Symposium article
`
`
`
`Annals ofOnco/ogy 12 (Suppl. 1): 575—579, 2001.
`© 2001 Kluwer Academic Publishers. Printed in the Netherlands
`
`Future directions in the adjuvant treatment of breast cancer:
`The role of trastuzumab
`
`I. Smith
`
`Royal Marsden Hospital. Sutton, UK
`
`Summary
`
`Current evidence shows that adjuvant cytotoxic or hormonal
`therapy increases the disease-free and overall survival of
`patients. The analysis by the Early Breast Cancer Trialists’
`Collaborative Group (EBCTCG) showed that anthracyeline/
`cyclophosphamide (AG-containing regimens are more effec-
`tive than those without AC, providing an 11% greater reduc—
`tion in the risk of death compared with non-AC-containing
`regimens. In addition, paclitaxel and docetaxel have significant
`anti-tumor activity in previously treated patients and sequential
`treatment with paclitaxel may further reduce the risk of recur—
`rence and improve survival. Tamoxifen is eflective in reducing
`the risk of recurrence and death in patients with estrogen
`receptor (ER)-positive tumors. The addition of tamoxifen to
`combination chemotherapy in patients with ER-positive tumors
`
`further reduces the risk of recurrence and improves survival.
`Debate on the effectiveness of tamoxifen in HERZ-positive
`patients is currently underway. A number of trials are in
`progress or planned to investigate the use of the anti-HERZ
`monoclonal antibody trastuzumab (Herceptin) in the adjuvant
`setting. These include a National Surgical Adjuvant Breast and
`Bowel Project (NSABP) adjuvant trial (AC —> paclitaxel vs.
`AC —> paclitaxel + trastuzumab) and an Intergroup study (AC
`—» paclitaxel vs. AC —» paclitaxel + trastuzumab vs. AC —>
`paclitaxel —> trastuzumab). Results from these trials will deter-
`mine whether this novel therapy has a survival benefit in early
`breast cancer.
`
`Key words: adjuvant breast cancer, anthracyclines, CMF,
`HERZ, Herceptin, taxanes, trastuzumab
`
`Introduction
`
`There has been significant progress in systemic adjuvant
`therapy of early breast cancer over the past 10—20 years.
`Evidence shows that the addition of adjuvant hormonal
`and chemotherapy increases disease-free and overall
`survival for many women.
`The key sources of evidence at present are the reviews
`by the Early Breast Cancer Trialists’ Collaborative Group
`(EBCTCG) [1—3]. These provide 10-year follow-up data
`from a large group of randomized trials, providing
`perhaps one of the most powerful collections of data
`available in medicine at this time.
`
`This report provides aWhe EBCTCG’s
`extensive analysis of trial data, together with a discussion
`of the recent advances in and current recommendations
`
`for adjuvant care of breast cancer. The potential impli-
`cations of new biologic agents, such as the antivl-lERZ
`monoclonal antibody trastuzumab (Herceptin), for the
`future of adjuvant therapy are also discussed.
`
`Adjuvant chemotherapy for breast cancer
`(EBCTCG overview)
`
`The EBCTCG reviewed a total of 47 trials in over 18,000
`women, examining the role of chemotherapy vs. no
`
`chemotherapy, together with a further 11 trials examin-
`ing 6000 patients entered into trials of anthracyclines vs.
`non-anthracycline chemotherapy. The key findings from
`this review of adjuvant chemotherapy are summarized
`in Table 1.
`
`Overall, chemotherapy provided a 23% proportional
`decrease in recurrence rate and a 15% decrease in all-
`
`cause mortality in comparison to no chemotherapy. This
`resulted in statistically significant absolute decreases in
`recurrence and death for all age groups studied up to 69
`years. There were only small numbers of patients studied
`over the age of 69 years, making comparisons in this
`group difficult.
`Risk reductions for axillary lymph nodevpositive and
`-negative women were proportionally the same. Absolute
`
`Tbble l. Adjuvant chemotherapy — age and nodal status [3].
