NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Cancer (UK). Advanced Breast Cancer: Diagnosis and Treatment. Cardiff (UK): National Collaborating Centre for Cancer (UK); 2009 Feb. (NICE Clinical Guidelines, No. 81.)

Cover of Advanced Breast Cancer

Advanced Breast Cancer: Diagnosis and Treatment.

Show details

Chapter 4Systemic disease-modifying therapy

4.1. What is the choice of 1st line treatment for patients with metastatic breast cancer, endocrine therapy or chemotherapy?

Short summary

Only one paper was appraised for this topic. A high quality systematic review (Wilcken et al., 2006) examined ten RCTs of chemotherapy vs endocrine therapy, the most recent of which was published in 1995 (even though Cochrane databases were searched as recently as October 2006).

Chemotherapy and endocrine therapy were equal in terms of overall survival but tumour response was variable between studies. No data were presented for quality of life (QOL) or adverse events but, in narrative form, the reviewers stated that in the majority of studies chemotherapy had resulted in higher levels of toxicity (predominantly nausea, vomiting and alopecia) but that it was not clear in which direction QOL had been affected as the results were conflicting.

PICO question

POPULATIONINTERVENTIONCOMPARISONOUTCOME
Individuals with metastatic breast cancerAny conventional chemotherapyAny endocrine therapy
  • Tumour response
  • Time to progression
  • Time to treatment failure
  • Overall survival
  • Toxicity
  • QOL

NB The search strategy developed from this PICO table and used to search the literature for this question can be found in Appendix A

Full evidence statement

Only one paper was appraised for this topic. A high quality systematic review (Wilcken et al. 2006) examined ten RCTs of chemotherapy vs endocrine therapy, the most recent of which was published in 1995 (even though Cochrane databases were searched as recently as October 2006). The definition of chemotherapy included any conventional cytotoxic chemotherapy with or without colony stimulating factors but excluded cytokines, monoclonal antibodies and high-dose chemotherapies requiring stem cell support. Endocrine therapies included anti-oestrogens, oestrogens, androgens, aromatase inhibitors, progestagens and ovarian or adrenal ablations but excluded corticosteroids when given alone.

The analysis for overall survival comprised 692 women and indicated that there was no significant difference between chemotherapy and endocrine therapy (HR = 0.94 (95%CI: 0.79–1.12, P = 0.5). Survival after one or two years was, similarly, not significantly different between comparators (HR =1.03 (95%CI: 0.74–1.43) and HR = 0.98 (95%CI: 0.72–1.34) respectively).

In terms of tumour response to treatment (n = 817 patients), statistical analysis appeared to favour endocrine therapy (RR = 1.25 (95%CI: 1.01–1.54, P = 0.04)) but since the two major contributing studies presented findings in opposite directions, the between studies heterogeneity was significant and hence this result should be viewed with caution. However, since the more up-to-date chemotherapeutic and endocrine agents differ from those commonly used at the times when the included studies were published (i.e. pre-1995) these results may not accurately reflect the comparison between modern regimes anyway. It was the reviewers’ opinion that this might cause the value of endocrine therapy in improving survival to have been underestimated.

No data were presented for quality of life or adverse events but, in narrative form, the reviewers stated that in the majority of studies chemotherapy had resulted in higher levels of toxicity (predominantly nausea, vomiting and alopecia) but that it was not clear in which direction QOL had been affected as the results were conflicting.

The reviewers concluded that for women with MBC, in whom hormone receptors are present, a policy of treating first with endocrine therapy rather than chemotherapy is recommended except in the presence of rapidly progressive disease.

Reference

  1. Wilcken N, Hornbuckle J, Ghersi D. Chemotherapy alone versus endocrine therapy alone for metastatic breast cancer. Cochrane Database of Systematic Reviews. 2006 Reviews. [PubMed: 12804433]

Evidence table

Question: Endocrine therapy or chemotherapy as first line treatment?

Created by: Karen Francis on 22/05/2007

Wilcken et al. (2006)

Health Economic Summary

The GDG did not consider this topic a health economic priority; therefore the cost-effectiveness literature on this topic has not been reviewed.

4.2. What is the most effective hormone treatment for (1) women and (2) men with metastatic breast cancer?

4.2.1. Women with metastatic breast cancer

Short summary

The evidence base for this topic comprises one guideline (Eisen et al., 2004), five systematic reviews (Mauri et al., 2006; Gibson et al., 2007; Ferretti et al., 2006, Klijn et al., 2001 and Crump et al., 1997), five RCTs (Chia et al. 2008, Mouridsen et al. 2007, Taylor et al., 1998, Klijn et al., 2000 and Goss et al. 2007) a pooled analysis of RCT data (Howell et al., 2005) and a small, low quality comparative study (Catania et al. 2007). The number of study participants exceeded 32,000 women, the majority of whom were post-menopausal with metastatic breast cancer. Most of the papers were of moderate to high quality, although the guideline did review non-published abstracts.

Pre-menopausal women with metastatic breast cancer experienced no significant difference in tumour response or survival between ovarian ablation and tamoxifen as first-line therapy. Atamestane and toremifine as first-line combination therapy resulted in similar tumour response and survival compared with letrozole alone.

Fulvestrant and exemestane showed equivalent efficacy for women that had previously received non-steroidal AIs for the treatment of advanced breast cancer. However, the lack of a placebo comparator meant that it was not possible to assess the true clinical activity of either agent. Limited evidence also suggested that fulvestrant conferred short term benefit to heavily pre-treated women with metastatic disease by postponing the requirement for chemotherapy. An equivalence analysis of pooled data (Howell et al. 2005) from two trials showed that fulvestrant and anastrozole were not significantly different from one another in their effects on overall survival. Study participants given fulvestrant reported fewer incidences of joint pain.

Good evidence showed that there was significant clinical benefit, increased progression-free survival and ~13% reduction in the risk of death with third generation AIs compared with standard endocrine therapy (the analyses included all treatment lines). No individual AI was better than another in this regard. Very limited evidence suggested that there was no significant difference between the AIs and standard therapy in patient reported quality of life. However, more gastrointestinal symptoms and hot flushes were associated with AI therapy compared to standard endocrine therapy but there were fewer reports of blood clots and vaginal bleeding.

A moderate quality systematic review (Klijn et al. 2001) and meta-analysis of data from four RCTs (one unpublished) concluded that combination therapy with LHRH agonists, buserelin or goserelin, combined with tamoxifen produced significant improvements in tumour response, reduction in the risk of death (~22%) and disease progression (~30%) than LHRH agonist monotherapy. Lack of methodological detail suggests caution in the interpretation of these results.

One RCT (Klijn et al. 2000) compared buserelin alone versus tamoxifen alone versus the two agents combined. Tumour response was not significantly different between combined and monotherapies unless data from patients with stable disease for > 6 months was included. The re-analysis showed a superior response for the combined therapy compared with tamoxifen but not LHRH. Combined therapy significantly improved actuarial survival at 5 and 7 years, together with overall survival and progression-free survival compared with monotherapy with either buserelin or tamoxifen.

A second RCT (Taylor et al. 1998) compared goserelin with surgical ovarian ablation (ovariectomy). The authors found that the outcomes for tumour response, overall survival and failure free survival were not significantly different between treatments and concluded that either treatment could reasonably be offered to patients and their physicians. The study was terminated prematurely due to poor accrual, believed to be because of the unwillingness of patients to be randomised to the surgical arm.

PICO question

POPULATIONINTERVENTIONCOMPARISONOUTCOME
Women with metastatic breast cancer
  • Aromatase inhibitors
  • Progestagens
  • Stilboestrol
  • Testosterone
  • Ovarian ablation
  • LHRH agonists
Each with each other
  • Time to progression
  • Overall survival
  • Toxicity
  • QOL
  • Response rate

NB The search strategy developed from this PICO table and used to search the literature for this question can be found in Appendix A

Full evidence summary

The evidence base for this question comprises one guideline, five systematic reviews and three RCTs. The total number of participants exceeds 28,000 and the majority of women were post-menopausal and with metastatic breast cancer. Only the review examining the issue of ovarian ablation versus tamoxifen included patients who were pre-menopausal. All the reviews were of good quality.

Mauri et al. (2006) presented a high quality systematic review of 23 RCTs in which a total of 8,503 women participated. These women were described as having advanced breast cancer (locally advanced, recurrent or metastatic disease). There were no inclusion/exclusion criteria regarding existing endocrine or menopausal status The included studies compared all generations of aromatase inhibitors (aminoglutethimide, formestane, fadrozole, vorozole, letrozole, exemestane and anastrozole) with other endocrine therapies (tamoxifen, megestrol acetate and medroxyprogesterone acetate).

The only data used in the meta-analysis related to overall survival and the results were presented for each generation of aromatase inhibitor. This approach revealed that only the third generation AIs provided a statistically significant survival benefit (HR = 0.87 (95%CI: 0.82–0.93) P<0.001) and reduced mortality (HR = 0.91 (95%CI: 0.86–0.96) P = 0.001) when compared with standard endocrine therapy. There was no significant between studies heterogeneity (I2 = 5%). The authors concluded that even in first line treatment of MBC this group of AIs might be preferable to tamoxifen.

This was a thorough, high standard analysis but which may have invited an element of bias by excluding all but published, peer reviewed papers and by not formally scoring included studies for quality. Whilst this may have led to an overestimation of benefit in favour of the intervention, the lack of detail about intention-to-treat and cross-over effects within the included trials may conceal a dilution of such benefit so that it would be hard to say that, if bias existed, it would favour the intervention or comparator.

Another very good quality systematic (Cochrane) review (Gibson et al, 2007) analysed data from 30 RCTs in which 10,054 women participated. The patient group included only post-menopausal women with advanced or metastatic breast cancer but excluded those who only had locally advanced disease. Aromatase inhibitors (aminoglutethimide, formestane, anastrozole, exemestane, fadrozole, letrozole and vorozole) were matched against megestrol acetate, tamoxifen, fulvestrant, medroxyprogesterone acetate and hydrocortisone. The primary outcome was overall survival. Secondary outcomes of interest included progression-free survival, clinical benefit, tumour response and adverse events.

When all aromatase inhibitors were included in the analysis, there was a significant survival benefit compared with standard endocrine therapy (HR for OS = 0.89 (95%CI: 0.82–0.96) with no significant between studies heterogeneity (I2 = 0%). When third generation AIs were analysed separately there was also a significant survival benefit with no between studies heterogeneity (I2 = 0%) (HR = 0.88 (95%CI: 0.88–0.96) which was strikingly similar to the result obtained by Mauri et al. (2006). However, progression-free survival was not significantly enhanced by the third generation AIs (HR for PFS = 0.92 (95%CI: 0.75–1.13) (I2 = 88.3% P<0.0001). High levels of heterogeneity between studies may confound the apparently significant advantages in clinical benefit seen with current AIs (OR = 0.79 (95%CI: 0.64–0.96) I2 = 64%) or tumour response (OR = 0.78 (95%CI: 0.64–0.95) I2 = 47.7%) although a random effects model was used to analyse data where appropriate.

Adverse events were thoroughly analysed and reported but, once again, the findings were confounded by significant heterogeneity between studies such that the results would have to be regarded with caution. Generally, use of AIs was associated with higher incidences of nausea, vomiting, and hot flushes compared with megestrol acetate and of diarrhoea and rash when compared with tamoxifen, but fewer incidences of thromboembolic events and vaginal bleeding, depending on the particular comparator. The authors conclude that there is a considerable survival advantage, particularly with the more modern AIs compared with standard endocrine therapies. Unfortunately, between-studies heterogeneity does interfere with the interpretation of significant findings in favour of AIs for all outcomes but overall survival.

Howell et al. (2005) presented RCT data from a study of fulvestrant and anastrozole, designed not to show superiority of one agent over the other but to show equality in terms of survival outcomes. Patients were post-menopausal women with advanced or metastatic breast cancer with positive endocrine status and/or proven responsiveness to endocrine therapy. The study arms were shown to be of equivalence in terms of overall survival (HR = 0.98 (95%CI: 0.84–1.15) P=0.809) the authors having set an upper limit for the confidence interval of < 1.25 to prove their hypothesis. Data were pooled from two RCTs (trials 0020 and 0021) neither of which had, at the time of this data analysis, published their own survival analysis. The authors concluded that fulvestrant was at least as effective as anastrozole in terms of overall survival and patients had reported less joint disorders.

Ferretti et al. (2006) presented a systematic review of 7 trials (2,787 participants) which had compared AIs against standard endocrine therapy. The data analyses were by sub-group: non-steroidal AIs (fadrozole, letrozole, anastrozole and vorozole); all AIs (as preceding group plus formestane and exemestane) and third generation AIs only. The primary outcomes were tumour response rate and time to progression with secondary outcomes of overall survival and clinical benefit.

Unlike the systematic reviews by Mauri et al. (2006) and Gibson et al. (2007) the results of this analysis suggested that there was no significant survival benefit in treatment with AIs with regard to overall survival however the methodology employed to produce this result was different. Rather than selecting median survival times and using these to extrapolate the efficacy of intervention against comparator, Ferretti et al. (2006) extracted the proportion of patients surviving at a single (6 month) time point and combined these data from included studies. The result is therefore only relevant at that time point and does not inform about the efficacy of the intervention on survival overall. This methodology was also used to assess the time to progression but significant heterogeneity between the included studies negated the findings of this outcome, together with those for tumour response and clinical benefit. The analyses showed that third generation AIs conferred a significant benefit compared with standard endocrine therapy with regard to TTP and clinical benefit whether analysed using fixed- or random-effects models but that between-studies heterogeneity is still significant.

Klijn et al. (2001) presented a moderate quality systematic review and meta-analysis of data from 4 RCTs, one of which was unpublished and uncited. The included trials investigated the efficacy of LHRH agonists, goserelin or buserelin, with or without tamoxifen in the treatment of pre-menopausal women with advanced breast cancer. The analytical methodology was not adequately described; the effects model used (i.e. fixed or random) was not stated and between studies heterogeneity was not expressed as a Q or I2 statistic hence the validity of combining these particular studies is uncertain. The results suggested superior overall survival (HR = 0.78 (95%CI: 0.63–0.96) P=0.02) progression-free survival (HR = 0.70 (95%CI: 0.58–0.85) P=0.0003) and tumour response (OR = 0.67 (95%CI: 0.46–0.96) P=0.03) for the LHRH agonist + tamoxifen combination.

A systematic review by Crump et al. (1997) examined a comparison of tumour response and survival between ovarian ablation (surgical or by radiation) and tamoxifen for the first line treatment of metastatic breast cancer in pre-menopausal women. There were four included studies with a total of 220 participants. The meta-analysis was well prepared and the reviewers followed up participants and contacted study authors with the aim of data completion (by the time of their analysis there were ten survivors). The data collection and meta-analysis were thorough but the review itself was less than optimally systematic with little of the methodology explained. All study participants had estrogen +ve (or unknown) tumour status and were pre-menopausal with previously untreated metastatic disease.

There was a deliberate cross-over design of included studies in order to investigate whether the response to the first treatment would prove a prognostic factor, predicting the response to the other treatment administered following disease progression. Whilst there was no significant difference in tumour response, disease progression or survival between ovarian ablation and tamoxifen as first line therapies, the response to either of them was found to predict the subsequent response to the other. With this in mind the authors felt that there was no evidence to suggest that irreversible ovarian ablation, with attendant side effects, would be suitable as first line therapy but that the prior response to tamoxifen might predict its suitability as second line treatment.

The Cancer Care Ontario practice guideline ‘The role of aromatase inhibitors in the treatment of postmenopausal women with metastatic breast cancer’ was constructed around a high quality systematic review (Eisen et al., 2004). The patients and settings included in the trials were diverse and not reported in detail. The patient group included was specified as postmenopausal women with stage IV (metastatic) breast cancer. Abstract data were included in the meta-analysis of first line treatment, caution must be taken when considering this evidence because abstract evidence can be limited and left to interpretation by the reviewer.

There were 3 RCTs comparing anastrozole with tamoxifen, one of letrozole versus tamoxifen and 1 of exemestane versus tamoxifen as first-line therapy for metastatic breast cancer. Treatment with selective aromatase inhibitors was associated with higher objective response rates and prolonged time to progression compared to tamoxifen, but definitive survival and quality-of-life data were not available. The toxicity profile of the aromatase inhibitors was acceptable.

There were 3 RCTs comparing letrozole to megestrol acetate or aminoglutethimide, 2 of anastrozole versus megestrol acetate, and 1 of exemestane versus megestrol acetate as second-line hormonal therapy for metastatic breast cancer. Women eligible for these trials included those who relapsed during or were within 6 months of completion of adjuvant anti-estrogen therapy and those who progressed on first-line anti-estrogen therapy for metastatic disease. Treatment with selective aromatase inhibitors was associated with equivalent or better objective response rates and time to progression, and a superior toxicity profile, compared to megestrol acetate or aminoglutethimide. Two individual trials and a meta-analysis of individual-patient data from four trials detected a modest but statistically significant survival advantage for aromatase inhibitors, compared to control. There were no consistent differences in measures of quality of life between aromatase inhibitors and control therapy in randomized trials. There were no significant differences between doses of anastrozole of 1.0 and 10 mg, but two of three trials detected significantly higher survival rates with letrozole 2.5 mg compared to 0.5 mg.

A non-blinded RCT of letrozole versus anastrozole (abstract only) detected a statistically significant increase in response rate with letrozole compared to anastrozole as second-line treatment but no difference in time to progression. No survival or quality-of-life data are available from this trial.

Data from three phase II trials indicate that exemestane therapy, as third- or greater-line hormonal therapy, was associated with modest but appreciable rates of objective response and is well tolerated. There are no data from clinical trials of other aromatase inhibitors in this setting.

Taylor et al. (1998) presented data from a moderate quality RCT comparing the LHRH agonist goserelin with surgical ovariectomy for pre-menstrual women with metastatic breast cancer. Tumour response was not significantly different between studies (no odds ratio or P value given) and neither was goserelin shown to be superior to surgery with respect to overall survival (HR = 0.80 (95%CI: 0.53–1.20) P=0.006) or failure-free survival (HR = 0.73 (95%CI: 0.51–1.04). This trial was designed as an equivalence study i.e. to show that the arms were not significantly different from one another, hence the significant P values shown allow rejection of the alternate hypothesis i.e. 50% improvement between goserelin compared with surgery. Early termination of accrual, due to difficulty in recruiting women to be randomised to surgery, resulted in under-powering of the statistics. Nonetheless, the authors felt that both treatments were of equal efficacy and should be offered as a valid choice to patient and physician.

