A Repurposed Cancer Drug Rebuilds Bone After Menopause, in Mice

As men and women age, bone mass and skeletal strength gradually decline. Although some degree of bone loss is a normal consequence of ageing, accelerated loss of bone mass can lead to osteoporosis, a metabolic bone disorder characterised by reduced bone mass, deterioration of bone structure and increased fracture risk. Globally, osteoporosis affects approximately one in three women and one in five men over the age of 50. Women are particularly susceptible to bone loss following menopause because reduced oestrogen levels disrupt normal bone remodelling. Therefore, post-menopausal osteoporosis represents a significant clinical challenge, highlighting the need for new therapeutic approaches.

Bone is continuously remodelled through the coordinated activity of osteoblasts, which are responsible for bone formation, and osteoclasts, which resorb old or damaged bone. RUNX2 is a transcription factor that is widely recognised as a key regulator of osteoblast differentiation and skeletal development. However, RUNX2 is also involved in cancer progression and metastasis. CADD522 is a small-molecule inhibitor of RUNX2-DNA binding that was originally investigated as a potential anticancer agent. Previous studies showed that CADD522 could reduce cancer-induced bone disease, prompting researchers at the University of East Anglia to investigate whether RUNX2 inhibition could also protect against post-menopausal bone loss.

The researchers used an ovariectomy (OVX)-induced mouse model to mimic post-menopausal oestrogen deficiency. Thirty-two five-week-old female C57BL/6J mice were randomly assigned to either OVX surgery or sham surgery. Each group was subsequently divided into vehicle-treated and CADD522-treated groups. CADD522 was administered at 25 mg/kg three times per week for eight weeks, while control animals received the vehicle. Clinical signs were monitored throughout the study.

Microcomputed tomography (μ-CT) analysis of the tibias demonstrated that CADD522 increased bone volume in both control and OVX mice. In OVX mice, CADD522 also increased bone surface density and preserved trabecular bone number, indicating improved preservation of bone microarchitecture. Furthermore, circulating procollagen type 1 N-propeptide (P1NP), a biomarker of collagen and bone formation, was increased in OVX-CADD522 mice. In contrast, β-carboxy-terminal telopeptide of type I collagen (βCTX), a marker of bone resorption, remained unchanged. These findings were consistent with a net anabolic effect of CADD522 rather than a conventional antiresorptive effect.

Histological analysis using tartrate-resistant acid phosphatase (TRAP) staining showed differences in the number and distribution of osteoclasts. Interestingly, OVX-CADD522 mice displayed a marked increase in TRAP-positive multinucleated osteoclasts distributed more continuously along trabecular surfaces. This suggests that CADD522 does not simply suppress bone resorption but may instead alter the dynamics of bone remodelling and restore a more balanced remodelling process.

The study also investigated the effects of CADD522 on adiposity and lipid metabolism. CADD522 reduced body weight and inguinal fat mass specifically in OVX mice and reduced adipocyte accumulation within the bone marrow. Fatty-acid profiling demonstrated widespread changes in the lipid landscape, particularly within the phospholipid fraction. Treatment shifted several fatty-acid profiles towards those observed in non-OVX control animals, suggesting partial restoration of menopause-associated lipid alterations. However, the researchers noted that these lipidomic findings were based on a limited number of animals and therefore demonstrate associations rather than proving a causal mechanism.

To determine whether CADD522 directly interacts with its proposed molecular target, the researchers performed a cellular thermal shift assay (CETSA). The assay demonstrated dose-dependent engagement of RUNX2 by CADD522, with an EC50 of approximately 116 nM. This supports the conclusion that the observed skeletal effects were associated with direct RUNX2 engagement.

The finding that inhibition of RUNX2 improves bone architecture is initially unexpected because RUNX2 is an important regulator of osteoblast differentiation and skeletal development. However, the researchers propose that the role of transcription factors can differ between developing and mature tissues. Their results suggest that RUNX2 may contribute to pathological bone remodelling in the adult skeleton following oestrogen deficiency. Inhibition of RUNX2 therefore appears to enhance bone formation and preserve trabecular architecture without globally suppressing bone turnover.

Overall, this study demonstrates that CADD522 has promising therapeutic potential for post-menopausal bone loss in a preclinical mouse model. RUNX2 inhibition by CADD522 increased bone volume, preserved trabecular microarchitecture and increased the bone-formation marker P1NP, while also reducing peripheral and bone-marrow adiposity and modifying lipid metabolism. Importantly, the study provides evidence of direct RUNX2 target engagement, oral bioavailability and favourable short-term tolerability. However, the findings remain preclinical: the experiments used a single ovariectomy model, and longer-term studies are needed to determine whether the skeletal benefits are sustained and whether they ultimately translate into reduced fracture risk. Therefore, CADD522 should not yet be considered an established treatment for osteoporosis. Still, the findings provide a strong rationale for further investigation and optimisation of RUNX2 inhibition as a potential therapeutic strategy for post-menopausal osteoporosis.

Reference

Ersek A, Kim MS, Suelzu C, Piec I, Matsangos A, Nurmemmedov E, Bull EC, Bartlett T, Dunn R, Dick JR, Lapidus R, Copeman DM, Tang JCY, Crichton PG, Pontifex MG, Fraser WD, Horwood NJ, Passaniti A, Green D. RUNX2 inhibitor CADD522 improves bone microarchitecture and lipid metabolism in post-menopausal bone loss. NPJ Drug Discov. 2026 Sep 7;3(1):38. doi: 10.1038/s44386-026-00076-z. PMID: 42706322; PMCID: PMC13550584.