Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Tamoxifen for early breast cancer

Background
Review
Results
Duration and oestrogen receptor status
Nodal status
Duration of effect
Dose
Age
Benefits versus harm
Comments

Bandolier was asked to nominate and précis some of the great systematic reviews and meta-analyses. The papers may be famous but few people actually know the results. That's not easy, because great systematic reviews and meta-analyses will often be large, and detailed, and resistant to précis. But, for our sins, we've decided to have a try at one of the great meta-analyses, that investigating the use of tamoxifen for early breast cancer [1].

Background


Early breast cancer is when all detectable cancer is restricted to the breast or local lymph nodes (node positive cancer). Undetected disease may be present, of course, and these micrometastases might develop into new cancers in the same breast, the contralateral breast, or at distant sites. Tamoxifen is known to have some effect on reducing the rate of recurrence. The details of for how long tamoxifen needs to be taken to have the best effect, in what types of cancer it has the best effects, and the relationship between benefit through reduced recurrence of breast cancer and adverse consequences like higher rates of endometrial cancer require meta-analysis.

Review


This review sought information on all trials in early breast cancer begun before 1990 comparing adjuvant tamoxifen versus no such treatment. Trials were divided into three categories of average duration of tamoxifen use, of one, two and more than two years. Information on each individual woman, on age, menopausal status, tumour spread to nodes, and results of any oestrogen receptor measurements on the primary tumour were obtained, as well as treatment details and information on outcomes of recurrence and mortality. It is unlikely that randomised trials were conducted but not included, and the analysis was by individual patient data, rather than pooling mean data from individual trials.

Results


There were 63 trials identified, and information from 55 of them, with nearly 37,000 women was available for analysis. Fourteen trials (9,000 women) examined one year duration of tamoxifen, 32 trials (19,000 women) two years, and nine trials (8,000 women) more than two years duration, with a median of five years.

Duration and oestrogen receptor status


In women with oestrogen receptor-poor tumours, tamoxifen was not effective for any duration for recurrence (Figure 1) or mortality. For women who were oestrogen receptor positive longer duration of tamoxifen produced bigger benefits in reduced recurrence (Figure 2) and mortality. Results for recurrence as first event and mortality were broadly similar in magnitude and effect of tamoxifen.

Figure 1: Recurrence as first event in women with oestrogen receptor negative tumour


 

Figure 2: Recurrence as first event in women with oestrogen receptor positive tumour

 



Nodal status


Tamoxifen was equally effective for women who were node negative as for those who were node positive.

Duration of effect


Benefits in recurrence were apparent early on, within the first year. After five years there was no further divergence. Benefits in mortality were not apparent in the early years, but appeared to continue to increase at least up to 10 years of follow up.

Dose


The dose of tamoxifen used is usually 20 mg, or 30-40 mg. There was no difference in results with different doses.

Age


Tamoxifen was effective in all women over 50 years. Women under 50 years receiving five years of treatment benefited (Figure 3), but there was less benefit for women under 50 with two years, and especially one year, of tamoxifen.

Figure 3: Recurrence as first event in women aged less than 50 years



Benefits versus harm


Table 1 lists the benefits in terms of recurrence and death over 10 years for 1000 women who were oestrogen receptor and node positive. It also shows the effects in terms of contralateral breast cancer, colorectal cancer, endometrial cancer and endometrial cancer death, plus death from any cause other than breast or endometrial cancer. Clearly the benefits greatly outweigh the harm.

Table 1: Benefits of five years of tamoxifen treatment in women with oestrogen positive tumour and who were node positive, compared with other outcomes

  Number of women out of 1000 affected over 10 years
Event Tamoxifen Control Difference
Oestrogen receptor and node positive women treated with five years of tamoxifen
Recurrence as first event 403 555 152
Death 386 495 109
       
Contralateral breast cancer 23 32 9
Colorectal cancer 7 7 0
Endometrial cancer 6 2 -4
Endometrial cancer death 1.7 0.4 -1
Death from any cause other than breast or endometrial cancer 59 59 0

Comments


No précis can convey the wealth of detail in this meta-analysis, detail that can be used to deliver the best care to women with early breast cancer. There is little doubt, though, that these analyses point the way to even better treatment for women with breast cancer.

It demonstrated that measuring oestrogen receptor status of the primary tumour is critical because it determines optimal treatment. It demonstrated that longer duration treatment with tamoxifen provides better results, especially in women under 50. It demonstrated the benefits for women with node negative disease as well as those with node positive disease. And it demonstrated the overwhelming balance in favour of benefit over harm. It's just a shame that one woman in five is unable to use tamoxifen.

What is better is that a few years from now we will have an updated meta-analysis, likely to include information on a further 6,000 women, many with longer duration of tamoxifen use.

Reference:

  1. Early Breast Cancer Trialists' Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet 1998 351: 1451-1467.
previous or next story in this issue