This news story reports the initial results of the Study of Tamoxifen and Raloxifene Trial (STAR) for post-menopausal women at high risk of developing breast cancer. The report greatly overestimates the true number of women who might safely benefit from chemoprevention with either tamoxifen (Nolvadex) or raloxifene (Evista). 9 million women do not fall into the category of very high risk of developing breast cancer; it is closer to 2 million women, and even this is a high estimate. The report also suggests that women currently taking tamoxifen would and should want to immediately switch to raloxifene based on the results of this study, which might not be prudent for many women and would be considered an off-label use for this osteoporosis medication. More importantly, the story fails to mention that Eli Lilly was previously fined for promoting the drug as chemoprevention for breast cancer without FDA approval. Additionally, this type of hormone therapy should be only used in estrogen-receptor positive, post-menopausal women, which the story fails to mention.
The STAR trial was designed as a 5-year double-blind clinical trial in which post-menopausal women at higher than average risk of developing breast cancer were randomly assigned to take tamoxifen or raloxifene (Evista) as chemoprevention for invasive breast cancer. Because of the initial positive results of the STAR trial, namely, there were fewer uterine cancers and blood clots in the raloxifene group, the trial was unblinded at 4 years. The story does report that these results did not reach statistical significance, however, the study design and early unblinding due to these non significant results was not mentioned. Both drugs in this study had other side effects typical of this type of hormone therapy, such as bothersome hot flashes and vaginal discharge. Results of this trial and further analysis of the data have yet to be published in a peer-reviewed medical journal and the story could have made clear that decisions about switching hormone medications should wait until this data is carefully reviewed.
Tamoxifen is currently the only drug with FDA approval for both treatment and prevention of breast cancer in high-risk women. Raloxifene is currently approved to prevent osteoporosis, and while it appears to have similar benefits to tamoxifen in the prevention of invasive breast cancer, it has not been studied for as long as tamoxifen and does not appear to prevent as many cases of non-invasive breast cancers as tamoxifen, which are more common.
It is important to stress that the women in this trial were at higher than average risk of developing the disease due to age and breast biopsies and/or a family history of breast cancer. For women at low to moderate risk for developing breast cancer, the risk of developing side effects outweighs the preventive benefits from taking tamoxifen or raloxifene. Women at high risk of developing breast cancer who are also at increased risk for developing blood clots in the legs or uterine cancers are not typically advised to use these hormone therapies as chemoprevention.
These were some of the possible points of emphasis missed in this story.
No mention of the cost of treatment for women outside clinical trials. On average, raloxifene is about $75 for a month’s supply, while tamoxifen is about $100.
Some quantitative relative risk benefit mentioned of taking hormone therapy for chemoprevention. Absolute risk reduction not given. Despite this, we realize this is a TV story with limited time (although they certainly can devote as much time as they wish), so we will give a satisfactory score. But in the future, we urge that absolute terms be given. (See “absolute vs. relative risk” on the home page under “Things You Should Know About Research Stories.”)
Does mention that the difference in side effects was not statistically significant for uterine cancer and blood clots. Mentions risk of cataracts and reduced incidence with Evista. Mentions harms of hormone therapy, such as an increased risk of blood clots and uterine cancers. Does not mention that raloxifene, like tamoxifen, has bothersome side effects such as hot flashes and vaginal discharge, which are reported as comparable and mild to moderate in the NCI press release of the initial results of this trial.
No mention of study design. No quantitative evidence presented on comparison of benefit or side effects of either drug, and no mention that the study was unblinded early. Tamoxifen and raloxifene were shown to reduce breast cancer risk by 50%, however, in the STAR trial, tamoxifen protected women better from non-invasive breast cancer such as DCIS (ductal carcinoma in situ) and LCIS (lobular carcinoma in situ). While 50% is listed as a RRR, it is incomplete: In one large U.S. trial (N=13,338) women were followed for 7 years and there was a 43% reduction in invasive breast cancer (25 per 1000 in the tamoxifen group vs. 42 of 1000 women in the placebo group). The risk of DCIS was reduced by 37% (10 in 1000 women in the tamoxifen group vs. 16 in 1000 women in the placebo group) (Fisher, Costantino et al. 1998; Fisher, Costantino et al. 2005). As a result of these trials, tamoxifen was approved as the ONLY chemopreventive drug for healthy women at high risk for developing breast cancer.
Evidence of disease mongering. 9 million women do not fall into the category of “very high risk” of developing breast cancer, it is closer to 2 million women, an even this is a high estimate (Freedman, A. N., B. I. Graubard, et al. (2003). “Estimates of the number of US women who could benefit from tamoxifen for breast cancer chemoprevention.” J Natl Cancer Inst 95(7): 526-32.). Chemoprevention is for a small number of women and clinical guidelines of the American College of Obstetricians and Gynecologists and the U.S. Preventive Services Task Force suggest that tamoxifen should only be offered to women with a 5-year projected risk for breast cancer of 1.6% in order to reduce their risk. Also, this medication is for women with estrogen-receptor positive breast cancer.
This study was partially funded by Eli Lilly and AstraZenaca, makers of Evista and Tamoxifen. There is no mention that any of the breast cancer experts cited have financial ties with these drug companies.
The story does mention another chemoprevention hormone drug, tamoxifen, but talks about how it has more serious side effects, and “many women will probably switch to raloxifene”. Does not mention need for mammograms (screening as early detection shown to improve outcomes) in post-menopausal women at a higher risk of breast cancer. Prophylactic mastectomy also an option for women at the highest risk of developing breast cancer, not for the typical post-menopausal woman (mostly for those who carry the BRAC1 or BRAC2 genes). In the STAR trial, the women who took raloxifene had a small, non-statistically significant, number of fewer blood clots and uterine cancers and there was a statistically significant difference in cases of cataracts, however, the risk of these side effects are very small, even with a regimen of tamoxifen. It should be noted that the results presented in this story are initial findings after 4 years of raloxifene. Tamoxifen is usually taken for 5 years and there is much more long-term clinical data on safety and efficacy for this drug. The study was unblinded early due to the improved side effect profile for raloxifene, which was not statistically significant.
The story mentions that Evista is currently available for use in post-menopausal women to prevent osteoporosis. Chemoprevention would be an off-label use. The story fails to mention that it is not yet approved for chemoprevention and Eli Lilly was fined last year for promoting the drug as chemoprevention for breast cancer w/o FDA approval. The news report suggests that most women could take the drug for this off-label use immediately. Also, this type of hormone therapy should be only used in estrogen-receptor positive, post-menopausal women. Not all high-risk women could benefit.
Notes that use of Raloxifene (Evista) as chemoprevention is new. Evista is already approved to prevent osteoporosis.
Does not appear to be directly from a press release. Breast cancer experts not affiliated with the NCI trial interpret the results.