Age-related bone loss affects one in five women and one in three men over 50. Existing drugs cut the risk of vertebral fractures by 40-70%, but do not eliminate it: falls happen because of muscle weakness, poorer balance, and worse vision. Geroprotectors are interesting because they could act on several links in this chain at once. Nearby in the archive sit studies of aging as loss of regulation and of radiation effects on tissues, but there are almost no direct neighbors on the topic of bone. The work stands off to the side: its closest neighbor is at a distance of 0.477, with nothing right next to it. This is the first systematic test of three geroprotectors on bone within a single trials program.
UM-HET3 mice are a genetically diverse line bred by crossing four inbred strains to better reflect human variability. Males and females received acarbose (1000 ppm) or rapamycin (14 ppm) in their food starting at 4 months of age, and 17α-estradiol (14.4 ppm) starting at 10 months, so as not to interfere with early hormonal development. Control groups got standard chow. The animals were euthanized at 12 or 22 months, the tibia was extracted and scanned by micro-CT at a resolution of 10.4 µm. Trabecular bone was analyzed in a 1 mm segment below the growth plate, and cortical bone in the middle of the diaphysis. Statistics were done in a Bayesian model, which allows effects to be estimated in small groups without p-values, using probabilities and credible intervals instead.
With age, the mice lost both trabecular and cortical bone. In females, trabecular bone volume fell by 6.69% by 12 months and by 8.06% by 22 months relative to the 4-month level; in males the decline was more gradual: 4.99% and 7.50%. Cortical bone thinned in both sexes: by 6.14% at 12 months and by 10.06% at 22 months, with no sex differences in the trajectory. The main result: in females, all three geroprotectors increased trabecular bone volume. At 22 months, rapamycin produced a 34% gain relative to age-matched controls, 17α-estradiol 47%, and acarbose 15%. In males, the effects were weak and statistically unconvincing. The gain came from an increase in the number of trabeculae, not their thickness, which points to preservation of existing architecture rather than active bone formation. Cortical bone responded modestly: in females at 22 months, rapamycin and 17α-estradiol increased the cortical area fraction by about 5%, but cortical thickness did not change, meaning the geometry shifted rather than new tissue accumulating.
This work differs from its neighbors in the archive in that it is the first to compare three geroprotectors at the same skeletal site, in both sexes, and at two ages. The neighboring studies deal with aging as loss of regulation and with radiation effects, but not with bone tissue. Here it is shown that the sex specificity of the skeletal response does not mirror the sex specificity of lifespan extension: 17α-estradiol extends lifespan mainly in males, yet preserves bone in females. This means the pathways through which the drugs act on survival and on bone remodeling are at least partly different. For geroprotective medicine this is an important caution: effects cannot be carried over from one sex to another or from one tissue to another.
The next step is to understand why the bone response is stronger in females. Possible mechanisms include suppression of inflammatory signals (IL-6, IL-1) that activate osteoclasts, or preservation of the pool of osteoblast precursors. It would be useful to test whether geroprotectors combine with classic anabolic drugs, such as parathyroid hormone, and how they affect bone strength under mechanical testing, not just microarchitecture. A separate question is whether a drug can be found that benefits both sexes at once.
The results matter for gerontology, osteoporosis, and translational medicine: they show that geroprotectors may be candidates for preventing age-related bone fragility, especially in postmenopausal women, but they require sex-specific evaluation.
Carry out mechanistic studies in cell cultures and animal models to determine which signaling pathways are responsible for the sex differences. Then test combinations of geroprotectors with existing anti-osteoporosis drugs.
In gerontology, it remains an open question why some tissues age faster than others and why drugs act selectively. This work shows that bone tissue may not follow the general patterns of organismal aging, and it underscores the need for tissue-specific biomarkers of aging.
UM-HET3 mice are "four-breed mutts": they are produced by crossing four inbred strains so that each individual is genetically unique, just like a person. That is exactly why they are used in the geroprotector testing program of the US National Institute on Aging.
In Kim Stanley Robinson's novel Aurora, colonists have to grapple with bone aging under low gravity, and geroprotectors are mentioned there as part of a combination therapy. Although the book is about space rather than earthly old age, the idea of combining drugs that slow the aging of different tissues echoes the conclusions of this work.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.