A novel continuous infusional 5-fluorouracil-based chemotherapy regimen compared with conventional chemotherapy in the neo-adjuvant treatment of early breast cancer: 5 year results of the TOPIC trial

I. E. Smith1,*, R. P. A’Hern1, G. A. Coombes2, A. Howell3, S. R. Ebbs4, T. F. Hickish5, M. E. R. O’Brien6, J. L. Mansi7, C. B. Wilson8, A. C. Robinson9, P. A. Murray10, C. G. A. Price11, T. J. Perren12, R. W. Laing13 and J. M. Bliss2

1 Royal Marsden Hospital; 2 Institute of Cancer Research, London and Sutton; 3 Christie Hospital, Manchester; 4 Mayday University Hospital, Croydon; 5 Royal Bournemouth Hospital; 6 Kent Oncology Centre, Maidstone; 7 St George’s Hospital, London; 8 Addenbrooke’s Hospital, Cambridge; 9 Southend Hospital, Southend-on-Sea; 10 Essex County Hospital, Colchester; 11 Bristol Oncology Centre, Bristol; 12 St James Hospital, Leeds; 13 St Luke’s Cancer Centre, Guildford, UK

Received 29 September 2003; revised 13 January 2004; accepted 15 January 2004


    ABSTRACT
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
Background:

To compare the efficacy of continuous infusional 5-fluorouracil (5-FU)-based chemotherapy against conventional bolus chemotherapy in the preoperative treatment of patients with large operable early breast cancer.

Patients and methods:

Four hundred and twenty-six women with histologically proven ³3 cm invasive early breast cancer were randomised to receive pre-operative infusional 5-FU 200 mg/m2 by daily 24 h continuous infusion via a Hickman line for 18 weeks with epirubicin 60 mg/m2 intravenous (i.v.) bolus on day 1 and cisplatin 60 mg/m2 i.v. bolus on day 1, both repeating 3-weekly (infusional ECisF), or conventional bolus doxorubicin 60 mg/m2 i.v. on day 1 and cyclophosphamide 600 mg/m2 i.v. on day 1, both repeating 3-weekly (AC), both schedules for six courses. Patients subsequently had local therapy (surgery or radiotherapy or both) and tamoxifen 20 mg orally daily as appropriate.

Results:

The 5 year results for AC and infusional ECisF, respectively, were as follows: overall response, 75% and 77%; complete clinical remission, 31% and 34%; pathological complete remission (pathCR), 16% for both; and pathCR with residual ductal carcinoma in situ (DCIS), 25% and 24%. Mastectomy rates were 37% and 34%, respectively. Five-year overall survival was 74% for AC and 82% for infusional ECisF (hazard ratio 0.76, 95% confidence interval 0.51–1.13; P = 0.18). Both treatments were well tolerated. Grade III/IV lethargy, vomiting, alopecia and plantar-palmar erythema were significantly greater for infusional ECisF; grade III/IV leucopenia was significantly greater for AC.

Conclusions:

Preoperative continuous infusional 5-FU-based chemotherapy is no more active than conventional AC for early breast cancer; with a median 5 year follow-up, the infusion-based schedule shows a non-significant trend towards improved survival.

Key words: breast neoplasm, chemotherapy, cisplatin, infusional 5-FU, neo-adjuvant, phase III


    Introduction
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
Neo-adjuvant (pre-operative) chemotherapy in patients with large potentially operable early breast cancer has the advantage of using the tumour as an in vivo measure of response to treatment [1], with a greater chance of avoiding mastectomy than with post-operative adjuvant chemotherapy [24]; survival outcome is at least as good as with conventional post-operative adjuvant therapy [38]. The pre-operative approach also offers the opportunity to compare new therapies against standard treatment in early breast cancer. We have shown that a novel combination of continuous infusional 5-fluorouracil (5-FU) with conventional cisplatin and epirubicin (infusional ECisF) was active against advanced breast cancer [9], and achieved a very high overall response rate of 98% and a complete remission rate of over 60% as pre-operative chemotherapy in patients with ≥3 cm potentially operable early breast cancer [10]. This schedule had been selected empirically on the basis of its efficacy in the treatment of advanced gastric carcinoma, where it had been shown to achieve a higher response rate than conventional chemotherapy, and with improved survival [11].

