Accurate overall survival (OS) prediction in non-small cell lung cancer (NSCLC) is crucial but challenging due to high-dimensional 3D computed tomography (CT) data, limited annotations, and time-to-event outcomes. Traditional 3D CNNs are computationally expensive and prone to overfitting on small datasets. We propose a lightweight framework that aggregates 2D CT slice embeddings via soft attention to form a 3D patient representation. In our approach, features are extracted with EfficientNetB0, and DeepHit models time-to-event survival. Validated on LUNG1 (415 patients), our method outperforms 3D ResNet (+0.077) and alternative aggregation strategies (+0.005) in time-dependent concordance index (Ctd-index). Furthermore, transfer learning from LUNG1 improves performance on the small private CLARO dataset (0.579 vs 0.503). This shows that 2D CNNs with soft attention provide a computationally efficient yet effective alternative to 3D CNN architectures for NSCLC OS prediction, with a substantially lower computational cost (54.3 GFLOPs vs. 2924.6 GFLOPs for ResNet3D-18).

Background:Adjuvant radiotherapy can be considered the standard of care for patients diagnosed with primary breast cancer who have undergone breast-conserving surgery. Left breast cancer receiving adjuvant deep inspiration breath hold (DIBH) were evaluated. A technique for monitoring patient position during DIBH is the surface guided radiation therapy (SGRT) as Vision AlignRT. The aim of this study was to estimate setup accuracy with AlignRT compared to AlignRT Advance in a cohort of patients undergoing DIBH treatment of breast cancer. Materials and methods:One hundred and fifty-four patients were studied. Patient positioning was informed using SGRT and verified by cone beam computed tomography (CBCT). Patients were divided into groups (G) based on treatment volume and version of Vision AlignRT. Comparison was performed by two tailed t-test. Data were deemed as statistically significant if p < 0.001. Results:The setup error for all patients treated with DIBH radiotherapy (RT) was calculated by CBCT. For left breast treatments, the interfractional displacement in the longitudinal and vertical directions was significantly (p < 0.001) reduced in patients positioned with the advanced version. For breast and supraclavicular lymph node chain and/or internal mammary chain treatments, interfractional displacement in the lateral and vertical directions was significantly (p < 0.001) reduced when the advanced version was used, while for longitudinal displacement the reduction was not significant. Conclusions:Positioning with AlignRT Advance is more accurate for either left breast only or left breast and supraclavicular lymph node chain and/or internal mammary chain treatments.

BACKGROUND:Stereotactic body radiotherapy (SBRT) has gained increasing interest in the treatment of locally advanced pancreatic cancer (LAPC), although its effectiveness has not been defined in randomized trials. This systematic review and meta-analysis aimed to compare clinical outcomes and treatment-related toxicity between SBRT and CFRT in LAPC. METHODS:This analysis was performed in accordance with PRISMA guidelines (PROSPERO: CRD420251128943). MEDLINE and Scopus were searched for comparative studies published between January 2015 and July 2025. Five retrospective studies comprising 768 patients fulfilled the eligibility criteria. Pooled hazard ratios (HRs) were calculated for overall survival (OS) and progression-free survival (PFS), while risk ratios (RRs) were estimated for severe (grade ≥ 3) acute toxicity using random-effects models. Study quality was evaluated using the ROBINS-I tool. RESULTS:No significant OS or PFS differences were observed between SBRT and CFRT. SBRT was associated with a lower incidence of severe acute toxicity. The overall risk of bias across studies was moderate. CONCLUSIONS:SBRT appears to achieve survival outcomes comparable to CFRT with a favorable acute toxicity profile in patients with LAPC. Nevertheless, the current evidence is limited by retrospective designs and heterogeneity, highlighting the need for prospective randomized trials to define the role of SBRT in this setting.

Accurate prognosis of Non-Small Cell Lung Cancer (NSCLC) is crucial for enhancing patient care and treatment outcomes. Despite the advancements in deep learning, the task of overall survival prediction in NSCLC has not fully leveraged these techniques, yet. This study introduces a novel methodology for predicting 2-year overall survival (OS) in NSCLC patients using CT scans. Our approach integrates CT scan representations produced by EfficientNetB0 with a soft attention mechanism to identify the most relevant slices for survival risk prediction, which are then analyzed by a risk-assessment network. To validate our method and ensure reproducibility, we employed the public LUNG1 dataset and a smaller private dataset. Our approach was compared to benchmark 3D networks and two variants of our methodology: on the LUNG1 it outperformed the competitors achieving a mean C^td -index of 0.584 over tenfold cross-validation. On the LUNG1 we also demonstrated the adaptability of our method with 5 other 2D backbones replacing the EfficientNetB0, confirming that our mechanism of combining 2D slice representations to construct a 3D volume representation is more effective for OS prediction compared to a traditional 3D approach. Finally, we used transfer learning on the private dataset, showing that it can significantly enhance performance in limited data scenarios, increasing the C^td -index by 0.076 compared to model without transfer learning.

