Long-Term Follow-Up of Women in Trials of Adjuvant Therapy for Breast Cancer: Is It Still Important?

The article by Hackshaw et al that accompanies this editorial includes important data on the long-term benefits of 5 years versus 2 years of tamoxifen use from a large randomized trial of women more than 50 years of age with early breast cancer. The median follow-up in this study is 10.1 years; 25% of the women were observed for more than 14 years. The trial results were last reported in 1996. Follow-up strategies included matching trial subjects with death data from the British National Health Service Information Center. This extensive follow-up revealed 1,103 recurrences, including 876 recurrences that were followed by death, and 621 cardiovascular events that included 236 cardiovascular deaths. This study, like others, shows the long-term positive effects of 5 years of tamoxifen use compared with 2 years, with effects on recurrence that continue for up to 10 years and on breast cancer deaths that continue beyond 10 years. Extended follow-up data from studies such as this one are crucial for describing the long-term history of early breast cancer treated with tamoxifen, but also for assessing other potential long-term effects of such breast cancer treatment. Perhaps because of the long follow-up period, the use of British National Health Service cause of death data, and the 50 years of age required for enrollment onto this trial, the investigators were also able to report a trend toward overall reduction in cardiovascular events (hazard ratio [HR], 0.90; P .18) and a nearly statistically significant reduction in cardiovascular deaths (HR, 0.79; P .08) when tamoxifen was used for 5 years compared with 2 years of use. Furthermore, the women who were between 50 and 59 years of age when they were enrolled onto the study were found to have an even greater reduction in cardiovascular events (HR, 0.65; P .005; P .046 for interaction between age and treatment group) and in cardiovascular deaths (HR, 0.41; P .02; P .11 for interaction) for 5 years versus 2 years of tamoxifen use. The effects of long-term tamoxifen use on cardiovascular events and deaths were as great as or greater than the effects on breast cancer. Although neither the cardiovascular death event and survival analysis nor subanalyses by decades of age were preplanned, and should therefore be regarded with some caution, these data are still of considerable interest. Tamoxifen was initially postulated in the 1970s to potentially have a major beneficial effect on cardiovascular events and deaths, given that it is known to reduce blood lipid levels, including cholesterol and lipoprotein A and B, and to affect endothelial function. Furthermore, other studies, including the National Surgical Adjuvant Breast and Bowel Project (NSABP) trial B-14, a trial from the Scottish Breast Cancer Group, and the 2005 Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) meta-analysis, have suggested reduction in cardiovascular deaths with tamoxifen treatment. However, this was not seen in the NSABP Breast Cancer Prevention Trial, perhaps because of the large proportion of younger women who were enrolled. Interestingly, one of the first studies to show a relationship between reduced cardiac events and deaths and the use of tamoxifen also used the technique of matching patients from clinical trials with national data registry causes of death. Similar use of this technique in Sweden produced the first definitive reports of the association between endometrial cancer and tamoxifen; information on endometrial cancer incidence and death was obtained from the national registry in relation to a large study of women who were randomly assigned to receive tamoxifen versus a control years earlier. It may well be that this technique represents a more powerful methodology for ascertaining this type of secondary end point. Furthermore, depending on the availability of the registry data, its accuracy with regard to the end points being sought, and the costs and ethical issues involved in linking patients from randomized clinical trials with such registries, this methodology may also prove less costly. Thus, long-term follow-up data from studies such as the trial described by Hackshaw et al add to pre-existing data that suggests that long-term therapy with tamoxifen may be beneficial in reducing cardiac events and deaths as well as breast cancer recurrence and death. The results of the EBCTCG tamoxifen overview for 2010 (in preparation) will contribute additional data. However, only those individual studies in which long-term follow-up and accurate cause of death ascertainment have been performed can contribute to a metaanalysis of secondary medical effects of tamoxifen such as cardiovascular events and deaths. Given that breast cancer is increasingly treated effectively in the adjuvant setting, the secondary effects of drugs used in its long-term treatment will become increasingly important. We now know from a variety of randomized trials that cardiovascular events and perhaps cardiovascular deaths are higher in women who are randomly assigned to receive aromatase inhibitors (AIs) as compared with tamoxifen, although lipid levels and cardiovascular events seem little affected in trials of AIs versus placebos. It may be that women can be selected for therapy with tamoxifen or AIs on the basis of their underlying risk of cardiovascular disease and the use of other cardioprotective approaches. However, data to guide us can only be obtained with long-term follow-up from ongoing and future adjuvant trials. JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L S VOLUME 29 NUMBER 13 MAY 1 2011

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.