The headlines from April 1991 were filled with hope, even hubris, in the battle against prostate cancer -- a disease that was killing 30,000 American men every year.
''New Blood Test Found to Detect Cancer," said one. ''Blood Test for Prostate Cancer Hailed," said another.
The optimism was understandable: A high level of a protein called prostate-specific antigen, or PSA, was a clear indicator that a biopsy was needed. No other cancer, it seemed, could be so easily detected in its early stages -- and thus be treated before it was too late.
But the test's early promise soon faded. Men with supposedly low PSA scores were found to have cancer; others with high PSA scores underwent biopsies, only to find that they were OK. When Stanford University professor Thomas Stamey, one of the test's strongest proponents, declared last year that the ''PSA era is over," it seemed that medical researchers had falsely raised the public's hopes.
But don't dismiss the PSA test just yet. Most doctors aren't. Some are even betting that it can tell us more than it is now.
''If we discard PSA, we're throwing out some very useful information," said Dr. Anthony D'Amico, a radiation oncologist at Brigham and Women's Hospital.
Led in part by doctors like D'Amico, urologists and oncologists are moving away from the simple notion that there is a magic PSA number, above which a man should get a biopsy and below which he should rest assured. Doctors are now more focused on how a man's PSA changes over time -- particularly how fast it's rising.
That information could not only determine who needs a biopsy and who doesn't. It might indicate what kind of treatment they need, or whether they need to be treated at all.
''It provides you with a pretty good prediction of a man's risk of prostate cancer, and more interestingly, it does a better job of finding the bad prostate cancer," said Dr. Ian M. Thompson, the chairman of urology at the University of Texas Health Science Center in San Antonio.
There is still a long way to go before the PSA can fulfill those hopes. Doctors don't even have conclusive evidence that PSA screening lowers the death rate from prostate cancer, although the death rate for the disease among American men dropped 3.6 percent between 1992 and 2002 -- the years during which the test became commonplace. Doctors hope that a couple of large studies, one in Europe and one in the United States, will confirm the link in a couple of years.
The problem is not so much the test itself, as it is prostate cancer's peculiar nature: It's very common, and it grows very slowly.
Most men develop cancer in the prostate -- the walnut-shaped gland below the bladder that helps make seminal fluid -- if they live into their 70s or 80s. But the malignancy usually grows so slowly that many men harbor it unknowingly and wind up dying of something else.
So the PSA test has led to a quadrupling of men diagnosed with the disease. But it has also led to countless biopsies, surgeries, and radiation treatments that were never necessary.
''I think there are many, many more people out there with prostate cancer than need to be treated for it," said Dr. Philip Kantoff, director of the Lank Center for Genitourinary Oncology at the Dana-Farber Cancer Institute. ''And we're not very good yet at making that distinction."
For the first decade of the PSA test, doctors used a cutoff point of 4. When some men with lower scores were found to have prostate cancer, doctors began lowering the cutoff score, as low as 2.5. But casting a wider net has inevitably meant more men getting biopsies that turn out negative and more men being told they have extremely slow-growing prostate cancer -- what doctors call ''overdiagnosis."
''Then you tell a man, 'You've got prostate cancer, but don't worry about it,' " said D'Amico, the oncologist at Brigham and Women's. ''Yeah, right. That's not going to go over well with any man."
Instead, D'Amico is trying to strike the right balance between detection and overdiagnosis by looking at a PSA's change over time, or ''velocity."
Based on a study of 1,100 men, he concluded that a 2-point rise in a man's PSA in one year is ''an ominous change." Of the men who experienced such an increase and then underwent either surgery or radiation treatment, 20 percent died within seven years -- ''and that's a huge death rate," D'Amico said.
''That 2-point rise is telling you that not only does he have the prostate cancer, but it's already left the prostate," he said.
Such men, he said, will not only need surgery or radiation, which treat cancer in the prostate itself; they will also need chemotherapy or hormone therapy, which combat cancers that have traveled somewhere else in the body.
As for using the PSA test to determine if a man needs a biopsy, D'Amico said he's still waiting for the results of a study to see what level of change should trigger that procedure. Right now, men in their 50s or 60s who experience more than a half-point rise in a year get biopsies; D'Amico suspects that doctors should look for a larger increase, somewhere just under a point.
But D'Amico has already gone out on a limb by advising that all men get PSA tests at age 35 to establish a benchmark.
''I can't go that far, based on what I know," said Kantoff, his colleague at Dana-Farber.
The American Cancer Society currently recommends that screening begin at age 50, or at age 45 for African-Americans or men with a family history of the disease.
Besides deciding when men should start getting a PSA test, the medical establishment also must figure out how often. Although the American Urologic Association encourages an annual test for men over age 50, the American Cancer Society doesn't recommend ''routine" PSA tests for men who don't have specific risk factors, especially a family history of the disease. But then there are cases like David Hamilton.
A 59-year-old Newburyport man, Hamilton began getting the PSA test every year at age 50 at the urging of his wife, who happened to work for a surgeon. For years, his PSA level held steady at 1.5. Then, in October of last year, it registered a 5.3. A biopsy confirmed he had cancer.
The first urologist he saw recommended surgery but also mentioned other options and encouraged him to get a second opinion. Hamilton, a reservationist for a ground transportation company, quickly realized how much uncertainty still surrounds his disease.
''There's so much you can read . . . and they're still learning about it," he said. ''So you have to hang your hat somewhere, and I got to the point where I said, 'I can't fool around with this much longer, I really need to make a decision to go forward.' "
Under D'Amico's care, he wound up undergoing radiation as well as hormone treatments to starve the tumor of testosterone, which fuels the growth of prostate cancer. The radiation treatments ended this month and the hormone treatments continue for another few weeks.
Still, one case does not make for sound medical policy. Prostate cancer experts, even as they laud D'Amico's research into PSA velocity, remain reluctant to make broad recommendations for screening and biopsies until more results come in.
Thompson, at the University of Texas, says doctors will likely have to consider factors such as the patient's family history, ethnicity, and even his anxiety about the disease, before deciding how often he should be screened and what PSA number signals danger.
''That's the challenge," Thompson said. ''People don't like nuanced information."