Abstract P1-13-03: Mature analysis of UK Taxotere as Adjuvant Chemotherapy (TACT) trial (CRUK 01/001); effects of treatment and characterisation of patterns of breast cancer relapse

Introduction: TACT, an investigator-led study in 4162 women with node positive (N+ve) or high risk node negative (N-ve) early breast cancer (EBC), is the largest taxane trial unconfounded by treatment (trt) duration. At principal analysis, with 5 years follow-up (fup), no evidence of improved disease-free survival (DFS) was observed by switching to 4 cycles of docetaxel (D) after 4 cycles of FEC (Ellis, Lancet 2009). Results were provocative in suggesting differential effects according to ER & HER2 status. Longer fup provides opportunity to detect emergence of late trt effects overall & within phenotypic subgroups & explore patterns of recurrence, by tumor characteristics.

Patients & methods: TACT recruited women with histologically confirmed completely resected invasive EBC from 104 centers (UK (103), Belgium (1)) between 02/2001 & 07/2003. Centers chose FEC (600/60/600 mg/m2 q3wk × 8) or E-CMF (E 100mg/m2 q3wk × 4 → CMF 100mg/m2 PO d1-14 or 600mg/m2 IV d1&8/40/600 mg/m2 q4wk × 4) as their control, reflecting standard UK practice. Patients (pts) were randomized to FEC-D (FEC q3wk × 4 → D 100 mg/m2 q3wk × 4) or control. 2523 pts were from FEC centers (FEC = 1265: FEC-D = 1258) & 1639 from E-CMF centers (E-CMF = 824; FEC-D = 815). Endocrine therapy was given for 5 years. Few pts received HER2 directed therapy; 589 pts had unknown HER2 status. Median fup is now 97.5 months; this analysis updates DFS & overall survival in the ITT population. It also explores patterns of relapse by phenotypic & clinical characteristics. Analyses of trt effect are stratified by ER status due to issues of non-proportionality of hazard associated with length of fup.

Results: DFS events have been reported for 1329 pts (FEC-D=640, Control=689) giving an unadjusted hazard ratio (HR) & 95%CI (stratified by control regimen & ER status) of 0.93 (0.83, 1.03) overall; p = 0.16 in favor of FEC-D & for ER+ve/HER2-ve of 0.99 (0.84, 1.17), for ER+ve/HER2+ve) 0.97 (0.73, 1.30), for ER-ve/HER2+ve 0.74 (0.53, 1.03), & ER-ve/HER2-ve 0.93 (0.73, 1.17). 1017 patients have died (FEC-D=500, Control=517); unadjusted HR=0.98 (95%CI: 0.86, 1.10); p = 0.69 with intercurrent deaths (prior to distant relapse) reported for 80 pts (FEC-D=40, Control=40).

Annual event rates show different pattern of disease relapse by phenotypic subgroup

Cardiac effects when using trial-derived monitoring protocols for adjuvant trastuzumab: Results from a retrospective multicenter UK audit

582

Background: Guidelines for cardiac scanning in adjuvant Trastuzumab (T) trials were developed due to clinical concerns regarding cardiotoxicity, and these same guidelines are now used in clinical practice. This was a retrospective audit to assess adherence to these guidelines and characterise the nature and timescale of problems experienced with adjuvant T in clinical practice. Methods: A retrospective review of MUGA/ECHO results was conducted in patients who received adjuvant T for breast cancer. Data was compiled from five UK cancer centres: Glasgow, Manchester, Cardiff, The Royal Free hospital and Newcastle. Results: A total of 424 patients received at least one dose of adjuvant T between September 2005 and January 2008, using a HERA trial schedule. There were 262 with detailed information regarding treatment delays/withdrawals for cardiac reasons. 12% (32/262 pts) were withdrawn from adjuvant trastuzumab treatment for cardiac reasons, with another 11% (29/262 pts) experiencing at least 1 delay in treatment. Analysis of cardiac scan results for the remaining 162 pts suggested a total cardiac delay/withdrawal percentage of 14.4% (61/424 pts). There was no age effect with 15% (50/338) of <65 year olds experiencing cardiac delays/withdrawals compared to 18% (9/50) of ≥65 year olds. Preherceptin cardiac scan results did appear to be predictive of subsequent problems, with 29% (38/132 pts) who had an ejection fraction (EF) of 40–59% experiencing cardiac delays/withdrawals compared to 10% (19/183 pts) with an EF of 60–69% and 4% (4/98 pts) with an EF ≥ 70%. Pretreatment with Doxorubicin was associated with an approximate doubling of cardiac problems (6/20 pts; 30%) compared to pretreatment with Epirubicin containing chemotherapy regimens. No temporal relationship existed between timing of cardiac delays/withdrawals and stage of T treatment. Conclusions: Significant numbers of patients are experiencing delays/withdrawals during adjuvant T treatment. In routine practice the withdrawal rates are more than twice those reported in the HERA trial population. There is a clear role for a re-assessment of the trial-derived cardiac management guidelines, which should be prospectively audited.

[Table: see text]