer seems to be a qualified "maybe," although several large-scale trials under way should provide better answers in the coming decade.
Radiation Oncologists "Versus" Urologists
In the absence of hard evidence, Barry noted, radiation oncologists and urologists overwhelmingly tend to recommend the therapy that they themselves deliver. Barry and a group of researchers reached this conclusion through a recent mail survey of physicians in both specialties (the results were published in the June 28, 2000 issue of the Journal of the American Medical Association). Dr. Barry first noted the areas of agreement that emerged from the survey:
- the use of PSA testing as a routine part of primary care: Most radiation oncologists and urologists agreed that men ages 50-74 should be tested routinely. Only about _ of each group thought men ages 40-49 should. (Disagreement emerged regarding men ages 75 and over. Urologists were less likely to recommend routine PSA testing for this group than radiation oncologists were.)
- treatment survival benefit for older men: Only a minority of specialists in both groups thought that treatment had a survival benefit for men with localized, moderately differentiated prostate cancer who otherwise had a life expectancy of less than ten years (i.e. in the 75 and over age group).
- risk of side effects associated with treatments: Both groups of specialists tended to share the opinion that side effects such as impotence and incontinence were most likely to happen with non-nerve-sparing prostatectomy, unilateral nerve-sparing prostatectomy, and bilateral nerve-sparing prostatectomy, followed by external radiation therapy and brachytherapy (use of radioactive implants). And this was the information they shared with patients.
- use of androgen deprivation (hormonal therapy) as a primary treatment: What Barry called a "substantial minority" in both groups saw this as a favorable option.
In spite of these areas of agreement, however, one key disagreement emerged, just as researchers expected it would. Urologists favored radical prostatectomy as the optimal treatment for moderately differentiated, localized prostate cancer, while radiation oncologists favored radiation therapy. For example, 93 percent of urologists chose radical prostatectomy as the preferred treatment option, while 72 percent of radiation oncologists believed that surgery and external beam radiotherapy were equivalent treatments. Furthermore, while 84 percent of radiation oncologists said that radical prostatectomy is overused as a primary treatment for prostate cancer, only 32 percent of urologists thought so. For most tumor grades and prostate-specific antigen levels, both specialty groups were significantly more likely to recommend the treatment in their specialty than the other treatment.
Is it simply that a specialist is biased toward the treatment with which he or she is most comfortable? Not necessarily, said Barry. "We don't know who is right," he said, "but there are fundamental differences in how specialists view the effectiveness of these therapies in relation to side effects." He noted that urologists thought radical prostatectomy was "still worth it," despite the perceived higher risk of side effects. Radiation oncologists didn't, and therefore were more likely to favor radiation therapy.
Even though these two groups of specialists diverged in their thinking about treatment, Barry noted that they seemed to be painting accurate pictures for patients when estimating the risk of side effects. Using data from the National Cancer Institute's Prostate Cancer Outcomes Study, which is looking at quality-of-life issues subsequent to treatment, Barry pointed out that these patients' actual experiences with side effects tended to match up with urologists' and radiation oncologists' predictions in his survey.
However, he was also quick to point out that side effects can vary according to where a procedure is done. With radical prostatectomy, Barry noted, side effects can range widely due to variations in technique and a "volume-outcome relationship" -- surgeons who perform more procedures are likely to produce better results. He cited a study by Dr. Walsh of Johns Hopkins University, which reported that the rates for recovery of urinary continence and sexual function were much better for patients treated by an experienced surgeon at a high-volume center (Urology 2000 Jan; 55(1): 58-61).
Unfortunately, there is not enough information available to say which treatment is best for the man with localized, moderately differentiated prostate cancer (meaning that he does not have the most aggressive form of the disease, as suggested by cell appearance). "There are few randomized clinical trials available yet for localized prostate cancer," Dr. Barry noted, but they are the best way to get more definite answers. At the same time, thanks to PSA testing, more and more men are being diagnosed with localized, early-stage prostate cancer. In the absence of hard data, those men will have to work with their physicians to make the most educated choice they can.
Does Early Treatment Make a Real Difference?
There's also the pressing question of whether the push toward finding and treating prostate cancer early in this "PSA era" really makes a difference in survival. Two large-scale trials -- the NCI's Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial and the European Randomized Study of Screening for Prostate Cancer -- likely will provide answers about the costs versus benefits of screening in the next several years. Dr. Barry noted that the dramatic spike in prostate cancer rates seen in the early 1990s was a direct result of the availability of PSA testing. "We doubled incidence simply by looking for it earlier," he said. Mortality rates remained approximately the same between 1987 and 1997, he added.
He noted that one of the biggest difficulties researchers face in studying PSA-era outcomes is allowing sufficient time for follow-up. Since men are being diagnosed so much earlier that they were in the pre-PSA era, they must be followed longer after treatment, too -- particularly if we want to make accurate comparisons between pre-PSA and PSA-era data. "There are big differences in lead time [before diagnosis]," Barry said, "and we have to take these into account." The biggest problem with some of the data that is available, he said, is "insufficient follow-up." Researchers simply have not been able to follow PSA-era men long enough to make accurate judgments.
Barry is now collaborating with other researchers on a comparative study of men in Seattle versus those in Connecticut. He explained that Seattle adopted early detection and aggressive treatment for prostate cancer early in the PSA era (1987-90), while Connecticut did not. The study has been following a group of men since 1987. So far, researchers have found that Seattle men were twice as likely to have a prostate biopsy, twice as likely to be diagnosed with prostate cancer, and six times as likely to have radical prostatectomy. They were also getting more radiation therapy. So far, however, researchers have not seen an earlier impact on mortality rates in Seattle versus Connecticut. "It may be that we haven't followed the men long enough," Barry noted.
These results are very different from those published by Barsch et al. in a recent issue of the Journal of Urology. The researchers looked at 65,000 men ages 45 to 75 in Tyrol, a state in western Austria that adopted widespread PSA screening in 1993. In fact, two-thirds of the men had been tested by 1996. The researchers reported that population-based prostate cancer mortality rates had dropped 32 percent in 1997 and 42 percent in 1998 versus the rates for 1986-1990. They also found that the rates dropped faster than in the rest of Austria. "But it seems awfully early to see such benefit," Barry noted.
These sharply contrasting results helped Barry demonstrate his point that more research and more follow-up are needed before we will know whether finding prostate cancer early and treating it actually impacts mortality.
"The chances of being told you have prostate cancer are definitely higher," Barry said. "I would like to see the day when you can be definitively told that your likelihood of surviving is higher as well."