An analysis of data from a phase III clinical trial that was unblinded 5 years ago indicates that in some breast cancer patients, use of the aromatase inhibitor letrozole (Femara) after 5 years of adjuvant therapy with tamoxifen has survival benefits even if begun several years after completing tamoxifen.

Conducted by the National Cancer Institute of Canada Clinical Trials Group, the trial, known as MA.17, was unblinded in 2003 after an interim analysis demonstrated that letrozole, begun within 3 months after 5 years of tamoxifen, reduced breast cancer recurrence risk compared to placebo. Participants in the placebo arm of MA.17 were then offered the opportunity to begin 5 years of adjuvant treatment with letrozole. MA.17 participants were post-menopausal women with early-stage, ER-positive disease.

This new analysis from MA.17, released online on March 10 by the Journal of Clinical Oncology (JCO), compared the 1,579 women in the placebo arm who began using letrozole after the trial was unblinded with the 804 women who chose not to. The analysis revealed that letrozole improved disease-free survival by 63 percent and distant (or metastatic) disease-free survival by 61 percent. The median time of letrozole initiation after tamoxifen completion was 2.8 years, with a range from 1 to 7 years.

Questions about the optimal approach to adjuvant therapy in this patient population linger, the study's lead author, Dr. Paul Goss from Harvard Medical School, and colleagues noted, because even after 5 years of adjuvant tamoxifen, there continues to be a significant risk of late recurrence and death from breast cancer.

Although interpretation of the findings is "made difficult because this therapeutic intervention was self-selected, not randomly allocated," they continued, the women who took letrozole had disease characteristics that put them at higher recurrence risk than those who did not, suggesting letrozole was responsible for the beneficial effects.

Letrozole use was also associated with an increased risk of bone fractures and osteoporosis (5.2 and 5.3 percent, respectively) compared with those on placebo (3.1 and 1.6 percent). In a related editorial in JCO, Drs. Nancy U. Lin and Eric P. Winer from Dana-Farber Cancer Institute noted that there is a chance that the findings could be understating those comorbid effects.