`
`Absolute benefit
`
`Recurrence (%)
`
`Monality (%)
`
`< 50 years of age
`Node negative
`Node positive
`50—69 years of age
`Node negative
`Node positive
`
`10
`IS
`
`6
`5
`
`6
`12
`
`6
`2
`
`Genentech 2115
`
`Hospira v. Genentech
`|PR2017-OO737
`
`
`
`
`
`910;;"IKlnfnoon?”JQLUHM112filo'sltzumoprOJXO’auouue//:duq1110.1}popeolumog
`
`
`
`
`
`Genentech 2115
`Hospira v. Genentech
`IPR2017-00737
`
`

`

`76
`
`risk reductions in recurrence and mortality are shown in
`Table 1. These were greatest in node-positive patients
`<50 years of age with reductions in recurrence and
`mortality of 15% and 12%, respectively. For women over
`the age of 50 years risk reductions were less, with an
`absolute gain in survival of 2%-6%. It becomes a value
`judgement in this group of women as to whether such
`benefits are worthwhile.
`
`Significant benefit continued even after five years for
`those women aged <50 years who had chemotherapy
`compared with controls in both recurrence rate and
`overall survival. The mortality benefits after five years
`were also seen in the 50—69-year age group. Thus for
`controls and treated women surviving to five years, prog-
`nosis remained better for those who had been treated
`
`with chemotherapy (Table 2).
`
`721ble 2. Continued benefit from adjuvant chemotherapy with time.
`
`Time from randomization
`(years)
`
`Proportional risk reduction compared
`with control
`
`Recurrence
`M)
`
`Mortality (any cause)
`(W
`
`0—4
`2 5
`
`39
`2|
`
`22
`34
`
`Table 3. Addition of paclitaxel to AC significantly improves disease-free
`and overall survival in the adjuvant setting. (Adapted with permission
`from Henderson et a1. [5].)
`
`AC
`(W
`
`86
`95
`
`AC -> paclitaxel
`(W
`
`P-value
`
`90
`97
`
`0.008
`0.039
`
`Are anthracycline: more eflective as adjuvant treatment?
`
`Disease-free survival
`Overall survival
`
`
`
`
`
`
`
`910g‘1K|llfuo()[BH.totultM1nfito‘sleumoprOJXO'auouue/fiduqLUDJJpoptzoluMoq
`
`In the EBCTCG overview, anthracycline-containing
`chemotherapy gave statistically superior recurrence-free
`and overall survival compared with non-anthracycline
`regimens, although the magnitude of the benefit was
`small (absolute survival improvement of 2%—3%). In the
`10 randomized trials included in the overview,
`five
`showed increased disease-free survival and three showed
`increased overall survival. None showed inferior results
`
`for anthracyclines. Toxicity was somewhat greater for
`anthracycline-containing combinations.
`In a more recent trial, an anthracycline regime was
`compared with
`classic CMF (cyclophosphamide—
`methotrexate—S-fluorouracil) in premenopausal women
`[4]. There was a statistically significant absolute survival
`increase of 7% at five years for those who received
`anthracyclines. However, in the anthracycline arm, 50%
`had grade 2 or more nausea and vomiting compared
`with 25% in the CMF arm. Similarly, stomatitis was
`45% vs. 25% and alopecia 97% vs. 40%, respectively.
`There was no difi’erence in the myelotoxicity between the
`two arms. Cardiotoxicity is a potential problem, but this
`is rare when recommended cumulative doses are not
`
`exceeded. It did not occur in any of 351 patients given
`anthracycline in this trial. Quality of life scores were
`lower during chemotherapy for the anthracycline group,
`but were the same by six months after chemotherapy.
`
`The role of taxanes in adjuvant care
`
`In a recent trial involving 3170 women, half of whom
`were randomized to receive four cycles of paclitaxel
`given three times weekly after completing four cycles of
`adjuvant doxorubicin and cyclophosphamide, use of
`paclitaxel reduced the recurrence rate by 22% and the
`death rate by 26% at 18 months compared to doxorubicin—
`cyclophosphamide alone [5] (Table 3). At a three-year
`analysis the results were unchanged. These results sound
`more impressive than they really are: the absolute reduc-
`tion in mortality is so far only 2% (P = 0.0390), but this
`
`3170 node-positive patients (62% pre-menopausal). AC x 4 —> 1 patch-
`taxel 175 mg/m2 every 3 weeks x 4. Eighteen-month analysis. Grade 3
`toxicity for paclitaxel 25% WBC 1, 5% neuropathy, 5% pain.
`
`may enlarge or decrease with further follow-up. Further
`trials using both adjuvant paclitaxel and docetaxel await
`completion and analysis. New studies will examine the
`role of sequence and dose densification of the taxanes
`as well as their role in node-negative, high-risk tumors.
`Anthracyclines followed by paclitaxel has become a
`standard treatment for node-positive disease in the USA
`but not yet in Europe.
