GLP-1 and Bone: Turnover and Density (2026)

Here is the uncomfortable fact almost no one mentions when you start a GLP-1: losing weight — by any method — tends to lower your bone density. That is not unique to Ozempic, Wegovy, Mounjaro or Zepbound. Decades of weight-loss research show that when fat comes off, bone usually comes off too, because the skeleton sheds the mass it no longer has to carry (Liu 2025 ^[1]; Villareal 2006 ^[2]). So the real question for GLP-1 users is sharper than “do these drugs hurt bone?” It is: during the bone loss that weight loss always causes, do GLP-1 drugs protect the skeleton, add to the damage, or simply ride along with it? The honest answer in 2026 is that the evidence is genuinely mixed and still thin. Short-term studies in volunteers suggested GLP-1s might boost bone formation and protect bone (Iepsen 2015 ^[4]); but a year-long randomized trial in higher-risk adults found the opposite — more bone breakdown and measurably lower bone density on semaglutide than placebo (Hansen 2024 ^[5]). This article walks through what bone turnover markers actually measure, what the trials found, and what it means if you are older or postmenopausal. For the broader fracture-outcome picture, see GLP-1s and fracture risk.

The honest summary

• Weight loss itself lowers bone density. In a randomized trial, six months of calorie-restriction weight loss reduced bone density at the hip and spine, while weight loss from exercise alone did not — the bone loss tracks the lost weight, not the method (Villareal 2006^[2]). A meta-analysis of weight-loss trials confirmed dietary weight loss reduces hip bone density (Soltani 2016^[3]).
• Bone turnover markers tell the mechanism. CTX (C-terminal telopeptide) rises when bone is being broken down (resorption); P1NP (procollagen type I N-propeptide) rises when bone is being built (formation). Reading the two together shows whether the skeleton is in net gain or net loss (Anastasilakis 2025^[9]).
• Early GLP-1 data looked protective. In weight-reduced women, adding liraglutide for a year raised the formation marker P1NP by 16% and prevented the bone loss the control group experienced (Iepsen 2015^[4]).
• A year-long trial in higher-risk adults looked harmful. In adults at increased fracture risk (mostly postmenopausal women), once-weekly semaglutide raised the resorption marker CTX by ~166 ng/L versus placebo and produced lower hip and spine bone density at 52 weeks (Hansen 2024^[5]).
• A short older-adult trial found no difference. A 20-week pilot in older adults with prediabetes/diabetes found no detectable change in bone density or turnover markers from semaglutide-assisted weight loss — but it was small and brief (Dinkla 2025^[6]).
• Net read: the protective signal has not held up in the longer, higher-risk trial. Bone turnover during GLP-1 weight loss seems to be uncoupled in at least some people — resorption up, formation not — which is the pattern that erodes bone (Hansen 2024^[5]; Anastasilakis 2025^[9]).
• This matters most for older adults and postmenopausal women, who start with less bone reserve. The countermeasures are the same ones that protect bone during any weight loss: adequate protein, resistance training, vitamin D/calcium, and bone-density monitoring in higher-risk people.

First principle: weight loss lowers bone density — on any drug or no drug

Before blaming any medication, start with the baseline truth: intentional weight loss reduces bone mass. The skeleton is a load-bearing structure that adapts to the forces placed on it. When you carry less weight, the mechanical signal that tells bone to stay dense weakens, and bone is partly resorbed. There are other contributors too — lower calcium absorption, shifts in fat-derived and gut hormones, and reduced fat mass that was previously a source of estrogen (Liu 2025^[1]).

The cleanest demonstration came from a randomized trial that compared two ways of losing the same amount of weight. Adults randomized to lose weight by calorie restriction had measurable drops in bone density at the hip and spine; adults randomized to lose the same weight by exercise did not lose bone — because exercise supplies the mechanical loading that calorie restriction removes (Villareal 2006^[2]). A later meta-analysis of weight-loss trials reached the same conclusion: dietary weight-loss approaches reduce hip bone density, and the magnitude tends to track the amount of weight lost (Soltani 2016^[3]).

What bone turnover markers actually measure

Bone density (measured by DEXA) changes slowly — it can take a year or more to register a meaningful shift. Bone turnover markers are blood tests that move within weeks and reveal what the skeleton is doing right now. Two matter most here:

• CTX (C-terminal telopeptide of type I collagen) is a resorption marker. It is a fragment released when osteoclasts break down old bone. Rising CTX means more bone is being dismantled.
• P1NP (procollagen type I N-propeptide) is a formation marker. It is released when osteoblasts lay down new bone collagen. Rising P1NP means more bone is being built.

In a healthy skeleton these two are coupled — resorption and formation rise and fall together so bone is constantly renewed without net loss. Trouble appears when they uncouple: if CTX (breakdown) goes up but P1NP (building) does not follow, bone is being removed faster than it is replaced, and density falls. Watching CTX and P1NP together is how researchers tell whether a weight-loss drug is sparing bone, eroding it, or neutral (Anastasilakis 2025^[9]).

The optimistic early evidence: GLP-1s may build bone

The hopeful case rests on a well-designed mechanistic trial. Iepsen and colleagues took obese women who had already lost about 12% of their body weight on a low-calorie diet, then randomized them to a year of the GLP-1 drug liraglutide or no drug, with both groups maintaining the weight loss. The result: in the liraglutide group the bone formation marker P1NP rose by 16%, while it fell slightly in the control group, and liraglutide prevented the bone density loss that otherwise follows weight loss (Iepsen 2015^[4]). The proposed mechanism is that GLP-1 signaling — directly or via gut and thyroid C-cell pathways — can stimulate bone-forming cells. A small randomized trial of the GLP-1 drug exenatide in obese, antipsychotic-treated patients similarly found no harmful effect on bone turnover markers or density over three months (Eriksson 2018^[7]).

