RESUMEN
Background: To identify multiparametric magnetic resonance imaging (mp-MRI)-based radiomics features as prognostic factors in patients with localized prostate cancer after radiotherapy. Methods:From 2011 to 2016, a total of 91 consecutive patients with T1-4N0M0 prostate cancer were identified and divided into two cohorts for an adaptive boosting (Adaboost) model (training cohort: n = 73; test cohort: n = 18). All patients were treated with neoadjuvant endocrine therapy followed by radiotherapy. The optimal feature set, identified through an Inception-Resnet v2 network, consisted of a combination of T1, T2, and diffusion-weighted imaging (DWI) MR series. Through a Wilcoxon sign rank test, a total of 45 distinct signatures were extracted from 1,536 radiomics features and used in our Adaboost model. Results:Among 91 patients, 29 (32%) were classified as biochemical recurrence (BCR) and 62 (68%) as non-BCR. Once trained, the model demonstrated a predictive classification accuracy of 50.0 and 86.1% respectively for BCR and non-BCR groups on our test samples. The overall classification accuracy of the test cohort was 74.1%. The highest classification accuracy was 77.8% between three-fold cross-validation. The areas under the curve (AUC) of receiver operating characteristic curve (ROC) indices for the training and test cohorts were 0.99 and 0.73, respectively. Conclusion:The potential of multiparametric MRI-based radiomics to predict the BCR of localized prostate cancer patients was demonstrated in this manuscript. This analysis provided additional prognostic factors based on routine MR images and holds the potential to contribute to precision medicine and inform treatment management.
RESUMEN
BACKGROUND: Image-guided radiation therapy (IGRT) is the preferred method for curative treatment of localized prostate cancer, which could improve disease outcome and reduce normal tissue toxicity reaction. IGRT using cone-beam computed tomography (CBCT) in combination with volumetric-modulated arc therapy (VMAT) potentially allows smaller treatment margins and dose escalation to the prostate. The aim of this study was to compare the difference of dosimetric diffusion in conventional IGRT using 7-field, step-and-shoot intensity-modulated radiation therapy (IMRT) and hypofractionated IGRT using VMAT for patients with localized prostate cancer. METHODS: We studied 24 patients who received 78 Gy in 39 daily fractions or 70 Gy in 28 daily fractions to their prostate with/without the seminal vesicles using IMRT (n = 12) or VMAT (n = 12) for prostate cancer between November 2013 and October 2015. Image guidance was performed using kilovoltage CBCT scans equipped on the linear accelerator. Offline planning was performed using the daily treatment images registered with simulation computed tomography (CT) images. A total of 212 IMRT plans in conventional cohort and 292 VMAT plans in hypofractionated cohort were enrolled in the study. Dose distributions were recalculated on CBCT images registered with the planning CT scanner. RESULTS: Compared with 7-field, step-and-shoot IMRT, VMAT plans resulted in improved planning target volume (PTV) D95% (7663.17 ± 69.57 cGy vs. 7789.17 ± 131.76 cGy, P < 0.001). VMAT reduced the rectal D25 (P < 0.001), D35 (P < 0.001), and D50 (P < 0.001), bladder V50 (P < 0.001), D25 (P = 0.002), D35 (P = 0.028), and D50 (P = 0.029). However, VMAT did not statistically significantly reduce the rectal V50, compared with 7-field, step-and-shoot IMRT (25.02 ± 5.54% vs. 27.43 ± 8.79%, P = 0.087). CONCLUSIONS: To deliver the hypofractionated radiotherapy in prostate cancer, VMAT significantly increased PTV D95% dose and decreased the dose of radiation delivered to adjacent normal tissues comparing to 7-field, step-and-shoot IMRT. Daily online image-guidance and better management of bladder and rectum could make a more precise treatment delivery.