Klijn et al. (2000) provided details of a three-arm RCT comparing buserelin with tamoxifen and with buserelin + tamoxifen combined therapy for pre-menopausal women with locally advanced or metastatic breast cancer. Tumour response was not significantly different between the combined therapy and either LHRH alone (OR = 0.56 (95%CI: 0.24–1.30) nsd) or tamoxifen (OR = 0.42 (95%CI: 0.17–1.06) nsd). When data for patients with stable disease of more than 6 months duration was added to the data for tumour response, a significant clinical benefit was demonstrated when combined therapy was compared with tamoxifen (OR = 0.31 (95%CI: 0.127–0.753) no P value) but not LHRH alone. Combined therapy was also shown to result in significant improvements in overall survival (compared with tamoxifen: HR = 1.63 (95%CI: 1.03–2.59) P=0.029) (compared with LHRH: HR = 1.95 (95%CI: 1.23–3.1) P=0.006) and progression free survival (compared with tamoxifen: HR = 1.5 (95%CI: 1.01–2.24) P=0.047) (compared with LHRH: HR = 1.65 (95%CI: 1.09–2.49) P=0.008). Actuarial survival was also significantly higher for combined treatment at both 5 years and 7 years. Unfortunately, this study was underpowered (premature termination) and the authors gave no details of allocation, randomisation or of blinding or otherwise during tumour assessment. The possibility of bias is therefore strong and the study results and conclusions should be regarded with in mind.

References
  1. Crump M, Sawka CA, DeBoer G, Buchanan RB, Ingle JN, Forbes J, Meakin JW, Shelley W, Pritchard KI. An individual patient-based meta-analysis of tamoxifen versus ovarian ablation as first line endocrine therapy for premenopausal women with metastatic breast cancer (DARE structured abstract) Breast Cancer Res Treat. 1997;44:201–210. [PubMed: 9266099]
  2. Eisen A, Pritchard K, Johnston M, Oliver T. Role of aromatase inhibitors in the treatment of postmenopausal women with metastatic breast cancer. Curr Oncol. 2004;11(2):41–52.
  3. Ferretti G, Bria E, Giannarelli D, Felici A, Papaldo P, Fabi A, Di CS, Ruggeri EM, Milella M, Ciccarese M, Cecere FL, Gelibter A, Nuzzo C, Cognetti F, Terzoli E, Carlini P. Second- and third-generation aromatase inhibitors as first-line endocrine therapy in postmenopausal metastatic breast cancer patients: a pooled analysis of the randomised trials. Br J Cancer. 2006;94(12):1789–1796. [PMC free article: PMC2361349] [PubMed: 16736002]
  4. Gibson LJ, Dawson CL, Lawrence DJ, Bliss JM. Aromatase inhibitors for treatment of advanced breast cancer in postmenopausal women Cochrane Database of Systematic Reviews: Reviews, Cochrane Database of Systematic Reviews. 1. Vol. 2007. Chichester (UK): John Wiley & Sons, Ltd; 2007. [PubMed: 17253488]
  5. Howell A, Pippen J, Elledge RM, Mauriac L, Vergote I, Jones SE, Come SE, Osborne CK, Robertson JF. Fulvestrant versus anastrozole for the treatment of advanced breast carcinoma: a prospectively planned combined survival analysis of two multicenter trials. Cancer. 2005;104:236–239. [PubMed: 15937908]
  6. Klijn JGM, Beex LV, Mauriac L, van Zijl JA, Veyret C, Wildiers J, Jassem J, Piccart M, Burghouts J, Becquart D, Seynaeve C, Mignolet F, Duchateau L. Combined treatment with buserelin and tamoxifen in premenopausal metastatic breast cancer: a randomized study. J Natl Cancer Inst. 2000;92:903–911. [PubMed: 10841825]
  7. Klijn JGM, Blamey RW, Boccardo F, Tominaga T, Duchateau L, Sylvester R. Combined Hormone Agents Trialists’ Group and the European Organization for Research and Treatment of Cancer. Combined tamoxifen and luteinizing hormone-releasing hormone (LHRH) agonist versus LHRH agonist alone in premenopausal advanced breast cancer: a meta-analysis of four randomized trials.[see comment] J Clin Oncol. 2001;19:343–353. [PubMed: 11208825]
  8. Mauri D, Pavlidis N, Polyzos NP, Ioannidis JP. Survival with aromatase inhibitors and inactivators versus standard hormonal therapy in advanced breast cancer: meta-analysis. J Natl Cancer Inst. 2006;98(18):1285–1291. [PubMed: 16985247]
  9. Taylor CW, Green S, Dalton WS, Martino S, Rector D, Ingle JN, Robert NJ, Budd GT, Paradelo JC, Natale RB, Bearden JD, Mailliard JA, Osborne CK. Multicenter randomized clinical trial of goserelin versus surgical ovariectomy in premenopausal patients with receptor-positive metastatic breast cancer: an intergroup study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 1998;16:994–999. [PubMed: 9508182]

Evidence tables

Question: What is the most effective hormone treatment for women with metastatic breast cancer?

Created by: Karen Francis on 27/06/2007

Klijn et al. (2001)

Included studies in Klijn et al. (2001)

1.
Jonat W, Kaufmann M, Blamey RW, et al. A randomised study to compare the effect of the luteinising hormone releasing hormone (LHRH) agonist goserelin with or without tamoxifen in pre- and peri-menopausal patients with advanced breast cancer. Eur J Cancer. 1995;31:137–142. [PubMed: 7718316]
2.
Klijn JGM, Beex, Mauriac L, et al. Combined treatment with buserelin and tamoxifen in premenopausal metastatic breast cancer: a randomised study. J Natl Cancer Inst. 2000;92:903–911. [PubMed: 10841825]
3.
Unpublished study (no citation)
4.
Boccardo f, Rubagotti A, Perrotta A, et al. Ovarian ablation versus goserelin with or without tamoxifen in premenopausal patients with advanced breast cancer: results of a multicentric Italian study. Ann Oncol. 1994;5:337–342. [PubMed: 8075030]

Klijn et al. (2000)

Howell et al. (2005)

Taylor et al. (1998)

Crump et al. (1997)

Mauri D et al. (2006)

Gibson et al. (2007)

Ferretti et al. (2006)

Eisen A et al. (2004)

References included in the Eisen A et al. (2004) systematic review

  1. Bonneterre J, Buzdar A, Nabholtz J-M, et al. Anastrozole is superior to tamoxifen as first-line therapy in hormone receptor positive advanced breast carcinoma. Cancer. 2001;92:2247–58. [PubMed: 11745278]
  2. Bonneterre J, Thurlimann B, Robertson JF, et al. Anastrozole versus tamoxifen as first- line therapy for advanced breast cancer in 668 postmenopausal women: results of the tamoxifen or arimidex randomized group efficacy and tolerability study. J Clin Oncol. 2000;18:3748–57. [PubMed: 11078487]
  3. Buzdar A, Douma J, Davidson N, et al. Phase III, multicenter, double-blind, randomized study of letrozole, an aromatase inhibitor, for advanced breast cancer versus megestrol acetate. J Clin Oncol. 2001;19:3357–66. [PubMed: 11454883]
  4. Buzdar AU, Jonat W, Howell A, et al. Anastrozole, a potent and selective aromatase inhibitor, versus megestrol acetate in of postmenopausal women with advanced breast cancer: results of overview analysis of two phase III trials. J Clin Oncol. 1996;14:2000–11. [PubMed: 8683230]
  5. Buzdar AU, Jones SE, Vogel CL, Wolter J, Plourde P, Webster A. for the Arimidex Group. A phase III trial comparing anastrozole (1 and 10 milligrams), a potent and selective aromatase inhibitor, with megestrol acetate in postmenopausal women with advanced breast carcinoma. Cancer. 1997;79:730–9. [PubMed: 9024711]
  6. Dirix L, Piccart MJ, Lohrisch C, et al. Efficacy of and tolerance to exemestane versus tamoxifen in 1st line hormone therapy of postmenopausal metastatic breast cancer patients: A European Organisation for the Research and Treatment of Cancer (EORTC Breast Group) phase II trial with Pharmacia and Upjohn [abstract] Proc Am Soc Clin Oncol. 2001 abstract #114.
  7. Dombernowsky P, Smith I, Falkson G, et al. Letrozole, a new oral aromatase inhibitor for advanced breast cancer: double-blind randomized trial showing a dose effect and improved efficacy and tolerability compared with megestrol acetate. J Clin Oncol. 1998;16:453–61. [PubMed: 9469328]
  8. Gershanovich M, Chaudri HA, Campos D, et al. Letrozole, a new oral aromatase inhibitor: randomised trial comparing 2.5 mg daily, 0.5 mg daily and aminoglutethimide in postmenopausal women with advanced breast cancer. Ann Oncol. 1998;9:639–45. [PubMed: 9681078]
  9. Jonat W, Howell A, Blomqvist C, et al. A randomised trial comparing two doses of the new selective aromatase inhibitor anastrozole (Arimidex) with megestrol acetate in postmenopausal patients with advanced breast cancer. Eur J Cancer. 1996;32A:404–12. [PubMed: 8814682]
  10. Jones S, Vogel C, Arkhipov A, et al. Multicenter, phase II trial of exemestane as third-line hormonal therapy of postmenopausal women with metastatic breast cancer. Aromasin Study Group. J Clin Oncol. 1999;17:3418–25. [PubMed: 10550136]
  11. Kaufmann M, Bajetta E, Dirix LY, et al. Exemestane is superior to megestrol acetate after tamoxifen failure in postmenopausal women with advanced breast cancer: results of a phase III randomized double-blind trial. J Clin Oncol. 2000;18:1399–411. [PubMed: 10735887]
  12. Lonning PE, Bajetta E, Murray R, et al. Activity of exemestane in metastatic breast cancer after failure of nonsteroidal aromatase inhibitors: a phase II trial. J Clin Oncol. 2000;18:2234–44. [PubMed: 10829043]
  13. Messori A, Cattel F, Trippoli S, Vaiani M. Survival in patients with metastatic breast cancer: analysis of randomized studies comparing oral aromatase inhibitors versus megestrol. Anticancer Drugs. 2000;11:701–6. [PubMed: 11129731]
  14. Milla-Santos A, Milla L, Rallo L, Solano V. Anastrozole vs. tamoxifen in hormone dependent advanced breast cancer. A phase II randomized trial [abstract] Breast Cancer Res Treat. 2000;63 abstract #173.
  15. Milla-Santos A, Milla L, Rallo L, Solano V. Phase III trial of anastrozole vs. tamoxifen in post menopausal patients with hormone-dependent advanced breast cancer [abstract] Eur J Cancer. 2001;37(suppl 5):4.
  16. Mouridsen H, Gershanovich M, Sun Y, et al. Superior efficacy of letrozole versus tamoxifen as first-line therapy for postmenopausal women with advanced breast cancer: results of a phase III study of the International Letrozole Breast Cancer Group. J Clin Oncol. 2001;19:2596–606. [PubMed: 11352951]
  17. Mouridsen H, Sun Y, Gershanovich M, et al. First-line therapy with letrozole (Femara) for advanced breast cancer prolongs time to worsening of Karnofsky Performance Status compared with tamoxifen [abstract] Breast Cancer Res Treat. 2001;69 abstract #458.
  18. Nabholtz JM, Buzdar A, Pollak M, et al. Anastrozole is superior to tamoxifen as first-line therapy for advanced breast cancer in postmenopausal women: results of a North American multicenter randomized trial. J Clin Oncol. 2000;18:3758–67. [PubMed: 11078488]
  19. Rose C, Vtoraya O, Pluzanska A, et al. Letrozole (Femara) vs. anastrozole (Arimidex): second-line treatment in postmenopausal women with advanced breast cancer [abstract] Proc Am Soc Clin Oncol. 2002;22 abstract #131.
  20. Thurlimann B, Paridaens R, Serin D, et al. Third-line hormonal treatment with exemestane in postmenopausal patients with advanced breast cancer progressing on aminoglutethimide: a phase II multicentre multinational study. Exemestane Study Group. Eur J Cancer. 1997;33:1767–73. [PubMed: 9470830]
  21. Weinfurt KP, Wait SL, Boyko W, Schulman KA. Psychosocial quality of life in a phase III trial of letrozole [abstract] Proc Am Soc Clin Oncol. 1998;17:108a.
  22. From updated search

    1. Mouridsen H, Gershanovish M, Sun Y, Perez-Carrion R, Boni C, Monnier A, et al. Phase III study of letrozole versus tamoxifen as first-line therapy of advanced breast cancer in postmenopausal women: Analysis of survival and update of efficacy from the international letrozole breast cancer group. J Clin Oncol. 2003;21:2101–9. [PubMed: 12775735]
    2. Nabholtz JM, Bonneterre J, Buzar A, Robertson JFR, Thurlimann B. Anastrozole (Armidex) versus tamoxifen as first-line therapy for advanced breast cancer in postmenopausal women: survival analysis and updated safety results. Eur J Cancer. 2003;39:1684–9. [PubMed: 12888362]

Updated evidence (4.2)

Summary

Three RCTs (Chia et al. 2008, Mouridsen et al. 2007 and Goss et al. 2007) and a small comparative study (Catania et al. 2007) were identified to update the evidence on endocrine therapy for metastatic breast cancer. The total patient number across all trials was 2,522.

Chia et al. (2008) presented RCT data from the EFECT study, comparing i.m. fulvestrant with oral exemestane in women whose metastatic breast cancer had progressed on non-steroidal AI therapy. The authors concluded that the drugs had shown equivalent clinical benefit (~32% OR = 1.03 (95%CI: 0.72–1.487) in a good proportion of women, were both well tolerated with similar adverse events and quality of life reported. Since the route of administration differs in the two treatments it was suggested that patients may make a choice of therapy on this factor.

Mouridsen et al. (2007) described a large phase III double blinded, double dummy RCT comparing letrozole (LET) with tamoxifen (PO25 study) as first line therapy for women with advanced and/or metastatic breast cancer. LET significantly improved OS vs TAM over the first 24 months. The advantage was maximal at 14 months and significant for several time points beforehand (every 6 months to 24 months). However, at 36 months the advantage was lost since the survival curves crossed at this point, a phenomenon that could not be explained with certainty. Time to progression was superior in the LET arm (HR = 0.72 (P<0.0001)). The authors concluded that LET had demonstrated overall superiority over tamoxifen as first line therapy and with few reported adverse events.

Goss et al. (2007) presented a high quality paper which described a double blinded, double dummy phase III trial comparing atamestane + toremifine with letrozole in the first line treatment of advanced and/or metastatic breast cancer. There were no significant differences between study arms in any measured outcome including time to progression, response, survival or adverse events. The authors concluded that the combined therapy could not be supported in preference to letrozole but suggested that more potent SERMs and selective ER down-regulators might be combined with aromatase inhibitors to provide a more complete oestrogen blockade.

Catania et al. (2007) described a small comparative, but non-randomised, study in which women with metastatic breast cancer received fulvestrant, either after disease progression on a previous hormone therapy or as maintenance therapy after chemotherapy. All women had been heavily pre-treated and received only small benefit from this additional therapy (median TTP was 3 months across all participants). The authors felt, however, that fulvestrant could safely be used in such patients in order to lengthen the time before chemotherapy would be required and hence might improve the quality of life for that short period.

References
  1. Catania C, Ascione G, Adamoli L, De Pas T, Medici M, Franceschelli L, Verri E, Magni E, Sanna G, Torrisi R, Goldhirsch A, Nole F. Fulvestrant in heavily pre-treated patients with advanced breast cancer: results from a single compassionate use programme centre. Breast Cancer Res Treat. 2007;106:97–103. [PubMed: 17260095]
  2. Chia S, Gradishar W, Mauriac L, Bines J, Amant F, Federico M, Fein L, Romieu G, Buzdar A, Robertson JFR, Brufsky A, Possinger K, Rennie P, Sapunar F, Lowe E, Piccart M. Double-blind, Randomized placebo controlled trial of fulvestrant compared with exemestane after prior nonsteroidal aromatase inhibitor therapy in postmenopausal women with hormone receptor-positive, advanced breast cancer: Results from EFECT 19 28. J Clin Oncol. 2008;26:1664–1670. [PubMed: 18316794]
  3. Goss P, Bondarenko IN, Manikhas GN, Pendergrass KB, Miller WH, Langecker P, Blanchett D. Phase III, double-blind, controlled trial of atamestane plus toremifene compared with letrozole in postmenopausal women with advanced receptor-positive breast cancer. J Clin Oncol: official journal of the American Society of Clinical Oncology. 2007;25:4961–4966. [PubMed: 17971594]
  4. Mouridsen HT. Letrozole in advanced breast cancer: the PO25 trial. Breast Cancer Res Treat. 2007;105(Suppl 1):19–29. [PMC free article: PMC2001219] [PubMed: 17333340]
Evidence tables

Question: What is the most effective hormone treatment for metastatic breast cancer?

Created by: Karen Francis on 24/06/2008

Chia et al. (2008)

Mouridsen (2007)

Goss et al. (2007)

Catania et al. (2007)

Health Economic Summary

Evidence Summary

This question yielded a relatively large evidence base so the review criteria were tightened to include those studies that were most relevant to the decision problem; thus only studies taken from the perspective of the UK NHS were reviewed. A total of five studies met the stricter inclusion criteria from an initial search which identified 358 papers. No additional papers were identified in an update search. None of the economic evaluations compared hormone therapy with a ‘do-nothing’ alternative, probably due to the fact that hormone therapy in postmenopausal women with advanced breast cancer is standard clinical practice. Neither did any of the evaluations compare all the relevant interventions against each other.

The three older studies evaluate various third-generation aromatase inhibitors (AIs) against megestol as second-line treatment which was the standard hormone therapy at the time. The more recent studies evaluate letrozole against tamoxifen as first-line treatment, in line with current clinical practice.

All studies presented cost-effectiveness analyses (results in terms of cost per life years gained) and the two Karnon papers also presented cost-utility analyses (results in terms of cost per QALYs gained). Since we are investigating the use of AIs in the treatment of patients with advanced breast cancer, a consideration of quality of life is particularly important.

All studies used modelling techniques to model the decision problem over a lifelong time horizon. This meant included the costs and health benefits associated with subsequent treatment. All papers used RCTs to inform the clinical data and costs from nationally published sources. The Karnon and Jones (2003) and the Nuijten (1999) analysis used a similar model structure that was more comprehensive than the other models, using a Markov process and allowing for various clinical pathways subsequent to hormone treatment. Expert opinion was ascertained using formal methods of elicitation in these studies.

