BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

Tamoxifen As A Breast Cancer Preventive -- Mixed Results In Long-Term Follow-Up

This article is more than 9 years old.

Yesterday researchers updated long-term findings of a large, randomized trial evaluating tamoxifen in its capacity and safety as a breast cancer preventive. The International Breast Cancer Intervention Study-I IBIS-I, begun in 1992, has prospectively tracked over 7,150 women deemed to be at high risk for developing breast cancer. The women were assigned to take either tamoxifen or a placebo for five years.

Jack Cuzick, the John Snow Professor of Epidemiology at Queen Mary University in London, presented the IBIS-I findings.* Now with a median follow-up of 16 years, the trial shows an extended, dramatic reductive effect of tamoxifen on the incidence of breast tumors. The Lancet Oncology published the extended results and an accompanying editorial.

The findings come with a few major caveats, though. With such long follow-up, there has not yet been a meaningful difference observed, either way, in breast cancer-related or overall mortality. Worrisome, too, is a slightly increased number of estrogen-receptor negative tumors observed in the tamoxifen arm of the study. Although those findings may be due to chance, they warrant careful analysis, Cuzick said.

Most of the women entered the trial based on high risk assessed based on family history; in some cases, women were enrolled after a biopsy revealed a benign breast lesion associated with increased breast cancer risk. The women had a mean age of approximately 51 years at the enrollment between 1992 and 2001. As of May 2014, a total of 601 breast cancers were detected among the women in the study, including 251 among 3,579 in the tamoxifen group (7.0 percent) and 350 among 3,575 (9.8 percent) in the placebo group, as reported in the Lancet Oncology paper. This difference was highly statistically significant (p<0·0001).

By following the Kaplan-Meier curves for the cumulative incidence of tumors over 20 years, the investigators found breast cancer rates of 12.3 vs. 7.8 percent in the placebo vs. tamoxifen groups, respectively. This difference suggest that the number needed to treat (NNT) to prevent one case of breast cancer within 20 years would be 22 women. If the investigators exclude the DCIS cases from this calculation, the NNT to prevent a case of invasive breast cancer is 29 women.

Tamoxifen, a partial estrogen antagonist, was given at a standard dose of 20 milligrams by mouth per day. Those assigned to take tamoxifen did so for only five years on study. Remarkably, after a median of 16 years on study, approximately 75 percent of the women who have not developed breast cancer remain “blinded” as to whether they received tamoxifen or the placebo. The long-term reduction in breast cancer was maintained years after a finite course of treatment, Cuzick emphasized.

In the first ten years of the IBIS-I study, most of the breast tumors “spared” by tamoxifen were estrogen-receptor positive (ER+) and noninvasive ductal carcinoma in situ (DCIS) cases. At 16 years, the pattern of reduced ER+ cases held, but the diminution of DCIS didn’t last. Notably, the number of estrogen-negative breast cancers increased in the tamoxifen arm at 16 years, although this trend didn’t reach statistical significance.

Other cancers did arise among both patient groups. Those included endometrial tumors (a well-established toxicity of tamoxifen); these arose in 29 and 20 women who received tamoxifen and placebo, respectively. An unanticipated result is that non-melanoma skin cancers developed disproportionately among women on tamoxifen: those number were 116 and 84, respectively.

Deaths were more frequent in women assigned to receive tamoxifen, although the differences did not meet statistical significance. Based on the abstract* presented at the San Antonio meeting, 182 women (5.1 %) died on the tamoxifen arm, and 166 women (4.6 %) died on the placebo arm of the study. This difference was not statistically significant. Deaths from breast cancer were also more frequently seen in the tamoxifen arm (31 vs. 26 women). Again, this trend did not meet statistical significance. Five deaths were seen from endometrial cancer, all on the tamoxifen arm.

schematic, structure of tamoxifen (Wikimedia image)

Tamoxifen was manufactured exclusively by AstraZeneca (Nolvadex) at the start of this trial in the 1990s. The drug came off patent around 2002. A month’s supply of a generic tamoxifen can be purchased on-line for under $40, although recent shortages have been reported. Two of the study authors report speaking for AstraZeneca, and directly or indirectly receiving grant support from that company and/or Novartis.

*San Antonio Breast Cancer Symposium abstract #S3-07

Follow me on LinkedInCheck out my website