Special
Report
1995 Oxford
Breast Cancer Overview-Preliminary Outcomes
John
Horton, MB, ChB, FACP
Comprehensive Breast Cancer Program
H. Lee Moffitt Cancer Center & Research Institute
Introduction
Clinicians and investigators
alike have appreciated and used the results that emanated from the first report
of the Early Breast Cancer Trialists' Collaborative Group.[1] Another meeting
was held in Oxford, UK, on September 22-23, 1995. Preliminary infor mation has
been presented at several recent meetings. The following represents a collation
from two of these presentations. The results reported here are preliminary,
and some differences in the results reported in the formal publication should
be expected. The data on tamoxifen also can be 
interpreted in conjunction with the Clinical Announcement on November 30, 1995,
from the National Cancer Institute[2] in which data from the NSABP B-14 and
the Scottish Tamoxifen Trial indicate that a five-year duration of adjuvant
tamoxifen is probably optimal.
Data in the current meta-analysis
represent updates of over 250 trials performed by 144 groups! An intent-to-treat
methodology was used, and the dominant outcome was the relative reduction in
risk of failure, which represents the yearly percentage reduction in a patient's
risk of dying over a span of 15 years. The overview included data on b oth node-negative
and node-positive patients and did not discriminate among different combination
chemotherapy programs. Key results are summarized in Table 1.
Oophorectomy
The known adjuvant benefit
of this intervention in younger women was confirmed, although no additional
benefit accrued from oophorectomy if women received cytotoxic chemotherapy.
At 20 years, no excess of cancer-related deaths in the oophorectomy-treated
cohort was seen, and there was no effect on the incidence of contralateral breast
cancer.
Tamoxifen
New information is now available
on the effects of tamoxifen treatment as an adjuvant.
Hormone receptors: The
data are now clear that adjuvant tamoxifen for estrogen-receptor (ER)-poor breast
cancer is ineffective in reducing mortality risk in women regardless of age.
Its use with ER-rich (>10 fmol) tumors reduced the risk of dying by 14% overall
in those under age 50 and by 27% in those age 50 and older.
Duration of therapy:
The effect of duration of adjuvant tamoxifen treatment for breast cancer varies
depending on the age of the groups studied. For those age 50 and over, two years
of treatment is better than one and is almost as effective in reducing risk
of death as five years of treatment. It seems unlikely that longer durations
of treatment will produce more benefit. However, in the younger group, one year
of treatment produced little effect, while those with more than two years (usually
five) of treatment experienced a 27% reduction of risk of mortality. Thus, at
least two (and more likely five) years of treatment are indicated for women
under age 50.
Effects with chemotherapy:
Tamoxifen adds to the effects of chemotherapy in reducing the risk of mortality
in all age groups with ER+ tumors. It produces no such effect in ER tumors.
Non-breast cancer mortality:
At 15 years, the mortality curves for tamoxifen and no-tamoxifen non-breast
cancer deaths are identical. This result does not support the hypothesis that
putative changes in cardiovascular risk or bone density due to ta moxifen would
translate to a lower mortality risk.
Contralateral breast
cancer: Unlike treatment with oophorectomy, tamoxifen treatment 
produced effects on contralateral breast cancer incidence that may be dependent
on duration. Data from 30,000 women showed a reduction in the incidence of contralateral
breast cancer with an odds ratio of 35% +- 6%. For a treatment duration of less
than two years, the odds ratio was 19%; for two years, 29%; and for more than
five years, 47%.
Endometrial cancer: In
17,000 women, 78 cases with 21 deaths have occurred in women who took tamoxifen,
in contrast to 21 cases and two deaths in those who did not. The effect of duration
of tamoxifen treatment on outcomes is shown in Table 2.
Large bowel cancer:
No increased risk of large bowel cancer was shown, except that which could be
expected due to the longer survival of the tamoxifen-treated group.
Chemotherapy
The overall annual reduction
of risk of mortality over 15 years was 18% +- 3%. Little difference was seen
in the type of combination therapy used. The largest percentage of risk reduction
was in the node-negative group (27%), with a 14% figure for both the 1-to-3
node group and the 4+ node group.
Benefit decreased with increasing
age. Benefit for those over 60 years of age was barely significant and was not
apparent in those 70 years of age or older. Administration of tamoxifen in addition
to chemotherapy in patients with ER+ tumors led to an additional 12% risk reduction.
Deaths from non-breast cancer
causes were identical at 15 years in the chemotherapy-treated (including anthracycline-treated)
and untreated groups.
Conclusions
These data suggest several
inferences for clinical practice.
Tamoxifen is an ineffective
adjuvant treatment in patients with ER tumors at any age. It is effective alone
and also adds to benefit from cytotoxic therapy at all ages in patients with
ER+ tumors. At least two years of treatment with tamoxifen is needed in the
under-50 age group. There may be little advantage to a treatment duration of
longer than five years in the 50-and-older age group.
The effect of adjuvant cytotoxic
therapy diminishes with increasing age, with essentially no benefit after age
70, and oophorectomy does not provide additive benefit to chemotherapy.
Oophorectomy, tamoxifen,
and combination cytotoxic chemotherapy have no effect on mortality from causes
unrelated to breast cancer at 15 years.
References
- Systemic treatment of
early breast cancer by hormonal, cytotoxic, or immune therapy: 133 randomised
trials involving 31,000 recurrences and 24,000 deaths among 75 women. Early
Breast Cancer Trialists' Collaborative Group. Lancet. 1992;339:1-15,71-85.
- Adjuvant Therapy of Breast
Cancer - Tamoxifen Update. Bethesda, Md: National Cancer Institute; November
30, 1995.
Back
to Cancer Control Journal Volume 3 Number 1