Independent external validation of predictive models for urinary dysfunction following external beam radiotherapy of the prostate: Issues in model development and reporting

External Validation of a Predictive Model of Urethral Strictures for Prostate Patients Treated With HDR Brachytherapy Boost

“…The fitted model parameters ( Table 2) well-represented the AHRO observed data, as shown by the calibration plot ( Figure 2B), and internal calibration bootstrapped results (Figure 3). The AUC was of the order of 0.64, which is of the order of values obtained for most models based on dose features alone (42). An outlier was observed in the 80 Gy EUD dose group, believed to be associated with the little variability of urethral doses for patients in this dose group (Figures 2A,B).…”

External validation of an NTCP model for acute esophageal toxicity in locally advanced NSCLC patients treated with intensity-modulated (chemo-)radiotherapy

“…This model was internally validated and the area under the receiver operating curve (AUC) was 0.84 (0.82 after correction for optimism) indicating good discriminative power of the model. Nonetheless, as reproducibility (model performance on new samples from the same target population), and transportability (model performance on samples from different but related populations) of well internally validated prediction models can still be poor, external validation is needed to assess 'generalizability' of the NTCP model to external patient cohorts [21][22][23][24].…”

Multi-atlas-based segmentation of prostatic urethra from planning CT imaging to quantify dose distribution in prostate cancer radiotherapy

“…Urinary toxicity is a challenging issue, not only due to the variety of associated irritating or obstructive symptoms, but also owing to the limitations of dose descriptors and difficulties identifying the regions at risk responsible for those symptoms [4][5][6]. The bladder, for example, presents the largest inter-fraction shape variations, causing geometric and dose uncertainties that limit the possibility of accurately modeling the dose-volume response concerning GU toxicity [4,7,8,9 ].…”