Janine O'Leary Cobb

Last Spring, the U.S. National Cancer Institute (NCI) announced preliminary results from STAR, a large clinical trial comparing the effect of raloxifene and tamoxifen on the risk of developing breast cancer in postmenopausal women considered to be at higher-than-average risk. The trial was funded by the NCI and Eli Lilly, manufacturer of raloxifene, currently marketed under the brand name Evista and approved for prevention and treatment of osteoporosis.

According to the NCI, both drugs cut the number of new cases of breast cancer in half, but women taking raloxifene experienced fewer serious side effects such as blood clots and cancer of the uterus. The NCI stressed this benefit in its announcement, stating "For many women, raloxifene's benefits will outweigh its risks in a way that tamoxifen's benefits do not." Independent experts (i.e., the National Women's Health Network and Breast Cancer Action San Francisco) are more cautious about the results in the belief that the breast cancer benefit of raloxifene may only be useful to a very small number of women.

A total of 19,747 women took either tamoxifen (brand name Nolvadex) or raloxifene (Evista) for up to five years, without knowing which medication they were being given. All the women were postmenopausal, over half had already had hysterectomies, and all were considered at higher than average risk of developing breast cancer. During the trial, 163 women on tamoxifen and 167 women on raloxifene developed breast cancer. According to NCI, about 320 cases of breast cancer were expected in both groups if no drug treatment were taken. Both drugs led to a drop in breast cancer cases of roughly 50%, which is similar to the benefit found for tamoxifen when it was tested against placebo in a previous trial.

However, in earlier trials, tamoxifen caused a significant number of problems and researchers hoped that raloxifene would be safer. According to the NCI, raloxifene is somewhat better than tamoxifen, although not strikingly so. Twenty-three women on raloxifene developed uterine cancer, compared with 36 women on tamoxifen; and 100 women taking raloxifene experienced serious blood clots, compared with 141 women on tamoxifen.

These results show that raloxifene is significantly less likely than tamoxifen to cause serious side effects — but experts question whether this is likely to make a significant difference. Of the nearly 10,000 women who took raloxifene for up to five years, only about 30 benefitted once the serious risks are taken into account. While some women may appreciate this marginal benefit, we wonder whether it should be viewed as a major advance in preventing breast cancer.

NCI's announcement doesn't include information that women and health care providers need to evaluate the overall risks and benefits of these drugs. Moreover, the results have not yet been reviewed by experts who weren't involved in the study (a requirement before the research can be published in a major medical journal), nor has the Food and Drug Administration (FDA) examined the records of the study to determine whether the agency will approve raloxifene for breast cancer risk reduction.

We don't know whether or not raloxifene — or tamoxifen for that matter — truly prevents breast cancer, or just delays it. NCI's earlier trial of tamoxifen compared to a placebo followed women for only two years after they stopped taking their pills, not nearly long enough to know whether tamoxifen's benefit is lasting. This trial can't answer that question, either.

Many of the details of other benefits and risks of tamoxifen and raloxifene were not released by NCI. For example, we don't know the specifics about how many women were able to avoid a bone fracture, how much both drugs increased the risk of stroke, and how many women died while they were part of the study.

The NWHN recommends that wo-men currently taking tamoxifen to reduce their risk of breast cancer should wait until the results of the STAR trial have been reviewed by the FDA before considering switching to raloxifene. FDA reviewers have access to all the information generated by the researchers and can provide an objective assessment of the results. Women who currently take bisphosphonates (drugs such as Fosamax, Actonel, Boniva) for osteoporosis also shouldn't rush to switch. While there are still some questions about the long-term effects of bisphosphonates on bone structure, it seems clear that they do not increase the risk of stroke, blood clots or cancer.

Healthy women who are not at higher than average risk of either breast cancer or osteoporosis should be especially skeptical of any efforts to persuade them to start taking raloxifene. NWHN believes it's very likely that the ultimate results of the STAR trial will show that raloxifene isn't quite as bad as tamoxifen, but it's still not safe enough for most women to use. And we've learned that it's important to wait until enough information is available for women to be able to make truly informed choices. In 1998, NCI announced the early results of the tamoxifen trial as a breakthrough in cancer prevention. Over a year later, the institute quietly published an article with the detailed risk benefit numbers and concluded that tamoxifen would harm more healthy women than it helped, except for the most high risk women. We urge women and their health care providers not to get caught up in the hype, and to wait for the facts.

— adapted from a report from the National Women's Health Network (NWHN), Washington, D.C.