Hip fractures were less common among those given romosozumab, and bone mineral density increased significantly as well.
Vertebral fractures were reduced by 36% in the romosozumab group relative to the alendronate group, and clinical fractures by 27% (P less than .001 for both).
Overall, the number of adverse events for the more than 7,500 patients in the safety population was similar between those receiving romosozumab and either placebo or alendronate, said Scott Wasserman, MD, vice president of global development for Amgen.
However, in Trial 142, which included patients who were slightly older and on more cardiovascular medications at baseline than in Trial 337, MACE – defined as cardiovascular death, MI, and stroke – occurred more frequently among those taking romosozumab, driven primarily by increased cardiac and cerebral ischemic events occurring within the first 12 months of beginning the study drug. At 12 months, the romosozumab arm saw 41 instances of MACE, compared with 22 in the alendronate arm to produce the HR of 1.87 (2.0% vs. 1.1%).
With regard to the imbalance in MACE seen in Trial 142, both the FDA and presenters for the sponsor entertained the notion that alendronate may have been somewhat protective for cardiovascular events. Although there is some biologic plausibility for a cardioprotective event for bisphosphonates, alendronate is highly specific for bone activity and the preponderance of previous studies have not shown such cardioprotection, the FDA, sponsors, and committee members all agreed.
Marc Sabatine, MD, the Lewis Dexter, MD Distinguished Chair in Cardiovascular Medicine at Harvard Medical School, Boston, was available to answer questions on behalf of the study’s sponsor. He noted at several points during the meeting just how few total cardiovascular events were seen overall in the romosozumab trial. The overall small numbers, he said, made it very difficult to distinguish whether the smaller number of MACE seen in the alendronate arm of Trial 142 were a true safety signal or just “a play of chance.”
Dr. Marc S. Sabatine
Bruce Jancin/Frontline Medical News
Almost all the committee’s questioning and discussion centered on this potential increased cardiovascular risk. Amgen, in discussion with the FDA, had agreed to a black box warning designed to wave off prescribing romosozumab to those at increased risk for cardiovascular disease, focusing on those with a history of MI or stroke.
Additionally, the sponsor proposed a postmarketing real-world observational study to track incidence of MACE in those receiving romosozumab, comparing them with those receiving standard of care for osteoporosis.
Several committee members pointed out a problem with the proposed safety mitigation scheme: By conducting a postmarketing observational study for a drug that has a black-box warning to exclude those at high risk for MACE, the chance of detecting an actual cardiovascular safety problem plummets. “This is the textbook example of when observational studies struggle or are virtually guaranteed to fail,” noted Tobias Gerhard, PhD, a pharmacoepidemiologist at Rutgers University, New Brunswick, N.J.
On the other hand, noted several committee members, pausing to conduct a premarketing randomized, controlled trial would keep a beneficial drug away from a population in need. A postmarketing randomized, controlled trial, even a simple trial, still presents challenges, some of the committee acknowledged. Dr. Khosla voiced the opinion that “A randomized, controlled trial is virtually impossible.”
The committee, which was charged with discussing, but not voting on, what additional data should be obtained – and when – to sort out the cardiovascular safety question, was approximately evenly divided in the matter of whether an observational or registry-based trial, or a controlled trial, would be the best path forward.
Committee member Robert A. Adler, MD, put a realistic frame around the debate. “As an endocrinologist, I deal with nuances every day. I really think the kind of clinician who is going to be using this drug is used to dealing with benefits and risks and trying to tailor treatment to a given patient,” said Dr. Adler, professor of internal medicine and epidemiology at Virginia Commonwealth University, Richmond.
In its proposed indication, Amgen defined the population of menopausal women at high risk of fracture as those with a history of osteoporotic fracture, multiple risk factors for fracture, or patients who have failed or are intolerant to other available osteoporosis therapy. Romosozumab would be given as a once-monthly subcutaneous injection of 210 mg for a period of 12 months, to be followed by antiresorptive therapy.
Most of the participants in the clinical trials resided outside the United States, primarily because of the difficulty of recruiting clinical trial participants in the United States, Amgen officials said during their presentation. However, neither the time to the first positively adjudicated MACE nor bone mineral density responses at month 12 differed significantly across the various geographic regions where clinical trial sites were located, said Rachel Wagman, MD, the executive medical director of global clinical development for Amgen. Still, several committee members called for postmarketing data to focus on U.S. patients.
Romosozumab was approved for marketing in Japan on Jan. 8, 2019; approval is also being sought in Europe .
The FDA usually follows the recommendations of its advisory panels.