Predicting tumor evolution during radiotherapy is a clinically critical challenge, particularly when longitudinal changes are driven by both anatomy and treatment. In this work, we introduce a Virtual Treatment (VT) framework that formulates non-small cell lung cancer (NSCLC) progression as a dose-aware multimodal conditional image-to-image translation problem. Given a CT scan, baseline clinical variables, and a specified radiation dose increment, VT aims to synthesize plausible follow-up CT images reflecting treatment-induced anatomical changes. We evaluate the proposed framework on a longitudinal dataset of 222 stage III NSCLC patients, comprising 895 CT scans acquired during radiotherapy under irregular clinical schedules. The generative process is conditioned on delivered dose increments together with demographic and tumor-related clinical variables. Representative GAN-based and diffusion-based models are benchmarked across 2D and 2.5D configurations. Quantitative and qualitative results indicate that diffusion-based models benefit more consistently from multimodal, dose-aware conditioning and produce more stable and anatomically plausible tumor evolution trajectories than GAN-based baselines, supporting the potential of VT as a tool for in-silico treatment monitoring and adaptive radiotherapy research in NSCLC.

Accurate survival prediction in non-small cell lung cancer (NSCLC) requires integrating clinical, radiological, and histopathological data. Multimodal deep learning (MDL) can improve precision prognosis, but small cohorts and missing modalities limit its clinical applicability, as conventional approaches enforce complete-case filtering or imputation. We present a missing-aware multimodal survival framework that combines computed tomography (CT), whole-slide histopathology images (WSI), and structured clinical variables for overall survival modeling in unresectable stage II-III NSCLC. The framework uses foundation models (FMs) for modality-specific feature extraction and a missing-aware encoding strategy that enables intermediate multimodal fusion under naturally incomplete modality profiles. By design, the architecture processes all available data without dropping patients during training or inference. Intermediate fusion outperforms unimodal baselines and both early and late fusion strategies, with the trimodal configuration reaching a C-index of 74.42. Modality-importance analyses show that the fusion model adapts its reliance on each data stream according to representation informativeness, shaped by the alignment between FM pretraining objectives and the survival task. The learned risk scores produce clinically meaningful stratification of disease progression and metastatic risk, with statistically significant log-rank tests across all modality combinations, supporting the translational relevance of the proposed framework.

Major pathological response (pR) following neoadjuvant therapy is a clinically meaningful endpoint in non-small cell lung cancer, strongly associated with improved survival. However, accurate preoperative prediction of pR remains challenging, particularly in real-world clinical settings characterized by limited data availability and incomplete clinical profiles. In this study, we propose a multimodal deep learning framework designed to address these constraints by integrating foundation model-based CT feature extraction with a missing-aware architecture for clinical variables. This approach enables robust learning from small cohorts while explicitly modeling missing clinical information, without relying on conventional imputation strategies. A weighted fusion mechanism is employed to leverage the complementary contributions of imaging and clinical modalities, yielding a multimodal model that consistently outperforms both unimodal imaging and clinical baselines. These findings underscore the added value of integrating heterogeneous data sources and highlight the potential of multimodal, missing-aware systems to support pR prediction under realistic clinical conditions.

Predicting the longitudinal evolution of tumors during radiotherapy is a complex and clinically critical challenge in medical imaging analysis, especially when models are conditioned solely on imaging data. This work introduces a Virtual Treatment (VT) framework that formulates non-small cell lung cancer (NSCLC) tumor progression as a multimodal conditional image-to-image translation paradigm, where the generative process is driven not only by past CT scans but also by treatment-related and patient-specific clinical information. Using a private longitudinal dataset of 222 NSCLC patients with 895 CT scans acquired during radiotherapy, it is shown that incorporating diverse modalities such as delivered dose increments, demographic characteristics, histological diagnosis, and tumor staging (cT, cN) improves the ability of generative models to predict plausible future anatomical states. A benchmarking study is conducted across four families of generative models including 2D GANs, 2.5D diffusion models, and fully 3D latent diffusion architectures, and showing that multimodal conditioning is essential for capturing patient-specific radiobiological dynamics. Unlike traditional image-based synthesis methods, the proposed VT framework leverages these complementary modalities to refine the generative trajectory, enabling the model to forecast future CT scans based on both anatomical evolution and clinical context, thereby reflecting realistic treatment-response patterns. Across quantitative metrics, tumor volumetric evaluation, and statistical tests, diffusion-based models conditioned on the combination of demographic and tumor-related features achieved the most stable and dose-aware tumor evolution forecasts. These findings underscore the importance of multimodal information for generative modeling in longitudinal oncology and position VT as a promising tool for in-silico treatment monitoring and adaptive radiotherapy support in NSCLC.

