Worst Pills, Best Pills

An expert, independent second opinion on more than 1,800 prescription drugs, over-the-counter medications, and supplements

Tamoxifen (Nolvadex) for Breast Cancer Prevention?

Worst Pills, Best Pills Newsletter article February, 2003

A new study published in the medical journal The Lancet (September 14, 2002) titled “First results from the International Breast Cancer Intervention Study (IBIS-1)” documents breast cancer incidence in women who took either tamoxifen or placebo (a “sugar pill”) for four years. IBIS-1 is the most recent of four large randomized, placebo-controlled trials aimed at seeing if giving tamoxifen to “high risk” but healthy women could lower breast cancer risk. The four studies completed, thus far,...

A new study published in the medical journal The Lancet (September 14, 2002) titled “First results from the International Breast Cancer Intervention Study (IBIS-1)” documents breast cancer incidence in women who took either tamoxifen or placebo (a “sugar pill”) for four years. IBIS-1 is the most recent of four large randomized, placebo-controlled trials aimed at seeing if giving tamoxifen to “high risk” but healthy women could lower breast cancer risk. The four studies completed, thus far, are the National Surgical Adjuvant Breast and Bowel Project P-1 Study (NSABP-P1) (U.S.), the Royal Marsden Hospital Study (London, UK), the Italian National Study (Italy) and most recently, IBIS, conducted primarily in the UK, Australia and New Zealand. These studies all began in 1992, with the exception of the Royal Marsden trial, which began as a pilot study in 1986.

Effectiveness of Tamoxifen Trials (Risk Reduction)

The reduction of breast cancer incidence shown in Table 1 is the “absolute” reduction, i.e., the difference in breast cancer incidence between women on placebo and women on tamoxifen, treated for the times indicated. The low absolute reduction in breast cancer risk (around 1 percent or less) means that between 77 (NSABP trial) and 500 healthy women (Royal Marsden trial) would have to take tamoxifen from four to seven years to prevent one case of breast cancer.

Medical journals, as well as most of the media, have presented a misleadingly rosy picture of tamoxifen’s effectiveness by relating not the reduction in absolute risk but only the reduction in relative risk. In the NSABP-P1 trial, for example, the authors stated that the reduction in risk (for getting breast cancer) by taking tamoxifen was 50 percent but never mentioned that the absolute risk was reduced only 1.3 percent (from 2.6 percent to 1.3 percent, Table 1). Likewise, the IBIS trial authors stated that tamoxifen produced a 32 percent risk reduction without mentioning anywhere that the absolute risk reduction was less than 1 percent over 4.2 years. The two European trials stated that they found no reduction in risk at all.

Safety (Adverse Events Due to Tamoxifen)

The other part of the picture that needs to be addressed is the increase in risk for having one of the adverse events that can occur when taking tamoxifen. The risks about which we know the most are those that occur after only a few years exposure, a relatively short time if one is concerned about causing cancer. Given that tamoxifen is known to be a human carcinogen (i.e., it is known to cause cancer in people) and that cancer can take 10 to 20 years to develop, it is clear that we have only a partial picture of tamoxifen safety from trials that last four to seven years.

New adverse events become known over time. Adenocarcinoma of the uterus has been known for many years as an adverse effect of tamoxifen use, but only this year the manufacturer issued a new warning: patients on tamoxifen are also at an increased risk for uterine sarcoma, a different type of uterine cancer with a poorer prognosis and shorter survival time than the adenocarcinoma. Other adverse events that were more common in women taking tamoxifen included deep vein thrombosis (blood clots in the veins), pulmonary emboli (blood clots in the lungs), stroke and cataracts. Bothersome, but less serious, adverse events due to tamoxifen were hot flashes and vaginal discharges.

Public Citizen published an analysis of the largest prevention trial, NSABP P-1 (Worst Pills Best Pills June 1999), showing that over four years, the decrease in new breast cancer cases (benefit) was almost exactly offset by an increase in serious adverse events (risks). In that case, we felt that at the minimum, women should have a Medication Guide with these risks spelled out clearly.