None of the studies used the current discounting recommendation of 3.5% for both health benefits and costs; many of the studies used differential discount rates. By using a lower discount rate for health benefits these studies will have overestimated future health benefits of the interventions which would result in higher incremental cost effectiveness ratios than have been reported. However since the time horizon is not long (lifetime perspective yet never more than 6 years) this effect is not likely to change the conclusions from the studies.

All baseline ICERs for the comparison between letrozole or anastrozole and tamoxifen were below £5,075 per life year gained and £9,200 per QALY. Similar results were obtained for letrozole, anastrozole or exemestane versus megestrol with a maximum ICER of £9,667 per life year. All of these results were tested to varying degrees of sophistication with sensitivity analysis and were robust to all scenarios presented. However a major limitation of the studies was that all were supported by the pharmaceutical industry. Since not all assumptions were tested, bias from this source cannot be ruled out. In addition none of the studies compared third-generation aromatase inhibitors against each other, so there is no evidence as to which AI is most cost-effective, in either the first- or second-line setting.

An independent analysis would be useful, especially if it incorporated indirect comparison methods to compare all the interventions of interest against each other. This was not undertaken as part of the economic work for this guideline since it was felt that the evidence showed all the baseline ICERs for new AIs in first- or second-line fall within an acceptable level of cost-effectiveness; thus independent modelling on this topic was not considered a high priority.

References
  1. Drummond M, Thompson E, Howell A, Jonat W, Buzdar A, Brown J. Cost-effectiveness implications of increased survival with anastrozole in the treatment of advanced breast cancer. Journal of Medical Economics. 1999;2:33–43.
  2. Karnon J, Johnston SR, Jones T, Glendenning A. A trial-based cost-effectiveness analysis of letrozole followed by tamoxifen versus tamoxifen followed by letrozole for postmenopausal advanced breast cancer. Ann Oncol. 2003;14(11):1629–33. [PubMed: 14581270]
  3. Karnon J, Jones T. A stochastic economic evaluation of letrozole versus tamoxifen as a firstline hormonal therapy: For advanced breast cancer in postmenopausal patients. Pharmacoeconomics. 2003;21(7):513–525. [PubMed: 12696991]
  4. Lindgren P, Jonsson B, Redaelli A, Radice D. Cost-effectiveness analysis of exemestane compared with Megestrol in advanced breast cancer - A model for Europe and Australia. Pharmacoeconomics. 2002;20(2):101–108. [PubMed: 11888362]
  5. Nuijten M, Meester L, Waibel F, Wait S. Cost effectiveness of letrozole in the treatment of advanced breast cancer in postmenopausal women in the UK. Pharmacoeconomics. 1999;16(4):379–397. [PubMed: 10623366]
Evidence Tables

Question: In patients with metastatic breast cancer, which hormone treatment is most cost-effective?

Created by: Sarah Willis and Christoph Rollig on 11/09/07 (updated by Sarah Willis on 16/07/08)

Karnon and Jones et al. (2003)

Karnon & Johnston et al. (2003)

Lindgren et al. (2002)

Drummond et al. (1999)

Nuijten et al. (1999)

Drummond Checklist (of Included Studies)

4.2.2. Men with metastatic breast cancer

Short summary

Three papers (Kantarjian et al. 1983, Patel et al. 1984 and Lopez et al. 1985a) presented case series of men who had received a great variety of endocrine therapies, including surgery. None of the treatments were highlighted for specific analysis and the numbers of each patient sub-group are too low to make a summary of any value.

Otherwise, there were eight retrospective case series (El Omari-Alaoui 2002, Giordano 2002, Harris et al. 1986, Lopez 1985b & 1993, Patterson et al. 1980 and Ribeiro 1976 & 1983) which reviewed data from case files of male patients treated for breast cancer. The papers spanned nearly three decades and involved 321 males - four papers were from the United Kingdom. None of the studies were comparative and, although of low quality, represent probably the best available evidence on this topic.

Very limited evidence (n=5) (Harris et al. 1986) suggested that aminoglutethimide may be suitable therapy for men with advanced breast cancer who have been previously orchidectomised. Diethylstilboestrol therapy was effective for men with soft tissue disease but failed to elicit a significant tumour response in those with more widespread metastatic breast cancer.

Limited evidence suggests that cyproterone was an effective therapy in some men but there were no factors by which response could be predicted and the treatment resulted in impotence and loss of libido for many patients. Androgen blockade with buserelin did not appear to enhance the response but may have prevented response flare. A very limited case series (n=5) (Harris et al. 1986) showed that anastrazole therapy did not result in a positive response in ER-positive males with metastatic breast cancer.

Two poor quality studies (Ribeiro 1983 and Patterson et al. 1980) reviewed data on treatment with tamoxifen. Some patients were included in both studies. The authors reported objective response rates from 37.5% to 48% and response duration from 1 month to 5 years. Where endocrine status was known, only the ER-positive sub-group was associated with favourable tumour response. Few adverse events were reported.

PICO question

POPULATIONINTERVENTIONCOMPARISONOUTCOME
Men with metastatic breast cancer
  • Aromatase inhibitors
  • GnRH agonists
  • Stilboestrol
  • Anti-androgens
  • Orchidectomy
Each with each other
  • Time to progression
  • Overall survival
  • Toxicity
  • QOL
  • Response rate

NB The search strategy developed from this PICO table and used to search the literature for this question can be found in Appendix A

Evidence summary

There were eleven retrospective case series (El Omari-Alaoui 2002, Giordano 2002, Harris et al. 1986, Kantarjian et al. 1983, Lopez 1985a, 1985b & 1993, Patel et al. 1984, Patterson et al. 1980 and Ribeiro 1976 & 1983) which reviewed data from case files of male patients treated for breast cancer. The papers spanned nearly three decades and involved 292 males - four papers were from the United Kingdom. None of the studies were comparative and, although of low quality, represent probably the best available evidence on this topic.

In 1976, Ribeiro presented data from 58 men with advanced breast cancer who were given first line oral diethylstilboestrol at three different doses at a single UK centre between 1942 and 1972. An objective response rate of 38% was seen in men who had predominantly soft tissue disease. The response duration for complete responders was 7 years with a median survival of 6 years and 10 months. Non-responders were men with widespread metastatic disease who had a median OS of just over one year.

Ribiero (1983) and Patterson et al. (1980) retrospectively reviewed data from men with breast cancer who had been given 20 mg to 40 mg tamoxifen per day for unknown durations. Nine of the men were included in both studies. The authors reported objective response rates of 37.5% and 48% respectively with response duration in a very wide range, between one month and five years. Survival outcomes were not reported in either paper. Few tumours were assessed for endocrine status but in those men in which this had been undertaken, only ER +ve status was associated with tumour response. Few adverse events were reported.

Harris et al. (1986) presented five case reports on men with advanced breast cancer who were given aminoglutethimide. One man, the only subject who had been orchidectomised, responded positively to the treatment for 14 months. The other four men failed to respond and yet, although their oestrogen levels were not reduced by this aromatase inhibitor, they all subsequently responded to tamoxifen.

Lopez et al. (1985b) reported on a small series of men (n=10) with recurrent or progressive breast cancer who were given cyproterone twice a day for an unknown duration. The objective response rate was 70% and the median duration for responders was 8 months. The anti-androgenic activity of the drug resulted in impotence and there were also reports of gynaecomastia, weight gain and tiredness. The response to treatment was not predicted by any identifiable factor, including hormone levels since response did not always correlate with measured testosterone, oestradiol or gonadotropins. Lopez et al. (1993) presented further case reviews in which men (n=11) had received cyproterone with a GnRH analogue, buserelin until disease progression. The objective response rate was 63.6% with median response duration for responders of 11.5 months. Even in those men with stable disease, pain and symptoms were alleviated. Median overall survival was 18.5 months. Both drugs were well tolerated but, again, impotence and loss of libido were commonly reported. The tumour responses were independent of measured hormone levels and the relationship between therapy and endocrine status was not explored (although five men with a positive response were ER +ve/ PR −ve).

Giordano et al. (2002) reported on five men who had received anastrazole for the treatment of metastatic breast cancer. Each case was discussed in narrative synthesis. None of the men experienced an objective response although 3/5 had a temporary stabilisation of disease and two men derived a clinical benefit (stable disease > 24 weeks). All men were ER +ve.

Three groups (Kantarjian et al. 1983, Patel et al. 1984 and Lopez et al. 1985a) presented data from men who had received various therapies for advanced breast cancer including orchidectomy, adrenalectomy, hypophysectomy, aminoglutethimide, androgens, oestrogens, corticosteroids, medroxyprogesterone, tamoxifen and stilboestrol. Individual therapies were not discussed but rather endocrine therapy as a whole was assessed for its effect on survival. In most cases the number of patients receiving any one treatment is so low that summarizing the outcomes is not meaningful.

References
  1. El Omari-Alaoui H. Male breast cancer. A report of 71 cases. Cancer/Radiotherapie. 2002;6:349–351. [PubMed: 12504771]
  2. Giordano SH. Efficacy of anastrozole in male breast cancer. Am J Clin Oncol. 2002;25:235–237. [PubMed: 12040279]
  3. Harris AL, Dowsett M, Stuart-Harris R, Smith IE. Role of aminoglutethimide in male breast cancer. Br J Cancer. 1986;54:657–660. [PMC free article: PMC2001494] [PubMed: 3778807]
  4. Kantarjian H, Yap HY, Hortobagyi G, Buzdar A, Blumenschein G. Hormonal therapy for metastatic male breast cancer. Arch Intern Med. 1983;143:237–240. [PubMed: 6824391]
  5. Lopez M. Cyproterone acetate in the treatment of metastatic cancer of the male breast. Cancer. 1985;55:2334–2336. [PubMed: 3986737]
  6. Lopez M, Di LL, Lazzaro B, Papaldo P. Hormonal treatment of disseminated male breast cancer. Oncology. 1985;42:345–349. [PubMed: 2933617]
  7. Lopez M. Combined treatment with buserelin and cyproterone acetate in metastatic male breast cancer. Cancer. 1993;72:502–505. [PubMed: 8319180]
  8. Patel JK, Nemoto T, Dao TL. Metastatic breast cancer in males. Assessment of endocrine therapy. Cancer. 1984;53:1344–1346. [PubMed: 6198069]
  9. Patterson JS, Battersby LA, Bach BK. Use of tamoxifen in advanced male breast cancer. Cancer Treat Rep. 1980;64:801–804. [PubMed: 7427964]
  10. Ribeiro GG. The results of diethylstilboestrol therapy for recurrent and metastatic carcinoma of the male breast. Br J Cancer. 1976;33:465–467. [PMC free article: PMC2025078] [PubMed: 1268034]
  11. Ribeiro GG. Tamoxifen in the treatment of male breast carcinoma. Clin Radiol. 1983;34:625–628. [PubMed: 6673881]

4.3. Combination versus sequential or single chemotherapy regimes

4.3.1. Combination vs sequential chemotherapy

Short summary

Evidence for comparing single chemotherapy with sequential chemotherapy comprised five RCTs (Creech et al., 1979, Chlebowski et al., 1979, Sledge et al., 2003, Smalley et al., 1976 and Baker et al., 1974) and one observational study (Chlebowski et al., 1989). The older studies were not always very stringently reported.

Two small, poor quality trials (Baker et al. 1974 and Creech et al. 1979) found no significant difference in tumour response, response duration, time to progression or overall survival when chemotherapy agents were given together or sequentially (on disease progression). Two other studies (Chlebowski et al. 1979 and Smalley et al. 1976) and a retrospective analysis (Chlebowski et al. 1989) of their data showed that whilst combined therapy resulted in superior tumour response and apparently significantly longer median overall survival, follow-up revealed that long term survival was no different between study arms.

One large RCT (Sledge et al. 2003) demonstrated that combining anthracycline and taxane, rather than giving the drugs sequentially in either order, resulted in a better tumour response and superior time to progression but did not improve median overall survival.

Consistently, adverse events due to combined therapy were reported as being more numerous or of greater severity than those experienced with single agents.

PICO question

POPULATIONINTERVENTIONCOMPARISONOUTCOME
Patients with metastatic breast cancer receiving chemotherapy Same agents in combination vs sequentially
  • Time to progression
  • Overall survival
  • Response rate
  • Toxicity

NB The search strategy developed from this PICO table and used to search the literature for this question can be found in Appendix A

Full evidence summary

The majority of papers identified from keywords report studies that assess the benefits of single drug interventions when compared with drug combinations e.g. drug A versus drugs A+B (but do not assess the impact of drug A followed by drug B at disease progression versus drugs A+B). These studies often have no control over subsequent treatment given after removal from study and so take the therapy under examination out of the context of the whole patient pathway. These papers are therefore excluded (including the very good Cochrane systematic review by Carrick et al. (2005) ‘Single Agent versus Combination Chemotherapy for Metastatic Breast Cancer’).

The remaining evidence base consists of five RCTs and one observational study. The studies in themselves are not bad in design but rather less than stringent in reporting the statistical analyses, randomisation and other features that are expected of more contemporary studies. Because the lack of such information carries with it the possibilities of introduced bias they are graded 1- or 1. A brief summary of outcomes are shown in table 4.3.1.1 below.

Baker et al. (1974) reported a small RCT comparing combined therapy of 5′-fluorouracil (5′-FU) cyclophosphamide, and vincristine with the same agents given sequentially on disease progression. The quality of the paper was poor for several reasons most of which relate to the likelihood of bias in, for example, the allocation of patients or the lack of an intent to treat analysis (although some patients’ data had been excluded from the results due to death from toxicity). The authors reported that there were no significant differences between the two regimes with respect to tumour response, the duration of that response or of mean survival. Unfortunately these conclusions were not supported with statistical evidence. Since patients in the combination arm had a higher incidence of visceral disease and so perhaps a poorer prognosis, the outcomes of combined treatment may possibly have been undervalued in this trial.

Creech et al. (1979) presented a small RCT of patients randomised to receive four chemotherapeutic drugs in combination (5′-fluorouracil (5′-FU) cyclophosphamide, methotrexate and adriamycin – CAMF) or four of the drugs (CMF) in combination with the fifth (adriamycin - A) given on disease progression. For the purposes of analysis the patients were also stratified into high and low risk categories according to ambulatory status, symptoms and degree of visceral disease. The tumour response was not significantly different between arms or between high and low risk groups in either arm. The statistics in support of these findings were not reported. For those patients responding to therapy in either treatment arm, there was no significant difference in time to disease progression or in median overall survival between the CAMF and CMF + A responders. Outcomes were unsurprising since it was found that patients who responded to treatment or who were stable survived for longer than patients who failed to respond and patients in high risk groups had poorer survival than low risk patients, regardless of treatment arm. The main conclusion was that the side effects outweighed the advantages, if any of combined therapy with adriamycin but that this drug would be better left as a response to treating disease progression after initial therapy.

Sledge et al. (2003) provides the only recent publication relating to this question. The RCT was substantial (n = 731) in size and patients were randomised into three groups receiving either doxorubicin and paclitaxel together (A+T) doxorubicin until progression then paclitaxel (A)or paclitaxel until disease progression then doxorubicin (T). Unfortunately the randomisation method was not discussed which means that bias cannot be excluded thus downgrading the quality of the paper to some degree and data were not well presented but, in other respects, the study appears to have been thorough. The tumour response rate and the time to treatment failure were statistically significantly lower between either of the sequential arms when compared with the combined therapy but did not differ from each other. There were no significant differences between the length of overall survival between arms, however some of the treatment-related effects appear to have been more severe in the combination arm and the authors point this factor our in their summary. They conclude that adverse events affecting the quality of life of patients have to be carefully balanced against any apparent advantage in the tumour response, particularly since the overall survival appears not to be enhanced by combining the chemotherapy agents.

Smalley et al. (1976) and Chlebowski et al. (1979) conducted very similar trials testing five-drug combinations against the same drugs in sequence each given on disease progression. The studies differ in their choice of fifth drug, either triiodothyronine or vincristine, but the other four drugs are the same (5′-FU, cyclophosphamide, methotrexate and prednisone). Smalley tried two methods of combining therapy, either as a continuous administration or by giving the treatment intermittently within a 28-day cycle, and found that the continuous administration of combined therapy resulted in a significantly higher tumour response rate than that experienced by patients in the sequential arm. Additionally the combined therapy group had a much higher median survival rate and at this point the study was closed to recruitment, presumably on ethical grounds. The authors recommended combination therapy despite finding that overall survival was not significantly different between combination and sequential treatment.

Chlebowski et al. (1979) used a life table analysis to find the projected survival outcomes for their treatment groups. Given the statistically superior tumour response rate and response duration for the combination therapy, it was predicted that this regime would provide a significantly better overall survival, especially for patients with a poorer prognosis. However, analysis of the full patient data showed that there was no significant difference in survival between the arms. Chlebowski et al. (1989) returned to this and another (Smalley et al., 1976) study and combined the data for all 222 patients. As the two RCTs were very similar in patient demographics and interventions this seems to be a valid exercise. It was found that there was no significant difference in overall survival between the 129 patients treated with combination therapy and the 93 patients treated with sequential therapy. There was also no significant difference in overall survival for patients without liver metastases (64%) or without liver or lung metastases (30%) between the two arms. However, patients with liver metastases (36%) had statistically longer survival with combination treatment (P < 0.05).

Both trials individually reported better tumour response and time to progression data and the studies have since been interpreted to indicate that combination therapy is superior to single agents taken sequentially. The authors say, however, that whilst this might be true for the sub-set of patients with liver involvement it is probably not true for the majority of other MBC patients. They also point out that chemotherapy-related deaths were three times higher with combination therapy and that such toxicity, as well as affecting the quality of life of patients in the interim might also affect longer term survival of those who had made initially good responses.

Some of these data illustrate a limitation of the randomised controlled trials offering evidence for this question. Two sub-sets of patients were identified that either died very quickly soon after entering a study or who survived well beyond the study end. It could be argued that neither group were affected by their treatment allocation hence only about half the patients in a trial were properly testing the interventions. Since these patients occupy the middle section of a survival curve, they will contribute to the median values. If patients aren’t completely and fully followed up to the end of their lives the survival analysis can over estimate the value of an intervention or exaggerate the difference between two interventions. This explains why, on further examination, overall survival does not differ in the long run between combination and sequential therapy but only in the short duration of the study and follow-up which may be taken out of context of the patient pathway. The more aggressive combined treatment with associated side effects would seem only to favour patients with a very poor prognosis and heavy tumour burden.

