SAN ANTONIO – In sharp contrast with previous findings, zoledronic acid failed to improve disease-free survival when added to adjuvant chemotherapy for women with stage II/III breast cancer in a highly anticipated phase III trial.
Dr. Robert Coleman
"In terms of primary analysis and primary end point, this is a negative trial," Dr. Robert Coleman said during a press briefing on Dec. 8 at the annual San Antonio Breast Cancer Symposium. "It is highly unlikely that this conclusion will change with further follow-up."
The randomized, open-label AZURE (Adjuvant Treatment with Zoledronic Acid in Stage II/III Breast Cancer) trial involved 3,360 women with stage II/III breast cancer at 174 participating centers. Based on the results, Novartis, maker of zoledronic acid (Zometa), announced that it is withdrawing applications for the drug’s approval as an adjuvant breast cancer treatment in the United States and Europe. The company said it will evaluate its plans based on the data, which did include one positive finding.
Older women who had undergone menopause at least 5 years earlier showed a significant 29% improvement in overall survival with the osteoporosis drug added to standard therapy. This was not true for premenopausal or perimenopausal women, however, and no benefit was seen in disease-free survival, the primary end point.
The National Cancer Research Network in the United Kingdom conducted the study, with partial funding by an academic grant from Novartis. Several study authors reported financial relationships with a number of pharmaceutical companies, including Novartis.
Patients were randomized to either a standard adjuvant therapy alone or in combination with the study drug. In the latter arm, a 4-mg dose of zoledronic acid was given every 3-4 weeks for the first 6 months (six doses). For months 6-30, zoledronic acid was given every 3 months (eight doses). After 30 months, zoledronic acid was given every 6 months (five doses). Treatment with zoledronic acid lasted for 5 years.
Patients were included if they had stage II/III breast cancer; the population included node-positive adjuvant patients and T3/T4 or confirmed node-positive neoadjuvant patients. The women could not have any evidence of metastasis. They had to have a complete primary tumor resection and a Karnofsky performance score of at least 80.
Patients were excluded if they had received bisphosphonate treatment in the last year, bone disease (including osteoporosis) at study entry, a serum creatinine level greater than 1.5 upper limit of normal, significant ongoing dental problems or planned dental surgery, or other malignancies.
At a median follow-up of 59 months, there was no difference in disease-free survival overall (adjusted hazard ratio 0.98, P = .79) or invasive disease-free survival (adjusted HR = 0.98, P = .73).
"Clearly our results are very different from those published by the ABCSG [the Austrian Breast & Colorectal Cancer Study Group] investigators [study 12]," said Dr. Coleman, professor of medical oncology at the University of Sheffield (England), referring to an earlier study that suggested the bone drug might also prevent breast cancer recurrence (N. Engl. J. Med. 2009; 360:679-91).
The ABCSG-12 study included more than 1,800 premenopausal women with hormone receptor–positive, early-stage breast cancer. Following complete resection and hormone therapy – including goserelin treatment to suppress ovarian function and induce menopause – women were treated with or without zoledronic acid for 3 years. The researchers demonstrated that the addition of 3 years of zoledronic acid therapy to hormonal therapy following surgery improved disease-free survival by 32% (HR .68, P = .009).
"In AZURE, ER-positive premenopausal women [the same population as in the ABCSG XII trial] are showing no benefit – in fact, there’s almost a disadvantage with zoledronic acid – and that’s clearly different from ABCSG 12," said Dr. Coleman.
The AZURE researchers did extensive planned subgroup analyses. "One clearly stands out from all the others. That is the treatment effect on disease-free survival according to menopausal status," Dr. Coleman noted.
The AZURE trial demonstrated a trend toward longer overall survival for all women on zoledronic acid, but this did not achieve significance (adjusted HR = 0.85, P = .07). However, there was a clear overall survival benefit for women at least 5 years out from menopause on zoledronic acid. These women had a 29% reduction in the risk of death (P = .017). There was no difference between the control and treatment groups among pre- and perimenopausal women (adjusted HR 1.01, P = .93).
Dr. Coleman speculated that lower levels of estrogen had a role in the responses of older women to zoledronic acid. In the earlier Austrian trial, he noted, younger women were forced into early menopause with hormonal therapy, and that may have played a similar role in the responses seen in that trial.