Based on our phase II results, the Trial of Preoperative Infusional Chemotherapy (TOPIC) was set up to compare pre-operative infusional ECisF with conventional bolus doxorubicin and cyclophosphamide (AC) in a phase III multicentre randomised trial. A first analysis has already been presented in abstract form [12] and the 5-year median follow-up results are presented here.


    Patients and methods
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
Patients
Patients with operable ≥3 cm histologically proven invasive early breast cancer were entered into this trial. Histological diagnosis was established by core needle biopsy. Patients with locally advanced or inflammatory breast cancer were excluded. Further inclusion criteria were as follows: age ≤70 years; World Health Organization (WHO) performance status 0–1; ability to give informed consent; pre-treatment white blood cell (WBC) count ≥3 x 109/l; platelet count >150 x 109/l; glomerular filtration rate >60 ml/min, as assessed by EDTA; and no evidence of metastatic disease.

Patients were recruited from July 1995 until February 1999 from 17 centres as follows (in order of recruitment): Royal Marsden Hospital; Christie Hospital, Manchester; Mayday University Hospital, Croydon; Royal Bournemouth Hospital; Kent Oncology Centre; St George’s Hospital, London; Addenbrooke’s Hospital, Cambridge; Southend Hospital; Essex County Hospital, Colchester; Bristol Oncology Centre; Yorkshire Breast Group; St Luke’s Cancer Centre, Guildford; Royal Free Hospital, London; Salisbury Hospital; University College Hospital; St Bartholomew’s Hospital; Ipswich Hospital.

Treatment regimen
Patients were randomised to receive either doxorubicin–cyclophosphamide (AC) or epirubicin–cisplatin and continuous infusional 5-FU (infusional ECisF) as follows. AC: doxorubicin 60 mg/m2 and cyclophosphamide 600 mg/m2 bolus on day 1, repeating 3 weekly for six courses; infusional ECisF: 5-FU 200 mg/m2 daily in a 24 h continuous infusion by ambulatory pump via Hickman line for 18 weeks; epirubicin 60 mg/m2 bolus on day 1, repeating 3 weekly and cisplatin 60 mg/m2 i.v. in a 1 h 250 ml saline infusion on day 1, repeating 3 weekly with full intravenous hydration, both drugs for six courses. Carboplatin was substituted for cisplatin in the event of renal toxicity, peripheral neuropathy or persistent emesis. The carboplatin dose was calculated according to renal function [13] with an AUC = 5. All patients receiving infusional ECisF were started on oral low dose warfarin 1 mg daily while the Hickman line remained in situ.

Dose adjustments for toxicity
If WBC counts were <3.0 x 109/l and/or platelets <100 x 109/l then treatment with epirubicin and cisplatin was delayed 1 week (5-FU was continued) and treatment was given at full dose. If there was a 2-week delay then there was a dose reduction of 25% for all drugs. If there was a >2-week delay then there was a 50% reduction of all drugs.

Patients with any grade of plantar-palmar erythema received pyridoxine 50 mg orally three times daily throughout treatment. For grade ≥3 mucositis or plantar-palmar syndrome, 5-FU was interrupted for 1–2 weeks until healing and then recommenced with a 25% reduction in the dose of 5-FU, but with 100% dose of the other drugs.

Local treatment
Conservative surgery with axillary resection or mastectomy was carried out post-chemotherapy based on clinical tumour size after treatment, backed up by ultrasound reassessment when available. Until 1997, surgeons reviewing a patient who had achieved a complete clinical remission had the option of no surgery but an immediate referral for radiotherapy to breast and axilla. This option was subsequently discontinued because of potential concerns about long-term local recurrence. Data on these patients have been published separately [14].

