Despite the concerns of some members of the Oncologic Drugs Advisory Committee (ODAC) and the opposition of several advocacy groups, the panel recommended that the Food and Drug Administration approve two cancer-related indications for Evista (raloxifene), Eli Lilly's selective estrogen-receptor modulator (SERM).
ROCKVILLE, MarylandDespite the concerns of some members of the Oncologic Drugs Advisory Committee (ODAC) and the opposition of several advocacy groups, the panel recommended that the Food and Drug Administration approve two cancer-related indications for Evista (raloxifene), Eli Lilly's selective estrogen-receptor modulator (SERM).
By a vote of 8 to 6 with one abstention, ODAC supported marketing approval for the drug's use for reducing the risk of invasive breast cancer in postmenopausal women with osteoporosis, and by a 10-to-4 vote with one abstention, for reducing the risk of invasive breast cancer in postmenopausal women at high risk of breast cancer.
FDA previously approved Evista for the prevention and treatment of osteoporosis in postmenopausal women. The need for a separate breast cancer prevention indication for women with osteoporosis puzzled some committee members. "I don't understand, really, what is to be gained by the first indication," said Michael Link, MD, professor of pediatrics and chief of hematology/oncology at Stanford University School of Medicine.
"It's a very competitive environment," replied Richard Pazdur, MD, director of FDA's Office of Oncology Drug Products, a remark that drew a hearty laugh from the committee and audience.
The meeting was also punctuated by a rare rebuke to patient advocates from a member of the panel (see box), and questions about whether at-risk women will accept Evista as a preventive agent at a higher rate than they have tamoxifen.
The relatively low usage of of tamoxifen since FDA approved the drug for reducing the risk of breast cancer in high-risk women has generally been attributed to the concerns of women about the risks of endometrial cancer and thrombolic adverse events associated with its use.
RUTH, STAR, and MORE/CORE
To support its New Drug Application, Lilly presented data from three placebo-controlled trials and the active-control Study of Tamoxifen and Raloxifene (STAR) trial, which was conducted without a placebo arm. "The most important data supporting the proposed new indications come from the RUTH [Raloxifene Use for the Heart] and STAR trials," Dr. Pazdur told the committee members.
In RUTH, which compared Evista with placebo in 10,101 postmenopausal women at risk for major coronary events, there were 40 invasive breast cancers in the treatment group and 70 in the placebo arm, a 44% decrease and an absolute risk difference of -1.16. Stage IIa or lower breast cancers were diagnosed in 83% of the Evista-treated women and in 92% of those getting placebo.
Compared with placebo, the Evista group had a significantly higher incidence of fatal strokes, deep vein thrombosis and pulmonary embolism, hot flashes, leg cramps, and peripheral edema.
In RUTH, 5,044 women were treated with Evista every day for a median of 5 years to prevent 30 invasive breast cancers, almost all stage I or II, Dr. Pazdur said. "Described another way," he said, "862 women must be treated for 1 year to prevent an invasive breast cancer in 1 woman." He also noted that "only ER-positive breast canacer are reduced."
STAR Results
STAR, a 5-year, randomized, phase III, multinational, double-blind trial, involved 19,747 postmenopausal women with a projected 1.66% or higher 5-year probability of developing breast cancer or a prior history of lobular carcinoma in situ treated by excision only. The trial's primary endpoint was the occurrence of invasive breast cancer.
The efficacy results from STAR showed a similar number of invasive breast cancer cases in the tamoxifen and Evista groups (168 and 173, respectively). Neither treatment showed superiority, compared with the other.
The tamoxifen-treated patients had a significantly higher incidence of deep vein thrombosis, pulmonary embolism, endometrial hyperplasia, cataracts, and hysterectomy, compared with Evista.
MORE, which randomized 7,705 women, was designed primarily to study the effect of Evista on the risk of fracture in postmenopausal women with osteoporosis, with breast cancer as a secondary endpoint (median follow-up, 3.95 years). Evista patients had a significant 71% decrease in the incidence of invasive breast cancer, compared with placebo.
CORE followed a subset of MORE participants who continued their assigned therapy for an additional 4 years; the study showed a significant 56% decrease in the incidence of invasive breast cancer, compared with placebo.
In the three randomized studies, there was a small incidence of noninvasive breast cancer, with no difference in incidence between raloxifene and placebo.
The STAR trial's lack of a placebo arm caused consternation among committee members, who argued it would have helped them to better evaluate the risk-benefit ratio of Evista. One pointed out that approving a drug for prevention purposes in high-risk but otherwise healthy womenthe second proposed indicationcarried with it an even stronger responsibility to ensure that the benefits of the drug outweigh its risks.
FDA officials said that for all breast cancers, the noninferiority analysis indicated that Evista could potentially lose up to 47% of the preventive effect (invasive and noninvasive) of tamoxifen vs placebo seen in the NSABP P-1 trial, which served as the pivotal study for the approval of Nolvadex (the tamoxifen brand now off-patent and off the market) as a breast cancer preventive.
"The lack of a placebo arm in the STAR trial prevents one from knowing what the tamoxifen effect would have been in this trial," said biostatistician David Harrington, PhD, of the Dana-Farber Cancer Institute. "That is why any active-control trial that uses information from prior trials has to be analyzed very carefully, because you are making a leap across trials. So you need a parachute when you make that leap and to make sure you've looked at all the forces of bias that might cloud the picture."
In their discussions prior to voting, committee members expressed a spectrum of views on the studies' findings. "I think that raloxifene clearly has benefits, and I think the side effects are somewhat exaggerated," said Michael Perry, MD, director, Division of Hematology/Medical Oncology, University of Missouri. "The confidence levelsif you look at some of these side effectsI don't think they are statistically significant. The benefit is not as great as we would like, the side effects are more than we would like, but overall I think approval is justified."
Ronald Richardson, MD, a consultant in medical oncology at the Mayo Clinic, Rochester, Minnesota, commented: "I have to confess I am troubled by all of this. I think that tamoxifen and Evista, in reality, both have minimal activity for either of these indications. They've got toxicities that are significant. The long-term side effects aren't fully known but certainly seem to persist over time. I'm particularly troubled with regard to stroke risk, and I find that the number of patients needed to treat for some benefit is astounding."
Others expressed concern about whether tamoxifen and Evista simply delay the occurrence of cancer rather than prevent it. "I'm worried that probably the natural history of the disease is not the same in osteoporotic women as in other women, and I'm concerned about extrapolating the data to high-risk women who do not have osteoporosis," said Joanne Mortimer, MD, professor of clinical medicine and medical director of the Moores UCSD Cancer Center. "Secondly, I'm consistently troubled by the lack of a decrease in ductal carcinoma in situ (DCIS), which is a precursor for a major cancer. Why in the MORE study, in the RUTH study, and in the STAR study was there no benefit of raloxifene for decreased incidence of DCIS?"