Biparametric MRI of the prostate radiomics model for prediction of pelvic lymph node metastasis in prostate cancers : a two-centre study.

Exploring the value of adding correlation analysis (radiomic features (RFs) of pelvic metastatic lymph nodes and primary lesions) to screen RFs of primary lesions in the feature selection process of establishing prediction model. A total of 394 prostate cancer (PCa) patients (263 in the training group, 74 in the internal validation group and 57 in the external validation group) from two tertiary hospitals were included in the study. The cases with pelvic lymph node metastasis (PLNM) positive in the training group were diagnosed by biopsy or MRI with a short-axis diameter ≥ 1.5 cm, PLNM-negative cases in the training group and all cases in validation group were underwent both radical prostatectomy (RP) and extended pelvic lymph node dissection (ePLND). The RFs of PLNM-negative lesion and PLNM-positive tissues including primary lesions and their metastatic lymph nodes (MLNs) in the training group were extracted from T2WI and apparent diffusion coefficient (ADC) map to build the following two models by fivefold cross-validation: the lesion model, established according to the primary lesion RFs selected by t tests and absolute shrinkage and selection operator (LASSO); the lesion-correlation model, established according to the primary lesion RFs selected by Pearson correlation analysis (RFs of primary lesions and their MLNs, correlation coefficient > 0.9), t test and LASSO. Finally, we compared the performance of these two models in predicting PLNM. The AUC and the DeLong test of AUC in the lesion model and lesion-correlation model were as follows: training groups (0.8053, 0.8466, p = 0.0002), internal validation group (0.7321, 0.8268, p = 0.0429), and external validation group (0.6445, 0.7874, p = 0.0431), respectively. The lesion-correlation model established by features of primary tumors correlated with MLNs has more advantages than the lesion model in predicting PLNM.

Li C ，Hu J ，Zhang Z ，Wei C ，Chen T ，Wang X ，Dai Y ，Shen J ... -  《BMC MEDICAL IMAGING》

Prediction of pelvic lymph node metastasis in prostate cancer using radiomics based on T(2)-weighted imaging.

Liu X ，Zhang Y ，Sun Z ，Wang X ，Zhang X ，Wang X ... -  《-》

Preoperative prediction of pelvic lymph nodes metastasis in prostate cancer using an ADC-based radiomics model: comparison with clinical nomograms and PI-RADS assessment.

To develop and test radiomics models based on manually corrected or automatically gained masks on ADC maps for pelvic lymph node metastasis (PLNM) prediction in patients with prostate cancer (PCa). A primary cohort of 474 patients with PCa who underwent prostate mpMRI were retrospectively enrolled for PLNM prediction between January 2017 and January 2020. They were then randomly split into training/validation (n = 332) and test (n = 142) groups for model development and internal testing. Four radiomics models were developed using four masks (manually corrected/automatic prostate gland and PCa lesion segmentation) based on the ADC maps using the primary cohort. Another cohort of 128 patients who underwent radical prostatectomy (RP) with extended pelvic lymph node dissection (ePLND) for PCa was used as the testing cohort between February 2020 and October 2021. The performance of the models was evaluated in terms of discrimination and clinical usefulness using the area under the curve (AUC) and decision curve analysis (DCA). The optimal radiomics model was further compared with Memorial Sloan Kettering Cancer Center (MSKCC) and Briganti 2017 nomograms, and PI-RADS assessment. 17 (13.28%) Patients with PLNM were included in the testing cohort. The radiomics model based on the mask of automatically segmented prostate obtained the highest AUC among the four radiomics models (0.73 vs. 0.63 vs. 0.70 vs. 0.56). Briganti 2017, MSKCC nomograms, and PI-RADS assessment-yielded AUCs of 0.69, 0.71, and 0.70, respectively, and no significant differences were found compared with the optimal radiomics model (P = 0.605-0.955). The radiomics model based on the mask of automatically segmented prostate offers a non-invasive method to predict PLNM for patients with PCa. It shows comparable accuracy to the current MKSCC and Briganti nomograms.