Radical radiotherapy was routinely offered to all patients who had received conservative surgery and was given according to the standard policy of the unit concerned. The axilla was included only if it had not been surgically treated. Radiotherapy after mastectomy was optional, based on the policy of the individual unit, but had to be consistent for both treatment arms. Radiotherapy was planned to commence within 4 weeks of the completion of surgery (or chemotherapy if surgery was to be omitted).

Tamoxifen
According to local practice at the time of the trial, all patients were offered adjuvant tamoxifen 20 mg orally daily starting on around day 1 of the first course of chemotherapy for 2–5 years. Estrogen receptor (ER) and progesterone receptor (PgR) status were not available for all patients, again based on practice at the time.

Pre-treatment evaluation
All patients had a histological diagnosis of invasive breast cancer established by core needle biopsy. The primary was assessed by bidimensional clinical measurement. Other pre-treatment assessment included mammography, full peripheral blood count, plasma urea and electrolytes, serum liver function tests, renal function using 51Cr-EDTA clearance for patients receiving infusional ECisF, clotting screen, chest X-ray, performance status assessment using standard WHO criteria and an electrocardiogram. More elaborate staging investigations were carried out only where clinically indicated to exclude suspicion of metastatic disease.

Follow-up
Following completion of chemotherapy and local treatment, patients were assessed 3 monthly for 2 years, 6 monthly until 5 years post-randomisation and annually thereafter.

Assessment of response
Primary tumours were assessed at each 3 week visit for chemotherapy and again 3 weeks after the last course. Response was assessed according to standard WHO criteria [15] as follows: complete clinical response (CR) was defined as the clinical disappearance of palpable disease in the breast for ≥4 weeks; partial response (PR) was defined as ≥50% decrease in the product of the tumour’s two longest perpendicular diameters for a duration of ≥4 weeks; progressive disease (PD) was defined as a ≥25% increase in the product of one or more of the measurable lesions of the appearance of a new lesion(s). In addition, two additional criteria were defined in this study: minor response (MR) was defined as a <50% decrease in the product of the tumour’s longest dimension and its widest perpendicular measurement for a duration of ≥4 weeks; stable disease (SD) was defined as no change in size or a <25% increase in the product of the measurable disease.

Relapse-free survival and survival
Time to first relapse was measured from the date of randomisation to the date of first recorded local or distant relapse. In those patients who died without relapsing, death was counted as an event. Survival was likewise measured from the date of randomisation.

Assessment of toxicity
Patients were assessed for toxicity after each course of treatment according to standard WHO criteria [15].

Randomisation procedure
Patients were stratified according to institution and age (<50 and ≥50 years) prior to randomisation. Randomisation was carried out centrally by telephone at the Institute of Cancer Research–Clinical Trials and Statistics Unit (ICR-CTSU).

End points and statistical considerations
The primary end points of this trial were survival and relapse-free survival. Accrual was planned to include at least 400 patients to allow the detection of a survival improvement of 15% with a power of 90% and a two-sided significance level of 5%.

Secondary end points included clinical response rate, pathological response rate, incidence of mastectomy in each arm and differences in standard WHO toxicities.

The patient characteristics of the two treatment arms were compared by means of the chi-squared test with Yates correction for categorical data, or the Mann–Whitney test for quantitative or ordinal data. Objective response rates were compared by means of the Mann–Whitney test for trend. Survival curves were constructed by the Kaplan–Meier method [16] and compared by means of the log rank test [17]. All P values were two-sided.

An independent data monitoring committee reviewed the trial yearly with the application of conservative stopping rules. First analysis was carried out as planned in April 2000 for the comparison of response rates and toxicity, and results were presented at the American Society for Clinical Oncology 2000 [12]. An analysis in December 2002 was used for this publication. Sufficient events have been observed to be able to reliably detect a 40% relative reduction in the relapse-free survival event rate (90% power, 5% significance level).

Ethical considerations
The protocol was approved by the local ethics committee of each collaborating institution, and written informed consent was obtained from all patients.


