In:
Cancer Research, American Association for Cancer Research (AACR), Vol. 72, No. 24_Supplement ( 2012-12-15), p. P5-18-02-P5-18-02
Kurzfassung:
BACKGROUND: Disease-free survival (DFS) is often a primary endpoint of randomized trials of adjuvant therapies for breast cancer, but long-term follow-up of DFS and especially overall survival (OS) remain important. When the primary DFS results favor the experimental arm, patients (pts) assigned to the control group may select the option to crossover to receive the experimental treatment via protocol amendment. Such “selective crossover” disturbs the integrity of the randomized comparison for any efficacy endpoints that rely on further follow-up. Selective crossover, which is motivated by positive results having been observed in the current trial, is distinct from so-called “unplanned crossover,” which refers to non-adherence to protocol. In this abstract, we discuss the consequences of selective crossover for trials evaluating adjuvant trastuzumab, using the HERA (HERceptin Adjuvant) trial as an example, and present a variety of alternative analysis approaches. METHODS: HERA enrolled 5102 women with HER2-positive early breast cancer who had completed all surgery and (neo)adjuvant chemotherapy to compare 1 or 2 years of trastuzumab treatment vs observation. After a positive first interim analysis at 1y median follow-up (MFU) showed that 1 year of trastuzumab significantly improved DFS vs observation [MJ Piccart-Gebhart et al; NEJM 2005], event-free patients in the observation group were offered crossover to receive trastuzumab. 885 (52%) of the 1698 pts in the observation group selectively crossed over to trastuzumab. RESULTS: Previously reported intention-to-treat (ITT) analysis of HERA at 4y MFU showed a decreasing effectiveness of trastuzumab with respect to DFS compared with those at 2y MFU [L Gianni et al, Lancet Oncol 2011; I Smith et al, Lancet 2007]. In addition, the ITT analysis of OS at 4y MFU showed little effect of trastuzumab, while the analysis artificially censoring follow-up in the observation group at the time of selective crossover showed a substantial OS advantage for trastuzumab. The dependent censored analysis of OS is clearly biased in favor of trastuzumab because data for pts who remain event-free can be censored at the time of crossover, while data for the sicker pts in the observation group (those who relapsed) cannot be censored due to crossover. The issues related to the ITT and dependent censored analyses will be reviewed and discussed. Alternative analytic approaches designed to estimate the treatment effect that would have been observed had there been no selective crossover will be presented. The methods include the inverse probability of censoring weighted (IPCW) approach, and randomization-based estimators under the accelerated failure time model. HERA data to about 8y MFU (available fall 2012) will be used to illustrate approaches. CONCLUSION: Alternative methods addressing selective crossover are required to estimate the trastuzumab effect for updated analyses of DFS and OS for HERA, and for any other large randomized trial with positive interim results. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P5-18-02.
Materialart:
Online-Ressource
ISSN:
0008-5472
,
1538-7445
DOI:
10.1158/0008-5472.SABCS12-P5-18-02
Sprache:
Englisch
Verlag:
American Association for Cancer Research (AACR)
Publikationsdatum:
2012
ZDB Id:
2036785-5
ZDB Id:
1432-1
ZDB Id:
410466-3
