Prostate cancer (PCa) is the second most common malignant tumor in men and the fifth leading cause of cancer death worldwide [1]. Bone metastasis (BM) is the main complication and cause of death in patients with PCa, and the 5-year survival rate in patients with BM is only 30% [2,3]. Therefore, early prediction of BM can promote prognosis and select an appropriate treatment.

Numerous studies have demonstrated that prostate-specific antigen (PSA) and Gleason score (GS) are major predictors of BM in PCa [[4], [5], [6]]. PSA is a widely used tumor marker for screening and early detection of PCa. However, the sensitivity and specificity of PSA for predicting BM in patients with PCa are unsatisfactory since benign diseases such as prostatitis can also cause an increase in PSA, while some poorly differentiated PCa may lead to a decrease in PSA [7]. On the other hand, GS, the most common classification standard for histological grading of PCa, is generally obtained through transrectal ultrasound-guided biopsy. However, as the biopsy is an invasive procedure, it is not always well tolerated by patients. Accordingly, it is necessary to identify a non-invasive method that could accurately predict the risk of BM in patients with PCa.

Multi-parameter magnetic resonance imaging (MRI) is considered to be the most effective non-invasive examination method for diagnosing and evaluating PCa. The Prostate Imaging Reporting & Data System (PI-RADS) consensus guideline recommends using diffusion-weighted imaging (DWI) sequence for identifying PCa in the peripheral zone [8]. However, DWI may be affected by microcirculation perfusion and cannot accurately reflect the diffusion of water molecules in the tissue, which can be precisely measured using intravoxel incoherent motion (IVIM) imaging. IVIM can quantitatively evaluate the effect of pure water molecule diffusion and microcirculation perfusion using the multi-b-value bi-exponential model [9]. Previous studies have indicated that IVIM parameters have potential value in making the diagnosis, aggressiveness assessment, and treatment response monitoring of PCa [[10], [11], [12]].

Amide proton transfer-weighted (APTw) imaging is a new type of endogenous chemical exchange saturation transfer technique that reflects the content of proteins and peptides in tissues. It is widely used to diagnose neurological diseases such as brain tumors and cerebrovascular diseases [13]. Several prostate APTw studies have demonstrated the potential value of APTw values in the diagnosis and risk assessment of PCa. In their study, Jia et al. [14] reported that APT value can discriminate between PCa and peripheral zone noncancerous tissues. Takayama et al. [15] and Yin et al. [16] found that APT value is closely associated with GS of PCa. Guo and colleagues [17] showed that APT value could differentiate between transition zone PCa and benign prostatic hyperplasia. However, so far, no study has investigated the feasibility of APTw and IVIM for predicting BM in patients with PCa.

The purpose of this study was to explore the value of 3D APTw and different diffusion models (DWI, IVIM with different fitting algorithms) in predicting BM in patients with PCa and the correlation between the two techniques.