BETHESDA, MdInterim analysis of the National Surgical
Adjuvant Breast and Bowel Project (NSABP) B-28 trial has shown no additional
benefit to adding four cycles of paclitaxel (Taxol) after the standard
doxorubicin (Adria-mycin)/cyclophosphamide (AC) regimen in the adjuvant
treatment of node-positive breast cancer.
The data were presented at the NIH Conference on Adjuvant
Therapy for Breast Cancer, which also included a debate over the use of
adjuvant taxanes in patients with invasive breast cancer.
B-28 closed in 1998 after accruing 3,060 patients. Interim
analysis was performed last summer after 269 deaths and 561 recurrences, said
Eleftherios P. Mamounas, MD, medical director of the Cancer Center at Aultman
Hospital, Canton, Ohio,
Patients with node-positive primary breast cancer were
stratified according to number of positive nodes, tamoxifen (Nolvadex)
treatment, and type of surgery, and were randomized to two treatment groups:
four cycles of AC or four cycles of AC followed by four cycles of paclitaxel.
AC was given in the standard dose: Adriamycin 60 mg/m2 plus
cyclophosphamide 600 mg/m2. Paclitaxel 225
mg/m2 was administered as a 3-hour infusion every 21 days for four cycles.
Tamoxifen 20 mg was given for 5 years concurrent with
chemotherapy for patients age 50 and over, and for patients under age 50 who
were positive for estrogen receptors (ER) or progesterone receptors (PR).
The intent-to-treat analysis included 1,525 eligible patients
in the AC arm and 1,520 in the AC/paclitaxel arm. Median follow-up was almost 3
Tumor size was 2 cm or smaller in 60% of patients and more than
4 cm in about 9%. Positive nodes numbered up to three in 70% of patients, four
to nine in 26%, and 10 or more in 4%. About two thirds of patients were ER or
PR positive, and 84% received tamoxifen.
More than 98% of patients in both treatment arms completed all
four cycles of AC; 75% also completed the four additional cycles of
chemotherapy with paclitaxel; 90% received at least one cycle of paclitaxel.
Overall toxicity was acceptable for the adjuvant setting, with 27% of AC
patients and 36% of AC/paclitaxel patients developing some toxicity of grade 3
or greater. With paclitaxel, the rate of neurotoxicity grade 3 or higher was
Interim analysis revealed no significant differences in the
treatment arms in terms of disease-free survival and overall survival. There
were 282 recurrences with AC and 269 with AC/paclitaxel, for a relative risk of
.93. There were 133 deaths in the AC group and 136 in the AC/pac-litaxel group,
for a relative risk of 1.0, Dr. Mamounas reported.
Adjusted for number of positive nodes and tamoxifen use,
disease-free survival at 36 months was 81% for both treatment groups and
overall survival was 92% for the AC arm and 90% for AC plus paclitaxel, he
Sites of treatment failure were also similar between the arms,
with no differ-ences in the rates of ipsilateral tumor recurrence, other local
recurrences, regional recurrences, or visceral recurrences. Likewise, there
were no differences in contralateral breast cancers and other second primary
In patients not receiving tamoxifen, a trend emerged showing
benefit with the addition of paclitaxel. In this subgroup there was a 14%
reduction in the odds of recurrence and a 25% reduction in the odds of death.
Specifically, there were 80 events for AC and 73 for AC/paclitaxel, and 52
deaths vs 39, respectively.
This finding was not statistically significant, although the
number of patients in this group (237 in each arm) was small, he said. "We
formally tested for interactions involving tam-oxifen and found none that were
statistically significant for disease-free or overall survival," he said.
Dr. Mamounas concluded that the results of this interim
analysis of NSABP
B-28 do not, as of yet, show benefit for the addition of paclitaxel to the
standard AC regimen. There is no statistically significant interaction between
treatment effects and tamoxifen use, though there is a trend toward more
benefit in patients not receiving tamoxifen, he said.
Commentary: Opinions Differ
The main outcome in B-28 differs from that of the earlier CALGB
9344 trial, which showed a survival benefit with the addition of four cycles of
paclitaxel after four cycles of AC.
