Overall survival data were available for eleven trials. Pooled analyses of the trial results indicated that taxane-containing regimens improved overall survival, with a relative risk reduction of 19% compared to non-taxane-containing control groups (p<0.00001).2 The absolute risk reduction for taxane-containing regimens was ~2.6% compared to non-taxane containing regimens.2
Disease-free survival data were available for eleven trials. Pooled analyses of the trial results indicated that taxane-containing regimens improved disease-free survival, with a relative risk reduction of 19% compared to non-taxane-containing control groups (p<0.00001).2 The absolute risk reduction for taxane-containing regimens was ~4.1% compared to non-taxane containing regimens.2
Post-hoc analysis of the data by various sub-groups of trials did not alter the overall or disease-free survival estimates,2 see Table 2 (all remained statistically significant).
The Cochrane meta-analysis2 found statistically significant overall survival and disease-free survival benefits in side-by-side post-hoc analyses that were not altered by scheduling or duration of chemotherapy. Taxane treatment was given either concurrently with anthracycline or sequentially in the trials. Meta-analysis2 results investigated trial data where there was an addition of a taxane to the control chemotherapy regimen and where the taxane was substituted for part of the control chemotherapy regimen. Optimal scheduling of adjuvant taxane-containing chemotherapy regimens is unclear from the available trial results. The recommendation on scheduling reflects that regardless of how a taxane is used in a chemotherapy regimen, there are overall survival and disease-free survival benefits for women. Results of ongoing trials may provide further evidence for optimal scheduling.
Questions regarding dose density were outside the scope of the Cochrane meta-analysis.
There were no available trial data about taxanes and HER2 status at the time of the Cochrane review and meta-analysis2 (May 2007). Early and retrospective analysis of tissue from the CALGB26 study (published after March 2007) indicates that the addition of paclitaxel to an adjuvant chemotherapy regimen improved outcomes in HER2 positive women. Further trial results are required to determine definitive recommendations regarding taxanes and HER2 positive early breast cancer patients. This guideline relates specifically to taxanes in adjuvant chemotherapy regimens. It is acknowledged that taxanes have a key role in adjuvant trastuzumab (Herceptin ) trials.
The twelve trials included in the Cochrane review used a variety of control chemotherapies and different dosing and scheduling of the taxane drug. Toxicities need to be considered on a trial-by-trial basis to interpret tolerability of each taxane-containing regimen.
Pooled analysis of six trials reporting data on cardiotoxicity showed no difference in the risk of cardiotoxicity between taxane-containing and non-taxane-containing regimens. However, for trials in which the use of a taxane resulted in a reduction8,13 or omission of anthracycline5, the risk of cardiotoxicity was reduced in the taxane-containing arm (OR 0.38, 95% CI 0.15-0.98, p=0.05).2
Pooled analysis from seven trials demonstrated a significant increase in febrile neutropenia associated with the taxane-containing regimens (p<0.0001).2 The risk was particularly high in the trials that administered the taxane concurrently with an anthracycline, rather than in those that administered sequential taxane and anthracycline treatment.2
Other adverse events
Grade III/IV nausea and/or vomiting were less common in the taxane-containing regimens (OR 0.55, 95% CI 0.39-0.77, p=0.0006).2 There was no difference shown between treatment arms for both grade III/IV fatigue or grade III/IV stomatitis. There was no difference between treatment groups in the number of cases of secondary leukaemia or myelodysplasia reported (25 from taxane-containing regimens, 23 from control regimens). Treatment-related deaths were uncommon and there was no difference between taxane-containing and non-taxane-containing groups (seven treatment-related deaths in each treatment group). There were insufficient and inconsistently reported data for meta-analysis of neurotoxicity or nail changes; however, both toxicities were reported with greater frequency in the taxane-containing arm where reported, with the exception of one study12 that contained vinorelbine in the control arm and reported greater neurotoxicity in this arm.
Formal analyses of quality of life for patients receiving taxanes are limited, with only two trials reporting on quality of life data. One trial6 reported that both treatment arms experienced a transient reduction in the quality of life score, with a greater reduction in the taxane-containing regimen. However, at first follow-up, both treatment arms had returned to base-line. The other trial10 did not demonstrate any difference in quality of life scores between the two treatment arms, either at the beginning or at the end of chemotherapy. Further research is required to determine the short and long-term effects of taxanes on quality of life.
Evidence is available from eight randomised trials16 23 assessing the neoadjuvant use of taxanes for early breast cancer. Only one of the trials16 was large enough to detect survival differences (N=1605), the remaining trials were small, enrolling between 35 and 365 participants. The trials show no significant difference between taxane-containing and non-taxane-containing neoadjuvant chemotherapy regimens for overall, disease-free or relapse-free survival.16 18,22 There was a trend that taxane-containing regimens achieved higher clinical and pathological response rates compared to non-taxane-containing regimens,17 23 however, only one trial reported a statistically significant increase.16 Breast conserving therapy is performed at least as often after taxane-containing neoadjuvant chemotherapy compared to non-taxane-containing neoadjuvant chemotherapy.16,18,21,22 Toxicity outcomes are not consistently reported in all trials and often the number of events are small, therefore it is difficult to determine whether the toxicity profiles differ significantly between taxane and non-taxane-containing neoadjuvant regimens. The most commonly, and consistently, reported outcome was febrile neutropenia, with four trials reporting higher rates in the taxane-containing arms compared to the control arms.16-18,22 Further information is needed to determine the optimal role of neoadjuvant taxanes for treatment of early breast cancer.