Early Breast Cancer Trialists' Collaborative Group (EBCTCG) overview of adjuvant tamoxifen

Tamoxifen had been in clinical use for early breast cancer since the 1970s, and several randomized trials had confirmed that it reduced recurrence. What those individual trials could not do was quantify the long-term survival benefit with precision, particularly beyond five years, or reliably identify which subgroups gained and which did not. The Early Breast Cancer Trialists' Collaborative Group, founded at Oxford with Richard Peto as its statistical architect, existed specifically to answer questions of that kind by pooling raw patient-level data from every eligible trial rather than working from published summaries.

The 2005 Lancet overview assembled individual patient data from 194 randomized trials of adjuvant chemotherapy or hormonal therapy that had begun by 1995; its comparison of about five years of tamoxifen with none included about 15,000 women, followed for up to 15 years. In patients with estrogen receptor-positive tumors, five years of tamoxifen reduced the annual breast cancer death rate by 31% (standard error 3%). That proportionate reduction was largely the same whatever the patient's age, progesterone receptor status or use of chemotherapy, though the absolute gain varied with baseline risk.

The survival benefit kept accumulating for at least 15 years after diagnosis, well beyond the five-year treatment period. Earlier overviews had documented the five-year and ten-year gains; the 2005 data showed that annual breast cancer death rates, and tamoxifen's proportional reduction in them, were similar in years 0 to 4 and 5 to 14, so the cumulative reduction in mortality was more than twice as big at 15 years as at 5. In ER-negative tumors, tamoxifen offered no meaningful benefit, a result that reinforced receptor testing as a prerequisite before prescribing adjuvant endocrine therapy.

The overview also compared durations: five years of tamoxifen was significantly more effective than one to two years. Combining its estimates, the collaborators calculated that six months of anthracycline-based chemotherapy followed by five years of tamoxifen would roughly halve 15-year breast cancer mortality for a middle-aged woman with ER-positive disease. None of the trials used aromatase inhibitors, which Mitch Dowsett of the Royal Marsden Hospital and colleagues compared with tamoxifen in a 2010 meta-analysis.

The group used the same method, collecting and harmonizing raw data from research groups worldwide, for later overviews of radiotherapy (2005 and 2011), polychemotherapy (2012) and aromatase inhibitors (2015). Its 2011 update of the tamoxifen analysis, with 20 trials and 21,457 women, found breast cancer mortality reduced by about a third throughout the first 15 years in ER-positive disease and little or no effect in ER-negative disease.