References
  1. Baker LH, Vaughn CB, Al SM. Evaluation of combination vs. sequential cytotoxic chemotherapy in the treatment of advanced breast cancer. Cancer (Philad) 1974;33(2):513–518. [PubMed: 4812768]
  2. Chlebowski RT, Irwin LE, Pugh RP, Sadoff L, Hestorff R, Wiener JM, Bateman JR. Survival of patients with metastatic breast cancer treated with either combination or sequential chemotherapy. Cancer Res. 1979;39(11):4503–4506. [PubMed: 387216]
  3. Chlebowski RT, Smalley RV, Weiner JM, Irwin LE, Bartolucci AA, Bateman JR. Combination versus sequential single agent chemotherapy in advanced breast cancer: associations with metastatic sites and long-term survival. The Western Cancer Study Group and The Southeastern Cancer Study Group. Br J Cancer. 1989;59(2):227–230. [PMC free article: PMC2246985] [PubMed: 2649130]
  4. Creech RH, Catalano RB, Harris DT, Engstrom PF, Grotzinger PJ. Low dose chemotherapy of metastatic breast cancer with cyclophosphamide., adriamycin., methotrexate., 5-fluorouracil (CAMF) versus sequential cyclophosphamide., methotrexate., 5-fluorouracil (CMF) and adriamycin. Cancer. 1979;43(1):51–59. [PubMed: 367574]
  5. Sledge GW, Neuberg D, Bernardo P, Ingle JN, Martino S, Rowinsky EK, Wood WC. Phase III trial of doxorubicin., paclitaxel., and the combination of doxorubicin and paclitaxel as front-line chemotherapy for metastatic breast cancer: an intergroup trial (E1193) J Clin Oncol. 2003;(4):588–592. [PubMed: 12586793]
  6. Smalley RV, Murphy S, Huguley CM Jr, Bartolucci AA. Combination versus sequential five-drug chemotherapy in metastatic carcinoma of the breast. Cancer Res. 1976;36(11 Pt 1):3911–3916. [PubMed: 975039]

Evidence tables

Question: Sequential or combination chemotherapy for improved outcomes?

Created by: Karen Francis on 06/06/2007

Chlebowski et al. (1979)

Baker et al. (1974)

Creech et al. (1979)

Sledge et al. (2003)

Smalley et al. (1976)

Chlebowski et al. (1989)

4.3.2. Combination vs single chemotherapy

Short summary

Evidence for comparing single chemotherapy with combined chemotherapy comprised one very high quality systematic review (n > 7,000 study participants) (Carrick et al., 2005) a more modest systematic review (Takeda et al., 2007) three RCTs (Eijertsen et al., Pacilio et al., 2006 and Martin et al., 2007) and two post-study papers published from the pivotal trial by O’Shaughnessy et al., (2002) (Leonard et al., 2006 and Miles et al., 2004).

Good evidence suggests that the relative risk of death was significantly reduced for patients given combined chemotherapy agents compared with single drugs as first- or second-line treatment. The advantage was greatest for those studies in which the combination therapy did not include the comparator agent. Combined therapies containing anthracyclines or alkylating agents were significantly better at reducing the relative risk of death whereas taxanes did not improve survival as part of a combined therapy.

RCT evidence from three trials showed that first-line treatment with combined therapies including an anthracycline and/or taxane compared with the same anthracycline or taxane, provided no survival advantages but were associated with higher levels of adverse events. Quality of life outcomes were equivocal. Similarly, a small RCT compared second-line (or higher) combined therapy of vinorelbine and gemcitabine with vinorelbine alone and reported no significant difference in overall survival between arms but more adverse events with combined therapy. In contrast, a post-study analysis of long term patient outcomes from a trial of capecitabine (CAP) and docetaxel (DOC) vs DOC alone showed that either combined or sequential therapy with the two agents was significantly better in terms of survival than receiving DOC alone.

Although considerable data were published within systematic reviews about comparison of adverse events and quality of life between combined and single agent regimes the findings were equivocal across studies.

PICO question

POPULATIONINTERVENTIONCOMPARISONOUTCOME
Patients with metastatic breast cancer receiving chemotherapy Same agents
  • in combination
  • vs singly
Specific combinations;
  • Overall survival
  • Quality of life

NB The search strategy developed from this PICO table and used to search the literature for this question can be found in Appendix A

Full evidence summary

The evidence base for this topic is not extensive. One very high quality systematic review (Carrick et al., 2005) examines a wide range of single versus combined therapy regimes whilst another, very limited, systematic review (Takeda et al., 2007) presents findings from a (yet to be published) trial of gemcitabine (GEM) and paclitaxel (PAC) vs PAC, by extracting data from three (possibly non-peer reviewed) abstracts. Two modest quality RCTs presented data from comparisons of epirubicin (EPI) and vinorelbine (VIN) vs EPI (Eijertsen et al., 2003) and GEM and VIN vs VIN (Martin et al., 2007). Two further papers were published from the trial by O’Shaughnessy et al. (2002) of capecitabine (CAP) and docetaxel (DOC) vs DOC: Leonard et al. (2006) presented a retrospective analysis on the effects of within trial drug dose reductions on overall survival and Miles et al. (2004) published a follow-up paper reporting the influence of CAP and DOC as post-study therapies on overall survival in light of the original treatment allocation.

All but two papers (Eijertsen et al., 2003 and Martin et al., 2007) report improved survival for combined therapies when compared to monotherapies. Both studies in which VIN was combined with another agent failed to show an advantage over a monotherapy regime and Carrick et al. (2005) also questioned whether or not taxanes in combination were superior to taxanes given alone.

Carrick et al. (2005) reviewed single vs combination chemotherapies either where the single agent also formed part of the combined therapy regime or where the comparators were completely different. The analyses comprised thirty-seven trials (with over 7,000 participants) of first and second line drugs including anthracyclines, taxanes, anti-metabolites and alkylating agents.

Across all studies, survival data were available for 86% of the randomised participants and results showed that the relative risk for patients given combined therapies was 88% of that for those patients given single drug regimes (95% CI: 0.83–0.94) (P < 0.0001). These results were completely unchanged when only data for those women receiving first-line therapy were analysed. Looking only at those studies comparing a single agent with a combination therapy which included that agent, the advantage of combined therapy was lower, at 91% (95% CI: 0.85–0.98) (P = 0.02) whilst single agents compared with combination regimes NOT including the single agent were significantly more favourable to the combined therapy, at 83% (95% CI: 0.74–0.92) (P = 0.0003).

Amongst the sub-analyses for drug types where the single agent was also included in the combined therapy regime, anti-metabolites in combination showed the greatest advantage over a single anti-metabolite (HR = 0.65 (95% CI: 0.50–0.86) (P = 0.003) but there was significant between trials heterogeneity which may adversely affect the soundness of these results. Combination therapies containing anthracyclines or alkylating agents were all significantly superior in terms of the relative risk (HR of 95% and 91% respectively) but taxanes were thought not to be advantageous to survival when added to a combined therapy (whether or not that combination also included a taxane).

Within this review, quality of life (QOL) was measured and reported in only nine studies using several scales assessing mainly mood, pain, nausea, vomiting, diarrhoea, hair loss, loss of appetite and social functioning. Of these, four studies reported statistically significant differences between combined and single agent therapy but with mixed findings. One trial (Heidemann et al., 2002) found that patients receiving mitoxantrone (MTX) experienced less hair loss, nausea and vomiting than those on fluorouracil, EPI and cyclophosphamide (FEC). Joensuu et al. (1998) found that patients were less physically distressed after six months of EPI monotherapy, including less nausea than those who received FEC followed by VIN. However, Nabholtz et al. (1999) reported that patients in the DOC arm of a trial had less nausea and vomiting but that patients in the other arm (mitomycin plus VIN) had poorer social functioning. Finally, Simes et al. (1994) found that patients in a combination arm (CMFP vs MTX) reported better QOL for pain, mood and nausea & vomiting over the first three months of the trial but lower QOL with regard to hair loss compared with single agent MTX.

Takeda et al. (2007) reviewed a trial of GEM and PAC vs PAC which is yet to be published, extracting the data from three meeting abstracts (O’Shaughnessy et al., 2003, Albain et al., 2004 and Moinpour et al., 2004). Since these reports are incomplete and may not have been peer reviewed the conclusions should be viewed with caution as the likelihood of bias is strong. With a median follow-up of 15.6 months the reported median OS for combined therapy is 18.5 months (95% CI: 16.5–21.2) and for PAC monotherapy is 15.8 months (95% CI: 14.4-17.4). Unfortunately there is no associated P-value and hence statistical significance is not known. Conversely, Kaplan-Meier analysis suggests a considerable survival benefit to GEM + PAC with a relative risk of 78% regardless of adjustments for baseline variation. Reports of QOL outcomes differ between the abstracts but no data were given for these preliminary findings.

Eijertsen et al. (2003) presents data from a phase III RCT (n = 387) comparing VIN and EPI with EPI monotherapy as first line therapy for MBC. With a fairly long follow-up of about 42 months per arm, the majority of patients in both arms had experienced disease progression. The median OS for the combined therapy was 19.1 months compared with 18 months for EPI monotherapy – although confidence intervals were not given the P-value (0.5) confirmed the lack of significant difference between treatments. Whilst QOL was not an outcome for this study, the incidence of adverse events were significantly higher with combined therapy which together with survival data caused the authors to recommend the combined therapy only in patients with rapidly progressive disease. Unfortunately, this trial failed to give details of allocation and randomisation and hence bias cannot be ruled out.

Martin et al. (2007) described a smaller (n = 252) trial comparing VIN and GEM with VIN monotherapy for MBC patients who had previously received anthracyclines and taxanes. Data were analysed at a point where over 80% of participants had experienced disease progression. Just over half of the patients in both study arms received the intervention as second line therapy, a third as third line therapy and the rest as first line treatment for MBC. The median OS for combined therapy was 16.4 months (95% CI: 11.6–21.1) and for monotherapy 15.9 months (95% CI: 12.6–19.1) and there was no significant difference between the two groups (P = 0.8). Quality of life was not an outcome in this trial. With a slightly increased risk of adverse events and lack of significance in improvement in survival (or disease-free progression) the authors made no recommendation for combined therapy.

In 2002, O’Shaughnessy published results from a large (n = 511) RCT of DOC and CAP compared to DOC monotherapy in which the authors found the combined therapy to be significantly superior in terms of survival. In a follow-up paper, Miles et al. (2004) reported on the influence of post-study treatment and showed that patients in the DOC monotherapy arm who had subsequently received single agent CAP had a survival advantage over patients receiving any other post-study therapy (median OS = 21 month (95% CI: 15.6–27.6) vs 12.3 months (95% CI: 10.5–14) a hazard ratio of 0.5 (P < 0.005). Patients originally allocated to combined therapy but who subsequently received either single agent CAP or single agent DOC showed similar survival rates: median OS 18.3 months (95% CI: 14.5–23.4) vs 15.8 months (95% CI: 9.9–21.5) P = 0.2. The authors concluded that a clear survival advantage existed for patients receiving either CAP and DOC combined therapy or sequential treatment with both agents. QOL was not reported.

Leonard et al. (2006) also presented a paper on data from the O’Shaughnessy trial. The authors performed a retrospective analysis looking at dose reductions within the study period and their influence on survival. Unfortunately, the relatively low numbers in this sub-group analysis weaken the findings statistically and data were only poorly presented hence this does not provide strong evidence. There was no significant difference between the median OS for patients who had received reduced drug dosage of CAP and DOC compared to those study participants who had received at least four cycles of therapy with both drugs at full dose (16.3 months compared with 13.1 months). QOL was not reported.

References
  1. Albain KS, Nag S, Calderillo-Ruiz G, Jordaan JP, Llombart A, Pluzanska A, Pawlicki M, Melemed AS, O’Shaughnessy J, Reyes JM. Global phase III study of gemcitabine plus paclitaxel (GT) vs. paclitaxel (T) as frontline therapy for metastatic breast cancer (MBC): First report of overall survival (abstract) Ann Meet Proc Am Soc Clin Oncol. 2004;5
  2. Carrick S, Parker S, Wilcken N, Ghersi D, Marzo M, Simes J. Single agent versus combination chemotherapy for metastatic breast cancer Cochrane Database of Systematic Reviews: Reviews, Anonymous Cochrane Database of Systematic Reviews. 2. Vol. 2005. Chichester (UK): John Wiley & Sons, Ltd; 2005. [PubMed: 15846660]
  3. Ejlertsen B, Mouridsen HT, Langkjer ST, Andersen J, Sjöström J, Kjaer M. Scandinavian Breast Group. Phase III study of intravenous vinorelbine in combination with epirubicin versus epirubicin alone in patients with advanced breast cancer: a Scandinavian Breast Group Trial (SBG9403) J Clin Oncol. 2004;(12):2313–2320. [PubMed: 15197192]
  4. Heidemann E, Minckwitz GV, Holländer N, Souchon R, Clemens M, Mahlke M, Eggeling B, Marseille A, Krekeler G, Kaufmann M. Mitoxantrone plus docetaxel vs single agent mitixanthrone in metastatic breast cancer (MBC): Results of a multicenter randomized trial (abstract) Ann Meet Proc Am Soc Clin Oncol. 2004;36
  5. Leonard R, O’Shaughnessy J, Vukelja S, Gorbounova V, Chan-Navarro CA, Maraninchi D, Barak-Wigler N, McKendrick JJ, Harker WG, Bexon AS, Twelves C. Detailed analysis of a randomized phase III trial: can the tolerability of capecitabine plus docetaxel be improved without compromising its survival advantage? Ann Oncol. 2006;17(9):1379–1385. [PubMed: 16966367]
  6. Martín M, Ruiz A, Muñoz M, Balil A, García-Mata J, Calvo L, Carrasco E, Mahillo E, Casado A, García-Saenz JA, Escudero MJ, Guillem V, Jara C, Ribelles N, Salas F, Soto C, Morales-Vasquez F, Rodríguez CA, Adrover E, Mel JR. Spanish Breast Cancer Research Group. Gemcitabine plus vinorelbine versus vinorelbine monotherapy in patients with metastatic breast cancer previously treated with anthracyclines and taxanes: final results of the phase III Spanish Breast Cancer Research Group (GEICAM) trial. Lancet Oncol. 2007;8(3):219–225. [PubMed: 17329192]
  7. Miles D, Vukelja S, Moiseyenko V, Cervantes G, Mauriac L, Van HG, Liu WY, Ayoub JP, O’Shaughnessy JA. Survival benefit with capecitabine/docetaxel versus docetaxel alone: analysis of therapy in a randomized phase III trial. Clin Breast Cancer. 2004;5(4):273–278. [PubMed: 15507172]
  8. Moinpour C, Wu J, Donaldson G, Liepa A, Melemed A, O’Shaughnessy J, Rappold E, Albain K. Gemcitabine plus paclitaxel (GT) versus paclitaxel (T) as first-line treatment for anthracycline pre-treated metastatic breast cancer (MBC): Quality of life (QoL) and pain palliation results from the global phase III study (abstract) Ann Meet Proc Am Soc Clin Oncol. 2004;32
  9. Nabholtz JM, Senn HJ, Bezwoda WR, Melnychuk D, Deschenes L, Douma J, Vandenberg TA, Rapoport B, Rosso R, Trillet-Lenoir V, Drbal J, Molino A, Nortier JW, Richel DJ, Nagykalnai T, Siedlecki P, Wilking N, Genot JY, Hupperets PS, Pannuti F, Skarlos D, Tomiak EM, Murawsky M, Alakl M, Aapro M. Prospective randomized trial of docetaxel versus mitomycin plus vinblastine in patients with metastatic breast cancer progressing despite previous anthracycline-containing chemotherapy. 304 Study Group. J Clin Oncol. 1999;17(5):1413–1424. [PubMed: 10334526]
  10. O’Shaughnessy J, Miles D, Vukelja S, Moiseyenko V, Ayoub JP, Cervantes G, Fumoleau P, Jones S, Lui WY, Mauriac L, Twelves C, Van HG, Verma S, Leonard R. Superior survival with capecitabine plus docetaxel combination therapy in anthracycline-pretreated patients with advanced breast cancer: phase III trial results. J Clin Oncol. 2002;20(12):2812–2823. [PubMed: 12065558]
  11. O’Shaughnessy J, Nag S, Calderillo-Ruiz G, Jordaan J, Llombart A, Pluzanska A, Pawlicki M, Reyes JM, Sekhon J, Albain KS. Gemcitabine plus paclitaxel (GT) versus paclitaxel (T) as first-line treatment for anthracycline pre-treated metastatic breast cancer (MBC): interim results of a global phase III study (abstract) Proc Am Soc Clin Oncol. 2003;7
  12. Takeda AL, Jones J, Loveman E, Tan SC, Clegg AJ. The clinical effectiveness and cost-effectiveness of gemcitabine for metastatic breast cancer: a systematic review and economic evaluation. Health Technol Assess. 2007;11(19):iii–ix. [PubMed: 17462169]

Evidence tables

Question: Single vs combined chemotherapy

Created by: Karen Francis on 29/10/2007

Ejlertsen et al. (2003)

Martín et al. (2007)

Leonard et al. (2006)

Miles, D et al. (2004)

Carrick et al. (2005)

References to papers in the Carrick et al. (2005) systematic review

  1. Heidemann E, Stoeger, et al. Is first-line single-agent mitoxantrone in the treatment of high-risk metastatic breast cancer patients as effective as combination chemotherapy? No difference in survival but higher quality of life was found in a multicenter randomized trial. Ann Oncol. 2002;13(11):1717–1729. [PubMed: 12419743]
  2. Joensuu H, Holli K, Heikkinen M, Suonio E, Aro AR, Hietanen P, Huovinen R. Combination chemotherapy versus single-agent therapy as first- and second-line treatment in. J Clin Oncol. 1998;16(12):3720–3730. [PubMed: 9850014]
  3. Nabholtz JM, Senn HJ, Bezwoda WR, Melnychuk D, Deschenes L, Douma J, Vandenberg Prospective randomized trial of docetaxel versus mitomycin plus vinblastine in patients with. J Clin Oncol. 1999;17(5):1413–1424. [PubMed: 10334526]
  4. Simes RJ, Gebski V, Coates AS, Forbes J, Harvey V, VanHazel G, Tattersall MHN, Abdi E, Brigham B. Quality of life (QOL) with single agent mitozantrone (MTZ) or combination chemotherapy (CMFP) for advanced breast cancer: a randomised trial. Proc Am Soc Clin Oncol. 1994;13:73.

