Annals of Oncology 11: 505-507. 2000
Editorial
Tamoxifen for early breast cancer: Better late than never
"There are many thousands ofpreviously untreated women,
with hormone-sensitive disease, who could still be offered
tamoxifen, and for whom there would still be worthwhile
benefit."
Worldwide each decade, about eight million women
are diagnosed with breast cancer, and about half of
these women will die from their disease [1]. Since breast
cancer is so common, the reliable demonstration that
practicable and widely available treatments can produce
even a moderate improvement in long-term survival
should lead to some hundreds of thousands of women
being treated accordingly and the avoidance or delay of
many thousands of deaths world-wide each year. The
overviews, by the Early Breast Cancer Trialists' Collaborative Group (EBCTCG), of the worldwide randomised
evidence on the treatment of breast cancer, have established reliably that, for women with potentially hormonesensitive - i.e., oestrogen receptor (ER)-positive-disease,
tamoxifen produces definite reductions in recurrence
and breast cancer death [2]. In almost all of the studies
reviewed, tamoxifen treatment started soon after surgery.
The report by Delozier et al. in this issue of Annals of
Oncology [3] is therefore important in that it demonstrates that, even if tamoxifen was omitted after surgery,
the later introduction of tamoxifen produces survival
benefits that appear to be comparable in size with those
that can be achieved with earlier treatment.
This finding is of considerable practical importance
because the most recent review of randomised trials of
tamoxifen [2] shows that tamoxifen is of value to a
broader group of women than originally thought, with
the corollary that many women who could benefit from
such treatment have not been receiving it. For example,
the latest overview found that age and nodal status
appeared to have little or no effect on the expected
benefit from tamoxifen. Yet, an international survey of
prescribing practice has shown that age and nodal status
are often used as the main factors influencing use of
tamoxifen (unpublished). Further, there was added
benefit when tamoxifen was given in addition to chemotherapy compared with if chemotherapy was given alone.
Again, however, because of theoretical concerns that
tamoxifen might be less effective if given concurrently
with chemotherapy, many women receiving chemotherapy have not also been offered tamoxifen. So, there are
many thousands of previously untreated women, with
hormone-sensitive disease, who could still be offered
tamoxifen, and for whom there would still be worthwhile benefit. It is important that all clinicians treating
breast cancer are aware of this because the late introduction of tamoxifen, up to 10 years after surgery, in
younger women and women with node negative disease,
who have not so far been treated, could avoid many
thousands of deaths per year world-wide.
Where do we now stand with tamoxifen? The
EBCTCG reviewed individual patient data on 37,000
women in 55 trials of tamoxifen versus no tamoxifen,
with an average of about 10 years follow-up. There was
no clear benefit for women with oestrogen receptor (ER)negative tumours but, for women with potentially hormone-sensitive disease, tamoxifen produced definite
reductions in recurrence and breast cancer death throughout the first decade after surgery. Five years of tamoxifen
was of greater benefit than just one or two years in terms
of recurrence prevention and survival. For every 100
women treated with 5 years of tamoxifen, about 16
relapses and 8 deaths are prevented (Figure 1).
In the trials in the overview, half of the women did
not receive any tamoxifen initially. When these women
relapsed, however, most of them would then have been
prescribed tamoxifen. These 'control' women were compared with those who had received adjuvant tamoxifen
from the time of surgery. In terms of time to recurrence,
these trials compare tamoxifen vs. no tamoxifen. But in
terms of survival, the randomised comparison is not of
tamoxifen vs. no tamoxifen, but rather of immediate vs.
delayed tamoxifen, as tamoxifen provides benefit not
only amongst those who received tamoxifen initially,
but also amongst those who received it when they
relapsed. This explains why the 10-year difference between the two groups is only half as big for survival as
for recurrence.
Five years of tamoxifen: Survival benefits thirty times
greater than the risks
The overview now has reliable information on the adverse as well as the beneficial effects of 5 years tamoxifen
on cause specific 10-year survival. Tamoxifen does
increase the risk of endometrial cancer about threefold
but few of these cancers are fatal and so the absolute
risk is small. There is also a small increase in the risk of
pulmonary embolism. However, the reduction in breast
cancer deaths with tamoxifen greatly outweighs these
risks (Table 1).
How long should tamoxifen treatment continue?
The conclusions in the overview, that five years of
tamoxifen was superior to one or two years treatment,
506
Table I Benefits and risks of tamoxifen [2].
100
71-7%
Cause of death
Ten-year mortality among 1000 women
Breast cancer
Endometrial cancer
All other cancers
Pulmonary embolus
Other causes of death
~ 80 per 1000/eiver deaths
~ 2 per 1000 more deaths
No difference
~ 1 per 1000 more deaths
No difference
2p < 0-00001
57-7%
E 40
8300 women: 14 % difference in breast cancer
recurrence (SD 1-2)
•a
£
%
20
10+
tamoxifen because the trials of this question that have
already closed (NSABP B-14 [7], the ECOG studies [8],
and the Scottish trial [9] of five years versus longer) were
not large enough to look at this reliably, even with longer
follow-up. So, even the next Overview in September
2000 will leave substantial uncertainty as to whether
treatment should routinely continue beyond the fifth
year. Until trials such as ATLAS and aTTom that are
currently randomisingfiveyears versus longer hormonal
therapy have entered and followed very large numbers of
women, this question will remain unanswered [10].
