BETHESDA, MdRapid advances in managing prostate
cancer over the last decade have created dilemmas for clinicians as they
attempt to determine which treatment is best for each patient, said Patrick
Walsh, MD, director, Department of Urology, Johns Hopkins University School of
Medicine. Such dilemmas will not be resolved fast enough through randomized
trials, he said.
"We’re facing not a light at the end of a tunnel, but a
tunnel at the end of the light," Dr. Walsh quipped during his presentation
at a conference sponsored by the National Cancer Institute and the Society of
The use of prostate-specific antigen (PSA) provides not only
earlier diagnosis at curable stages but also objective criteria for earlier
identification of treatment failure, permitting multiple points for
intervention. "There are multiple options for treatment and multiple
definitions of success," he said, "too many for anyone to say which
is best, and too many to evaluate now in randomized trials."
The dilemma is not just about choosing the best treatment to
cure the disease; physicians must also determine if the patient’s disease is,
in fact, curable. Is the patient well enough to be cured? Will the treatment be
effective? Will the potential added years of life be quality years?
Even increased awareness of the disease and treatment options
can create problems, he noted. When a well-known public figure is diagnosed
with the disease, people may assume that the treatment the famous individual
receives is the best treatment for every patient.
Dr. Walsh called prostate cancer "a moving target,"
with treatment options and outcomes changing over the years. Data from 1982 to
1999, covering 2,400 men, show that prior to the advent of PSA testing (1982 to
1989), only 30% of men undergoing a radical prostatectomy at Johns Hopkins had
organ-confined disease. Today, that figure is 80%.
Furthermore, Dr. Walsh said, those diagnosed more recently have
higher actuarial survival. In men who underwent surgery from 1982 to 1988, at
10 years, 66% had an undetectable PSA; from 1989 to 1999, 80% had a PSA of less
than 0.2 ng/mL at 10 years.
This change over time has created problems in comparing
therapies, he said. Interpretation of study results has to be adjusted not only
for disease stage but also for the era of diagnosis. Further, he said, it is
hard to randomize patients between aggressive treatments like radical
prostatectomy and brachytherapy. How, then, can valid comparisons be made among
treatment modalities? he asked.
Usually, such evaluation calls for randomized controlled
trials, Dr. Walsh said. However, conducting such trials in prostate cancer will
require far too many treatment arms and will take too much time before results
are available. "Randomized controlled trials are important, and eventually
should be done, once we know what should be compared, but a nonrandomized
comparison will provide the quickest answer," he said.
Dr. Walsh called for a "blue-ribbon registry" to
enroll patients and record data. Those involved would have to agree on the
selection of patients; definition of cancer control ("whether undetectable
PSA or a certain nadir, or ASTRO or other criteria"); and instruments for
evaluating quality of life, both pre- and postoperatively. "We should
enlist all centers of excellence and entrust the registry to collate and
compare data," he said, adding that a multidisciplinary team would do the
final evaluation of treatment alternatives.