MRI-Derived Tumor Characteristics As Prognostic Indicators in Prostate Cancer With Bone Metastases: A Preliminary Study.

Background While magnetic resonance imaging (MRI) is well-established for local assessment of prostate cancer, its prognostic utility in the metastatic setting remains poorly defined. We investigated whether pretreatment MRI findings can predict survival and disease progression in prostate cancer patients with bone metastases. Methods Of the 2,314 patients with pathologically proven prostate cancer between 2014 and 2021 in our hospital, this retrospective study finally included 19 patients with bone metastases who underwent prostate MRI within six months before diagnosis. Clinical and radiological parameters were analyzed, including Gleason score, prostate-specific antigen (PSA) density, clinical T stage, maximum tumor diameter (MTD), and normalized mean apparent diffusion coefficient (nADCmean). Time-dependent receiver operating characteristic curve analysis was used to determine the optimal cut-points for continuous variables with p-values <0.2. Univariate survival and disease progression analyses were performed using the Kaplan-Meier method and the log-rank test. Univariate Cox proportional hazards regression analyses were performed to estimate the hazard ratio. Results The median age was 69 years (range, 54-91 years). Gleason scores were 4 +4 = 8, 4 + 5 = 9, 5 + 4 = 9, and 5 + 5 = 10 in 12, three, three, and one patients, respectively. Median follow-up duration was 45 months. Seven patients died, and 12 experienced disease progression. MTD ≥49 mm was significantly associated with shorter overall survival compared to MTD <49 mm (median survival, 51 vs 95 months; p = 0.015). A higher nADCmean (≥0.19) was a significant prognostic factor for disease progression (p = 0.031). Gleason score, PSA density, and clinical T stage did not significantly discriminate outcomes in this metastatic cohort. Conclusions In prostate cancer patients with bone metastases, larger maximum tumor diameter on pretreatment MRI predicts poorer overall survival and may be more useful than clinical T stage for prognostic stratification. Higher normalized apparent diffusion coefficient (ADC) was associated with disease progression. Histopathological characteristics that raise ADC despite high Gleason score such as relatively low cell density, heterogeneous morphology, and cribriform pattern might have led to this result. These MRI-derived parameters could help guide treatment planning and risk assessment in metastatic prostate cancer. However, further studies are necessary to validate these findings due to the limitation of small sample size in this study.