SEATTLE — Researchers who have established the level of bone mineral density (BMD) change that can serve as a surrogate endpoint in anti-osteoporosis drug clinical trials have now also determined the clinical benefit of such levels through the number of patients needed to treat (NNT) to prevent different kinds of fractures. 

In recent years, development of new anti-osteoporosis drugs has stalled, due in part to the need for clinical trials of investigational drugs to demonstrate fracture risk reduction. That requirement has led to very high costs for clinical trials and a call to use total hip BMD (%THBMD) change at 24 months as a surrogate marker in order to streamline clinical trials and hasten the arrival of new therapies.

The Study to Advance BMD as a Regulatory Endpoint (SABRE), which is managed by the Foundation for the National Institutes of Health and supported by the American Society for Bone and Mineral Research (ASBMR), was established to study the question. Through a previous analysis of 16 clinical trials, the SABRE project team confirmed that the difference between active treatment and placebo in %THBMD at 24 months, also known as the surrogate threshold effect (STE), is a good surrogate for fracture risk. They determined that STE values of 1.43% for vertebral fractures and 2.01% for all clinical fractures could successfully predict the outcomes of past trials. 

“If you can be pretty sure that level of change in BMD would have allowed identification of all the trials that were successful, which it does, then in the future, there’s no reason why it wouldn’t predict those clinical trials that will be successful,” said Richard Eastell, MD, who presented the new study at the ASBMR 2025 Annual Meeting. Eastell is a professor of bone metabolism at the University of Sheffield, England.

The FDA approved a biomarker qualification plan for use of BMD in osteoporosis trials in 2022, and members of the SABRE project hope to hear from the agency by the end of this year, according to Mary Bouxsein, PhD, who is a professor of orthopedic surgery at Harvard Medical School, Boston. However, the FDA wanted to know the expected clinical benefit of a given %THBMD value. 

Eastell described SABRE’s efforts to answer that question by establishing the NNT for over 3 years from the %THBMD value at 24 months. The researchers analyzed data from 25 clinical trials of anti-osteoporosis drugs, the most common of which were bisphosphonates (13 trials). Others were selective estrogen receptor modulators (four studies), denosumab (one study), cathepsin K inhibitor (one), anabolic drugs (four), and menopausal hormone therapy (two). They calculated the absolute risk reduction for fractures in each case and took the inverse of that value to produce the NNT. 

For each trial, their calculations included the 3-year fracture rate in the placebo arm, the NNT, and the first quartile, median, and third quartile for fracture rate. They obtained NNT values ranging between 29 and 67 for vertebral fractures and between 73 and 152 for all clinical fractures. The latter NNT values depended heavily on the fracture rate in a trial’s placebo group.

Within the first quartile, the placebo group’s 3-year risk was 5.9% (NNT = 67) for vertebral fractures and 7.7% (NNT = 152) for all clinical fractures. In the third quartile, the 3-year risk was 13.6% (NNT = 29) for vertebral fractures and 16.0% for all clinical fractures (NNT = 73). 

“The NNTs are lower if you have a high decrease in BMD, and also if you have a high fracture risk in the placebo group,” Eastell said. As a result, clinical trials in patients at high risk for fracture might be expected to have NNTs at the STE of around 29 for vertebral fracture and 73 for clinical fractures, which is similar to previous clinical trials, according to Eastell. 

Changing ‘the Opportunities for Pharma to Produce New Drugs for Osteoporosis’

“Moving from doing trials with fractures as the endpoint to a surrogate marker, which could be BMD after 24 months, is going to change the opportunities for pharma to produce new drugs for osteoporosis. We’ve had no new drugs since 2019 because the companies won’t make that investment,” said Bente Langdahl, MD, PhD, who co-moderated the session where the research was presented and was asked to comment on the study. She is a clinical professor endocrinology and internal medicine at Aarhus University in Denmark.

Historically, the need for fracture risk reduction as an endpoint grew out of clinical trials of fluoride treatments, according to Langdahl. The fluoride was incorporated into bone, but unexpectedly it led to poor bone quality that actually increased the risk of fracture. “That was a scare. Before that, you could have a drug approved by BMD changes, but because of that specific drug, FDA and EMA [European Medicines Agency] and all the other regulatory agencies said, ‘that’s not good enough.’ But now, so much more is done before the studies with respect to bone quality analysis in animals, so we know it’s not fragile bone they [build]. So maybe now is the correct time” to reestablish surrogate markers, she said.

It isn’t a free pass to companies, though. “They will have to do all the preclinical [research], the animal studies, and the bone quality studies in different animals. That’s still there, but I think now it has all come together. The companies are just waiting. They have drugs that have been in phase 1, and some in phase 2, and they’re just waiting to learn how to design their big phase 3 clinical trials,” Langdahl said.

There is optimism that the FDA will approve the surrogate endpoint, according to Bouxsein. In addition to communicating its intent, the agency in July approved the use of BMD as the primary endpoint for Entera Bio’s upcoming phase 3 clinical trial of its parathyroid hormone fragment EB613, which the company bills as the first oral, once-daily anabolic therapy for osteoporosis. That “gives us great optimism that the FDA is going to come out with a positive ruling,” Bouxsein said.

The SABRE project is currently funded by the ASBMR, Amgen, Angitia, Entera Bio, the FDA, Radius, and UCB. Langdahl has consulted for Amgen, Angitia, Astellas, AstraZeneca, Entera Bio, Gedeon Richter, Mereo, and UCB. Bouxsein reported consulting for Angitia and Keros. Eastell disclosed having financial relationships with Alexion, Amgen, Angitia, AstraZeneca, Atman Pharma, Biocon, CL Biologics, Cureteq, Grünenthal, Immunodiagnostic Systems, Osteolabs, Pharmacosmos, Radius, Samsung, Sandoz, Theramex, and UCB. 

Jim Kling is a writer based in Bellingham, Washington.