Ultra-high dose rate FLASH Radiotherapy (FLASH-RT) has attracted wide attention because the well-known FLASH effect and the extremely short irradiation time. During FLASH-RT, high radiation doses and dose rate (usually thousands of times compared with conventional radiotherapy (CONV-RT)) are delivered to the tumor area. This novel irradiation technique shows a reduction of normal tissue injury (20-40%) in comparison to CONV-RT. Meanwhile, FLASH-RT maintaining comparable tumor killing effect as CONV-RT. With the progress of basic research on FLASH-RT in reducing radiation-induced injury to normal tissues, clinical trials of FLASH-RT have been carried out across the world. To date, there is no consensus in China focused on the exploration of clinical transformation and application of electron FLASH-RT. Therefore, the China Anti-Cancer Association Radiation Oncology Committee and the Chinese Medical Doctor Association Radiation Oncology Physician Committee gathered a group of experts together to develop this consensus statement. The authors discuss their current views on electron FLASH-RT, demonstrate the unresolved questions, provide insights for the further application of this technology in clinical practice.

Background: We hypothesize generative adversarial networks (GAN) combined with self-attention (SA) and aggregated residual transformations (ResNeXt) perform better than conventional deep learning models in differentiating hepatocellular carcinoma (HCC). Attention modules facilitate concentrating on salient features and suppressing redundancies, while residual transformations can reuse relevant features. Therefore, we aim to propose a GAN+SA+ResNeXt deep learning model to improve HCC prediction accuracy.

Methods: 228 multiphase CTs from 57 patients were retrospectively analyzed with local IRB's approval, where 30 patients were pathologically confirmed with HCC and the rest 27 were non-HCC. Pre-processing of automatic liver segmentation and Hounsfield unit (HU) normalization was performed, followed by deep learning training with five-fold cross validation in a conventional 3D GAN, a 3D GAN+A, and a 3D GAN+A+ ResNeXt, respectively (training: testing ∼ 4:1). Area under receiver operating characteristics curves (AUROC), accuracy, sensitivity and specificity of HCC prediction were evaluated.

Results: Results showed the proposed method had larger AUROC (95%), better accuracy (91%) and sensitivity (93%) with acceptable specificity (88%) and prediction time (0.04s). Deep GAN with attentions and residual transformations for HCC diagnosis using multiphase CT is feasible and favorable with improved accuracy and efficiency, which harbors clinical potentials in differentiating HCC from other benign or malignant liver lesions.

Background: Ki-67 is a key marker of tumor proliferation. This study aimed to develop machine learning models using single- and multi-parameter MRI radiomic features for the preoperative prediction of Ki-67 expression in primary central nervous system lymphoma (PCNSL), aiding prognosis and individualized treatment planning.

Methods: A retrospective analysis of 74 patients was conducted using MRI scans, including T1, contrast-enhanced T1, T2, T2-FLAIR, DWI, and ADC sequences. Patients were categorized into high-expression (Ki-67 > 70%) and low-expression (Ki-67 ≤ 70%) groups. Tumor volumes of interest (VOIs) were manually delineated by radiologists, and 851 radiomic features were extracted using 3DSlicer. After preprocessing, including bias field correction and normalization, feature selection was performed using SelectKBest and ANOVA. Eight machine learning classifiers, including Logistic Regression, Random Forest, and SVM, were applied to single- and multi-parameter datasets.

Results: Multiparameter models, particularly Naive Bayes and Logistic Regression, demonstrated superior predictive performance (AUC: 0.78, 0.73; AP: 0.90, 0.83) compared to single-parameter models. Decision curve analysis highlighted that Logistic Regression provides the highest net benefit, followed by Naive Bayes.

Conclusion: Multiparameter MRI models are more accurate and stable for predicting Ki-67 expression in PCNSL, supporting clinical decision-making.

Purpose: The response of various detectors in small fields from a variety of treatment machines has been studied and is summarized in IAEA TRS-483. However, data for the novel RefleXion system remains largely unexplored. This study measured the output correction factors of multiple detectors in small fields for a clinical RefleXion unit.

Methods: The RefleXion machine consists of a binary multi-leaf collimator and two pairs of Y-jaws with clinical openings of 1 and 2 cm. The reference dosimetry is applied to the 10 × 2 cm² clinical-reference field, and the output factors of different fields are presented relative to the clinical-reference field. The responses of detectors Edge, Razor, Micro-Diamond, A14SL, CC01 and CC03 in rectangular fields from 1.25 × 1 to 20 × 2 cm² on the RefleXion unit was studied at a depth of 10 cm in an IBA Blue-Phantom-Helix with a 85 cm source-to-surface distance. Gafchromic EBT4 film data in a solid-water phantom were used as the reference to obtain correction factors for the detectors.

Results: In the fields of the 2 cm jaw, all 6 detectors showed similar responses to the film reference within around 0.5% except at the first field width (1.25 cm), where the Edge and Micro-Diamond exhibited over-response and the CC13 showed the volume effect of ion chambers. In the fields of the 1 cm jaw, the Edge and Micro-Diamond had responses close to the film and the same over-response at small field-widths. Significant deviations of the CC13 (∼4%) and the A14SL (∼2.5%) from the film were present over the whole range of field widths.

Conclusions: The small field output correction factors of 6 kinds of detectors were determined for a RefleXion system, conforming to the formalism in TRS-483. All detectors except CC13 fulfil the 5% correction limit recommended by the TRS-483 for output factor measurement.

