A somewhat different study population participated in the two other tamoxifen studies, a factor that, say some authorities, may help account for the very different results seen.4,5 In a Scottish study, 2,471 women with strong family histories of breast cancer were randomized to receive either tamoxifen, 20 mg per day, or placebo, and the trial was planned to last 8 years. By the time of the interim analysis, at 70 months, the investigators believed they would be more than 90% likely to be capable of detecting a 50% reduction in cancer incidence, as was seen in the Breast Cancer Prevention Trial, if indeed their patient population was responding in a similar fashion.

This was not the case, however. “From the 47% reduction in the frequency of contralateral new breast cancers in women on adjuvant tamoxifen after treatment of primary breast cancer, we anticipated that giving tamoxifen to healthy high-risk women would produce at least an equivalent result,” the investigators wrote. “We are therefore surprised to see no overall reduction in the occurrence of breast cancer in participants randomised in our trial to tamoxifen.” A total of 34 and 36 cases of new breast cancer was seen among the tamoxifen-treated women and the placebo group, respectively.

In the Italian study, a group of 5,408 women who had undergone hysterectomy were randomized to receive tamoxifen, 20 mg/d or a placebo, and the trial was intended to last 5 years. Similar statistically insignificant results were seen in a preliminary analysis at 46 months, with 19 and 22 cases in the tamoxifen and placebo groups, respectively. No deaths were seen in either group, and these authors particularly emphasized that much longer follow-up will be needed, not only to achieve sufficient power for a more valid analysis, but also to evaluate the effects of tamoxifen on mortality rates.

Both groups of investigators and other authorities have advanced a number of possible explanations for the discrepancies in their results:

Age at entry. In the British study, the mean age was 47 years and two thirds of participants were premenopausal. The mean age in the Italian study was 51 years.

Family history. In the Breast Cancer Prevention Trial, 76% of participants had a first-degree relative with breast cancer, compared with 96% in the British study and only 15% in the Italian study. The effect of tamoxifen on breast cancers that may have a genetic component is particularly difficult to determine, and many experts believe this factor may be the most important aspect of the lack of efficacy seen in the UK study.

Concurrent HRT. Unlike the Breast Cancer Prevention Trial, in the two European studies, concurrent hormone replacement therapy (HRT) was permitted. In the UK study, for example, 26% of participants were on HRT.

Compliance. The Italian study was characterized by poor compliance, with 26% of participants dropping out of the trial.

Length of follow-up. Median follow-up in the British study was 70 months, considerably longer than the follow-up period in the other studies, with a median of 55 months in the Breast Cancer Prevention Trial and 46 months in the Italian trial. This finding prompted the author of an editorial that accompanied the publication of the two European studies to suggest another possible explanation: “The substantial preventive effect seen in the large Breast Cancer Prevention Trial with its relative short follow-up is actually a treatment effect on early occult breast cancers, one that may disappear with longer follow-up.“6

This question as to whether tamoxifen is simply suppressing tumors already present, rather than preventing cellular transformations, remains one of the most difficult to answer. Most breast cancers are slow growing and only become clinically apparent 7 to 10 years after the initial cellular changes, and if the data showing preventive effects with tamoxifen actually are reflecting temporary effects on subclinical tumors, authorities ask, what will happen when that treatment is stopped? Will the tumors have become tamoxifen resistant? Will other untoward events begin to occur, such as more invasive uterine or ovarian cancers? Until longer-term study results are available, they caution, these safety concerns must be included in clinical decision-making with regard to tamoxifen for breast cancer prevention.

The US study
The Breast Cancer Prevention Trial was a double-blind, randomized, placebo-controlled trial that included 13,388 women considered to be at risk for breast cancer because of age (60 years or older), a history of lobular carcinoma in situ, or a 5-year Gail-model-predicted risk of 1.66% or greater. The average 5-year risk of the Breast Cancer Prevention Trial participants, according to the Gail model, was 3.2%. Cumulative incidence rates through 69 months of follow-up were 43.4 versus 22.0 for the tamoxifen and placebo groups, respectively. The overall risk reduction of 49% was accompanied by decreased risk in all age groups, with a reduction of 44% in women under 50 years and 51% in women older than 50 years.

As expected, a significant increase in rates of invasive endometrial cancer was seen among tamoxifen-treated women in the Breast Cancer Prevention Trial, almost entirely among women older than 50. Endometrial cancer developed in 36 women in the tamoxifen group in the BCPT, compared with 15 in the placebo group. Also of concern was an increased incidence of Pulmonary embolism, which occurred three times more frequently in the tamoxifen group than in the placebo group; and stroke, which was seen twice as often in the active treatment group.

By Trudy L. Bush, PhD, Steven R. Cummings, MD, and Clifford A. Hudis, MD

References

1. Struewing JP, Hartge P, Wacholder S, et al. The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. N Engl J Med. 1997;336:1401-1408.

2. Peto J, Collins N, Barfoot R, et al. Prevalence of BRCA1 and BRCA2 mutations in patients with early-onset breast cancer. J Natl Cancer Inst. 1999;91:943-949.

3. Hartmann LC, Schaid DJ, Woods JE, et al. Efficacy of bilateral prophylactic mastectomy in women with a family history of breast cancer. N Engl J Med. 1999;340:77-84.

4. Wisen A, Weber BL. Prophylactic mastectomy - the price of fear. N Engl J Med. 1999;320:137-138.