Takeda et al. (2007)

References included in Takeda et al. (2007) review

  1. Albain KS, Nag S, Calderillo-Ruiz G, Jordaan JP, Llombart A, Pluzanska A, et al. Global phase III study of gemcitabine plus paclitaxel (GT) vs. paclitaxel (T) as frontline therapy for metastatic breast cancer (MBC): first report of overall survival. J Clin Oncol. 2004;22:5S.
  2. Moinpour C, Wu J, Donaldson G, Liepa A, Melemed A, O’Shaughnessy J, et al. Gemcitabine plus paclitaxel (GT) versus paclitaxel (T) as first line treatment for anthracycline pre- treated metastatic breast cancer (MBC): quality of life (QoL) and pain palliation results from the global phase III study. J Clin Oncol. 2004;22:32S.
  3. O’Shaughnessy J, Nag S, Calderillo-Ruiz G, Jordaan J, Llombart A, Pluzanska A, et al. Gemcitabine plus paclitaxel (GT) versus paclitaxel (T) as first-line treatment for anthracycline pretreated metastatic breast cancer (MBC): interim results of a global phase III study (abstract 25) Proc Am Soc Clin Oncol. 2003;22:7.

Updated evidence (4.3.2)

Summary

Only one additional paper was identified comparing combined and single agent therapies. A poor quality, low patient number RCT (Pacilio et al. 2006) gave patients either epirubicin and docetaxel combined or docetaxel monotherapy as first line treatment for metastatic breast cancer. All the women had received epirubicin in the neoadjuvant or adjuvant setting. There was no difference in outcomes for efficacy or survival but there were significantly more adverse events reported for the combined therapy arm (grade 4 leukopenia, and grade 3 nausea and stomatitis).

Reference
  1. Pacilio C, Morabito A, Nuzzo F, Gravina A, Labonia V, Landi G, Rossi E, De Maio M, D’Aiuto G, Botti G, Normanno N, Chiodini P, Gallo C, Perrone F, de Matteis A. Is epirubicin effective in first-line chemotherapy of metastatic breast cancer (MBC) after an epirubicin-containing adjuvant treatment? A single centre phase III trial. Br J Cancer. 2006;94:1233–1236. [PMC free article: PMC2361414] [PubMed: 16622454]
Evidence table

Question: Sequential or combination chemotherapy for improved outcomes?

Created by: Karen Francis on 18/07/2008

Pacilio et al. 2006

4.4. Vinorelbine as first or subsequent line therapy following anthracycline failure

Short summary

The level of evidence on the use of vinorelbine (VIN) as a monotherapy or in combination with other agents is generally of very poor quality consisting mainly of low patient number, non-comparative phase II trials or small RCTs. The majority of patients were believed to have had prior anthracycline therapy.

VIN monotherapy

One small, statistically underpowered RCT (Pajk et al. 2008) compared VIN with capecitabine (CAP) in a small number of heavily pre-treated women and reported no significant difference in response or survival outcomes but more adverse events (particularly neutropenia) in the VIN group. Two poor quality phase II studies evaluated VIN for women with metastatic disease (Udom et al., 2000 and Zelek et al., 2001) finding that as second or third-line treatment response rates of up to 41%, response duration of 4 months and time to progression of ~2.75 months were reported.

VIN combined therapy

Two poor to moderate quality RCTs tested VIN in combination with 5′-fluorouracil (5′-FU) vs docetaxel (DOC) (Bonneterre et al., 2002) or gemcitabine (GEM) vs VIN (Martin et al., 2007). VIN and 5′-FU combined resulted in similar treatment outcomes as DOC monotherapy but with a higher incidence of neutropenia. VIN and GEM resulted in superior progression-free survival, but not significantly different overall survival or response duration, compared with VIN alone.

Thirteen poor to moderate quality phase II, non-comparative, studies described VIN combined with: trastuzumab (TRZ) (Burstein et al., 2003, Chan et al., 2006, Jahanzeb et al., 2002, Bartsch et al., 2007, De Maio et al., 2007 and Catania et al., 2007), CAP (Ghosn et al., 2006 and Davis, 2007), DOC (Mayordomo et al., 2004), GEM (Ardavanis et al., 2007 and Colomer et al. 2006), 5′-FU (Stuart, 2008), mitozantrone (MTZ) (Onyenadum et al. 2007), cisplatin (CIS) followed by DOC (Shamseddine et al. 2006) and CAP followed by DOC (Ghosn et al. 2008).

For all phase II combination studies, the overall tumour response rates ranged from 33–75%, median overall survival from 13–35.8 months, median response duration from 2.6–17.5 months, median time to progression (reported in two studies) from 6.6–8.6 months and median progression-free survival (reported in two studies) from 9.6–9.9 months. The most commonly reported adverse events attributed to VIN were neutropenia, nausea and vomiting and alopecia.

PICO question

POPULATIONINTERVENTIONCOMPARISONOUTCOME
Patients with advanced breast cancerVinorelbine alone or in combination with other agentsSystemic therapy without vinorelbine
  • Tumour response
  • Progression-free survival
  • Overall survival
  • Symptom relief
  • QOL
  • Adverse effects

NB The search strategy developed from this PICO table and used to search the literature for this question can be found in Appendix A

Full evidence summary

i. Vinorelbine monotherapy

There were two poor quality, low patient number (n = 60), phase II studies evaluating the use of vinorelbine (VIN) as a second, third line (or higher) salvage therapy for MBC (Udom et al., 2000 and Zelek et al., 2001). At least 80% of patients had received previous anthracycline therapy.

Udom et al. (2000) reported a 35% (95%CI: 15–59%) overall response rate (ORR) for all patients but 32% for the anthracycline-treated sub-group. Efficacy and survival data were presented without statistical analyses: median response duration (RD) = 4 months and time to progression (TTP) = 2.75 months. The most commonly reported adverse event was neutropenia which ranged in severity from grade 1–3.

Zelek et al., (2001) presented a study in which VIN was given initially at a weekly dose of 30mg per m2 which had to be reduced to 25 mg per m2 after the first administration due to severe adverse events in the first six patients. The ORR = 25% (95%CI: 13–41%) and the median time to treatment failure (TTF) = 6 months (range: 2–18+ months). Nine patients experienced grade 4 neutropenia and three patients had neutropenic fever. Other grade 3/4 events included thrombocytopenia (n = 2), anaemia (n = 4) constipation (n = 3) and peripheral neuropathy (n = 3).

ii. Vinorelbine combined therapy

There were two poor to moderate quality RCTs testing VIN in combination with another drug as therapy for metastatic breast cancer. Bonneterre et al., (2002) randomised 176 patients to receive either docetaxel (DOC) or VIN plus 5′-fluorouracil (FUN) on a 21-day cycle. All patients had received prior anthracycline therapy either in the adjuvant or first line setting and were sub-grouped on the basis of being sensitive or resistant/refractory to the drug.

There were no significant differences between the two arms in respect of ORR, overall survival (OS) or time to progression (TTP). Median ORR DOC = 43% (95%CI: 32–53%) versus FUN = 38.9% (95%CI: 29–49%) median overall survival (OS) DOC = 16 months (no CI) versus FUN = 15 months (no CI) median TTP DOC = 6.5 months (95%CI: 5.5–8.4 months) versus FUN = 5.1 months (95%CI: 4.4–6.9 months). This also held true for the anthracycline resistant/refractory groups. One notable difference between arms was in the severity of adverse events - the FUN group lost five patients by the cut-off date, all believed to have died from treatment related events (sepsis, diarrhoea and liver failure). The most common adverse event was neutropenia (82% in DOC arm and 67% in FUN arm) followed by stomatitis (40% in FUN arm) and alopecia (44% in DOC arm).

This paper was not of good quality due to the lack of elaboration on methodology e.g. allocation, blinding, randomisation etc so appears to fail in the elimination of bias.

A second RCT (Martin et al., 2007) compared VIN alone with VIN plus gemcitabine (GEM) combined therapy as first, second or third line treatment for MBC. This was a better quality study in respect of randomisation but, as far as is known, the trial participants and reviewers were not blinded to treatment and some of the authors declared an interest in the company which manufactures one of the drugs being tested (GEM).

It is assumed that, since previous anthracycline and taxane treatment was required for study entry, all patients had been treated thus but this has not been clarified. The tumour response rate between arms was not significantly different (P = 0.093). Median ORR VIN = 26% (95%CI: 18–34%) versus VIN plus GEM = 36% (95%CI: 28–45%).

Log rank testing revealed that VIN plus GEM was significantly better than VIN monotherapy in respect of progression-free survival (P = 0.0028) but not in response duration or overall survival. Median PFS: VIN = 6 months (95%CI: 4.8–7.1 months) versus VIN plus GEM = 4 months (95%CI: 2.9–5.1 months) median OS VIN = 16.4 months (95%CI: 11.6–21.1 months) versus VIN plus GEM = 15.9 months (95%CI: 12.6–19.1 months) median RD VIN = 3.7 months (95%CI: 3.0–4.4 months) versus VIN plus GEM = 4.8 months (95%CI: 3.1–6.6 months).

The most common adverse events were neutropenia (61% in VIN + GEM arm and 44% in VIN arm) and febrile neutropenia (11% in VIN + GEM arm and 6% in VIN arm).

There were six phase II studies of varying quality but all with the shortcoming of having no comparator group. Three studies examined the efficacy and safety of a combined treatment of VIN with trastuzumab (T) (Burstein et al., 2003, Chan et al., 2006 and Jahanzeb et al., 2002) one study reported on VIN plus capecitabine (Ghosn et al., 2006) and one on VIN plus DOC (Mayordomo et al., 2004). The results for VIN plus T are broadly overlapping in terms of tumour response – no other parameter was reported in all three studies.

Generally the quality of the phase II studies was poor: non-independent tumour assessment (Mayordomo et al., 2004, Burstein et al., 2003 and Ghosn et al., 2006) lack of patient demographics (Burstein et al., 2003) poorly reported data (Mayordomo et al., 2004, Ghosn et al., 2006 and Jahanzeb et al., 2002). However, Chan et al. (2006) was well conducted and presented with sound statistical methodology, Kaplan-Meier analysis and a reasonable patient number.

The outcomes from all phase II studies are summarised in table 4.4.1 below which shows point estimates with or without 95% confidence intervals.

Table 4.4.1. Efficacy data for all studies where this information was given or could be extracted.

Table 4.4.1

Efficacy data for all studies where this information was given or could be extracted. Abbreviations: A anthracycline, A+T anthracycline plus taxane, ITT intention to treat, NR no reported, n number.

All the studies appraised for this question had a minimum 50% of patients pre-treated with anthracycline, either in the adjuvant or metastatic setting. Overall, the standard and quantity of evidence for either vinorelbine as a monotherapy or combined with another agent was weak.

References

  1. Bonneterre J, Roché H, Monnier A, Guastalla JP, Namer M, Fargeot P, Assadourian S. Docetaxel vs 5-fluorouracil plus vinorelbine in metastatic breast cancer after anthracycline therapy failure. Brit J Cancer. 2002;87:1210–1215. [PMC free article: PMC2408916] [PubMed: 12439707]
  2. Burstein HJ, Harris LN, Marcom PK, Lambert-Falls R, Havlin K, Overmoyer B, Friedlander RJ Jr, Gargiulo J, Strenger R, Vogel CL, Ryan PD, Ellis MJ, Nunes RA, Bunnell CA, Campos SM, Hallor M, Gelman R, Winer EP. Trastuzumab and vinorelbine as first-line therapy for HER2-overexpressing metastatic breast cancer: multicenter phase II trial with clinical outcomes, analysis of serum tumor markers as predictive factors, and cardiac surveillance algorithm. J Clin Oncol. 2003;21:2889–2895. [PubMed: 12885806]
  3. Chan A, Martin M, Untch M, Gil MG, Guillem-Porta V, Wojtukiewicz M, Kellokumpu-Lehtinen P, Sommer HL, Georgoulias V, Battelli N, Pawlicki M, Aubert D, Bourlard T, Gasmi J, Villanova G, Petruzelka L, Navelbine HP. Vinorelbine plus trastuzumab combination as first-line therapy for HER 2-positive metastatic breast cancer patients: an international phase II trial. Brit J Cancer. 2006;95:788–793. [PMC free article: PMC2360530] [PubMed: 16969343]
  4. Ghosn M, Kattan J, Farhat F, Younes F, Gasmi J. Phase II trial of capecitabine and vinorelbine as first-line chemotherapy for metastatic breast cancer patients. Anticancer Res. 2006;26:2451–2456. [PubMed: 16821631]
  5. Jahanzeb M, Mortimer JE, Yunus F, Irwin DH, Speyer J, Koletsky AJ, Klein P, Sabir T, Kronish L. Phase II trial of weekly vinorelbine and trastuzumab as first-line therapy in patients with HER2(+) metastatic breast cancer. The Oncologist. 2002;7:410–417. [PubMed: 12401903]
  6. Martín M, Ruiz A, Muñoz M, Balil A, García-Mata J, Calvo L, Carrasco E, Mahillo E, Casado A, García-Saenz JA, Escudero MJ, Guillem V, Jara C, Ribelles N, Salas F, Soto C, Morales-Vasquez F, Rodríguez CA, Adrover E, Mel JR. Spanish Breast Cancer Research Group. Gemcitabine plus vinorelbine versus vinorelbine monotherapy in patients with metastatic breast cancer previously treated with anthracyclines and taxanes: final results of the phase III Spanish Breast Cancer Research Group (GEICAM) trial. The Lancet oncology. 2007;8:219–225. [PubMed: 17329192]
  7. Mayordomo JI, Milla A, Morales S, Yubero A, Lorenzo A, Baena JM, Modolell A, Sanz J, Illarramendi J, Garcia MJ, Machengs I, Burillo MA, Tres A. Biweekly docetaxel and vinorelbine as first-line chemotherapy in metastatic breast cancer. Clin Breast Cancer. 2004 Jun 5:131–135. 2004. [PubMed: 15245617]
  8. Udom DI, Vigushin DM, Linardou H, Graham H, Palmieri C, Coombes RC. Two weekly vinorelbine: administration in patients who have received at least two prior chemotherapy regimes for advanced breast cancer. Eur J Cancer. 2000;36:177–182. [PubMed: 10741275]
  9. Zelek L, Barthier S, Riofrio M, Fizazi K, Rixe O, Delord JP, Le CA, Spielmann M. Weekly vinorelbine is an effective palliative regimen after failure with anthracyclines and taxanes in metastatic breast carcinoma. [see comment] Cancer. 2001;92:2267–2272. [PubMed: 11745280]

Updated evidence (4.4)

Summary

Eight extra papers were identified to update the evidence on the use of vinorelbine (VIN) as a monotherapy or in combination with other agents for the treatment of advanced breast cancer and where the patient has previously been treated with anthracycline or where it was contraindicated. All were phase II trials without comparators and therefore weak in their evidential value; many outcomes of interest were not reported or data were poorly presented.

Three prospective phase II studies (Bartsch et al., 2007, De Maio et al., 2007 and Catania et al., 2007) examined the combination of VIN plus trastuzumab (TRZ) in women who were Her2 +ve. Other phase II studies looked at the use of VIN combined with: capecitabine (CAP) (Davis, 2007), gemcitabine (GEM) (Ardavanis et al., 2007 and Colomer et al. 2006), 5′-Fluorouracil (5′-FU) (Stuart, 2008), cisplatin (then docetaxel) (Shamseddine et al. 2006) and mitozantrone (Onyenadum et al., 2007).

The outcomes from all phase II studies are summarised in table 4.4.2 below. Data are point estimates with or without 95% confidence intervals (or ranges (rge) where stated).

Table 4.4.2. Efficacy data for all studies where this information was given or could bec extracted.

Table 4.4.2

Efficacy data for all studies where this information was given or could bec extracted. Abbreviations: NR no reported, n number, PA prior anthracycline, AN anthracycline naïve, o oral

References
  1. Ardavanis A, Kountourakis P, Maliou S, Vassilakopoulou M, Basioukas S, Kyriakou F, Koumna S, Rigatos G. Gemcitabine and oral vinorelbine as salvage treatment in patients with advanced anthracycline- and taxane-pretreated breast cancer. Anticancer Res. 2007;27:2989–2992. [PubMed: 17695483]
  2. Bartsch R, Wenzel C, Altorjai G, Pluschnig U, Bachleitner-Hoffmann T, Locker GJ, Rudas M, Mader R, Zielinski CC, Steger GG. Results from an observational trial with oral vinorelbine and trastuzumab in advanced breast cancer. Breast Cancer Research & Treatment. 2007;102:375–381. [PubMed: 17028979]
  3. Catania C, Medici M, Magni E, Munzone E, Cardinale D, Adamoli L, Sanna G, Minchella I, Radice D, Goldhirsch A, Nole F. Optimizing clinical care of patients with metastatic breast cancer: a new oral vinorelbine plus trastuzumab combination. Ann Oncol. 2007;18:1969–1975. [PubMed: 17846020]
  4. Colomer R, Llombart-Cussac A, Tusquets I, Rifa J, Mayordomo JI, Ojeda B, Ciruelos E, Hornedo J, Vicente D, Cortes-Funes H. Biweekly gemcitabine plus vinorelbine in first-line metastatic breast cancer: efficacy and correlation with HER2 extracellular domain. Clinical & Translational Oncology: Official Publication of the Federation of Spanish Oncology Societes & of the National Cancer Institute of Mexico. 2006;8:896–902. [PubMed: 17169763]
  5. Davis AJ. Multicenter phase II study of combination chemotherapy with capecitabine and intravenous vinorelbine in patients with pretreated metastatic breast cancer. Asia-Pacific Journal of Clinical Oncology. 2007;3:37–43.
  6. De Maio ME, Pacilio C, Gravina A, Morabito A, Di RF, Labonia V, Landi G, Nuzzo F, Rossi E, Silvestro P, Botti G, Di BM, Curcio MP, Formichelli F, La VF, Staiano M, Maurea N, D’Aiuto G, D’Aiuto M, Thomas R, Signoriello G, Perrone F, de MA. Vinorelbine plus 3-weekly trastuzumab in metastatic breast cancer: a single-centre phase 2 trial. BMC Cancer. 2007;7:50. [PMC free article: PMC1832208] [PubMed: 17374151]
  7. Ghosn M, Kattan J, Farhat F, Younes F, Nasr F, Moukadem W, Gasmi J, Chahine G. Sequential vinorelbine-capecitabine followed by docetaxel in advanced breast cancer: long-term results of a pilot phase II trial. Cancer Chemother Pharmacol. 2008;62:11–18. [PubMed: 17717668]
  8. Onyenadum A, Gogas H, Markopoulos C, Bafaloukos D, Aravantinos G, Mantzourani M, Koutras A, Tzorakoelefterakis E, Xiros N, Makatsoris T, Fountzilas G, Kalofonos HP. Mitozantrone plus vinorelbine in pretreated patients with metastatic breast cancer. J Chemother. 2007;19:582–589. [PubMed: 18073159]
  9. Pajk B, Cufer T, Canney P, Ellis P, Cameron D, Blot E, Vermorken J, Coleman R, Marreaud S, Bogaerts J, Basaran G, Piccart M. Anti-tumor activity of capecitabine and vinorelbine in patients with anthracycline- and taxane-pretreated metastatic breast cancer: Findings from the EORTC 10001 randomized phase II trial. Breast. 2008;17:180–185. [PubMed: 17976988]
  10. Shamseddine AI, Otrock ZK, Khalifeh MJ, Yassine HR, Charafeddine M, bdel-Khalek Z, Chehal A, Bitar N, Jalloul R, Dheiny M, Dandashi A, Wehbeh M, El-Saghir NS. A clinical phase II study of a non-anthracycline sequential combination of cisplatin-vinorelbine followed by docetaxel as first-line treatment in metastatic breast cancer. Oncology. 2006;70:330–338. [PubMed: 17164589]
  11. Stuart NSA. Vinorelbine and Infusional 5-fluorouracil in Anthracycline and Taxane Pre-treated Metastatic Breast Cancer. Clin Oncol. 2008;20:152–156. [PubMed: 18083354]

4.5. Capecitabine and docetaxel as a combined therapy or capecitabine as a monotherapy

Short summary

The level of evidence on the use of CAP as a monotherapy is generally of poor quality consisting mainly of low patient number, non-comparative phase II studies. Evidence for capecitabine in combination with DOC consists of one good phase III RCT. As such, the findings from these studies should be interpreted with caution.