    Results
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
Patient characteristics
A total of 426 patients were entered into the trial between July 1995 and February 1999; 215 patients were randomised to receive the control arm (AC) and 211 patients to the experimental arm (infusional ECisF). Analysis was by intention-to-treat; two patients (one in each arm) did not receive treatment. Patient characteristics are given in Table 1. Details in summary for AC and infusional ECisF, respectively, were as follows: median age, 47 (range 25–66) and 46 (range 22–68) years; tumour size, 5 cm (range 3–15) and 5 cm (range 2–11); clinical node positivity, 42% and 45%; radiotherapy was given to 75% and 76% of patients. Median follow-up was 5 years (range 3.8–7.4).


View this table:
[in this window]
[in a new window]
 
Table 1. Patient characteristics
 
Response rate
Objective clinical response rates are given in Table 2 and, in summary, are as follows for AC and infusional ECisF, respectively: overall response, 75% and 77% [95% confidence interval (CI) 69– 81% and 71–83%]; complete clinical remissions, 31% (95% CI 28–41%) and 34% (95% CI 24–37%); minor responses, 17% and 16%; stable disease, both 6%; progressive disease, 2% and 1%.


View this table:
[in this window]
[in a new window]
 
Table 2. Response, axillary node involvement, relapse-free survival and overall survival
 
Pathological complete remission
Excision specimens following the completion of chemotherapy were available for 144 and 143 patients receiving AC and infusional ECisF, respectively. Pathological complete remission (pathCR) was 16% (95% CI 9% to 20%) for both arms; pathCR for invasive disease but with residual ductal carcinoma in situ (DCIS) was 25% (95% CI 18–32%) for AC and 24% (95% CI 17–31%) for infusional ECisF. (In a separate unpublished analysis of 551 patients treated with preoperative chemotherapy at the Royal Marsden Hospital, London, the survival for patients achieving a pathCR with DCIS was as good as for pathCR alone.)

Surgery after chemotherapy
Thirty-eight patients receiving AC and 37 patients receiving infusional ECisF who achieved complete clinical response after chemotherapy did not receive surgery but were referred directly for radiotherapy, as described above. Eighty-two per cent of patients in each arm had surgery, including 73% in each arm who had axillary nodal surgery. In these patients, axillary node involvement was found in 50% of those receiving AC and 51% receiving infusional ECisF.

Mastectomy rates
Thirty-seven per cent of patients were deemed to require mas-tectomy at surgical assessment following chemotherapy in patients treated with AC compared with 34% in patients receiving infusional ECisF (P = 0.19).

Relapse-free survival and overall survival
Relapse-free survival and overall survival, as assessed by life table analyses, are given in Figures 1 and 2. Five-year relapse-free survival was 63% for patients treated with AC and 62% for infusional ECisF (hazard ratio 1.05, 95% CI 0.77–1.44; P = 0.77). Five-year overall survival was 74% for patients treated with AC compared with 82% with patients treated with infusional ECisF (hazard ratio 0.76, 95% CI 0.51–1.13; P = 0.18). The cut-off date for these analyses was December 2002, when 155 relapses (or death without relapse) and 98 deaths had occurred.



View larger version (13K):
[in this window]
[in a new window]
 
Figure 1. Probability of relapse-free survival. AC, doxorubicin and cyclophosphamide; ECisF, continuous infusional 5-fluorouracil with conventional cisplatin and epirubicin.

 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 2. Probability of survival.

 
Hazard ratios
Hazard ratios with 95% CIs are given in Figure 3 for relapse-free survival, local recurrence, distant recurrence and overall survival. The odds ratio is given for complete response.



View larger version (13K):
[in this window]
[in a new window]
 
Figure 3. Hazard ratios with 95% confidence intervals for relapse-free survival (RFS), local recurrence (LR), distant relapse (DR) and overall survival (SUR). The odds ratio (OR) is shown for complete response.