International Breast Cancer Intervention Study (IBIS)

The IBIS trial tested a total of 7,152 women (3576/group) who were randomly assigned to either 20 mg/day of tamoxifen or placebo for four years. To enter this trial, women had to have an elevated risk factor for getting breast cancer. These risks included having a first-degree relative that developed breast cancer at or before the age of 50 and/or having at least two first- or second-degree relatives with breast cancer.

Death rates: The IBIS results have provided a further cautionary note to those coming from previous trials, for although breast cancer incidence decreased by 0.9 percent (the absolute rate), the death rate (deaths from any cause) more than doubled, from 0.31 percent in the placebo group to 0.70 percent in the tamoxifen-treated group. Interestingly, there were the same number of deaths from breast cancer (2) in each group, but the total number of cancer deaths doubled in the tamoxifen group compared to the placebo group (12 vs. 6).

Tamoxifen produced a reduction of 0.9 percent in cases of breast cancer (good) but increased the number of life-threatening adverse events (uterine cancer, blood clots, stroke) and deaths (bad). The result is that the percentage of women experiencing serious risks is greater than the percentage of women who benefit. In addition to the risks listed above, there was a significant increase in four gynecological procedures and seven gynecological symptoms.

A commentary accompanying the IBIS article, in the same issue of The Lancet, acknowledges that tamoxifen can reduce the number of breast cancer cases, yet “none of the trials have the power as yet to determine whether tamoxifen reduces breast cancer mortality.” (And isn’t this the whole reason to take tamoxifen anyway?) These authors continued that, in addition, “The IBIS trial adds a new concern about tamoxifen: an increase in all-cause mortality.”

The commentary states the obvious: “A cardinal requirement of chemopreventive agents is that they must be safe” and continues, “Tamoxifen clearly does not have a safety profile that would allow it...to have a large impact on the overall incidence of breast cancer.” The authors conclude with conditions that should be met in order for women to benefit from taking a drug for reduction of breast cancer risk: better risk assessment tools (to more clearly target those at risk), a good safety profile (so women don’t suffer serious adverse events) and a demonstrated reduction in breast cancer mortality. (We would hope that there should be a demonstrated reduction in all-cause mortality, as well — certainly not an increase.)

Comparison with Incidence of Lung Cancer

Since 1987, lung cancer has been the leading cause of cancer death among women, yet it is the fear of breast cancer that has permeated U.S. society. In the most recent year (2000) for which there is data, 67,600 women died of lung cancer (mostly from smoking) vs. 40,800 women who died of breast cancer. The Centers for Disease Control concluded that, “Smoking is now the leading known cause of preventable death and disease among women.” Yet, this is not where medical efforts are aimed.

Conclusion

In order that they suffer the least harm, women need to be fully informed about the results coming from these tamoxifen “prevention” trials. They need to realize that the current poor risk assessment tool coupled with the increase in possible serious illnesses or death is a poor foundation on which to base their hopes by taking tamoxifen for many years. A competent clinical breast exam coupled with a mammogram (for those over 50) offers a safer approach. Not smoking is clearly a number one health priority.

 Table 1. Total Breast Cancer Incidence in the Four Prevention Trials

TRIAL NAME 

Duration of Trial (Yrs) 

Breast Cancer Incidence in trial (%)

Reduction in Incidence (%) 

Placebo

Tamoxifen

 NSABP-P1

 4.0

 3.6

1.9 

 1.7 (statistically
significant)

 IBIS

 4.2

 2.8

1.9 

 0.9 (statistically
significant)

 Royal Marsden

 5.8

 2.9

2.7 

 0.2 (not significant)

 Italian National

 6.8

 1.7

1.3 

 0.4 (not significant)

 Table 2. Risks and Benefits of Tamoxifen Therapy in the IBIS Study 1

 

Placebo

Tamoxifen

Difference between tamoxifen and placebo

Benefit

Breast cancer incidence

2.8%

1.9%

-0.90%

 

 

 

 

Risks

Uterine cancer

0.14%

0.31%

+0.17%

Blood clots in lungs

0.28%

0.36%

+0.08%

Blood clot in veins

0.14%

0.67%

+0.53%

Stroke

0.31%

0.36%

+0.05%

Death from any cause

0.31%

0.70%

+0.39%

Sum of Serious Events and Deaths

+1.22