SAN FRANCISCOOral sodium clodronate appears to have delayed
progression of bone metastasis from prostate cancer in a randomized clinical
trial, although the results did not reach statistical significance. British
investigator David Dearnaley, MD, presented the preliminary results on behalf
of the Medical Research Council (MRC) Clinical Trials Unit at the 37th Annual
Meeting of the American Society for Clinical Oncology (ASCO) in San Francisco.
Despite longer time to progression and better survival for
those patients who received clodronate, Dr. Dearnaley said the gains were not
strong enough to support the growing use of bisphosphonates in prostate cancer
patients. Instead, he urged that the findings be used to design new trials with
more powerful agents (see related article).
"We believe these results are interesting and intriguing,
but inconclusive, and should be a motivation for further larger trials to
confirm or reject these findings, perhaps using more potent third-generation
bisphosphonates and an IV route of administration," said Dr. Dearnaley,
the Bob Champion Senior Lecturer, Royal Marsden NHS Trust and Institute of
Cancer Research, Sutton, England.
The investigation focused on clodronate because bisphosphonates
have been shown to slow bone metastasis in breast cancer and myeloma, and to
diminish bone pain in prostate cancer patients.
From 1994 to 1998, the trial enrolled 311 patients from 33
centers, including one in New Zealand. All had proven bone metastasis on a bone
scan and were either starting or responding to initial hormonal therapy. Median
follow-up was 43.5 months.
The trial was divided into two arms. A group of 159 patients
received 2,080 mg of oral sodium clodronate in four tablets daily for 3 years
or until reaching one of the primary endpoints of the study: symptomatic bone
progression or death. A balanced cohort of 156 men received four placebo pills
daily with the same guidelines.
As of early April, 220 of the 311 participants had reached a
primary study endpoint. The main reason for stopping treatment was symptomatic
bone progression, which occurred in 186 patients: 52% of the placebo arm and
43% of the clodronate arm. Actuarial curves show a difference of 4.3 months in
the median time to progression: 19.3 months for the placebo arm vs 23.6 months
for the clodronate arm.
Survival outcomes were also better for the patients on
clodronate. Their median survival was 36.8 months, compared with 28.4 months
for the placebo group. At 2 years, 69% of the clodronate group was still alive
vs 63% of the placebo group. Although 103 patients died in the control group10
more than in the clodronate groupDr. Dearnaley said the causes of death were
He noted that 15% of patients stayed on the trial medication
for 3 years. At 2 years, 49% of the clodronate group was event free, compared
with 41% of the control group.
Despite the data favoring the clodronate arm, Dr. Dearnaley
warned against a definitive interpretation of the results. "We’ve shown
an estimated 2-year reduction in symptomatic bone progression of prostate
cancer of 8.5%. The trial was designed to detect an 11% difference at 2 years,
so our result is not statistically significant," he said.
Gastrointestinal problems caused 9% of patients in the
clodronate group, but only 1% of patients in the placebo arm, to leave the
trial. About one third of patients in the clodronate group had their dose
modified, usually reduced rather than stopped.
Other adverse events were described as asymptomatic. These
included raised LDH levels and hypocalcemia in five patients. "I have to
say none of these side effects was what I would call serious," Dr.
One potentially important observation was that the time from
when a patient was diagnosed to when he entered the trial seemed to matter.
Patients who started taking clodronate earlier in their treatment had better
"It may be that we were already beginning to see patients
failing at the time we started the bisphosphonate," Dr. Dearnaley said.
"There is a strong indication that we ought to come in with
bisphosphonates on day 1 of hormone treatment, if we want to push this
hypothesis as far as it can go. I think this is important for any further
Another provocative finding was a relative reduction of
prostate-specific antigen (PSA) in the clodronate group. A 20-month analysis of
the data showed a median PSA level of 3.7 ng/mL in the clodronate group vs 6.8
ng/mL in the placebo arm. "There’s a suggestion of biological activity
based on the follow-up PSA level," he said, "and a suggestion of
greater activity if patients are started on the drug when newly
In response to a question, he agreed that, based on his data,
there is no reason to use bisphosphonates outside of a clinical trial in
patients with prostate cancer. "I think that’s very fair," he said.
". . . [even for pain] the data for using bisphosphonates in prostate
cancer are very weak."