 
Courses completed, treatment delays and dose reductions
Ninety-eight per cent of patients receiving AC completed at least three courses of chemotherapy compared with 96% of patients for infusional ECisF. The corresponding figures for completion of six courses were 86% and 72%, respectively (P <0.001). Details of treatment delays and dose reductions are given in Table 3. There was a significant increase in both treatment delays and in dose reductions for patients receiving infusional ECisF compared with AC.


View this table:
[in this window]
[in a new window]
 
Table 3. Treatment delays and dose reductions
 
Toxicity
Details of non-haematological toxicity for all 426 patients are given in Table 4. In general, both treatments were well tolerated but grade 3/4 toxicity occurred significantly more frequently with infusional ECisF than for AC for lethargy, nausea and vomiting, alopecia, thrombosis and plantar-palmar erythema; likewise, there was a trend towards increased grade 3/4 toxicity for mucositis and diarrhoea. Grade 3/4 Hickman line infections occurred in 16% of patients receiving infusional ECisF chemotherapy. Routine haematological toxicity data were not centrally collected on the basis that data on such toxicity have already been published for both schedules elsewhere. However, haematological toxicity data are available for 156 patients treated at the Royal Marsden Hospital, 77 receiving AC and 79 infusional ECisF. Grade 3/4 toxicity data on these patients are given in Table 5. The only significant difference that emerged between the two treatments was increased grade 3/4 white blood count suppression with AC compared with infusional ECisF (26% versus 10%; P = 0.01).


View this table:
[in this window]
[in a new window]
 
Table 4. Grade 3 or 4 non-haematological toxicity (%)
 

View this table:
[in this window]
[in a new window]
 
Table 5. Grade 3 or 4 haematological toxicity (%)
 

    Discussion
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
An encouraging number of new therapies for advanced breast cancer have emerged over the last few years and this trend is likely to continue. The large scale adjuvant trial is currently the standard method for assessing the efficacy of these in early breast cancer where they have the potential to achieve maximum benefit, but such trials take many years to complete. Neo-adjuvant/pre-operative trials of novel therapies offer a possible solution to this problem, if short-term surrogate markers can be shown to predict for long-term outcome. It is already established that clinical tumour response [5, 1821] and, even more so, pathological complete remission [5, 18, 19, 21] after neo-adjuvant chemotherapy predict for long-term survival. The next step is to establish from randomised trials whether differences in short-term response rates will predict for similar differences in disease-free survival and survival.

This schedule of continuous infusional 5-FU with cisplatin and epirubicin is an unusual one in breast cancer therapy and was based on several considerations. Single-agent continuous infusional 5-FU is an active and well tolerated regimen for previously treated patients with advanced breast cancer [22]: a series of small phase II studies have demonstrated an overall response rate of 29% [22], rising to 54% in one study [23], in patients previously treated with other forms of chemotherapy, including bolus 5-FU. Cisplatin, an ineffective drug in pretreated patients with advanced breast cancer, has nevertheless been shown to have activity of around 50% in three small trials of previously untreated patients [24].

The infusional ECisF combination was initially shown to have high activity in advanced gastric cancer and has recently been shown to have superior efficacy to conventional therapy in this tumour type [11]. Based on this, we found that the schedule was also very active against advanced breast cancer [9] and subsequently as pre-operative therapy [10].

Despite these encouraging initial results, and in contrast to results in advanced gastric cancer, this trial has shown no increase in clinical response rate or in pathological complete remissions for infusional ECisF compared with conventional AC. In this context our results with both infusional ECisF and with AC are very similar to results in the large National Surgical Adjuvant Breast and Bowel Project (NSABP) B-18 pre-operative chemotherapy trial, in which 749 patients were treated with pre-operative AC, with an overall response rate of 80%, and a complete response rate of 36% [2].