`
`Overview of tamoxifen in adjuvant breast cancer
`
`In the 1998 overview [2], individual patient data were
`reviewed from trials of tamoxifen vs. no tamoxifen
`
`started before 1990, with an average follow-up of 10
`years. Individual patient data were collected for 37,000
`women in 55 trials, nearly 8000 of whom had estrogen
`receptor (ER)-poor tumors (< 10 fmol/mg). This group
`did not benefit from tamoxifen and were excluded from
`
`further analysis. Of the remaining women, approximately
`18,000 had ER-positive tumors and 12,000 had unknown
`ER status, of whom two-thirds were assumed to have
`been ER positive. This large data set allows statistically
`powerful conclusions on many aspects regarding adju-
`vant tamoxifen usage.
`The influence of duration of tamoxifen use is shown
`
`in Table 4. This shows that five years’ tamoxifen treat-
`ment is better than a shorter duration of treatment; the
`risk reduction in mortality at one, two and five years for
`ER-positive women was 14%, 18% and 28%, respectively,
`and this difference was highly significant, 21’ < 0.00001.
`The proportional risk reductions for recurrence and
`mortality were approximately the same for node-positive
`and -negative patients. The absolute risk reduction was
`greater for node-positive than node-negative patients in
`
`

`

`Table 4. Effects of tamoxifen duration and ER status on mortality
`
`Duration (years)
`
`Proportional reduction in mortality (%)
`
`I
`2
`5
`
`ER poor
`
`ER unknown
`
`ER positive
`
`6
`l7
`—3
`
`l0
`l5
`2|
`
`l4
`l8
`28
`
`that their overall risk of recurrence and death is higher.
`The absolute reduction in mortality with five years of
`tamoxifen was 11% for node-positive and 6% for node-
`negative patients.
`The benefit for mortality and recurrence reduction
`continued into the second five years after randomization.
`ER-positive women who had a 50% reduction in their
`rate of relapse in the first five years after randomization
`continued to have a one-third reduction in relapse in the
`second five years compared with control. Interestingly,
`the proportional mortality reductions for the second five
`years were the same as for the first five years, showing
`prolonged continued benefit after discontinuation of
`tamoxifen.
`
`Tamoxifen in HER2-posilive cancers
`
`There is current concern about the role of adjuvant
`tamoxifen in patients with HERZ-positive cancers. In a
`20-year update of the Naples GUN trial which random-
`ized women to two years’ tamoxifen or no treatment,
`HERZ status was determined retrospectively in 245 of
`433 patients [6]. Overall, at a median follow-up of 14
`years, tamoxifen improved disease—free and overall sur-
`vival, but only in HERZ-negative patients in whom there
`was a 19% decrease in expected death rate in patients on
`tamoxifen. ln stark contrast there was a 57% increase in
`
`mortality for HERZ-positive patients on tamoxifen. In a
`multivariate analysis, HERZ positivity independently
`predicted for tamoxifen failure.
`In a large Scandinavian trial of two vs. five years of
`tamoxifen in ER-positive, early breast cancer patients,
`HERZ status was measured in 449 women in a post [we
`analysis [7]. Results showed 13% of those who were
`disease free two years after surgery had HERZ-gene
`amplification or protein overexpression on examination
`of their tumors. HERZ-negative patients had a 0.65
`relative risk of recurrence with five vs.
`two years of
`tamoxifen. HERZ-positive patients,
`in contrast, had a
`suggested increase in their relative risk, 1.9 (confidence
`interval (CI): 0.54-6.6) on tamoxifen, although this was
`not statistically significant. This suggests that prolonged
`tamoxifen therapy may not be beneficial in patients with
`ER-positive tumors if also HERZ positive.
`Contradictory data have been obtained from the
`CALGB 8541 trial
`in which tissue from ER-positive
`women who had received tamoxifen for IO years was
`assessed for HER2 status [8]. Results in this relatively
`small number of patients revealed that
`the five-year
`
`77
`
`Table 5. Adjuvant tamoxifen in HERE-positive patients. (Adapted
`with permission from Muss et al. [8].)
`
`Number of patients
`Patients receiving tamoxifen
`Five-year relapse-free survival
`with tamoxifen (%)
`with no tamoxifen (%)
`Reduction in risk of relapse (%)
`
`HERZ positive
`
`HER2 negative
`
`[86
`68 (37%)
`
`555
`263 (47%)
`
`74
`59
`42
`
`73
`58
`44
`
`CALGB 854| re-analyzed. No interaction of tamoxifen and HERZ
`status.