Purpose: To assess the efficacy of clinical radiomics models in predicting microsatellite instability-high status in endometrial cancer and to identify patients who may benefit from immunotherapy.

Materials and Methods: Two hundred and twenty-two patients with endometrial cancer who were consecutively admitted to Yantai Yuhuangding Hospital between January 2021 and April 2022 were retrospectively recruited, and 64 were excluded. Of the remaining 158 patients, 110 and 48 were randomly divided into the training and test sets, respectively. Regions of interest were delineated, and radiomic features were extracted from dynamic contrast-enhanced T1-weighted, T2-weighted, and apparent diffusion coefficient images. The intraclass correlation coefficients, Spearman correlation analysis, Mann–Whitney U test, and least absolute shrinkage and selection operator (LASSO) algorithm were employed for feature selection in radiomics models' development. Seven clinical risk factors were incorporated into the clinical models. Finally, the clinical-radiomics models integrating clinical risk factors and radiomic features were constructed. Clinical, radiomics, and clinical-radiomics models were developed in the training set using logistic regression (LR), random forest (RF), and support vector machine (SVM). The performance of the models was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analyses (DCA).

Results: Four clinical factors (progesterone receptor, tumor suppressor gene p53, diabetes, and carbohydrate antigen 153) and 15 radiomic features were recognized as key predictors of microsatellite instability-high status in endometrial cancer. The clinical-radiomics models developed using the SVM classifier exhibited the best performance in the test set, achieving an area under the curve (AUC) of 0.997, sensitivity of 1.000, specificity of 0.936, and accuracy of 0.952. DCA demonstrated that the SVM-based clinical-radiomics model achieved a greater net clinical benefit than the clinical and radiomics models across threshold probabilities ranging from 0 to 0.405 and 0.585 to 1, respectively.

Conclusion: The clinical-radiomics nomogram constructed using the SVM classifier exhibited robust predictive performance for microsatellite instability-high status in endometrial cancer. This nomogram may assist in identifying patients with endometrial cancer who are likely to benefit from immunotherapy, thereby providing a tool for personalized immune management.

Purpose: Colorectal cancer (CRC) is one of the leading causes of cancer-related death worldwide. Dihydroartemisinin (DHA) is an anti-malaria agent and recent evidence indicates a broad anti-cancer activity. Radiotherapy (RT) is the cornerstone of cancer treatment and often has synergistic effects when combined with chemotherapy or targeted therapy. We aim to investigate whether synergistic effect exists for RT combined with DHA in a preliminary animal study of CRC treatment.

Methods: Twenty-four BALB/c nude mice were subcutaneously injected with 5.0 × 10⁶ cells of the murine CRC cell line CT26. When tumor xenografts were formed (∼ 100 mm³), mice were randomly allocated into four groups (n = 6 per group) and the tumor-bearing mice were intra-abdominally injected at Day 7, 9, 11 with PBS (control), 6Gy irradiation (RT), DHA (50mg/Kg), and DHA with irradiation (DHA+RT). All RT was performed on a medical linear accelerator using collimated anterior posterior 6MV photon beam conformal to tumor xenografts. The tumor volume was measured using an electronic caliper and was calculated based on the length (L) and the width of the tumor using the formula V = (L x W²)/2. Tumor weight was also measured after mice sacrificed. Histological assay was conducted, including using gammaH2AX for DNA double strand breaks (DSB) analysis.

Results: The tumor weight was 2.4±0.8g, 2.5±0.6g, 0.4±0.2g, and 0.4±0.1g for the control, DHA, RT, and DHA+RT groups, respectively. Significant difference was observed between the control and RT groups, and between the control and DHA+RT groups (p<0.05). However, there was almost no difference between the RT alone and DHA+RT groups. The longitudinal change in tumor volume showed tumor progression inhibition in the RT and DHA+RT groups, but not so obvious in the DHA group, which was consistent with histological assay.

Conclusion: Similar treatment efficacy is observed in the RT alone and concurrent DHA+RT group. No significant difference in tumor volume or weight or tumor progression inhibition is observed between the RT alone and concurrent DHA+RT groups, demonstrating that DHA might not provide synergistic effect with RT for the proposed hypofractionated radiation dose regimen.

The approach to prophylactic neck irradiation in nasopharyngeal carcinoma (NPC) has undergone significant changes. For decades, prophylactic whole-neck irradiation has been the standard for all patients with NPC; however, it is linked to a high risk of late complications. Advances in imaging modalities have deepened understanding of the metastatic characteristics of the cervical lymph nodes (LN), prompting a shift towards sparing the uninvolved lower neck and medial retropharyngeal nodal region. This targeted approach has proven effective in controlling cervical LN recurrence as whole-neck irradiation while significantly reducing adverse effects. Currently, contouring of the neck lymphatic drainage clinical target volume (CTV) is being explored to eliminate the use of uniform cervical LN levels as a delineation boundary. Instead, the inferior boundary of the neck CTV is determined either by the distance from the lowest positive LN or two cervical vertebrae below the lowest positive LN, facilitating more individualized CTV delineation and prophylactic neck irradiation. Additionally, omitting lower-risk neck lymphatic drainage CTVs and irradiating only visible LNs in these areas are also being explored. This review examines the evolution of prophylactic neck irradiation for NPC, providing key insights into these advancements.