Summary Background Clinical trials have shown that trastuzumab, a recombinant monoclonal antibody against HER2 receptor, significantly improves overall survival and disease-free survival in women with HER2-positive early breast cancer, but long-term follow-up data are needed. We report the results of comparing observation with two durations of trastuzumab treatment at a median follow-up of 11 years, for patients enrolled in the HERA (HERceptin Adjuvant) trial. Methods HERA (BIG 1–01) is an international, multicentre, open-label, phase 3 randomised trial of 5102 women with HER2-positive early breast cancer, who were enrolled from hospitals in 39 countries between Dec 7, 2001, and June 20, 2005. After completion of all primary therapy (including, surgery, chemotherapy, and radiotherapy as indicated), patients were randomly assigned (1:1:1) to receive trastuzumab for 1 year (once at 8 mg/kg of bodyweight intravenously, then 6 mg/kg once every 3 weeks) or for 2 years (with the same dose schedule), or to the observation group. Primary endpoint is disease-free survival, and analyses are in the intention-to-treat population. Hazard ratios (HRs) were estimated from Cox models, and survival curves were estimated by the Kaplan-Meier method. Comparison of 2 years versus 1 year of trastuzumab is based on 366-day landmark analyses. This study is registered with ClinicalTrials.gov (NCT00045032). Findings Of the 5102 women randomly assigned in the HERA trial, three patients had no evidence of having provided written informed consent to participate. We followed up the intention-to-treat population of 5099 patients (1697 in observation, 1702 in 1-year trastuzumab, and 1700 in 2-years trastuzumab groups). After a median follow-up of 11 years (IQR 10.09–11.53), random assignment to 1 year of trastuzumab significantly reduced the risk of a disease-free survival event (HR 0.76, 95% CI 0.68–0.86) and death (0.74, 0.64–0.86) compared with observation. 2 years of adjuvant trastuzumab did not improve disease free-survival outcomes compared with 1 year of this drug (HR 1.02, 95% CI 0.89–1.17). Estimates of 10-year disease-free survival were 63% for observation, 69% for 1 year of trastuzumab, and 69% for 2 years of trastuzumab. 884 (52%) patients assigned to the observation group selectively crossed over to receive trastuzumab. Cardiac toxicity remained low in all groups and occurred mostly during the treatment phase. The incidence of secondary cardiac endpoints was 122 (7.3%) in the 2-years trastuzumab group, 74 (4.4%) in the 1-year trastuzumab group, and 15 (0.9%) in the observation group. Interpretation 1 year of adjuvant trastuzumab after chemotherapy for patients with HER2-positive early breast cancer significantly improves long-term disease-free survival, compared with observation. 2 years of trastuzumab had no additional benefit. Funding F Hoffmann-La Roche (Roche).

In computational science literature including, e.g., bioinformatics, computational statistics or machine learning, most published articles are devoted to the development of “new methods”, while comparison studies are generally appreciated by readers but surprisingly given poor consideration by many journals. This paper stresses the importance of neutral comparison studies for the objective evaluation of existing methods and the establishment of standards by drawing parallels with clinical research. The goal of the paper is twofold. Firstly, we present a survey of recent computational papers on supervised classification published in seven high-ranking computational science journals. The aim is to provide an up-to-date picture of current scientific practice with respect to the comparison of methods in both articles presenting new methods and articles focusing on the comparison study itself. Secondly, based on the results of our survey we critically discuss the necessity, impact and limitations of neutral comparison studies in computational sciences. We define three reasonable criteria a comparison study has to fulfill in order to be considered as neutral, and explicate general considerations on the individual components of a “tidy neutral comparison study”. R codes for completely replicating our statistical analyses and figures are available from the companion website http://www.ibe.med.uni-muenchen.de/organisation/mitarbeiter/020_professuren/boulesteix/plea2013.

Purpose Women receiving trastuzumab with chemotherapy are at risk for trastuzumab-related cardiac dysfunction (TRCD). We explored the prognostic value of cardiac markers (troponins I and T, N-terminal prohormone of brain natriuretic peptide [NT-proBNP]) to predict baseline susceptibility to develop TRCD. We examined whether development of cardiac end points or significant left ventricular ejection fraction (LVEF) drop was associated with markers' increases. Patients and Methods Cardiac marker assessments were coupled with LVEF measurements at different time points for 533 patients from the Herceptin Adjuvant (HERA) study who agreed to participate in this study. Patients with missing marker assessments were excluded, resulting in 452 evaluable patients. A primary cardiac end point was defined as symptomatic congestive heart failure of New York Heart Association class III or IV, confirmed by a cardiologist, and a significant LVEF drop, or death of definite or probable cardiac causes. A secondary cardiac end point was defined as a confirmed significant asymptomatic or mildly symptomatic LVEF drop. Results Elevated baseline troponin I (> 40 ng/L) and T (> 14 ng/L), occurring in 56 of 412 (13.6%) and 101 of 407 (24.8%) patients, respectively, were associated with an increased significant LVEF drop risk (univariate analysis: hazard ratio, 4.52; P < .001 and hazard ratio, 3.57; P < .001, respectively). Few patients had their first elevated troponin value recorded during the study (six patients for troponin I and 25 patients for troponin T). Two patients developed a primary and 31 patients a secondary cardiac end point (recovery rate of 74%, 23 of 31). For NT-proBNP, higher increases from baseline were seen in patients with significant LVEF drop. Conclusion Elevated troponin I or T before trastuzumab is associated with increased risk for TRCD. A similar conclusion for NT-proBNP could not be drawn because of the lack of a well-established elevation threshold; however, higher increases from baseline were seen in patients with TRCD compared with patients without.