`
`relapse—free survival and reduction in risk of relapse
`were similar in both HERZ-positive and -negative pa-
`tients (Table 5). Survival curves for HERZ overexpres-
`sors were identical to those for HERZ-negative patients.
`At present, therefore, the role of adjuvant tamoxifen
`in HER2-positive patients remains uncertain.
`
`Trastuzumab: A new therapeutic option in adjuvant
`breast cancer
`
`Amplification of the HERZ gene confers a poorer out-
`look for both node—positive and -negative patients with
`early breast cancer. HERZ status is increasingly being
`considered as one of a number of prognostic factors in
`early breast cancer when determining the most appro-
`priate chemotherapy regimen. New biologically targeted
`therapies are also being investigated in the adjuvant
`setting. One such novel agent is the anti-HER2 mono-
`clonal antibody, trastuzumab, which has been shown to
`produce a significant survival benefit in combination
`with chemotherapy (AC or paclitaxel) in HERZ-positive
`metastatic breast cancer patients when given as first-line
`treatment (9—! I]. Furthermore, trastuzumab is generally
`well tolerated and non-cross—resistant with chemother-
`
`apeutic agents.
`A problem in considering trastuzumab as adjuvant
`therapy is the unexpected risk of developing cardiac
`dysfunction, particularly when combined with the anthra-
`cycline doxorubicin. This would require careful monitor-
`ing in patients with early breast cancer who have poten~
`tially curable disease. It has been suggested that the issue
`of cardiotoxicity could be avoided by simply using
`trastuzumab with agents other than doxorubicin. Despite
`this possibility,
`there are several compelling reasons
`for using anthracyclines in the adjuvant setting. Firstly,
`follow—up results from the CALGB 8541 study, examin-
`ing dose intensification of CAF (cyclophosphamide—
`doxorubicin—S-FU), have revealed that patients receiving
`the high and moderate (standard) doses of CAF survive
`longer than those receiving the low dose [12]. Secondly,
`examination of patients’ HERZ status within this trial
`revealed that those with tumors expressing high levels of
`the HERZ protein had a significantly worse survival if
`treated with moderate or low-dose intensity CAF chemo—
`therapy, compared with standard dose intensity (Figure l)
`
`
`
`9[oz‘1KlnfuoolnH.IOuIIIM1t:flto‘s[eumofp103x0'ououuta/fiduqLllOJJpopeolumoq
`
`
`
`
`
`

`

`
`
`
`
`9103‘rKlnfuoolaHtown,“filo'S[BUJHO[DJOJXO‘OUOUUE/fiduqit:ulOJJpopeolumoq
`
`
`
`
`
`
`
`CM HE
`
`n2
`
`+
`
`HR 0.495
`
` F
`
`\‘s\Control HER2+
`
`20 years
`
`1m
`
`80
`
`
`
`— High (n = 93)
`- - - Moderate (n = 91)
`------- Low (n: as)
`
`
`
`Overallsurvival
`
` P < 0.001 high vs. low dose
`
`or moderate dose
`a
`i
`is
`a
`to
`12
`Years
`
` o
`
`
`
`
`
`Disease-treesurvival(%)
`
`Figure I. High HER2 expression — disease-free survival according
`to the CAF treatment group. (Reproduced by permissron of Oxford
`University Press from Thor et al. [l3].)
`
`Figure 2 Overall survival benefit from adjuvant CMF in node-positive,
`HER2-positive disease [[5].
`
`that patients with tumors
`[13]. These results suggest
`expressing high levels of the HER2 receptor protein
`may require dose-intensive anthracyclines.
`
`trastuzumab. This trial has roughly the same entry
`criteria to the NSABP trial,
`together with the same
`degree of cardiac monitoring.
`
`Planned trials with trastuzumab
`
`A number of trials are either planned or underway with
`trastuzumab in the early breast cancer setting. These are
`summarized below.
`
`National Surgical Adjuvant Breast and Bowel Project trial
`
`A National Surgical Adjuvant Breast and Bowel Project
`(NSABP) trial currently underway involves node-posi-
`tive, HER2-positive patients receiving four courses of
`AC followed by randomization to receive four courses of
`paclitaxel vs. the same treatment but with trastuzumab
`added weekly for one year. A three-week gap exists
`between patients finishing their course of AC and start-
`ing trastuzumab, and patients will undergo extensive
`cardiac monitoring to avoid the potential risk of cardio-
`toxicity associated with this combination.