Liu X ，Wang X ，Zhang Y ，Sun Z ，Zhang X ，Wang X ... -  《-》

Multiparametric MRI-based radiomics model to predict pelvic lymph node invasion for patients with prostate cancer.

To identify which patient with prostate cancer (PCa) could safely avoid extended pelvic lymph node dissection (ePLND) by predicting lymph node invasion (LNI), via a radiomics-based machine learning approach. An integrative radiomics model (IRM) was proposed to predict LNI, confirmed by the histopathologic examination, integrating radiomics features, extracted from prostatic index lesion regions on MRI images, and clinical features via SVM. The study cohort comprised 244 PCa patients with MRI and followed by radical prostatectomy (RP) and ePLND within 6 months between 2010 and 2019. The proposed IRM was trained in training/validation set and evaluated in an internal independent testing set. The model's performance was measured by area under the curve (AUC), sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV). AUCs were compared via Delong test with 95% confidence interval (CI), and the rest measurements were compared via chi-squared test or Fisher's exact test. Overall, 17 (10.6%) and 14 (16.7%) patients with LNI were included in training/validation set and testing set, respectively. Shape and first-order radiomics features showed usefulness in building the IRM. The proposed IRM achieved an AUC of 0.915 (95% CI: 0.846-0.984) in the testing set, superior to pre-existing nomograms whose AUCs were from 0.698 to 0.724 (p < 0.05). The proposed IRM could be potentially feasible to predict the risk of having LNI for patients with PCa. With the improved predictability, it could be utilized to assess which patients with PCa could safely avoid ePLND, thus reduce the number of unnecessary ePLND. • The combination of MRI-based radiomics features with clinical information improved the prediction of lymph node invasion, compared with the model using only radiomics features or clinical features. • With improved prediction performance on predicting lymph node invasion, the number of extended pelvic lymph node dissection (ePLND) could be reduced by the proposed integrative radiomics model (IRM), compared with the existing nomograms.

Zheng H ，Miao Q ，Liu Y ，Mirak SA ，Hosseiny M ，Scalzo F ，Raman SS ，Sung K ... -  《-》

PI-RADS version 2: Preoperative role in the detection of normal-sized pelvic lymph node metastasis in prostate cancer.

To analyze whether Prostate Imaging Reporting and Data System (PI-RADSv2) scores are associated with a risk of normal-sized pelvic lymph node metastasis (PLNM) in prostate cancer (PCa). A consecutive series of 221 patients who underwent magnetic resonance imaging and radical prostatectomy with pelvic lymph node dissection (PLND) for PCa were retrospectively analyzed under the approval of institutional review board in our institution. No patients had enlarged (≥0.8cm in short-axis diameter) lymph nodes. Clinical parameters [prostate-specific antigen (PSA), greatest percentage of biopsy core, and percentage of positive cores], and PI-RADSv2 score from two independent readers were analyzed with multivariate logistic regression and receiver operating-characteristic curve for PLNM. Diagnostic performance of PI-RADSv2 and Briganti nomogram was compared. Weighted kappa was investigated regarding PI-RADSv2 scoring. Normal-sized PLNM was found in 9.5% (21/221) of patients. In multivariate analysis, PI-RADSv2 (reader 1, p=0.009; reader 2, p=0.026) and PSA (reader 1, p=0.008; reader 2, p=0.037) were predictive of normal-sized PLNM. Threshold of PI-RADSv2 was a score of 5, where PI-RADSv2 was associated with high sensitivity (reader 1, 95.2% [20/21]; reader 2, 90.5% [19/21]) and negative predictive value (reader 1, 99.2% [124/125]; reader 2, 98.6% [136/138]). However, diagnostic performance of PI-RADSv2 (AUC=0.786-0.788) was significantly lower than that of Briganti nomogram (AUC=0.890) for normal-sized PLNM (p<0.05). The inter-reader agreement was excellent for PI-RADSv2 of 5 or not (weighted kappa=0.804). PI-RADSv2 scores may be associated with the risk of normal-sized PLNM in PCa.

Park SY ，Shin SJ ，Jung DC ，Cho NH ，Choi YD ，Rha KH ，Hong SJ ，Oh YT ... -  《-》