CAP monotherapy

Nine phase II studies (El Helw and Coleman, 2005, Fumoleau et al., 2004, Lee et al., 2004, Pierga et al., 2004, Reichardt et al., 2003, Wist et al., 2004, Sezgin et al. 2007, Venturini et al. 2007 and Yap et al. 2007) and one retrospective case series (Leonard et al., 2002) were identified. The majority of patients are believed to have been treated with anthracycline and taxane.

Across all studies, the overall tumour response rates ranged from 10–42%, median overall survival from 9.4–18.1 months, median response duration from 3.8–15.4 months and median time to progression from 3.5–6.6 months. The most commonly reported adverse event was hand-foot syndrome which at grade 3/4 occurred in up to 21% of patients.

CAP combined therapy

The evidence for combined therapy with CAP and DOC comprised one phase III RCT (Chan, 2005) three phase II studies (Mackey et al., 2002, Silva et al. 2008 and Mrozek et al. 2006) and a retrospective analysis of post-study data (Miles et al., 2004).

The RCT compared CAP and DOC with gemcitabine and DOC and reported no significant difference between study arms in overall response rate, median time to treatment failure or response duration. There were higher levels of hand-foot syndrome and diarrhoea in the CAP and DOC arm. The phase II studies offered poor quality and conflicting evidence on reduced doses of CAP and DOC reporting overall tumour response rates ranging from 44–50%, median overall survival of ~19 months (1 study), median response duration of ~ 9.1 months (1 study) and median time to progression of ~5.5 months (1 study). A post study analysis (Miles et al. 2004) of a pivotal RCT (O’Shaughnessy et al., 2002) confirmed a survival advantage with CAP and DOC, either combined or sequentially, when compared with either agent as monotherapy.

PICO question

POPULATIONINTERVENTIONCOMPARISONOUTCOME
Patients with advanced breast cancerOral capecitabine alone or in combination with docetaxel
  • Overall survival
  • Progression-free survival
  • Tumour response
  • Adverse events/toxicity
  • QOL
  • Symptom relief

NB The search strategy developed from this PICO table and used to search the literature for this question can be found in

Full evidence summary

i. Capecitabine monotherapy

No randomised controlled trials of capecitabine (CAP) monotherapy were identified. Six phase II uncontrolled studies, presented as full papers, and one retrospective case series were appraised. Of these, five studies reported that all patients (or the great majority) had been pre-treated with both anthracycline and a taxane.

Five studies (Fumoleau et al., 2004, Lee et al., 2004, Pierga et al., 2004, Reichardt et al., 2003 and Wist et al., 2004) report median overall survival (OS) and median time to progression (TTP) with 95% confidence intervals. As a group, these data are similar at about 4 months for median TTP but dissimilar for median OS. Reichardt et al. (2003) reports a median OS of 10.1 months with a very narrow 95%CI: (8.2–11.5). Fumoleau et al. (2004) had a similar patient number (n = 126) but here the point estimate of OS was much higher at 15.2 months (95%CI: 13.5–19.6).

Two studies (El-Helw and Coleman, 2005) recruited 57 patients of which only 30% had received prior treatment with both anthracycline and taxane. Although the data analysis was presented for this subgroup, the point estimates (median OS = 10 months, median TTP = 6 months) were without confidence intervals. The other study (Leonard et al., 2002), similarly, gave only point estimates of these parameters (median OS = 7.7 months, median TTP = 4.1 months).

Four studies (Lee et al., 2004, Pierga et al., 2004, Reichardt et al., 2003 and Wist et al., 2004) report the median response duration with 95% confidence intervals which overlap broadly at about 8 months. This parameter was not reported by Fumoleau et al. (2004). El-Helw and Coleman (2005) gave a figure of 10 months (no 95%CI). The overall response is very different between Reichardt et al. (2003) and Fumoleau et al. (2004).

Outcomes are summarised below, in Table 4.5.1 and data are shown as point estimates and 95% confidence intervals, where given.

Table 4.5.1. Abbreviations: OR overall response, TTP time to progression, OS overall survival, RD response duration, NR not reported.

Table 4.5.1

Abbreviations: OR overall response, TTP time to progression, OS overall survival, RD response duration, NR not reported.

Overall response (OR) rates were estimated as the sum of complete and partial responses to treatment. Median response duration was given as a point estimate and 95%CI in only four out of seven studies.

The studies generally showed poor methodological quality. Licensing information for CAP indicates monotherapy for patients failing treatment with both an anthracycline and a taxane. Only two studies (Lee et al., 2004 and Wist et al., 2004) recruited patients exclusively that had been pre-treated with both regimes, whilst two studies reported that the majority of patients had received both (Fumoleau et al., 2004: 96% anthracycline and 99% taxane) and (Reichardt et al., 2003: 93% anthracycline and 99% taxane) and the remaining three studies enrolled such patients as part of a wider population.

The lack of a control group in all studies means that the results should be treated with caution. The resultant lack of blinding which, whilst not crucial for an objective end-point such as death, is important for accurately gauging a more subjective outcome such as tumour response. No paper reported the use of independent assessors and, in many cases, no reference was made to the procedure at all and so the probability of bias is strong.

Comparing data from the TA 62 ‘Guidance on the use of capecitabine for the treatment of locally advanced or metastatic breast cancer’ it appears that the recommendations are unlikely to be significantly affected by the addition of these more up-to-date studies. Comparing ranges from TA62 with data from these studies:

  • Median OS in TA62: 8.1–15.2 months vs 9.4–18.1 months from papers in this summary
  • Median TTP in TA62: 2.8–6.2 months vs 3.5–4.9 months from papers in this summary
  • Median response duration in TA62: 5.0–8.3 months vs 4.5–8.9 months from papers in this summary
  • Overall response rates in TA62: 15–28% vs 14–28% from papers in this summary

Reporting of treatment side-effects was, in most cases, thorough and data are summarised below in Table 4.5.2 showing the five most common grade 3/4 events, expressed as a percentage of each patient population. Leonard et al. (2002) reported that ‘at least 20% of study patients had experienced adverse events, mostly mild to moderate’, commonly hand-foot syndrome, diarrhoea and nausea. Wist et al. (2004) did not differentiate between grade 2/3 events, which were reported as 35% hand-foot syndrome and 23% gastrointestinal toxicity.

Table 4.5.2

Table 4.5.2

Abbreviation: NR not reported

ii. Capecitabine combined with docetaxel

The current evidence for combination therapy comprised: one phase III RCT comparing CAP (1,250 mg per m2 twice daily for 14 days of a 21-day cycle) plus docetaxel (DOC) (75 mg per m2) given as an iv infusion on day 1 of that 21-day cycle against gemcitabine (1 g per m2) on days 1 and 8 of a 21-day cycle with docetaxel as above on day 1) (Chan, 2005).

One observational study which examined a reduced dose CAP (900 mg per m2) twice daily for 14 days of a 21-day cycle) plus a reduced dose DOC (30 mg per m2) given as an iv infusion once per week (Mackey et al., 2002).

One post-hoc analysis of a phase III RCT comparing CAP (1,250mg per m2 twice daily for 14 days of a 21-day cycle) plus DOC (75 mg per m2) as an iv infusion on day 1 of that 21-day cycle against DOC (100 mg per m2) given as an iv infusion on day 1 of a 21-day cycle (Miles et al., 2004).

The phase III RCT reported on 305 participants (gender not stated) of which only 76% of the 152 patients in the capecitabine/docetaxel (CD) treatment arm and 69% of the 153 patients in the gemcitabine/docetaxel (GD) comparator arm had received prior treatment with anthracycline. Patient characteristics were not detailed at all in the paper as preliminary results had been presented as an abstract at ASCO in 2005. However, access to this abstract was not possible and hence the information is not recorded in this summary.

There were no significant differences between the treatment and comparator arms in respect of overall response rate, median time to treatment failure or response duration. Overall survival was not reported. Grade 3/4 adverse events were thoroughly recorded, of which the statistically significant parameters were hand-foot syndrome (26% in the CD group against 0% in the GD group) and diarrhoea (17% in the CD group against 7% in the GD group).

The observational study showed that a regime of reduced capecitabine with reduced docetaxel resulted in median OS of 18.7 months (95%CI: 8.6–22.9). However, the patient number was very low (n = 20) and hence the study was, unfortunately, significantly underpowered. The most commonly reported grade 3 events were: hand-foot syndrome (30%), nail disorder (45%), diarrhoea (20%) and asthenia (30%). No grade 4 adverse events were reported. The low study number and lack of comparator make it impossible to determine whether or not the use of a reduced dose of capecitabine and docetaxel is as effective as the standard dose regime and whether or not the side effects are less severe.

The post-hoc analysis reported results of therapy given to patients that had been enrolled in a multi-centre study (O’Shaughnessy et al., 2001) comparing the combined therapy of capecitabine with docetaxel (CD) against single agent docetaxel (DOC). The original report concluded that CD was of higher efficacy than DOC: median OS: 14.5 months (95%CI: 12.3–16.3) vs 11.5 months (95%CI: 9.8–12.7) hazard ratio of 0.77 (P < 0.01).

In this follow-up paper, Miles et al. (2004) reported on the influence of post-study treatment and showed that patients enrolled into the CD arm who had subsequently received single agent CAP had a survival advantage over CD patients receiving any other post-study therapy (median OS: 21 month (95%CI: 15.6–27.6) vs median OS of 12.3 months (95%CI: 10.5–14) a hazard ratio of 0.5 (P < 0.005).

Additionally, the order of sequential therapy did not significantly affect outcome since patients who had been treated with DOC followed by CAP compared with patients who had received CAP then DOC had similar median OS (P = 0.2).

The evidence for capecitabine and docetaxel as a combination therapy remains centred on the RCT of O’Shaughnessy et al. (2002) and related papers which suggest an advantage of either combination or sequential therapy with these two agents over single agent docetaxel. In the trial of capecitabine and docetaxel against capecitabine and gemcitabine the survival statistics are not yet available but it appears that side effects may be an issue.

The literature search identified two ongoing trials which are currently recruiting...

NCT00083200. This randomized phase II trial is studying two different doses of capecitabine when given together with docetaxel to compare how well they work in treating women with locally advanced or metastatic breast cancer that has not responded to previous anthracycline-based chemotherapy (such as daunorubicin, doxorubicin, or epirubicin). PharmaNet.

NCT00077857. This study will compare the efficacy and safety of label dose Xeloda to that of a lower dose of Xeloda plus Taxotere in patients with locally advanced or metastatic breast cancer after failure of chemotherapy with an anthracycline. The anticipated time on study treatment is until disease progression and the target sample size is 100–500 individuals. Hoffman La-Roche.

References

  1. Chan S. Gemcitabine plus docetaxel versus capecitabine plus docetaxel for patients with anthracycline-pretreated metastatic breast cancer: a review of the results of a European Phase III trial. Ejc Supplements. 2005;3:17–21.
  2. El-Helw L, Coleman RE. Reduced dose capecitabine is an effective and well-tolerated treatment in patients with metastatic breast cancer. Breast. 2005;14:368–374. [PubMed: 16216738]
  3. Fumoleau P, Largillier R, Clippe C, Dieras V, Orfeuvre H, Lesimple T, Culine S, Audhuy B, Serin D, Cure H, Vuillemin E, Morere J-F, Montestruc F, Mouri Z, Namer M. Multicentre, phase II study evaluating capecitabine monotherapy in patients with anthracycline- and taxane-pretreated metastatic breast cancer. Eur J Cancer. 2004;40:536–542. [PubMed: 14962720]
  4. Lee SH, Lee J, Park J, Park SH, Lee KE, Lee SI, Nam E, Park JO, Kim K, Jung CW, Park YS, Yoon SS, Kang WK, Lee MH, Park K, Im YH. Capecitabine monotherapy in patients with anthracycline- and taxane-pretreated metastatic breast cancer. Med Oncol. 2004;21:223–231. [PubMed: 15456949]
  5. Leonard RC, Twelves C, Breddy J, Chaturvedi A, Hutcheon A, Salazar R, Cameron D. Capecitabine named-patient programme for patients with advanced breast cancer. The UK experience. Eur J Cancer. 2002;38:2020–2024. [PubMed: 12376207]
  6. Mackey JR, Tonkin KS, Koski SL, Scarfe AG, Smylie MGB, Joy AA, Au H, Bodnar DM, Soulieres D, Smith SW. Final results of a phase II clinical trial of weekly docetaxel in combination with capecitabine in anthracycline-pretreated metastatic breast cancer. Clinical Breast Cancer. 2004 Oct;5:287–292. [PubMed: 15507175]
  7. Miles D, Vukelja S, Moiseyenko V, Cervantes G, Mauriac L, Van HG, Liu WY, Ayoub JP, O’Shaughnessy JA. Survival benefit with capecitabine/docetaxel versus docetaxel alone: analysis of therapy in a randomized phase III trial. Clinical Breast Cancer. 2004;5:273–278. [PubMed: 15507172]
  8. O’Shaughnessy JA, Miles D, Vukelja S, Moiseyenko V, Ayoub JP, Cervantes G, et al. Superior survival with capecitabine plus docetaxel combination therapy in anthracycline-pre-treated patients with advanced breast cancer: phase III trial results. J Clin Oncol. 2002;20:2812–2823. [PubMed: 12065558]
  9. Pierga JY, Fumoleau P, Brewer Y, Zelek L, Martin D, Turpin FL, Goudier MJ, Gil-Delgado M, Baticle JL, Namer M, Chollet P, Sutherland W, Barats JC. Efficacy and safety of single agent capecitabine in pretreated metastatic breast cancer patients from the French compassionate use program. Breast Cancer Research & Treatment. 2004;88:117–129. [PubMed: 15564795]
  10. Reichardt P, von MG, Thuss-Patience PC, Jonat W, Kolbl H, Janicke F, Kieback DG, Kuhn W, Schindler AE, Mohrmann S, Kaufmann M, Luck HJ. Multicenter phase II study of oral capecitabine (Xeloda) in patients with metastatic breast cancer relapsing after treatment with a taxane-containing therapy. Ann Oncol. 2003;14:1227–1233. [PubMed: 12881384]
  11. Wist EA, Sommer HH, Ostenstad B, Risberg T, Bremnes Y, Mjaaland I. Oral capecitabine in anthracycline- and taxane-pretreated advanced/metastatic breast cancer. Acta Oncol. 2004;43:186–189. [PubMed: 15163168]

Updated evidence (4.5)

Summary

Additional poor quality evidence on the efficacy and safety of CAP monotherapy comprised three phase II studies (Sezgin et al., 2007, Venturini et al., 2007 and Yap et al., 2007). The treatment outcomes are summarised below, in Table 4.5.3 and data are shown as point estimates and 95% confidence intervals, where given, or data ranges if stated. The most commonly reported adverse events were hand-foot syndrome, diarrhoea and nausea & vomiting.

Table 4.5.3

Table 4.5.3

Abbreviations: ORR overall response rate, TTP time to progression, OS overall survival, RD response duration, NR Not reported

Two poor quality phase II studies (Mrozek et al., 2006, and Silva et al., 2008) offered limited evidence on CAP + DOC combined therapy which was given at different doses and time intervals between studies. The only outcome reported by both studies was tumour response rate which ranged from 44–50%. Neither study reported median overall survival. The most commonly reported adverse events were neutropenia, diarrhoea and nausea & vomiting.