`Objectives for the trial include improvement in disease-
`free and overall survival and also the determination of
`
`If an incidence of
`any major cardiac safety issues.
`cardiac dysfunction > 5% is observed in the first 1000
`patients the trial will be discontinued. Patients will also be
`stratified according to their prior exposure to tamoxifen.
`All
`those who are PER/progesterone receptor (PgR)
`positive will receive tamoxifen and those who are > 50
`years of age have the option of receiving tamoxifen
`regardless of their ER/PgR status.
`
`Alternatives to the combination of trastuzumab plus
`doxorubicin in the adjuvant setting
`
`to avoid any potential cardiotoxicity
`In an attempt
`following administration of trastuzumab plus doxorubi-
`cin, other therapeutic combinations are being considered.
`These options include trastuzumab plus CM F, epirubicin,
`liposomal doxorubicin, taxane—cisplatin (or carboplatin)
`or taxane—vinorelbine.
`
`Trastuzumab plus CMF
`
`Despite being one of the earliest cytotoxic options for
`adjuvant therapy in breast cancer, data available suggest
`that CMF is comparable to the anthracyclines in terms
`of survival benefit. Indeed, an analysis of 337 patients
`included in the original Milan trial of classical, dose-
`intensive CMF (CMF X 12 vs. control) revealed that
`those who were HER2 positive (l6%) had a significant
`overall survival benefit from CMF compared with the
`control patients who were HER2 positive (Figure 2)
`[14, 15]. These observations were important in view of
`previous data suggesting that adjuvant CMF did not
`produce any clinical benefit in HER2-positive patients.
`As this trial demonstrated an appreciable survival benefit
`from dose-intensive CMF, it can be suggested that other
`studies having failed to show any benefit may have
`employed a less-intensive regimen of CMF.
`
`Intergroup trial
`
`Other agents
`
`An Intergroup trial in progress has a very similar design
`to the NSABP trial as discussed‘above. The first two
`arms are identical, although the Intergroup trial includes
`a third arm in which trastuzumab is given after the four
`courses of paclitaxel, thereby increasing the period of
`time between receiving anthracyclines and receiving
`
`The anthracycline epirubicin is a potentially useful drug
`in the adjuvant setting and is much more widely used in
`Europe than the USA. Epirubicin is associated with a
`lower incidence of cardiotoxicity than doxorubicin, and
`an important German trial is addressing the incidence
`of cardiac dysfunction following epirubicin—cyclo-
`
`

`

`phosphamide~trastuzumab in patients with metastatic
`disease. Encapsulating the drug doxorubicin in liposomes
`has been proposed to limit
`its cardiotoxic effects,
`although clinical trials with Iiposomal doxorubicin are
`not yet advanced enough to make conclusions on its
`efiicacy and safety. Finally, the theoretical advantages of
`using different schedules of various cytotoxic agents,
`such as a taxane with cisplatin—carboplatin or a taxane
`with vinorelbine,
`is problematic at this time because
`these are not standard recognized schedules in advanced
`breast cancer.
`
`Conclusions
`
`Adjuvant chemotherapy improves survival in early breast
`cancer. In the USA, AC followed by paclitaxel is in-
`creasingly favored for high—risk patients.
`In Europe,
`FAC/FEC or CMF remain the standard.
`
`The humanized anti-HER2 monoclonal antibody
`trastuzumab is an attractive drug for adjuvant therapy
`because of its prolongation of survival
`in metastatic
`disease, lack of cross resistance to other agents, and low
`incidence of toxicity. In order to answer questions relat-
`ing to the issue of cardiotoxicity following administra-
`tion of a combination of trastuzumab and doxorubicin,
`a number of trials are underway in both metastatic
`patients and in the adjuvant setting. In addition, it is
`important to consider non-anthracycline combinations.
`Finally,
`the important
`issue of whether trastuzumab
`may confer a survival benefit to tamoxifen treatment in
`HERZ-positive patients, perhaps reversing the reduction
`in response previously observed with tamoxifen in a
`number of trials, is currently being considered.
`
`Note
`
`Professor Smith has reported that he serves on a Roche advisory board
`for Herceptin.
`
`References
`
`79
`
`Levine MN, Bramwell VH, Pritchard K1 et a1. Randomized trial of
`intensive cyclophosphamide, epirubicin, and fluorouracil chemo—
`therapy compared with cyclophosphamide, methotrexate, and
`fluorouracil in premenopausal women with node-positive breast
`cancer. National Cancer Institute of Canada Clinical Trials
`Group] Clin Oncol 1998; 16: 2651—8.
