Background: Androgen deprivation therapy (ADT) increases the risk of fractures. The addition of an androgen receptor pathway inhibitor (ARPI) to ADT (MAB) has shown to improve outcomes compared to ADT monotherapy in patients (pts) with metastatic castration-resistant PC (mCRPC), non-mCRPC (nmCRPC), metastatic hormone-sensitive PC (mHSPC), high-risk localized disease and high-risk biochemical relapse. There is no definitive evidence whether MAB increases the risk of fractures compared to ADT alone. Methods: We conducted a systematic review of all clinical trials assessing treatment with an ARPI plus ADT for pts with PC having placebo plus ADT as control arm, using the PubMed/Medline and Cochrane library databases. We also performed a meta-analysis to compare the risk of fractures of each ARPI versus placebo. Further we assessed the number of pts receiving bone protecting agents (BPA) in the studies selected. The comparison between ARPI and placebo in terms of risk of fractures was performed using odds ratio (OR) as meta-analytic outcome. Results: We identified 15 studies comprising 15183 pts (8638 treated with an ARPI and 6545 with placebo), of which 3 studies evaluated abiraterone, 3 apalutamide, 3 darolutamide and 6 enzalutamide. Each ARPI resulted in a statistically significant increase in the risk of fractures compared to placebo (see table), but without statistical differences among the different ARPIs since there is an overlap between the confidence intervals (CI) of the pooled outcome measure of different ARPIs compared to placebo. Only 7/15 studies reported the number (n°) of pts treated with a BPA of which 4 were in mCRPC, 2 in nmCRPC and one in the mHSPC setting. The highest percentage was found in studies including mCRPC pts, of which 35-50% received a BPA. In the studies in the nmCRPC setting, 10-11% of pts received a BPA while in the only mHSPC study reporting data on the use of BPA only two pts (0.4%) in the ARPI group received BPA and no patient in the placebo group. Conclusions: In our meta-analysis MAB resulted in a statistically significant increase in fracture risk compared to ADT regardless of the ARPI used. Data on the use of BPA should be properly reported in future clinical trials. Since long term MAB represents the standard of care in various settings of PC, the use of a BPA should be generally recommended. Dosing and frequency of BPA needs to be adapted according to the specific PC setting.