Purpose Reliable and accurate predictive models are necessary to drive the success of radiomics. Our aim was to identify the optimal radiomics-based machine learning method for isocitrate dehydrogenase (IDH) genotype prediction in diffuse gliomas. Methods Eight classical machine learning methods were evaluated in terms of their stability and performance for pre-operative IDH genotype prediction. A total of 126 patients were enrolled for analysis. Overall, 704 radiomic features extracted from the pre-operative MRI images were analyzed. The patients were randomly assigned to either the training set or the validation set at a ratio of 2:1. Feature selection and classification model training were done using the training set, whereas the predictive performance and stability of the model were independently assessed using the validation set. Results Random Forest (RF) showed high predictive performance (accuracy 0.885 ± 0.041, AUC 0.931 ± 0.036), whereas neural network (NN) (accuracy 0.829 ± 0.064, AUC 0.878 ± 0.052) and flexible discriminant analysis (FDA) (accuracy 0.851 ± 0.049, AUC 0.875 ± 0.057) displayed low predictive performance. With regard to stability, RF also showed high robustness against data perturbation (relative standard deviations, RSD 3.87%). Conclusions RF is a promising machine learning method in predicting IDH genotype. Development of an accurate and reliable model can assist in the initial diagnostic evaluation and treatment planning for diffuse glioma patients.

Thermotherapy has been known to be one of the most effective adjuvants to radiotherapy (RT) in cancer treatment, but it is not widely implemented clinically due to some limitations, such as, inadequate temperature concentrations to the tumor tissue, nonspecific and non-uniform distribution of heat. So we constructed arginine-glycine-aspartate peptides-conjugated gold nanorods (RGD-GNRs) that target the alpha(v) beta(3) Integrin (αvβ3) and investigate whether the photo-thermal effect of RGD-GNRs by near infrared radiation (NIR) could enhance the efficiency of RT in melanoma cancer cells. RGD-GNRs could be seen both on the surface of the cell membranes and cytoplasm of A375 cells with high expression of αvβ3. After exposed to 808 nm NIR, RGD-GNRs with various concentrations could be rapidly heated up. Compared to other treatments, flow cytometric analysis indicated that RT + NIR + RGD-GNRs increased apoptosis (p < 0.001) and decreased the proportion of cells in the more radioresistant S phase (p = 0.014). Treated with NIR + RGD-GNRs, the radiosensitivity was also significantly enhanced (DMFSF2: 1.41). Results of the current study showed the feasibility of using RGD-GNRs for synergetic RT with photo-thermal therapy. And it would greatly benefit the therapeutic effects of refractory or recurrent malignant cancers.

Multifunctional nanoprobes have great potential as effective radiosensitizers and drug carriers. RGD-modified gold nanorods could increase the uptake of nanoparticles via receptor-mediated endocytosis in integrin alphaV beta3-overexpressing breast cancer cells, which could enhance the effects of radiation on tumor cells, leading to further radiosensitization. The purpose of our study was to demonstrate that RGD-conjugated mesoporous silica-encapsulated gold nanorods significantly enhanced the sensitization of triple-negative breast cancer to megavoltage energy. The results indicated that RGD-conjugated mesoporous silica-encapsulated gold nanorod multifunctional nanoprobes could achieve radiosensitization in vitro and in vivo, which suggests the potential translation of this nanotechnology to clinical applications in tumor-targeting and selective therapy.

Peripheral lymphocyte subset variation predicts prostate cancer carbon ion radiotherapy outcomes

Purpose: To explore the value of the pre-treatment MRI radiomic features in individualized prediction of the therapeutic response of carbon ion radiotherapy (CIRT) for prostate cancer patients. Patients and methods: Twenty-three patients with localized prostate cancer treated by CIRT were enrolled for analysis. Prostate tumors were manually delineated on T2-weighted (T2w) images and apparent diffusion coefficient (ADC) maps acquired before CIRT. Abundant radiomic features were extracted from the delineations, which were randomly deformed to account for delineation uncertainty. The robust features were selected and then compared between patient groups of different CIRT responses. Support vector machine (SVM) was subsequently applied to demonstrate the role of the radiomic features to predict individualized CIRT response in the way of artificial intelligence. Results: Radiomic features from ADC had significantly higher intra-correlation coefficient (ICC) (0.71±0.28) than T2w features (0.60±0.31) (p<0.01), indicating higher robustness of ADC features against delineation uncertainty. More features were excellently robust in ADC (58.2% of all the radiomic feature candidates, compared to 41.3% in T2w). By combining the excellently robust radiomic features of T2w and ADC, SVM achieved high performance to predict individualized therapeutic response of CIRT, ie, area-under-curve (AUC) = 0.88. Conclusion: Radiomic features extracted from T2w and ADC images displayed great robustness to quantify the tumor characteristics of prostate cancer and high accuracy to predict the individualized therapeutic response of CIRT. After further validation, the selected radiomic features may become potential imaging biomarkers in the management of prostate cancer through CIRT.

To assess toxicity and quality-of-life (QOL) after carbon ion radiotherapy (CIRT) at the Shanghai Proton and Heavy Ion Center (SPHIC) and identify clinical factors that correlate with urinary, bowel and sexual function. Sixty-four patients with localized prostate cancer admitted from July 2015 to January 2018 underwent CIRT. At baseline and 5 time-points after radiotherapy, we assessed patients’ QOL using the 26-item edition of the Expanded Prostate Cancer Index-Composite (EPIC-26) Chinese version. Logistic regression was performed to identify clinical factors associated with acute genitourinary (GU) toxicity and relative QOL. By the end of CIRT, urinary irritation/obstruction temporarily declined (− 7.92 ± 1.76, p < .001). For urinary incontinence, bowel and sexual QOL, the scores remained stable at 2-year follow-up. The occurrences of acute Grade 1 and 2 GU toxicity were 20.3 and 10.9%, respectively, and of late Grade 1 and 2 GU toxicity were 3.1 and 1.6%, respectively. No acute or late gastrointestinal (GI) toxicity occurred. Transurethral resection of the prostate (TURP) was a risk factor that predicted a decline in urinary related QOL, and age made a difference to bowel-related QOL. For sexual QOL, castration status was a remarkable risk factor. An international prostate symptom score (IPSS) ≥8 increased the risk of Grade 1–2 acute GU toxicity 5.3-fold. Patients with prostate cancer had favorable QOL after CIRT. IPSS ≥8 was a risk factor to acute GU toxicity, and TURP predicted a decline in urinary QOL. Age was related to bowel QOL, and castration status was associated with sexual QOL. Carbon Ion Radiotherapy for the Treatment of Localized Prostate Cancer, NCT02739659 . Registered April 15, 2016.