Despite the similarity in clinical response rates, the 3-year survival at first analysis was 90% for ECisF compared with 80% for conventional AC (P = 0.04) [12]. However, survival was not the defined end point for this analysis; only 55 of the required 256 deaths having been observed at this stage. At the current 5 year analysis the survival trend in favour of infusional ECisF continues (82% versus 74% for AC) but is not significant (hazard ratio 0.76, P = 0.18). The number of deaths so far is still well short of that required to see an absolute survival difference of 15%; in retrospect, this was probably too ambitious a difference but was based on the fact that the inconvenience of continuous infusional 5-FU with cisplatin could only be justified by a large difference. In this context, we have been conducting a ‘sister’ adjuvant trial of infusional ECisF versus conventional bolus 5-FU, epirubicin and cisplatin (FEC). These two trials will allow a combined survival analysis of around 800 patients with increased power to detect whether infusional 5-FU is indeed associated with a survival advantage. If not, then it is worth noting that toxicity in the infusional regimen caused significantly more dose reductions and treatment delays than with conventional chemotherapy, which could have prevented a potential therapeutic gain.

Meanwhile data are emerging from other randomised pre-operative trials which also test the hypothesis that pre-operative short-term markers are surrogate for long-term outcome. These include the NSABP B-27 three-arm trial comparing neo-adjuvant AC with neo-adjuvant AC followed by docetaxel before surgery, and with neo-adjuvant AC then surgery then docetaxel; a similar German trial compares neo-adjuvant dose-dense doxorubicin and docetaxel twice weekly using granulocyte colony-stimulating factor support with a sequential schedule of conventional 3 weekly AC followed by docetaxel. Preliminary results from both trials show significantly higher clinical responses and pathological complete remissions for AC followed by docetaxel [25, 26], but long-term outcome data are not yet available.

Several smaller comparative trials of pre-operative chemotherapy are also currently running. An Aberdeen group have reported that patients responding to four courses of cyclophosphamide, vincristine, doxorubicin and prednisolone, and subsequently randomised to four further courses of docetaxel, achieved a higher overall response rate and pathological complete remission rate than those receiving four further cycles of the same chemotherapy [27]. In two small trials neo-adjuvant single-agent paclitaxel was compared with FAC [5-FU, doxorubicin (Adriamycin), cyclophosphamide] [28], and a standard FEC schedule (5-FU, epirubicin and cyclophosphamide) was compared with a combination of epirubicin and docetaxel [29]; neither showed any significant differences in clinical or pathological response rates, but both were under-powered to detect survival differences, which are more likely to come from larger adjuvant and neoadjuvant trials addressing the same question.

Whether differences in clinical response can predict for long-term outcome differences remains to be seen. Our belief is that short-term biological end points, including changes in apoptosis or proliferation immediately after treatment [30, 31], may prove more sensitive surrogates. Changes in proliferation, in particular, are relatively easy to measure and future trials of pre-operative medical therapy should consider including this as an additional end point.


    Acknowledgements
 
Our thanks as always go to Alison Norton for her hard work, skill and wise advice in the preparation of this manuscript.

List of investigators: I. Smith, W. Allum, N. Sacks, G. Gui (No. of patients = 127), Royal Marsden/Epsom; A. Howell (n = 58), Christie, Manchester; S. Ebbs (n = 34), Mayday; T. Hickish, A. Skene (n = 33), Royal Bournemouth; M. O’Brien, F. McKinna, D. Pickering (n = 27), Kent Oncology Centre; J. Mansi (n = 20), St George’s, London; C. Wilson (n = 20), Addenbrooke’s, Cambridge; A. Robinson (n = 19), Southend; P. Murray, W. Pratt, N. Davidson (n = 18), Essex County, Colchester; C. Price, E. Whipp (n = 16), Bristol Oncology Centre; R. Laing, C. Topham (n = 11), St Luke’s Cancer Centre, Guildford; A. Jones (n = 9), Royal Free; T. Iveson (n = 7), Salisbury; J. Ledermann, R. Stein (n = 6), UCH, London; C. Gallagher (n = 4), St Bartholomew’s, London; J. Levay (n = 3), Ipswich. For the Yorkshire Breast Group: T. Perren (n = 12), St James’, Leeds; D. Dodwell (n = 2), Cookridge.