Background/Objectives: Early progression (EP) occurs in a subset of patients with locally advanced pancreatic cancer (LAPC), limiting the clinical benefit of treatment, and it remains difficult to predict. Methods: We developed a multiparametric predictive model integrating baseline 18F-FDG PET/CT radiomic features with clinical and biological data. A total of 242 radiomic features were extracted from each imaging modality (CT and PET), including first-order, gray-level co-occurrence matrix (GLCM), and local binary pattern (LBP-TOP) features, and combined with PET-derived metrics and clinical variables. Model development included cross-validation procedures and rigorous feature selection, followed by the training of a two-level decision tree classifier. Results: The model achieved an accuracy of 80.7% and an area under the curve (AUC) of 0.83. Integrated analysis of CT and PET texture enabled the identification of patients at high risk of EP prior to treatment initiation. Conclusions: PET/CT-based radiomic biomarkers, combined with clinical data, can non-invasively capture tumor heterogeneity and improve risk stratification in LAPC, supporting more personalized therapeutic decision-making.

Aims:The most appropriate management of patients who have undergone curettage for a suspected low-grade chondrosarcoma (CS), which has subsequently been found to be of grade 2, remains unknown. We aimed to assess whether these patients have an increased risk of local recurrence and distant metastasis if followed up over time, compared to those who undergo further treatment soon after the diagnosis has been established. Methods:A retrospective study was undertaken which included 71 patients treated between January 2010 and December 2022 by intralesional curettage for a supposed low-grade CS, but who subsequently proved to have a histological grade 2 CS. Thereafter, patients either underwent further surgery (resection group) or follow-up (follow-up group). Results:The estimated local recurrence rate was 36.8% at five (95% CI 35.6 to 38.0) and 48.1% at ten years (95% CI 46.4 to 49.8), and was significantly higher in the follow-up group (48.4%, 95% CI 45.4 to 51.4) than in the resection group (9.6%, 95% CI 8.1 to 11.1) at five years (p = 0.005). Locally recurrent CS, considered as a time-dependent covariate, had an increased risk of metastasis (36.8% vs 2.5% at five years; p < 0.001) and a worse disease-specific survival (81.3% vs 100% at five years; p = 0.007). Conclusion:The optimal treatment strategy should be individualized based on the histological features of the tumour, tumour location, morbidity of resection, and patient-specific factors. We recommend that patients who have undergone unplanned surgery be treated in the standard manner. Observation may be appropriate in specific cases with a properly informed patient.

Backgrounds and aim: Protective effects of natural compounds have been suggested in the prevention and treatment of radiation-induced mucositis or bacterial infections. In this study, the protective effects of proanthocyanidin-rich grape seed extract (GSE) on bacterial Lipopolysaccharide (LPS) and radiation-induced epithelial barrier damage and Reactive Oxygen Species (ROS) production were investigated in an in vitro model. Methods: Human intestinal epithelial cells Caco-2, previously treated with LPS, GSE, or LPS + GSE, were irradiated with 10 Gy divided into five daily treatments. Epithelial barrier integrity and ROS production were measured before and after each treatment. Results: Irradiation, at different doses, significantly increased intestinal permeability and ROS production; pretreatment with GSE was able to significantly prevent the increased intestinal permeability (4.63 ± 0.76 vs. 15.04 ± 1.5; p < 0.05) and ROS production (12.9 ± 1.08 vs. 1048 ± 0.5; p < 0.0001) induced by irradiation treatment. When the cells were pretreated with LPS, the same results were observed: GSE cotreatment was responsible for preventing permeability alterations (5.36 ± 0.16 vs. 49.26 ± 0.82; p < 0.05) and ROS production (349 ± 1 vs. 7897.67 ± 1.53; p < 0.0001) induced by LPS exposure when added to the irradiation treatment. Conclusions: The results of the present investigation demonstrated, in an in vitro model, that GSE prevents the damage to intestinal permeability and the production of ROS that are induced by LPS and ionizing radiation, suggesting a potential protective effect of this extract on the intestinal mucosa during irradiation treatment.