The prognostic value of mean apparent diffusion coefficient measured with diffusion-weighted magnetic resonance image in patients with prostate cancer treated with definitive radiotherapy.

• Comprehensive evaluation of the entire prostate is required prior to performing definitive RT. • It is unknown whether pretreatment ADC can be used to predict outcomes following definitive RT. • A total of 503 prostate cancer patients treated with definitive RT were analyzed. • Lower tumor ADC values were found in patients with high-risk characteristics. • Lower ADC values and higher ISUP grades were associated with increased risk of BF and progression. To assess the correlation between initial tumor apparent diffusion coefficient (ADC) values and clinicopathological parameters in prostate cancer (PCa) patients treated with definitive radiotherapy (RT). Additionally, the prognostic factors for freedom from biochemical failure (FFBF) and progression-free survival (PFS) in this patient cohort were analyzed. The clinical data of 503 patients with biopsy-confirmed PCa were evaluated retrospectively. All patients had clearly evident tumors on diffusion-weighted magnetic resonance imaging (DW-MRI) for ADC values. Univariable and multivariable analyses were used to determine prognostic factors for FFBF and PFS. The median follow-up was 72.9 months. The 5-year FFBF and PFS rates were 93.2% and 86.2%, respectively. Significantly lower ADC values were found in patients with a high PSA level; advanced clinical stage; higher ISUP score, and higher risk group than their counterparts. Receiver operating characteristic (ROC) curve analysis revealed an ADC cut-off value of 0.737 × 10−3 mm2/sec for tumor recurrence. Patients who progressed had a lower mean ADC value than those who did not (0.712 ± 0.158 vs. 1.365 ± 0.227 × 10−3 mm2/sec; p < 0.001). There was a significant difference in 5-year FFBF (96.3% vs. 90%; p < 0.001) and PFS rates (83.8% vs. 73.5%; p = 0.002) between patients with higher and lower mean ADC values. The FFBF and PFS were found to be correlated with tumor ADC value and ISUP grades in multivariable analysis. Additionally, older age was found to be a significant predictor of worse PFS. Lower ADC values were found in patients with high-risk characteristics such as a high serum PSA level, stage or grade of tumor, or high-risk disease, implying that ADC values could be used to predict prognosis. Lower ADC values and higher ISUP grades were associated with an increased risk of BF and progression, implying that treatment intensification may be required in these patients. [ABSTRACT FROM AUTHOR]