PALM SPRINGS, Calif--Three major randomized trials comparing the antiestrogens
toremifene (Fareston) and tamoxifen (Nolvadex) in patients with advanced
breast cancer showed no significant differences in efficacy or side effects
profiles, Richard Gams, MD, of Ohio State University, said at the symposium.
These pivotal trials, which have served as the basis for the approval
of toremifene in countries around the world, show that the agent at a dose
of 60 mg/day is safe and effective for the treatment of advanced breast
cancer in postmenopausal women, Dr. Gams said. "Of course," he
added, "this begs the question of its use in the adjuvant setting,
which is the major use of antiestrogens in breast cancer."
In these studies, equivalence of the two antiestrogens was determined
by the Fleming method, developed by an FDA statistician. By this method,
two treatments are considered to be equivalent if the lower boundary of
the 95% confidence interval measuring the difference in response rates
is 10% or higher.
The studies consisted of a North American trial, which also included
patients from the United Kingdom, Australia, South Africa, and Mexico;
a study from the former Soviet Union (now called the Eastern European trial),
whose results were actually audited by the FDA; and the Nordic study, which
included patients from Finland, Sweden, Norway, and some Eastern European
In all three trials, patients were ER/PR positive or unknown, postmenopausal
(the North American trial also included perimenopausal patients), had good
performance status, and had not received prior antiestrogens for advanced
Patients were randomized to receive standard-dose toremifene (60 mg/day),
high-dose toremifene (200 or 240 mg/day), or a standard tamoxifen dose
(20 or 40 mg/day). The Nordic trial did not include the high-dose toremifene
arm. The two primary outcome variables were response rate and time to progression.
'Strikingly Similar Results'
"The most striking thing about the results is that they are similar
across all three trials," Dr. Gams said. The results of the North
American and Eastern European trials are "virtually identical,"
he said. In these two trials, the objective response rate (CR plus PR)
was about 20% for both toremifene, 60 mg/day, and tamoxifen. These response
rates were lower than anticipated from the literature, although the Nordic
study did show the anticipated 30% to 40% response rate.
Dr. Gams provided two possible explanations for the lower than expected
response rates: First, the studies were very rigorous in their definition
of response. "We insisted that all disease be measured and included
in the assessment, even it if meant repeat imaging," he said.
Second was the inclusion of a fairly substantial number of patients
who had bone-only disease, more so than seen in the typical oncology practice.
"Bone disease patients, while they may improve symptomatically, often
have a much lower objective response rate than patients with soft tissue
disease," he said. The investigators chose not to exclude bone-only
patients because of the difficulty in recruiting antiestrogen-naïve
Although the researchers looked for baseline differences in the patient
populations in the three studies, they could not explain why the Nordic
study showed higher response rates. "I would simply accept the fact
that when you do the same trial over and over, on a statistical basis it's
is highly unlikely that you will come up with exactly the same result each
Results in the high-dose toremifene arm showed increased side effects
and no significant improvement in response rate, Dr. Gams said, "so
we didn't feel that this was worth pursuing."
The weight of the evidence from all three trials, he said, is that tamoxifen
and toremifene provide very similar outcomes, "so we have to look
to the side effects profile or some other reason to make a choice of one
versus the other."
He pointed out that there were no differences between the two drugs
in quality of life or improvement in pain. "It is gratifying that
as many as 40% of patients had a reduction in their need for narcotic analgesics,"
In these studies, there were only two cases of endometrial carcinoma,
both of which occurred in women who had received tamoxifen. The investigators
did not routinely perform endometrial biopsies or otherwise specifically
look for endometrial cancers in these patients.
"Endometrial carcinomas occur fairly infrequently and would not
be the primary reason for making a choice between agents in patients with
advanced breast cancer," he said.