    FOOTNOTES
 
* Correspondence to: Prof. I. E. Smith, Royal Marsden Hospital, Fulham Road, London SW3 6JJ, UK. Tel: +44-20-7808-2751; Fax: +44-20-7352-5441; E-mail: ian.smith{at}rmh.nthames.nhs.uk Back


    REFERENCES
 Top
 ABSTRACT
 Introduction
 Patients and methods
 Results
 Discussion
 REFERENCES
 
1. Forrest AP, Levack PA, Chetty U et al. A human tumour model. Lancet 1986; 2: 840–842.[CrossRef][ISI][Medline]

2. Fisher B, Brown A, Mamounas E et al. Effect of preoperative chemotherapy on local-regional disease in women with operable breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-18. J Clin Oncol 1997; 15: 2483–2493.[Abstract]

3. Mauriac L, MacGrogan G, Avril A et al. Neoadjuvant chemotherapy for operable breast carcinoma larger than 3 cm: a unicentre randomised trial with a 124-month follow-up. Ann Oncol 1999; 10: 47–52.[Abstract]

4. Makris A, Powles TJ, Ashley SE et al. A reduction in the requirements for mastectomy in a randomized trial of neoadjuvant chemoendocrine therapy in primary breast cancer. Ann Oncol 1998; 9: 1179–1184.[Abstract]

5. Fisher B, Bryant J, Wolmark N et al. Effect of preoperative chemotherapy on the outcome of women with operable breast cancer. J Clin Oncol 1998; 16: 2672–2685.[Abstract]

6. Broet P, Scholl SM, de la Rochefordiere A et al. Short and long-term effects of survival in breast cancer patients treated by primary chemotherapy: an updated analysis of a randomized trial. Breast Cancer Res Treat 1999; 58: 151–156.[CrossRef][ISI][Medline]

7. Semiglazov VF, Topuzov EE, Bavli JL et al. Primary (neoadjuvant) chemotherapy and radiotherapy compared with radiotherapy alone in stage IIb–IIIa breast cancer. Ann Oncol 1994; 5: 591–595.[Abstract]

8. van der Hage JA, van de Velde CJ, Julien JP et al. Preoperative chemotherapy in primary operable breast cancer: results from the European Organization for Research and Treatment of Cancer trial 10902. J Clin Oncol 2001; 19: 4224–4237.[Abstract/Free Full Text]

9. Jones AL, Smith IE, O’Brien MER et al. Phase II study of continuous infusion fluorouracil with epirubicin and cisplatin in patients with metastatic and locally advanced breast cancer: an active new regimen. J Clin Oncol 1994; 12: 1259–1265.[Abstract]

10. Smith IE, Walsh G, Jones A et al. High complete remission rates with primary neoadjuvant infusional chemotherapy for large early breast cancer. J Clin Oncol 1995; 13: 424–429.[Abstract]

11. Waters JS, Norman A, Cunningham D et al. Long-term survival after epirubicin, cisplatin and fluorouracil for gastric cancer: results of a randomized trial. Br J Cancer 1999; 80: 269–272.[CrossRef][ISI][Medline]

12. Smith IE, A’Hern RP, Howell A et al. Preoperative continuous infusional ECisF (epirubicin, cisplatin and infusional 5-FU) versus conventional AC chemotherapy for early breast cancer: a phase III multicentre randomised trial by the TOPIC Trial Group. Proc Am Soc Clin Oncol 2000; 19: 84a (Abstr 320).