References
  1. Mrozek E, Ramaswamy B, Young D, Rhoades CA, Kendra K, Allen J, Moore T, Hauger M, Watson H, Merriman N, Nadella P, Villalona-Calero M, Shapiro CL. Phase II study of weekly docetaxel and capecitabine in patients with metastatic breast cancer. Clinical Breast Cancer. 2006;7:141–145. [PubMed: 16800973]
  2. Sezgin C, Kurt E, Evrensel T, Ozdemir N, Manavoglu O, Goker E. Efficacy of lower dose capecitabine in patients with metastatic breast cancer and factors influencing therapeutic response and outcome. South Med J. 2007;100:27–32. [PubMed: 17269522]
  3. Silva O, Lopes G, Morgenzstern D, Lobo C, Doliny P, Santos E, Abdullah S, Gautam U, Reis I, Welsh C, Slingerland J, Hurley J, Gluck S. A phase II trial of split, low-dose docetaxel and low-dose capecitabine: A tolerable and efficacious regimen in the first-line treatment of patients with HER2/neu-negative metastatic breast cancer. Clinical Breast Cancer. 2008;8:162–167. [PubMed: 18621613]
  4. Venturini M, Paridaens R, Rossner D, Vaslamatzis MM, Nortier JW, Salzberg M, Rodrigues H, Bell R. An open-label, multicenter study of outpatient capecitabine monotherapy in 631 patients with pretreated advanced breast cancer. Oncology. 2007;72:51–57. [PubMed: 18004077]
  5. Yap YS, Kendall A, Walsh G, Banerji U, Johnston SR, Smith IE, O’Brien M. Clinical efficacy of capecitabine as first-line chemotherapy in metastatic breast cancer--how low can you go? Breast. 2007;16:420–424. [PubMed: 17379519]

Evidence tables

Question: Capecitabine and docetaxel as a combined therapy or capecitabine as a monotherapy

Created by: Karen Francis on 14/07/08

Mrozek et al. (2006)

Sezgin et al. (2007)

Venturini et al. (2007)

Yap et al. (2007)

Silva et al. (2008)

4.6. Taxanes as first or second line therapy

Short summary

There was good quality evidence on the use of taxanes as first or second line monotherapy or in combination, comprising a high quality Cancer Care Ontario guideline (Verma et al., 2003), two good systematic reviews (Ghersi et al., 2005 and Bria et al., 2005) and four RCTs (Lin et al. 2007, Cassier et al. 2008, Bontenbal et al., 2005 and Jones et al., 2005). The total patient number exceeded 15,000.

Anthracycline naïve women did not derive any benefit from paclitaxel (PAC) as first line monotherapy compared with controls. A large systematic review (Verma et al. 2003) found that for anthracycline naïve patients, when taxanes were added to anthracycline based regimes, there were no significant differences in time to progression (TTP) or overall survival (OS) but tumour response was significantly improved. However, PAC and doxorubicin (DOX) combined therapy resulted in superior median OS and TTP compared with 5′-FU, DOX and cyclophosphamide (FAC) combined. There was no evidence to suggest a significant difference in quality of life between DOC and PAC when either was combined with anthracycline as first-line therapy. One moderate RCT (Bontenbal et al. 2005) demonstrated that DOX and DOC combined therapy in first line treatment of advanced disease resulted in superior tumour response and clinical benefit, when compared with FAC. Time to event analyses also showed significant reductions in the risk of death and time to progression with AT therapy compared to FAC but there were more reports of febrile neutropenia with FAC.

Meta-analysis demonstrated significant improvements in TTP, tumour response and time to treatment failure in favour of taxane containing regimes compared with non-taxane containing regimes and a borderline advantage in OS. However, statistical significance for OS and TTP was lost when only first-line therapy with taxanes was considered. Taxanes and taxane-containing regimes were reported to have a higher incidence of neurotoxicity and leukopenia but fewer cases of nausea and vomiting than controls.

PAC monotherapy was preferable to mitomycin in terms of TTP but not other outcomes. DOC monotherapy correlated with improved OS (compared with combined mitomycin and vinblastine) and improved TTP and tumour response compared with several other multi-agent therapies. Good RCT data (Jones et al. 2005) demonstrated a significant advantage in OS, TTP and response duration for patients on DOC versus PAC monotherapy although the tumour responses were similar. Another RCT (Cassier et al. 2008) found no significant differences in efficacy or survival outcomes between PAC and DOC as first line therapy combined with DOX then given as monotherapy.

PICO question

POPULATIONINTERVENTIONCOMPARISONOUTCOME
Patients with breast cancerPaclitaxel either alone or in combination with other drugs as part of a chemotherapy regimen

Docetaxel either alone or in combination with other drugs as part of a chemotherapy regimen
Standard chemotherapy regimen for advanced breast cancer (e.g. CMF, anthracyclines, mitozantrone, mitomycin)
  • Overall response
  • Progression-free survival
  • Overall survival
  • Symptom relief
  • QOL
  • Adverse events

NB The search strategy developed from this PICO table and used to search the literature for this question can befound in Appendix A

Full evidence summary

Taxanes alone or combined with other agents for first- or second-line chemotherapy

The evidence base for this update includes a high quality Cancer Care Ontario (CCO) guideline, a Cochrane systematic review, one other systematic review and three RCTs. There was minimal overlap in studies selected for inclusion in the different reviews. The total patient number across studies was in excess of 15,000.

The CCO guideline (Verma et al., 2003) examined the taxanes as first or second line therapy for metastatic breast cancer, reviewing data from seventeen RCTs (n = 5,689) which were identified by literature searches up to July 2002. The results were categorised by taxane and by anthracycline exposure as follows:

Paclitaxel in anthracycline-naïve patients

Six trials assessed survival, only one of which determined an advantage with the use of paclitaxel versus control - patients treated with paclitaxel plus doxorubicin experienced significantly superior median overall survival (OS) and median time to progression (TTP) compared with patients treated with 5′-fluorouracil, doxorubicin and cyclophosphamide. Two other studies reported statistically significant improvements in median progression-free survival (PFS), or time to treatment failure (TTF), when doxorubicin was compared with paclitaxel or combined paclitaxel and doxorubicin was compared with either agent alone. Three trials found higher response rates in parallel with these findings, all of which reached statistical significance. One study found that there was no significant difference in response rates between epirubicin with paclitaxel when compared with epirubicin and cyclophosphamide and, similarly, another study did not detect significant differences in either response rate or survival with these regimes.

Docetaxel in anthracycline-naïve patients

Out of four trials, only one reported on survival and tumour response and showed that there was no significant difference between doxorubicin and docetaxel given to patients who had received prior alkylating agents and no more than one line of chemotherapy for advanced or metastatic disease. Two trials reporting on TTP could detect no difference between regimens with or without docetaxel but three trials reported significantly higher response rates with such docetaxel containing therapies.

Paclitaxel in anthracycline-resistant patients

Two small phase II studies investigated differences between paclitaxel and mitomycin or capecitabine. In terms of response rate there was no significant differences between any comparators. Median TTP was significantly longer for patients on paclitaxel compared with mitomycin but not capecitabine and there was no difference in median OS between paclitaxel and mitomycin.

Docetaxel in anthracycline-resistant patients

Three trials compared docetaxel with other combined chemotherapies and a fourth trial examined docetaxel monotherapy compared with docetaxel and capecitabine combined. All participants had received prior anthracycline therapy. In two of the trials comparing docetaxel with non-docetaxel combined therapies, one detected a significant difference in median OS in favour of docetaxel (compared with mitomycin and vinblastine combined) and the other did not (compared with methotrexate and 5′-fluorouracil). Two of the three trials detected significantly longer times to disease progression and higher response rates with docetaxel than with multi-agent chemotherapy. The third trial’s data were incomplete in this respect. Compared with docetaxel monotherapy, treatment with docetaxel and capecitabine resulted in longer duration of survival, PFS and tumour response.

Eight of nine trials reporting on quality of life did not detect statistically significant differences between treatment groups on changes from baseline in measures of quality of life.

Key evidence (taken directly from Verma et al., 2003)

Anthracycline-naive patients

  1. Seven randomized trials assessed the use of paclitaxel in anthracycline-naïve patients and four randomized trials investigated the use of docetaxel in this setting.
  2. One randomized trial evaluated the use of single agent docetaxel versus doxorubicin. The trial reported a higher response rate and less febrile neutropenia, stomatitis, and nausea/vomiting with docetaxel than with doxorubicin monotherapy.
  3. Evidence from the three randomized trials of single-agent paclitaxel versus doxorubicin-based chemotherapy was conflicting.
  4. Paclitaxel or docetaxel, in combination with doxorubicin, was associated with higher response rates compared to standard anthracycline combinations in three randomized trials and longer time to disease progression and survival in one trial. Such therapy, however, was associated with higher rates of grade 3/4 neutropenia and neuropathy compared to standard anthracycline regimens.

Anthracycline-resistant patients

  1. Four randomized trials evaluated the use of docetaxel for anthracycline-resistant metastatic breast cancer and two small randomized trials investigated the use of paclitaxel in this setting.
  2. One of two small randomized trials detected improved time to progression with paclitaxel compared to non-taxane-containing chemotherapy. The other trial reported no significant difference in time to progression.
  3. Two of three randomized trials that compared docetaxel with non-taxane-containing chemotherapy detected improved response rates and time to progression with docetaxel, while the third reported no significant difference for these outcome measures. One trial also detected a significant survival advantage with docetaxel compared to mitomycin/vinblastine. The other trial that reported survival data did not detect a significant survival difference.
  4. One randomized trial that compared docetaxel plus capecitabine to docetaxel alone demonstrated a superior response rate, time to progression, and survival rate for the combination, with high rates of toxicity in both treatment arms.

The Cochrane review (Ghersi et al., 2005) was a high quality review of twenty-one eligible trials (n ~6,300) which compared taxane-containing therapies with non taxane-containing therapies for the treatment of advanced (metastatic) breast cancer. Data were sub-grouped by the nature of the study design e.g. single vs combined chemotherapy but were also presented across all studies by outcome. Data were not sub-grouped by taxane as per Verma et al. (2003).

The pooled data came from trials of very varied design - the statistical heterogeneity which was observed may have arisen, for example, because of the different treatment regimens in the comparator arms of individual studies. The authors’ conclusions that taxane-containing therapies are better than some, but not all, non taxane-containing therapies may be an acknowledgment of this heterogeneity. Using a random-effects model does not adjust for these differences which must therefore be borne in mind when considering the apparent significance of results. The data across all studies was presented only for assessable patients – an ITT analysis was performed for sub-groups.

Overall survival

Across all studies, the hazard ratio for OS was 0.93 (95%CI: 0.86–1.00 P = 0.05) which is of borderline significance in favour of taxane-combining regimens compared with non-taxane regimens. When only data from those studies comparing regimens for first-line treatment were combined, the findings lost statistical significance. There was no between-studies heterogeneity.

Time to progression

Across all studies, the hazard ratio for TTP was 0.92 (95%CI: 0.85–0.99 P = 0.02) – a significant advantage in favour of taxane-containing regimens but with the caveat that there was significant between-studies heterogeneity (I2 = 83.9%). Again, looking only at data for first-line therapy, the results lost significance but the between-studies heterogeneity remained (I2 = 75.0%).

Tumour response

The pooled data for assessable patients suggested a statistically significant advantage of taxane-containing regimens for tumour response (OR = 1.34 95%CI: 1.18–1.52 P < 0.00001) although the between-studies heterogeneity was also significant (I2 =72.99%). The data for first-line therapy showed similar results (OR = 1.28 95%CI: 1.10–1.50 P = 0.002) with significant heterogeneity between studies (I2 = 85.2%).

Time to treatment failure

Taxane-containing regimens showed a statistically significant advantage over non taxane-containing regimens with regard to the median TTF (HR = 0.79 95%CI: 0.71–0.88 P < 0.0001) and with no significant between-studies heterogeneity.

Data for all outcomes were analysed using a fixed effect model. Significant between studies heterogeneity may adversely affect the validity of any apparently significant results obtained. The authors, rather than re-applying the data to a random effects model, offered a plausible explanation for the observed heterogeneity. The principal reason given was the varying efficacy of comparator regimes, for example mitomycin, vinblastine and fluorouracil with vinorelbine which, in the authors’ opinion, may be regarded as sub-optimal in the treatment of breast cancer. The authors’ conclusion, that taxane-containing therapies are better than some, but not all, non taxane-containing therapies, is an acknowledgment of this heterogeneity.

The QOL and toxicity data were reported in tabular and in narrative form. Briefly, when data from studies examining taxane-containing combination therapies vs non taxane-containing controls were pooled, the taxane-containing arm reported significantly higher incidences of neurotoxicity and leukopenia but fewer of nausea and vomiting. Only neurotoxicity was significantly higher when taxane monotherapies were compared with non taxane-containing controls.

A second systematic review (Bria et al., 2005) pooled results from seven trials (n = 2085) that had compared taxane-containing, anthracycline based therapies with non taxane-containing, anthracycline based therapies. The primary outcome was TTP which showed no clear advantage for taxane- anthracycline combined therapy and, similarly, pooled data showed no significant difference between interventions and comparators for OS. However, tumour response was significantly different between arms in favour of taxane-containing therapies (P < 0.001) by either method of data analysis (using median time to event data or log of relative risk) as was the increase in CR of between 81–104% depending on method (P < 0.001).

The taxane-anthracycline combined therapy carried a significantly higher risk of neutropenia than comparator therapies, with a relative risk of between 2.82 and 3.44 (P < 0.001) but it was noted that there was also significant between-studies heterogeneity (P < 0.01). The authors concluded that, despite the higher haematological toxicity, the combination of an anthracycline and a taxane might be an active chemotherapy for patients with MBC. Although this was a very thorough analysis, much of the data for it came from abstracts i.e. unpublished works which were unlikely to have been peer-reviewed.

Bontenbal et al. (2005) presented the results from a moderate RCT which compared combined docetaxel and doxorubicin (AT) with FAC (5′-fluoruracil, doxorubicin and cyclophosphamide) in the first line treatment of women with histologically proven metastatic breast cancer. The study was terminated early due to poor accrual but the data were sufficient to demonstrate a tumour response (no odds ratio given but P=0.003) and clinical benefit (P=0.03) that were significantly different between study arms in favour of the AT regime. Time to event analyses also showed advantages with AT for median OS (HR = 1.43 (95%CI: 1.06–1.92) P=0.019) and median TTP (HR = 1.50 (95%CI: 1.13–1.98) P=0.004) values which remained significant even after adjustment for prognostic factors in a multivariate analysis. There were higher reported levels of febrile neutropenia with AT (33% vs 9%) versus FAC. Details of allocation and randomisation were omitted but tumour response was undertaken independently. The authors concluded that AT as first line therapy could result in superior tumour response and survival outcomes than FAC.

There were two studies which compared the taxanes as first or second line therapy in women with locally advanced or metastatic breast cancer. Jones et al. (2005) presented a RCT (n = 449) comparing docetaxel with paclitaxel as second-line therapy for women with advanced breast cancer (90% with MBC) who had received prior treatment with anthracyclines.

The tumour response in the ITT analysis was not significantly different between treatment arms (DOC = 32% vs PAC = 25%) but median OS (DOC = 15.4 months vs PAC = 12.7 months), median TTP (DOC = 5.7 months vs PAC = 3.6 months) and median RD (DOC = 7.5 vs PAC = 4.6 months) were all statistically significantly superior for the DOC arm participants.

The authors concluded that three-weekly DOC was superior to PAC in TTP, OS, RD and ORR (non-significant) and that the survival advantage was apparent despite higher toxicity in the DOC arm which had led to more dose reductions and withdrawals. The majority of the authors disclosed personal interests in Aventis (manufacturer of docetaxel).

Lin et al. (2007) presented a small study of 101 patients who received combination therapy of either docetaxel or paclitaxel with cisplatin for the first or second line treatment of MBC. Prior exposure to anthracycline was a study inclusion, either as neo-adjuvant or adjuvant therapy. Tumour response (DOC + CIS = 62.5% vs PAC + CIS = 42.6), median OS (DOC + CIS = 22.7 months vs PAC + CIS = 22.5 months) and median TTP (DOC + CIS = 9.8 months vs PAC + CIS = 6.5 months) all favoured docetaxel but the results were not statistically significant.

It is not possible to infer that the lack of clear advantage is necessarily because of low patient numbers but a larger study would perhaps have been more conclusive regardless of outcome. In addition, this study, although probably rigorous in its conduct was not well documented with regard to methodology so that although patients were said to have been randomised into two study arms, the complete lack of details about randomisation, allocation etc means that there is a strong possibility of bias.

References
  1. Bontenbal M, Creemers GJ, Braun HJ, de Boer AC, Janssen JT, Leys RB, Ruit JB, Goey SH, van dV, Kerkhofs LG, Schothorst KL, Schmitz PI, Bokma HJ, Verweij J, Seynaeve C. Phase II to III study comparing doxorubicin and docetaxel with fluorouracil, doxorubicin, and cyclophosphamide as first-line chemotherapy in patients with metastatic breast cancer: results of a Dutch Community Setting Trial for the Clinical Trial Group of the Comprehensive Cancer Centre. J Clin Oncol. 2005;23:7081–7088. [PubMed: 16192591]
  2. Bria E, Giannarelli D, Felici A, Peters WP, Nistico C, Vanni B, Cuppone F, Cognetti F, Terzoli E. Taxanes with anthracyclines as first-line chemotherapy for metastatic breast carcinoma. Cancer. 2005;103(4):672–679. [PubMed: 15637696]
  3. Ghersi D, Wilcken N, Simes J, Donoghue E. Taxane containing regimens for metastatic breast cancer Cochrane Database of Systematic Reviews: Reviews, Cochrane Database of Systematic Reviews. 2. Chichester (UK): John Wiley & Sons Ltd; 2005. [PubMed: 15846659]
  4. Jones SE, Erban J, Overmoyer B, Budd GT, Hutchins L, Lower E, Laufman L, Sundaram S, Urba WJ, Pritchard KI, Mennel R, Richards D, Olsen S, Meyers ML, Ravdin PM. Randomized phase III study of docetaxel compared with paclitaxel in metastatic breast cancer. J Clin Oncol. 2005;23(24):5542–5551. [PubMed: 16110015]
  5. Lin YC, Chang HK, Chen JS, Wang HM, Yang TS, Liaw CC. A Phase II Randomized Study of Two Taxanes and Cisplatin for Metastatic Breast Cancer after Anthracycline: A Final Analysis. Jpn J Oncol. 2007;37(1):23–29. [PubMed: 17172351]
  6. Verma S, Trudeau M, Pritchard K, Oliver T, Robinson P. the Cancer Care Ontario Practice Guidelines Initiative Breast Cancer Disease Site Group. Role of taxanes in the management of metastatic breast cancer. Curr Oncol. 2003;10(2):68–83.

