Big Data

At the 2025 ASCO Annual Meeting, artificial intelligence (AI) moved from theoretical promise to demonstrated clinical utility. Amid the surge of presentations on precision oncology, one theme became increasingly clear: as the number of actionable biomarkers increases, so does the complexity of sequencing treatments. AI may be the only tool agile enough to keep up. The biomarker pool is expanding rapidly. With tumor profiling now yielding dozens of potentially actionable signals, the sequencing question becomes central: Which target do we prioritize? What’s the optimal order of therapies? And how do we make decisions in the absence of head-to-head trials or long-term data?

The purpose of this research was to create and validate radiomic models based on machine learning that can effectively discriminate between primary non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC) in individuals with brain metastases (BMs) by utilizing high-dimensional radiomic characteristics derived from contrast-enhanced T1-weighted imaging (CE-T1WI).

A deep learning model was able to predict future lung cancer risk from a single low-dose chest CT scan, according to new research published at the ATS 2025 International Conference, and in American Journal of Respiratory and Critical Care Medicine. The model, called Sybil, which was originally developed using National Lung Screening Trial (NLST) data by investigators from the Massachusetts Institute of Technology and Harvard Medical School, could be used to guide more personalized lung cancer screening strategies.

María Jesús Ledesma, a researcher at UPM and CIBER-BBN, has developed an artificial intelligence algorithm capable of predicting which lung cancer patients will respond to immunotherapy. The tool is powered by real-world clinical data and routine blood tests, allowing it to predict treatment effectiveness and reduce adverse effects. The trailer, published in Cancer Immunology, Immunotherapy, represents a key step toward more accessible personalized medicine. The project was carried out in collaboration with leading hospitals and biomedical research centers in Spain.

Overdiagnosis in cancer care remains a significant concern, often resulting in unnecessary physical, emotional, and financial burdens on patients. Artificial intelligence (AI) has the potential to address this challenge by enabling more accurate, personalized cancer diagnoses and facilitating tailored treatment plans. Integrating AI with precision medicine can minimize unnecessary treatments and associated adverse effects by optimizing care strategies based on individual patient data. However, the integration of AI in oncology requires rigorous research and validation to ensure its effectiveness across diverse populations and clinical settings.

Según el Informe de EIT Health, aunque existen diferencias regionales, el sistema sanitario español en general ha alcanzado un alto nivel de madurez digital y se clasificó entre los cinco primeros de los 17 países de la Organización para la Cooperación y el Desarrollo Económicos (OCDE). Poseemos también sistemas sanitarios digitales altamente avanzados según el Índice de Salud Digital 2019 de la Fundación Bertelsmann.

La jefa del Laboratorio de Investigación Traslacional de la Fundación MD Anderson Cancer Center Madrid España, la doctora Gema Moreno-Bueno, ha afirmado que el Big Data y la Inteligencia Artificial (IA) permiten anticipar con una mayor precisión el comportamiento del cáncer.

Establishing an accurate prognosis remains challenging in older patients with cancer because of the population's heterogeneity and the current predictive models' reduced ability to capture the complex interactions between oncologic and geriatric predictors. We aim to develop and externally validate a new predictive score (the Geriatric Cancer Scoring System [GCSS]) to refine individualized prognosis for older patients with cancer during the first year after a geriatric assessment (GA).

The diagnostic performance of artificial intelligence (AI)-assisted endoscopy for digestive tumors remains controversial. The objective of this umbrella review was to summarize the comprehensive evidence for the AI-assisted endoscopic diagnosis of digestive system tumors. We grouped the evidence according to the location of each digestive system tumor and performed separate subgroup analyses on the basis of the method of data collection and form of the data. We also compared the diagnostic performance of AI with that of experts and nonexperts. For early digestive system cancer and precancerous lesions, AI showed a high diagnostic performance in capsule endoscopy and esophageal squamous cell carcinoma. Additionally, AI-assisted endoscopic ultrasonography (EUS) had good diagnostic accuracy for pancreatic cancer. In the subgroup analysis, AI had a better diagnostic performance than experts for most digestive system tumors. However, the diagnostic performance of AI using video data requires improvement.

Applying deep learning algorithms to mine ultrasound features of breast cancer and construct a machine learning model that accurately predicts Ki-67 expression level. This multi-center retrospective study analyzed clinical and ultrasound data from 929 breast cancer patients. We integrated deep features from the tumor and peritumoral areas to build a fusion model for predicting Ki-67 expression. The model underwent performance validation on both internal and external test datasets. Its accuracy as well as clinical usefulness were evaluated by diverse statistical metrics.

The integration of artificial intelligence (AI)–powered clinical tools is poised to transform oncology practice, and Scout, a customized large-language model (LLM), AI-powered and expert-trained search tool developed by OncLive®, could provide a streamlined method to aid in treatment decision-making and identifying clinical trial opportunities for patients with cancer, according to Joshua Feinberg, MD.

Glycolysis is recognized as a central metabolic pathway in the neoplastic evolution of gastric cancer, exerting profound effects on the tumor microenvironment and the neoplastic growth trajectory. However, the identification of key glycolytic genes that significantly affect gastric cancer prognosis remains underexplored. In this work, five machine-learning algorithms were used to elucidate the intimate association between the glycolysis-associated gene phosphofructokinase fructose-bisphosphate 3 (PFKFB3) and the prognosis of gastric cancer patients. Validation across multiple independent datasets confirmed the prognostic significance of PFKFB3. Further, we delved into the functional implications of PFKFB3 in modulating immune responses and biological processes within gastric cancer patients, as well as its broader relevance across multiple cancer types. Results underscore the potential of PFKFB3 as a prognostic biomarker and therapeutic target in gastric cancer.