If that pattern held broadly, GLP-1s would be unusual among weight-loss interventions: a way to lose weight while protecting the skeleton from the bone loss that weight loss normally causes. That is the claim you will sometimes see repeated online. The problem is that the longer, higher-risk trial did not reproduce it.

The cautionary evidence: uncoupled turnover and lower density

The most directly relevant trial to date is a 52-week, double-blinded, randomized, two-centre phase 2 study in 64 adults at increased fracture risk — mostly postmenopausal women, mean age 63 — given once-weekly semaglutide or placebo (Hansen 2024^[5]). The semaglutide group lost more weight (about 6.8 kg more than placebo), as expected. But the bone picture was unfavorable:

• Resorption rose. The bone-breakdown marker P-CTX was higher with semaglutide than placebo at 52 weeks — an estimated difference of about 166 ng/L.
• Formation did not. There was no difference in the bone-formation marker P1NP between groups (semaglutide 64.3 vs placebo 62.3 µg/L) — exactly the uncoupling that erodes bone.
• Density fell at the key fracture sites. Areal bone density was lower with semaglutide at the lumbar spine (estimated difference −0.018 g/cm²) and the total hip (−0.020 g/cm²); the femoral neck did not differ significantly.

This is essentially the opposite of the Iepsen liraglutide finding: instead of P1NP rising to protect bone, CTX rose while P1NP stayed flat, and density dropped at the hip and spine. A 2025 critical appraisal of anti-obesity medications and bone concluded that the reassuring early signal has not consistently held, and that the dominant driver of bone loss on these drugs is most plausibly the magnitude and speed of weight loss they produce, with the drug-specific bone effect still unsettled (Anastasilakis 2025^[9]). A separate 2025 study using high-resolution imaging in people with type 2 diabetes on oral semaglutide examined volumetric bone density and microarchitecture and likewise found changes that warrant monitoring rather than reassurance (Das 2025^[8]).

What the short, neutral trial adds

A 2025 pilot trial offers a middle data point. It randomized 20 older adults (mean age ~73) with prediabetes or diabetes to semaglutide plus lifestyle counseling or counseling alone, and measured bone density plus CTX and P1NP at baseline and 20 weeks. The semaglutide group lost more weight (−5.3% vs −0.89%), but the analysis found no significant difference in whole-body bone density or in either turnover marker between groups (Dinkla 2025^[6]). That sounds reassuring, but the trial was tiny and lasted only 20 weeks — bone density barely moves in that window, and 20 patients cannot detect a modest effect. It is best read as “no alarm in the short term in this small sample,” not as evidence of long-term safety.

What this means for higher-risk groups

The people who should pay attention are those with the least bone to spare: older adults and postmenopausal women, plus anyone with a prior fragility fracture, low body weight, or an existing osteoporosis/osteopenia diagnosis. They start with thinner bones, lose bone more readily during weight loss, and were exactly the group in which the year-long semaglutide trial showed lower density and higher resorption (Hansen 2024^[5]). The countermeasures are well established from the broader weight-loss literature and are worth building into any GLP-1 plan:

• Resistance training is the single best lever. In older adults losing weight, resistance exercise attenuated hip and femoral-neck bone loss better than aerobic exercise — it supplies the mechanical loading that weight loss removes (Beavers 2017^[10]; Villareal 2006^[2]).
• Protein intake. Adequate dietary protein supports both muscle and bone during weight loss; muscle loss and bone loss tend to travel together. (See common GLP-1 side-effect questions for the lean-mass picture.)
• Vitamin D and calcium. Calorie restriction can reduce calcium absorption, so meeting calcium and vitamin D targets matters more, not less, during weight loss (Liu 2025^[1]).
• Bone-density monitoring. For higher-risk patients, a baseline DEXA scan and follow-up after substantial weight loss lets you and your clinician catch meaningful decline early.
• Pace and total loss. Because bone loss scales with weight lost, the magnitude of total weight loss is itself a risk factor; for fragile patients, slower, monitored loss may be safer than the fastest possible.

If you already have osteoporosis or take a bone medication, that changes the calculus — see GLP-1s with osteoporosis and bisphosphonates. None of this is a reason to avoid a GLP-1 if you have a strong indication; obesity and diabetes carry their own fracture and health risks. It is a reason to protect bone proactively rather than assume the drug does it for you.

Bottom line

Any weight loss tends to lower bone density, and GLP-1 drugs cause large, fast weight loss — so some bone-density decline should be expected regardless of any direct drug effect (Liu 2025^[1]; Villareal 2006^[2]). The hope that GLP-1s actively protect bone — raising the formation marker P1NP and preventing loss — came from a year-long liraglutide maintenance trial in younger women (Iepsen 2015^[4]), but it did not reproduce in a year-long semaglutide trial in older, higher-fracture-risk adults, where the resorption marker CTX rose, formation did not, and hip and spine density fell (Hansen 2024^[5]). A short older-adult pilot found no detectable change but was too small and brief to be reassuring (Dinkla 2025^[6]). The honest verdict in 2026: the protective claim is unproven and the cautionary signal is real in the people most vulnerable to it. The practical move is to protect bone the way you would during any weight loss — resistance training, protein, vitamin D/calcium, and DEXA monitoring for higher-risk patients — and to discuss bone health with your clinician if you are older, postmenopausal, or have had a fracture.

This article is educational and is not medical advice. Every claim above is sourced to a peer-reviewed study, randomized trial, meta-analysis, or critical appraisal indexed in PubMed, verified against the live PubMed database before publication. Discuss your own bone-health plan with your prescriber, especially if you are older, postmenopausal, or have a history of fracture or osteoporosis.