Abstract GS2-06: Taxane with anthracycline versus taxane without anthracycline: An individual patient-level meta-analysis of 16,500 women with early-stage breast cancer in 13 randomised trials

Abstract

Abstract Background: Anthracycline and taxane-containing chemotherapy regimens reduce the rate of breast cancer recurrence by about one third compared to no chemotherapy. However, concerns about the increased risks of cardiac toxicity and leukaemia with anthracyclines have resulted in wider use of taxane chemotherapy schedules without anthracycline, in particular docetaxel-cyclophosphamide (DC). The benefits and risks of this approach have been assessed in several randomised trials but with conflicting results. Methods: We did an individual patient-level meta-analysis of data on 16,500 participants from 13 randomised controlled trials starting before 2010. Four compared 6 courses of DC with and without concurrent anthracycline (Anth), six compared sequential Anth then taxane (3 or 4 courses of Anth then 3 or 4 D, or vice versa) versus 6 courses of DC (resulting in a higher cumulative dose of taxane in the DC group), and 3 compared sequential Anth + D versus other taxane schedules. Primary outcomes were time to invasive breast cancer recurrence (distant, loco-regional, or new contralateral breast primary) and breast cancer mortality (by log-rank subtraction). Log-rank analyses were used to assess the first-event-rate ratio (RR) and confidence intervals. Pre-specified subgroup investigations included site of recurrence, age, ER/PR status, nodal status, tumor diameter, grade and HER2 status. Results: Overall, patients treated with an anthracycline and taxane combination averaged 18% lower rates of breast cancer recurrence (RR 0.82, 95%CI 0.75-0.90; p<0.0001) than those receiving a taxane schedule without anthracycline, equating to an absolute reduction of 3.1% (95%CI 1.4-4.8) in 10-year recurrence. The 10-year risk of death from breast cancer was reduced by 1.8% (RR 0.85, 95%CI 0.75-0.95; p=0.006) with no increase in deaths without recurrence. The proportional reduction in recurrence was greatest in trials of concurrent Anth + DC versus DC (RR 0.66, 95%CI 0.55-0.79; p<0.0001), where the only difference between arms was the addition of anthracycline (cumulative dose of doxorubicin≈300mg/m2). By contrast, in trials of sequential Anth + D versus the higher cumulative taxane dose DC regimen there was no significant benefit from Anth (cumulative dose of epirubicin≈300mg/m2): RR 0.93, 95%CI 0.80-1.07; p>0.1. In sub-group analyses of all trials, the benefit of anthracycline and taxane chemotherapy persisted throughout years 0-1, 2-4 and 5-9 with little data beyond year 10. There was a similar and highly significant proportional reduction in recurrence in ER-positive and in ER-negative disease and the RRs for recurrence did not differ significantly by any other pre-specified group including age, nodal status, tumor diameter, grade and HER2 status. There were no significant increases in deaths from cardiovascular disease or leukaemia, though longer follow-up is needed to fully assess risks. Conclusion: The addition of anthracycline to taxane chemotherapy, compared to taxane alone reduced the risk of breast cancer recurrence by 18% with larger proportional reductions in trials of DC ± concurrent Anth, in which the taxane dose was the same in both groups and the cumulative anthracycline dose was higher. Citation Format: Jeremy Braybrooke, Rosie Bradley, Richard Gray, Robert Hills, Zulian Liu, Hongchao Pan, Richard Peto, Joanne Blum, Xiaosong Chen, Bent Ejlertsen, Wolfgang Janni, Ulrike Nitz, Dennis Slamon, Masakazu Toi, Toru Watanabe, Sandra Swain, Jonas Bergh, on behalf of the Early Breast Cancer Trialists Collaborative Group. Taxane with anthracycline versus taxane without anthracycline: An individual patient-level meta-analysis of 16,500 women with early-stage breast cancer in 13 randomised trials [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr GS2-06.