Now at 52 months follow-up, the Cancer and Leukemia Group B
(CALGB) trial continues to show benefit after 901 recurrences and 589 deaths,
giving it "considerably more statistical power," said I. Craig
Henderson, MD, adjunct professor of medicine, University of California, San
Francisco. Dr. Henderson was principal investigator of CALGB 9344.
Dr. Henderson took the "pro" side in a debate of the
adjuvant use of taxanes held at the consensus conference. Martine J. Piccart,
MD, PhD, head of the Chemotherapy Department, Jules Bordet Institute, Brussels,
Belgium, took the "con" position.
Dr. Henderson noted a few important differences between CALGB
9344 and the current B-28 trial; for instance, the number of patients with one
to three positive nodes was 70% in NSABP but only 46% in CALGB, and 51% of
NSABP patients were over age 50 vs 60% of CALGB patients. But when the effects
of paclitaxel were analyzed according to whether patients received tamoxifen,
"the trials are strikingly similar," he said. "There has been a
fairly consistent pattern of less effect in patients treated with
Dr. Henderson maintained that taxanes are a beneficial addition
to the standard regimen, a factor that is most likely due to the addition of a
non-cross-resistant drug rather than to a longer treatment duration, he said.
In support, he cited data from M.D. Anderson, presented at ASCO
(Thomas et al, abstract 285), showing that patients receiving four cycles of
paclitaxel followed by four of FAC (fluorouracil, Adriamycin, cyclophosphamide)
had better disease-free survival than patients randomized to eight cycles of
At 4-year follow-up, disease-free survival was 81.5% with FAC
alone and 85.2% with FAC plus paclitaxel, a finding that was not statistically
significant. More than one quarter of the patients were node negative;
otherwise, the difference might have been larger, Dr. Hen-derson suggested.
"This is not significant by itself, except it is
supportive of the CALGB study without the confounding factors of that study.
The proportional reduction in recurrence was 26%," Dr. Henderson said.
He also cited, in support, a 158-patient study in which
patients crossed over to docetaxel (Taxotere) and achieved a highly significant
increase in pathologic complete response and clinical response. These studies
support the hypothesis that the benefits seen with the addition of a taxane are
not entirely due to the longer duration of chemotherapy, he said.
Both the CALGB and NSABP studies found little benefit among
patients with ER-positive tumors also treated with 5 years of tamoxifen.
"I don’t think we know what is gained by giving adjuvant
chemotherapy to a patient with an ER-positive tumor who is receiving optimal
endocrine therapy," he added. "In nontaxane regimens, it appears the
benefit for all forms of adjuvant chemotherapy may be smaller among those
patients with tumors that are receptor positive who have received adjuvant
In conclusion, Dr. Henderson maintained that the addition of
paclitaxel following a doxorubicin-containing regimen adds about the same
benefit as adding doxorubicin to one or more components of the CMF
(cyclophosphamide, methotrexate, 5-fluorouracil) regimen.
"This is a fair general statement regarding the currently
available data," he said. "Whether there is, in fact, a difference
that is unique to Taxol in patients who are receptor negative vs positive, or
tamoxifen treated or not treated, is a hypothesis generated by this
Dr. Piccart presented a more cautious view on adjuvant taxane
therapy. "Tax-anes do not yet represent the standard of care in women with
invasive breast cancer," she said. This has been the position taken by the
European Regulatory Agency, which, contrary to the FDA, "did not have
enough faith in the positive results of CALGB 9344 and did not license Taxol
for adjuvant treatment," she said.
Since then, she stressed, two other paclitaxel adjuvant breast
cancer trials have been reported, with negative results"the relatively
small M.D. Anderson trial presented at ASCO and the large NSABP B-28 trial
A single positive large trial is probably not enough to change
the standard of care in a complex disease like breast cancer, Dr. Piccart said.
"As a matter of fact," she added, "I think that the premature
announcement of a new treatment regimen will discourage other research
initiatives and may compromise our ability to find out which subset of patients
can really benefit from new treatments."
In its statement, the NIH Consensus panel called the data on
taxanes for node-positive breast cancer "inconclusive," and
recommended that taxanes not be used in node-negative patients outside of
randomized clinical trials.