Education And Debate

Adjuvant tamoxifen: how long before we know how long?

BMJ 1998; 316 doi: http://dx.doi.org/10.1136/bmj.316.7143.1518 (Published 16 May 1998) Cite this as: BMJ 1998;316:1518
  1. Daniel Rea, senior lecturer,
  2. Christopher Poole, McMillan senior lecturer in medical oncology with palliative care,
  3. Richard Gray, professor of medical statistics
  1. Cancer Research Council Institute for Cancer Studies, University of Birmingham, Birmingham B15 2TA
  1. Correspondence to: Dr Rea
  • Accepted 16 January 1998

Despite recent advances, we still need more information from large clinical trials to define the optimum duration of tamoxifen treatment after surgery for breast cancer. Tamoxifen is the most widely prescribed and arguably the most important anticancer drug in clinical use. It has been established beyond doubt that adjuvant tamoxifen after primary treatment for early breast cancer improves survival. 1 2 Tamoxifen also reduces the incidence of second primary breast cancers, 1 2 preserves bone mineral density,3 and may lower the incidence of coronary heart disease.4 Although tamoxifen is generally well tolerated, it can produce menopausal symptoms and, on occasion, it has serious side effects, including an increased risk of endometrial carcinoma. 1 2 However, the benefits of a few years of adjuvant tamoxifen treatment clearly outweigh the risks, and its widespread adoption is considered the major contributor to the recent fall in mortality from breast cancer in the United Kingdom.5

Summary points

  • Tamoxifen is the most widely prescribed anticancer angent; used as adjuvant therapy it is a major contributor to the recent decline in breast cancer mortality

  • Controversy remains over how long tamoxifen treatment should be continued

  • Continuing treatment beyond two to three years reduces the risk of recurrence, which may result in improved survival

  • Treatment beyond five years has not been adequately studied; there are conflicting results from three small studies

  • Large scale recruitment into ongoing studies is needed to resolve the uncertainty over the balance of benefits and risks of long term adjuvant tamoxifen

Evidence on duration

For how long should adjuvant treatment with tamoxifen be continued? Controversy remains. Recent reports from randomised studies comparing different treatment periods favour longer courses, but we are still far from defining an optimum. We do know from the preliminary findings of three European studies, in which more than 10 000 patients participated, that continuing treatment beyond two to three years reduces the risk of cancer recurrence by about a quarter.68 Longer follow up in these studies is required to establish whether survival is also improved.

Treatment beyond five years has not been adequately studied. None of the three studies (total 1700 patients) comparing adjuvant tamoxifen for five years with longer treatment have found a noticeable difference in survival—and they report conflicting results on recurrence. The National Surgical Adjuvant Breast Project (NSABP) B-14 study reported that survival free of disease was worse in patients treated for 10 years than in those treated for five years.9 A similar study from the Eastern Cooperative Oncology Group reported an opposite trend with fewer recurrences in those treated for longer than five years.10 However, a Scottish trial reported no appreciable differences in recurrence.11

Explaining inconsistency

These statistically inconsistent and puzzling results may be explained partly by the early closure of the NSABP study by its data monitoring committee. Based on the study's findings, a clinical announcement was issued by the National Cancer Institute recommending five years' adjuvant treatment as standard in breast cancer that had not spread to the lymph nodes.12 This recommendation, made before the European Cooperative Oncology Group's data became available, has been criticised as premature since it was based on a comparatively small number of individual events.13 Although 1000 women participated in NSABP B-14, only those with a good prognosis (no spread of cancer to the lymph nodes) were included, so the total number of adverse events was comparatively low.

There is further concern that follow up in all of these trials is not yet sufficiently long for any potential late benefits from prolonged tamoxifen treatment to register. The worldwide overview of adjuvant tamoxifen studies showed that benefit persisted for several years after the tamoxifen treatment had been stopped.2 Thus, delayed effects may have an important bearing on the eventual interpretation of these trials. Although further analysis of these trials is expected early in the next millennium, it is unlikely that this will produce definitive conclusions.

Need for reliable assessment

With an estimated one million women taking tamoxifen as adjuvant treatment, even small benefits would translate into many thousands of lives saved. The risks and benefits must be defined as accurately and as soon as possible. Even if new hormonal approaches eventually supersede tamoxifen, a reliable assessment of the optimal duration of tamoxifen treatment is relevant. At present, all we can really say with confidence is that two years' adjuvant tamoxifen treatment is less effective at preventing recurrence than five years' treatment. Further trials of tamoxifen duration are therefore appropriate and necessary.

Major trials, definitive answers

Two major trials—aTTom (adjuvant tamoxifen treatment offers more?) in the United Kingdom and ATLAS (adjuvant tamoxifen longer against shorter) internationally—are currently aiming to recruit at least 20 000 women in order to answer definitively the question on duration of adjuvant tamoxifen. These parallel trials are designed pragmatically with randomisation at the point when “substantial uncertainty” arises as to whether to stop or continue tamoxifen treatment. At this point women who give informed consent are randomised to stop or continue adjuvant tamoxifen for at least five more years. The trials are therefore able to accommodate divergent and evolving clinical opinion. Indeed recruitment into these trials has shown that the point at which doctors become uncertain about continuing treatment has shifted appreciably—from around two years towards five years—undoubtedly in response to publication of the recent European studies (figure).

Figure1

Duration of treatment with tamoxifen before patients were randomised into the aTTom trial

Trial entry

All doctors treating breast cancer patients should be aware of the uncertainty over how long adjuvant tamoxifen should be prescribed for. This now includes general practitioners since patients are often still taking tamoxifen when they are discharged from hospital follow up. When in doubt about stopping treatment, entry into a tamoxifen duration trial through a participating local breast unit should be considered. Only through widespread support for these studies can we resolve the complex balance of benefits and risks of long term adjuvant tamoxifen treatment.

Acknowledgments

Information about aTTom and ATLAS is obtainable from the aTTom Study Office, CRC Trials Unit, The Medical School, Birmingham, B15 2TA or ATLAS Study Office, CTSU, Radcliffe Infirmary, Oxford, OX2 6HE.

Funding: None.

Conflict of interest: None.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.