Evidence tables

Question: Taxanes as first or second line therapy

Created by: Karen Francis on 24/07/2007

Breast Cancer Disease Site Group et al. (2003)

References included in the Breast Cancer Disease Site Group et al. (2003) systematic review

  1. Paridaens R, Biganzoli L, Bruning P, et al. Paclitaxel versus doxorubicin as first-line single-agent chemotherapy for metastatic breast cancer: a European Organization for Research and Treatment of Cancer randomized study with cross-over. J Clin Oncol. 2000;18:724–33. [PubMed: 10673513]
  2. Bishop J, Dewar J, Toner G, et al. Initial paclitaxel improves outcome compared with CMFP combination chemotherapy as front-line therapy in untreated metastatic breast cancer. J Clin Oncol. 1999;17:2355–64. [PubMed: 10561297]
  3. Sledge G, Neuberg D, Ingle J, et al. Phase III trial of doxorubicin vs. paclitaxel vs. doxorubicin + paclitaxel as first line therapy for metastatic breast cancer: An Intergroup Trial [abstract] Proc Am Soc Clin Oncol. 1997;16:1a.
  4. Jassem J, Piekowski T, Puzaska A, et al. Doxorubicin and paclitaxel versus fluorouracil., doxorubicin., and cyclophosphamide as first-line therapy for women with metastatic breast cancer: final results of a randomized phase III multicenter trial. J Clin Oncol. 2001;19:1707–15. [PubMed: 11251000]
  5. Carmichael J. UKCCCR trial of epirubicin and cyclophosphamide vs. epirubicin and Taxol in the first line treatment of women with metastatic breast cancer [abstract] Proc Am Soc Clin Oncol. 2001;20 Abstract #84.
  6. Luck HJ, Thomssen C, Untch M, Kuhn W, Eidtmann H, du Bois A, et al. Multicentric phase III study in first line treatment of advanced metastatic breast cancer. epirubicin/paclitaxel vs epirubicin/cyclophosphamide. A study of the AGO Breast Cancer Group [abstract] Proc Am Soc Clin Oncol. 2000;19:73a.
  7. Biganzoli L, Cufer T, Bruning P, et al. Doxorubicin and paclitaxel versus doxorubicin and cyclophosphamide as first line chemotherapy in metastatic breast cancer: The European Organization for Research and Treatment of Cancer 10961 Multicentre Phase III Trial. J Clin Oncol. 2002;20:3114–21. [PubMed: 12118025]
  8. Chan S, Friedrichs K, Noel D, et al. Prospective randomized trial of docetaxel versus doxorubicin in patients with metastatic breast cancer. J Clin Oncol. 1999;17:341–54. [PubMed: 10561296]
  9. Nabholtz JM, Falkson G, Campos D, et al. A phase III trial comparing doxorubicin and docetaxel to doxorubicin and cyclophosphamide as first line chemotherapy for MBC [abstract] Proc Am Soc Clin Oncol. 1999;18:127a.
  10. Nabholtz JM, Paterson A, Dirix L, et al. A phase III randomized trial comparing docetaxel., doxorubicin and cyclophosphamide to FAC as first line chemotherapy for patients with metastatic breast cancer [abstract] Proc Am Soc Clin Oncol. 2001;20 abstract #83.
  11. Bonneterre J, Dieras V, Tubiana-Hulin M, et al. 6 cycles of epirubicin/docetaxel versus 6 cycles of 5FU epirubicin/cyclophosphamide as first line metastatic breast cancer treatment [abstract] Proc Am Soc Clin Oncol. 2001;20 abstract #163.
  12. Dieras V, Marty M, Tubiana N, et al. Phase II randomized study of paclitaxel versus mitomycin in advanced breast cancer. Semin Oncol. 1995;22(Suppl 8):33–9. [PubMed: 7638640]
  13. O’Reilly SM, Moiseyenko V, Talbot DC, et al. A randomized phase II study Of Xelodatm (capecitabine) vs paclitaxel in breast cancer patients failing previous anthracycline therapy [abstract] Proc Am Soc Clin Oncol. 1998;17:163a.27.
  14. Nabholtz JM, Senn HJ, Bezwoda WR, et al. Prospective randomized trial of docetaxel versus mitomycin plus vinblastine in patients with metastatic breast cancer progressing despite previous anthracycline-containing chemotherapy. J Clin Oncol. 1999;17:1413–24. [PubMed: 10334526]
  15. Sjostrom J, Blomqvist C, Mouridsen H, et al. Docetaxel compared with sequential methotrexate and 5-fluorouracil in patients with advanced breast cancer after anthracycline failure: a randomised phase III study with crossover on progression by the Scandinavian Breast Group. Eur J Cancer. 1999;35:194–201. [PubMed: 10615229]
  16. Hakamies-Blomqvist L, Luoma M, Sjostrom J, et al. Quality of life in patients with metastatic breast cancer receiving either docetaxel or sequential methotrexate and 5-fluorouracil. A multicentre randomised phase III trial by the Scandinavian breast group. Eur J Cancer. 2000;36:1411–7. [PubMed: 10899655]
  17. Bonneterre J, Roche H, Monnier A, et al. Taxotere versus 5-fluorouracil + Navelbine as second-line chemotherapy in patients with metastatic breast cancer (preliminary results) [abstract] Proc Am Soc Clin Oncol. 1997;16:162a.
  18. O’Shaughnessy J, Miles D, Vukelja S, et al. Superior survival with capecitabine plus docetaxel combination therapy in anthracycline-pretreated patients with advanced breast cancer: phase III trial results. J Clin Oncol. 2002;20:2812–23. [PubMed: 12065558]

Bontenbal et al. (2005)

Ghersi et al. (2005)

Bria et al. (2005)

Jones et al. (2005)

Lin et al. (2007)

Updated evidence 4.6

Summary

There was only one further paper to update the evidence on taxanes as a monotherapy or in combination with other agents. This RCT (Cassier et al., 2008) compared first line docetaxel (DOC) with paclitaxel (PAC), each combined with doxorubicin and then, after four cycles, given as monotherapy. The primary outcome was quality of life (QOL) which was found not to be significantly different between study groups despite the different toxicity profiles for the taxanes. Secondary outcomes including efficacy and survival were also not significantly different between study arms. Despite the trial being thoroughly conducted there were concerns that bias may have been introduced both from the lack of detail of methodology and also because the QOL questionnaires were self administered. Those women who were in poorer health may have been less likely to have complied with the assessment tool and have therefore excluded themselves from the analysis.

Reference
  1. Cassier PA, Chabaud S, Trillet-Lenoir V, Peaud PY, Tigaud JD, Cure H, Orfeuvre H, Salles B, Martin C, Jacquin JP, Agostini C, Guastalla JP, Perol D, Bachelot T. A phase-III trial of doxorubicin and docetaxel versus doxorubicin and paclitaxel in metastatic breast cancer: results of the ERASME 3 study. Breast Cancer Res Treat. 2008;109:343–350. [PubMed: 17611792]

Evidence tables

Question: Taxanes as first or second line therapy

Created by: Karen Francis on 24/07/2008

Cassier et al. (2008)

4.7. The management of patients with metastatic HER2+ breast cancer that are having ongoing treatment with biological therapy

Short summary

For patients undergoing therapy with a biological therapy who experience disease progression there was only limited evidence on trastuzumab (TRZ) which comprised a prospective post RCT study (Tripathy et al., 2004) five retrospective case series (Fountzilas et al., 2003, Gelmon et al., 2004, Garcia-Saenz et al., 2005, Montemurro et al., 2006 and Stemmler et al., 2005) and a phase II study (Bartsch et al., 2006).

Limited data from a post-RCT analysis (Tripathy et al. 2004) showed no significant improvements in safety or efficacy for women with disease progression who continued TRZ combined with different chemotherapies when compared with women in whom TRZ was given for the first time after their disease progressed on chemotherapy alone. Most case series also offered little evidence in support of continuing TRZ therapy beyond progression since, where relevant comparisons were made, no significant improvements were found for survival, efficacy or safety.

One retrospective case series (Garcia-Saenz et al. 2005) demonstrated a significant survival advantage for women who had received both first and second line therapy with TRZ but, taken from a non-randomised study, the data was open to strong selection bias. Weak phase II evidence (Bartsch et al. 2006) showed no significant difference in the length of time to progression between first, second or further lines of TRZ therapy which was interpreted as support for TRZ continuation.

PICO question

POPULATIONINTERVENTIONCOMPARISONOUTCOME
Patients with metastatic HER2 +ve breast cancer who are having ongoing treatment with a biological therapy and have disease progression Compare each intervention with each other
  • Disease response
  • Progression free survival
  • Overall survival
  • QOL
  • Time to progression
  • Toxicity
  • Patient tolerability
  • Cost effectiveness

NB The search strategy developed from this PICO table and used to search the literature for this question can be found in Appendix A

Full evidence summary

The evidence comprises a prospective cohort study from the USA, four retrospective case series from Europe (n = 3) and Canada (n = 1) and a phase II study from Europe. Non-comparative studies and retrospective data analyses are, by design, only very weak evidence compared with, for example a RCT or other controlled study.

Overall survival

Montemurro et al. (2006) studied a group of MBC patients who had been previously treated with trastuzumab (T). Of the original 184 patients, 132 experienced disease progression and of these, 71 stopped TRZ treatment and were given either a variety of chemotherapies (n = 61) or endocrine therapy (n = 10). 40 patients remained on T-based therapy (5 had TRZ + endocrine therapy, 23 had TRZ + chemotherapy and 12 had TRZ monotherapy). The median overall survival (OS) for those patients continuing TRZ was 30.1 months (95%CI: 26.9–33.3) compared with 30.2 months (95%CI: 20.5–39.9) for patients who had stopped T. For 21 patients with rapid disease progression and who were unable to be treated except in the palliative setting, median OS was 7.4 months (95%CI: 4.6–10.3).

Fountzilas et al. (2003) performed a retrospective analysis of data from patients with progressive MBC who continued on T, either as a monotherapy or in combination with a variety of chemotherapeutic drugs. They reported median OS (from the time of the first TRZ therapy) as 22.2 months (95%CI: 16.3–28.2). Without a comparator, it would be difficult to assess the relative worth of TRZ continuation in this setting. Gelmon et al. (2004) also performed a retrospective analysis of data from patients progressing on TRZ who continued either TRZ as a monotherapy or in combination with various chemotherapeutic agents.

The median OS for all patients was reported as 29 months (95%CI: 22.7–56). The problem with all retrospective studies is that they are prone to selection bias. Data were only chosen from patients who were fit enough to continue treatment after progression or who had not been forced to stop TRZ therapy for reasons other than disease progression i.e. toxicity. For these patients, TRZ continuation was presumably not an option and hence the introduction of bias – this should be considered if such studies are used to claim superior survival or tumour response after continuing TRZ therapy on disease progression.

Tumour progression

Montemurro et al. (2006) reported the median TTP for patients continuing TRZ therapy as 6.3 months (95%CI: 5–7.6) and for patients stopping TRZ therapy and continuing other treatment as 7.2 months (95%CI: 6.4–7.9).

Fountzilas et al. (2003) reports median TTP for each treatment group: TRZ monotherapy = 5.1 months (95%CI: 2.6–7.6), TRZ + PAC = 4.2 months (95%CI: 0–9.3), TRZ + VIN = 4.6 months (95%CI: 3.3–5.9), TRZ + GEM = 6.2 months (95%CI: 4.7–7.7 and TRZ + DOC = 4.1 months (95%CI: 0–11.6). The individual sub-group patient numbers are very low which does confound reliability.

Similarly, Gelmon et al. (2005) reported TTP by line and by sub-group: 1st line: TRZ monotherapy = 23 weeks (range: 3–73), TRZ + taxane = 24 weeks; 2nd line: monotherapy = 30.5 weeks (range: 18–68), TRZ + taxane = 24 weeks (range: 3–72) and TRZ + VIN = 26 weeks (range: 3–108). Evidence is also reported by Garcia-Saenz et al. (2005): 1st line: TRZ + chemotherapy = 6 months (range: 1–>39 months), 2nd line: TRZ + chemotherapy = 3 months (range: 1–>22) and 3rd line: TRZ + chemotherapy = 2 months (range: 1–>12).

Bartsch et al. (2006) supplied data on TTP with confidence intervals and comparative statistics: 1st line = 6 months (95%CI: 5.4–6.6), 2nd line = 6 months (95%CI: 5.36–6.64) and beyond 2nd line = 6 months (95%CI: 5.32–6.68) (no significant difference). The authors claim that the maintenance of TTP supports a case for continued TRZ therapy on progression but the patient numbers were low and the lack of a control group weakens the strength of this evidence.

Tumour response

Tripathy et al. (2004) reported on H0649, an extension to the pivotal trial (H0648) which was the main study in the NICE technology appraisal ‘Guidance on the use of trastuzumab for the treatment of advanced breast cancer’. H0648 patients were randomised to receive either TRZ with or without chemotherapy and, in this post-trial study, those patients experiencing disease progression either had their chemotherapy changed but remained on TRZ (group II, n = 93) or, if they had not previously received T, were offered TRZ usually with chemotherapy (group I, n = 154).

The overall response rate (ORR) was given as 14% (95%CI: 8.3–19.2) for group I compared with 11% (95%CI: 4.5–17) for group II. It’s difficult to interpret this result since one cannot know whether patients who had not previously received TRZ had an accelerated tumour response or whether patients continuing with TRZ did not benefit as much as with their first TRZ regime such that the two groups ‘caught up’ with one another. Patients with minor responses or stable disease were 18% (group I) and 12% (group II).

Montemurro et al. (2006) only presented the tumour response results for the first TRZ regime but not for continuing patients.

Fountzilas et al. (2003) in the retrospective analysis, reports ORR for each line of treatment: 2nd line ORR = 24%, 3rd line ORR = 14%, 4th line ORR = 19% and 5th line ORR = 8%. Gelmon et al. (2004) also presented data in this way: 1st line ORR = 41%, 2nd line ORR = 36% (TRZ monotherapy) 1st line ORR = 38%, 2nd line ORR = 38% (TRZ + taxane) and ORR = 27% (TRZ + VIN) as did the authors of a third retrospective case series (Garcia-Saenz et al., 2005): 1st line ORR = 39.7%, 2nd line ORR = 25.8%, 3rd line ORR = 12.5%, 4th line ORR = 0% (TRZ + chemotherapy). Bartsch et al. (2006) reported tumour response for 2 lines and beyond: 1st line ORR = 42.6%, 2nd line ORR = 25.9%, beyond 2nd line ORR = 30%.

Adverse events

Only half the studies reported adverse events. Tripathy et al. (2004) observed ‘severe’ events in 5% of the total patient population, including asthenia (11%), back pain (6%), headache (9%), general pain (6%), leukopenia (8%) and dyspnoea (7%) in patients who had previously not been exposed to TRZ and asthenia (10%), back pain (6%), headache (6%), general pain (10%), leukopenia (11%) and dyspnoea (2%) in patients continuing on TRZ therapy (group I). There were no events reported for one group alone. Additionally, there was cardiac dysfunction in 9% of group I patients and 2% of group II patients.

Fountzilas et al. (2003) also reported ‘severe’ events in 5% or more of the total patient population, including neutropenia (25%), fatigue (12.5%), thrombocytopenia (1.5%), infection (10%), peripheral neuropathy (9%), myalgia & arthralgia (5%), nausea & vomiting (6%), stomatitis (6%), diarrhoea (6%), constipation (6%) and oedema (6%). In addition there were 8 cases of cardiotoxicity and 3 of LVEF reduction.

Gelmon et al. (2006) reported grade 3 adverse events: nausea & vomiting (1.9%), neutropenia (25.9%), thrombocytopenia (3.7%), anaemia (11.1%), stomatitis (3.7%) and hand-foot syndrome (5.6%) and one grade 4 event of neutropenia (9.3%). There was no cases of heart failure reported and only patient experienced a drop in LVEF which normalised after treatment.

References

  1. Bartsch R, Wenzel C, Hussian D, Pluschnig U, Sevelda U, Koestler W, Altorjai G, Locker GJ, Mader R, Zielinski CC, Steger GG. Analysis of trastuzumab and chemotherapy in advanced breast cancer after the failure of at least one earlier combination: An observational study. BMC Cancer. 2006;6 [PMC free article: PMC1431554] [PubMed: 16539726]
  2. Fountzilas G, Razis E, Tsavdaridis D, Karina M, Labropoulos S, Christodoulou C, Mavroudis D, Gogas H, Georgoulias V, Skarlos D. Continuation of trastuzumab beyond disease progression is feasible and safe in patients with metastatic breast cancer: a retrospective analysis of 80 cases by the Hellenic Cooperative Oncology Group. Clin Breast Cancer. 2003 Jun;4:120–125. 2003. [PubMed: 12864940]
  3. Garcia-Saenz JA, Martin M, Puente J, Lopez-Tarruella S, Casado A, Moreno F, Grande E, az-Rubio E. Trastuzumab associated with successive cytotoxic therapies beyond disease progression in metastatic breast cancer. Clin Breast Cancer. 2005;6:325–329. [PubMed: 16277882]
  4. Gelmon KA, Mackey J, Verma S, Gertler SZ, Bangemann N, Klimo P, Schneeweiss A, Bremer K, Soulieres D, Tonkin K, Bell R, Heinrich B, Grenier D, Dias R. Use of trastuzumab beyond disease progression: observations from a retrospective review of case histories. Clin Breast Cancer. 2004 Apr 5:52–58. 2004. [PubMed: 15140285]
  5. Montemurro F, Donadio M, Clavarezza M, Redana S, Jacomuzzi ME, Valabrega G, Danese S, Vietti-Ramus G, Durando A, Venturini M, Aglietta M. Outcome of patients with HER2-positive advanced breast cancer progressing during trastuzumab-based therapy. Oncologist. 2006;11:318–324. [PubMed: 16614227]
  6. Tripathy D, Slamon DJ, Cobleigh M, Arnold A, Saleh M, Mortimer JE, Murphy M, Stewart SJ. Safety of treatment of metastatic breast cancer with trastuzumab beyond disease progression. J Clin Oncol. 2004;22:1063–1070. [PubMed: 15020607]

Evidence tables

Question: Ongoing treatment with a BRM

Created by: Karen Francis on 25/10/2006 16:17:14

Tripathy et al. (2004)

Fountzilas et al. (2003)

Gelmon et al. (2004)

Garcia-Saenz et al. (2005)

Bartsch et al. (2006)

Montemurro et al. (2006)

Updated evidence (4.7)

Summary

One retrospective study was identified to update the evidence on continuing trastuzumab beyond disease progression (Stemmler et al. 2005).

The case series demonstrated a significant survival advantage for women who had received both first and second line therapy with TRZ but, as a non-randomised retrospective study, the analysis was open to strong selection bias.

Reference
  1. Stemmler HJ, Kahlert S, Siekiera W, Untch M, Heinrich B, Heinemann V. Characteristics of patients with brain metastases receiving trastuzumab for HER2 overexpressing metastatic breast cancer. Breast. 2006;15:219–225. [PubMed: 16026983]

Evidence table

Question: Ongoing treatment with a BRM

Created by: Karen Francis on 22/07/2008

Stemmler et al. (2005)

Copyright © 2009, National Collaborating Centre for Cancer.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licenses issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

Bookshelf ID: NBK61872