13. Calvert AH, Newell DR, Gumbrell LA et al. Carboplatin dosage: prospective evaluation of a simple formula based on renal function. J Clin Oncol 1989; 7: 1748–1756.[Abstract]

14. Ring A, Webb A, Ashley S et al. Is surgery necessary after complete clinical remission following neoadjuvant chemotherapy for early breast cancer? J Clin Oncol 2003; 21: 4540–4545.[Abstract/Free Full Text]

15. Miller AB, Hoogstraten B, Staquet M et al. Reporting results of cancer treatment. Cancer 1981; 47: 207–214.[ISI][Medline]

16. Kaplan EL, Meier P. Non-parametric estimation from incomplete observation. J Am Stat Assoc 1958; 54: 457–481.

17. Peto R, Pike MC, Armitage P et al. Design and analysis of randomised clinical trials requiring prolonged observation of each patient. II. Analysis and examples. Br J Cancer 1976; 35: 1–39.[ISI]

18. Cameron DA, Anderson EDC, Levack P et al. Primary systemic therapy for operable breast cancer: 10 year survival data after chemotherapy and hormone therapy. Br J Cancer 1997; 76: 1099–1105.[ISI][Medline]

19. Bonadonna G, Valagussa P, Brambilla C et al. Primary chemotherapy in operable breast cancer: eight year experience at the Milan Cancer Institute. J Clin Oncol 1998; 16: 93–100.[Abstract]

20. Pierga J-Y, Mouret E, Dieras V et al. Prognostic value of persistent node involvement after neoadjuvant chemotherapy in patients with operable breast cancer. Br J Cancer 2000; 83: 1480–1487.[CrossRef][ISI][Medline]

21. Smith IE, Lipton L. Preoperative/neoadjuvant medical therapy for early breast cancer. Lancet Oncol 2001; 2: 561–570.[CrossRef][Medline]

22. Cameron DA, Gabra H, Leonard RCF. Continuous 5-fluorouracil in the treatment of breast cancer. Br J Cancer 1994; 70: 120–124.[ISI][Medline]

23. Huan S, Pazdur R, Singhakowinta A et al. Low-dose continuous infusion 5-fluorouracil: evaluation in advanced breast cancer. Cancer 1989; 63: 419–422.[ISI][Medline]

24. Smith IE, Talbot DC. Cisplatin and its analogues in the treatment of advanced breast cancer: a review. Br J Cancer 1992; 65: 787–793.[ISI][Medline]

25. National Surgical Adjuvant Breast and Bowel Project. The effect on primary tumor response of adding sequential taxotere to adriamycin and cyclophosphamide: preliminary results from NSABP B-27. Breast Cancer Res Treat 2001; 69: 210 (Abstr 5).

26. von Minckwitz G, Raab G, Schuette M et al. Dose-dense versus sequential adriamycin/docetaxel combination as preoperative chemotherapy (pCHT) in operable breast cancer (T2-3, N0-2,M0): primary end point analysis of the GEPARDUO study. Proc Am Soc Clin Oncol 2002; 21: 43a (Abstr 168).

27. Smith IC, Heys SD, Hutcheon AW et al. Neoadjuvant chemotherapy in breast cancer: significantly enhanced response with docetaxel. J Clin Oncol 2002; 20: 1456–1466.[Abstract/Free Full Text]

28. Buzdar AU, Singletary SE, Theriault RL et al. Prospective evaluation of paclitaxel versus combination chemotherapy with fluorouracil, adriamycin and cyclophosphamide as neoadjuvant therapy in patients with operable breast cancer. J Clin Oncol 1999; 17: 3412–3417.[Abstract/Free Full Text]

29. Luporsi E, Vanlemmens L, Coudert F et al. Six cycles of FEC 100 vs epirubicin–docetaxel as neoadjuvant chemotherapy in operable breast cancer patients: preliminary results of a randomised phase II trial of GIREC S01. Proc Am Soc Clin Oncol 2000; 19: 92a (Abstr 355).

30. Archer CD, Parton M, Smith IE et al. Early changes in apoptosis and proliferation following primary chemotherapy for breast cancer. Br J Cancer 2003; 89: 1035–1041.[CrossRef][ISI][Medline]

31. Shao Z-M, Li J, Wu J et al. Neo-adjuvant chemotherapy for operable breast cancer induces apoptosis. Breast Cancer Res Treat 1999; 53: 263–269.[CrossRef][ISI][Medline]