Abstract / Summary
Prostate cancer (PCa) and its treatments, notably androgen-deprivation therapy (ADT), are linked to accelerated bone loss, thereby elevating the risk of osteoporosis and fractures. Nevertheless, comprehensive data on the prevalence of osteoporosis and low bone mass (osteopenia) in men with PCa are still limited.
To systematically review and meta-analyze the prevalence of osteoporosis and low bone mass in adult men with PCa at all disease stages, and to investigate study-level factors such as ADT category and duration that might influence these estimates.
We conducted a systematic search of PubMed, Scopus, Web of Science, Embase, CINAHL, and ProQuest for studies published between January 1994 and October 2025. Population- or institution-based cross-sectional and cohort studies reporting the prevalence of osteoporosis or low bone mass among adult PCa patients were included. Osteoporosis and low bone mass were classified primarily according to BMD T-score thresholds of ≤  - 2.5 and >  - 2.5 to <  - 1.0, respectively, while study-specific definitions and ascertainment methods were recorded and considered in subgroup analyses. Risk-of-bias assessment was independently performed using the Joanna Briggs Institute checklist. Pooled prevalence estimates were calculated using random-effects meta-analysis, and heterogeneity was assessed using the I2 statistic. Subgroup, meta-regression, and sensitivity analyses were conducted to explore potential sources of heterogeneity.
Fifty-two studies comprising 91,913 men with PCa were included. All 52 studies contributed to the osteoporosis analysis (7,042 reported cases), and 42 studies contributed to the low bone mass analysis. The pooled prevalence of osteoporosis was 22.0% (95% CI: 19.9-24.1; I2 = 99.49%), and that of low bone mass was 41.7% (95% CI: 36.6-46.9; I2 = 95.85%). Pooled estimates varied across BMD ascertainment methods, outcome definitions, ADT categories, geographic regions, and study designs; the largest differences were methodological rather than clinical, and many subgroups contained only one or two studies, so these comparisons are exploratory. Meta-regression found no statistically significant links between publication year or ADT duration and overall prevalence. However, since ADT duration data were available for only 24 studies on osteoporosis and 20 on low bone mass, and residual heterogeneity remained above 92%, the lack of a detected association does not rule out a real connection. Sensitivity analyses yielded results similar to the main findings.