`Henderson 1C, Berry D, Demetri C et al. Improved disease-free
`and overall survival from the addition of sequential paclitaxel but
`not from the escalation of doxorubicin dose level in the adjuvant
`chemotherapy of patients with node-positive primary breast
`cancer. Proc Am Soc Clin Oncol I998; 17: 1013 (Abstr 390A).
`Bianco AR, De Laurentis M, Carlomagno C et al. Twenty-year
`update of the Naples GUN trial of adjuvant breast cancer therapy:
`Evidence of the interaction between c-erb-BZ expression and
`tamoxifen efficacy Proc Am Soc Clin Oncol 1998; 17: 97a (Abstr
`373).
`Nordenskjéld B, Hatschek T, Kallstrt’im A-C et al. Results of
`prolonged adjuvant tamoxifen therapy of breast cancer correlated
`to steroid receptor, S-phase and erb-BZ levels. Proc Am Soc Clin
`Oncol 1999; 18: 70a (Abstr 263).
`Muss H, Berry D, Thor A et a1. Lack of interaction of tamoxtfen
`use and erbB—2/Her-2/Neu expression in CALGB 8541: A random-
`ized adjuvant trial of three different doses of cyclophosphamide,
`doxorubicin and fluorouracil (CAF) in node-positive primary
`breast cancer. Proc Am Soc Clin Oncol 1999; 18: 68a (Abstr 256).
`Norton L, Slamon D, Leyland-Jones B et al. Overall survival (OS)
`advantage to Herceptin (H) in HERZ—overexpressing (HER2+)
`metastatic breast cancer (M BC). Proc Am Soc Clin Oncol 1999;
`18: 127a (Abstr 483).
`Slamon D, Leyland-Joncs B, Shak S et a1. Addition of Herceptino
`(humanized anti-HER2 antibody) to first line chemotherapy for
`HER2 overexpressmg metastatic breast cancer (HER2+/MBC)
`markedly increases anticancer activity: A randomized multina-
`tional controlled phase III trial. Proc Am Soc Clin Oncol 1998;
`17: 983 (Abstr 377).
`Eiermann W, on behalf of the International Herceptin Study
`Group. Trastuzumab combined Wllh chemotherapy for the treat-
`ment of HERZ-positive metastatic breast cancer: Pivotal trial
`data. Ann Oncol 2001; 12 (Suppl 1): $57—$62 (this supplement).
`Budman DR, Berry DA, Cirrincione CT et a1 Dose and dose
`intensity as determinants ofoutcome in the adjuvant treatment of
`breast cancer. The Cancer and Leukemia Group B. J Natl Cancer
`Inst 1998; 90: 1205—11.
`Thor AD, Berry DA, Budman DR et al. erbB-2, p53 and efficacy
`of adjuvant therapy in lymph node—positive breast cancer. J Natl
`Cancer Inst 1998; 90: 1346—60.
`Bonadonna G, Valagussa P, Rossi A et al. Ten-year experience
`with CM F—based adjuvant chemotherapy in resectable breast
`cancer. Breast Cancer Res Treat 1985; 5: 95—115.
`Menard S, Valagussa P, Pilotti S et al. Benefit of CMF treatment
`in lymph node-posuive breast cancer overexpressing HERZ. Proc
`Am Soc Clin Oncol 1999; 18: 69a (Abstr 257).
`
`10.
`
`12.
`
`13.
`
`14.
`
`15.
`
`1, Early Breast Cancer Trialists’ Collaborative Group. Effects of
`adjuvant tamoxifen and of cytotoxic therapy on mortality in early
`breast cancer, An overview of 6| randomized trials among 28,896
`women. N Engl .1 Med 1988; 319: 1681—92.
`2. Early Breast Cancer Trialists’ Collaborative Group. Tamoxtfen
`for early breast cancer: An overview of the randomized trials.
`Lancet 1998; 351: 1451—67.
`3. Early Breast Cancer Trialists’ Collaborative Group. Polychemo-
`therapy for early breast cancer: An overview of randomized trials.
`Lancet 1998; 352: 930—42.
`
`Correspondence to:
`1. Smith, MD
`Royal Marsden Hospital
`Downs Road
`Sutton
`Surrey SM2 SPT
`UK
`E-mail: ians@icr.ac.uk
`
`
`
`910Z‘[1([11fuoaluHJoulllm112[310'812LLmOprOJXO'OUOUUB/flduquioupopaolumoq
`
`
`
`
`
`