years
ATLAS and aTTom: Longer against shorter hormonal
treatment
100
8300 women: 6 2 % difference In survival
(SD1-1)
20
0
5
10+
Newer hormonal agents are currently being tested
such as aromatase inhibitors, newer anti-oestrogens, or
luteinising hormone releasing hormone (LHRH) analogues. These may have a place in future adjuvant
hormonal therapy either alone or in combination with
tamoxifen or other adjuvants. But, even if tamoxifen is
eventually superseded by some new (and currently unproven) hormonal agent, the question of duration will
persist and concerns hormonal therapy as a category of
adjuvant therapy. As such, the results of ATLAS and
aTTom would also be relevant to other hormonal agents,
and not just to tamoxifen. If, as seems likely to be the
case by the end of the year 2000, there emerges a general
consensus that 5 years of adjuvant hormonal treatment
is definitely better than just 2 years, then the question of
whether 10 years is superior to 5 years will become even
more pertinent, and will have to be answered.
years
Figure I. The effect of five years of tamoxifen compared with no
tamoxifen on recurrence and survival.
Important questions about best use of tamoxifen
remain unresolved
came from indirect comparisons between the treatment
effects in trials of different durations of tamoxifen versus
no tamoxifen. In addition, there have been several large
randomised trials directly comparingfivewith two years
of tamoxifen [4-6]. Data from both sources show lower
recurrence rates in women allocated five years of treatment. These trials will, however, require longer followup to assess the impact on survival.
There is, as yet, insufficient randomised evidence to
assess the effects of more than five years of adjuvant
The EBCTCG overview analyses provide the most comprehensive and reliable source of information on the size
of benefits to be expected, for different types of women,
from tamoxifen and other adjuvant therapies for early
breast cancer. All clinicians treating breast cancer should
be familiar with them. Widespread adoption of the more
effective therapies has already led to substantial falls
in breast cancer mortality over the last decade or so in
the UK and USA, with similar trends in other countries
(Figure 2) [II]. The most recent overview results show
that tamoxifen is of value to a broader group of women
507
c100
References
UK
o
S
75
o
s
SO
i
25
1950
1960
1970
1980
1990
2000
Year
Figure 2. Recent decreases in breast cancer mortality at ages 50-69 a in
UK and USA. "Mean of rates at ages 50-54, 55-59, 60-64, 65-69
(reproduced, with permission, from Peto et al. [11]).
than originally thought. At the same time, the EBCTCG
analyses have also highlighted the inadequacy of smallscale evidence and identified many important questions
about everyday use of tamoxifen that remain unresolved. In particular, previous studies of five years of
tamoxifen versus longer, have been too small to answer
this important question reliably [12]. Much larger-scale
evidence is needed and, to achieve this, an invitation to
participate in a relevant clinical trial should become part
of the standard management for most women with
breast cancer. Better late than never!
R. Gray,1 C. Davies2 & P. Perry1
'Clinical Trials Unit
University of Birmingham
2
Clinical Trial Service Unit
University of Oxford, UK
1. Pisani P, Parking DM, Ferlay J. Estimates of the Worldwide
mortality rate from eighteen major Cancers in 1985. Implications
for prevention and projections of future burden. Int J Cancer
1993; 55: 891-903.
2. EBCTCG. Tamoxifen for early breast cancer: An overview of the
randomised trials. Lancet 1998: 351: 1451-67.
3. Delozier T, Switsers O, Genot JY et al. Delayed adjuvant
tamoxifen: Ten-year results of a collaborative randomized controlled trial in early breast cancer (TAM-02 trial). Ann Oncol
2000; 11:515-9.
4. Swedish Breast Cancer Co-operative Group. Randomised trial of
two versus five years of adjuvant tamoxifen for postmenopausal
early stage breast cancer. Swedish Breast Cancer Cooperative
Group. JNCI1996; 88: 1543-9.
5. CRC Breast Cancer Trials Group. Preliminary results from the
Cancer Research Campaign trial evaluating tamoxifen duration
in women aged fifty years or older with breast cancer. JNCI 1996;
88: 1834-9.
6. Delozier T, Speilmann M et al. Short-term versus lifelong adjuvant tamoxifen in early breast cancer (EBC): A randomised trial
(TAM-01). Proc ASCO 1997, 16:128A (Abstr).
7. Dignam JJ, Bryant J.Wieand HS et al. Early stopping of a clinical
trial when there is no evidence of no treatment benefit: Protocol
B-14 of the National Surgical Adjuvant Breast and Bowel Project.
Controlled Clin Trials 1998; 19: 575-88.
8. Tormey D, Gray R et al. Post chemotherapy adjuvant tamoxifen
therapy beyond five years in patients with lymph node-positive
breast cancer. J Natl Cancer Inst 1996; 88: 1828-33.
9. Stewart HJ, Forrest AP et al. Randomised comparison of five
years of adjuvant tamoxifen with continuous therapy for operable
breast cancer The Scottish Cancer Trials Breast Group. Br J
Cancer 1996; 74: 297-9.
10. Peto R. Five years of tamoxifen - or more? J Natl Cancer Inst
1996; 88: 1791-3.
11. Peto R, Bovenham J, Clarke M et al. UK and USA breast cancer
deaths down 25% in year 2000 at ages 20-69 years. Lancet 2000;
355: 1822.
12. Rea D, Poole C, Gray R. Adjuvant tamoxifen: How long before
we know how long? BMJ 1998; 316: 1518-9 (Editorial).