Correspondence
Weekly Paclitaxel in the Adjuvant Treatment of Breast Cancer
N Engl J Med 2008; 359:310-311July 17, 2008
- Article
To the Editor:
Sparano et al. (April 17 issue)1 compare the efficacy of two different taxanes, given either weekly or every 3 weeks, in the adjuvant treatment of breast cancer. It is striking that there were no significant differences in disease-free survival between the group of patients receiving paclitaxel and the group receiving docetaxel or between the group receiving weekly treatment and the group receiving treatment every 3 weeks, whereas multivariate analysis with the use of a Cox proportional-hazards model showed significantly longer disease-free survival in the group receiving weekly paclitaxel than in the group receiving paclitaxel every 3 weeks. Clinicians will have difficulties in drawing a definite conclusion from the study because of this disparity, which might be associated with unrecognized bias.
1 ReferencesMasaharu Tsubokura, M.D.
Teikyo University, Ichihara 2990111, Japan
tsubokura-tky@umin.ac. jp Masahiro Kami, M.D.
University of Tokyo, Tokyo 1088639, JapanTsunehiko Komatsu, M.D.
Teikyo University, Ichihara 2990111, Japan1
Sparano JA ,Wang M ,Martino S , et al. Weekly paclitaxel in the adjuvant treatment of breast cancer. N Engl J Med 2008;358:1663-1671
Full Text | Web of Science | Medline
To the Editor:
In the study reported by Sparano et al., the total dose of paclitaxel in the group receiving paclitaxel weekly was higher than the total dose in the group receiving paclitaxel every 3 weeks (960 mg per square meter of body-surface area vs. 700 mg per square meter). This was not the case for patients who received docetaxel (420 mg per square meter in the group receiving weekly docetaxel vs. 400 mg per square meter in the group receiving docetaxel every 3 weeks). The higher dose intensity in the group receiving weekly paclitaxel may have contributed to the outcome.
Sercan Aksoy, M.D.
Omer Dizdar, M.D.
Kadri Altundag, M.D.
Hacettepe University, Ankara 06100, TurkeyAuthor/Editor Response
Tsubokura and colleagues question how a significant improvement in disease-free survival and overall survival could be observed in the group receiving weekly paclitaxel (as compared with the group receiving paclitaxel every 3 weeks) when there was no difference in the primary comparisons of taxane type (paclitaxel vs. docetaxel) and schedule (every 3 weeks vs. weekly). The trial used a two-by-two factorial design for the primary comparisons and was sufficiently powered to allow a secondary comparison of the standard-treatment group (paclitaxel every 3 weeks) with the three experimental groups (weekly paclitaxel, docetaxel every 3 weeks, and weekly docetaxel). As we state in the article, there was a significant interaction between docetaxel and the weekly schedule with regard to both disease-free survival (P=0.003) and overall survival (P=0.01); this interaction affected the primary but not secondary comparisons. The clinical relevance and importance of the significant improvement in the group that received weekly paclitaxel in a secondary comparison are not diminished by the failure to show a difference in the primary comparisons.
Aksoy and colleagues point out that the cumulative dose of taxane given weekly as compared with every 3 weeks was substantially higher with paclitaxel (960 vs. 700 mg per square meter) than with docetaxel (420 vs. 400 mg per square meter), implying that a benefit of weekly docetaxel may have been observed had a similar weekly dose of docetaxel (equivalent to 45 mg per square meter) been used. Patients treated with weekly docetaxel, as compared with patients treated with weekly paclitaxel, were more likely to have grade 3 or grade 4 toxic effects (45% vs. 28%), require dose modification (40% vs. 29%), and receive fewer than the 12 planned doses (25% vs. 12%), indicating that a docetaxel dose of more than the 35 mg per square meter used in our trial would not have been feasible.
Joseph A. Sparano, M.D.
Montefiore–Einstein Cancer Center, Bronx, NY 10461
jsparano@montefiore.org
