Researchers present the Virtual Biotech, a multi-agent AI system to inform drug-development decisions. By combining diverse biomedical and clinical evidence in a unified platform, the system may help identify therapeutic opportunities that would otherwise be missed. Drug discovery is a costly process – in both time and resources – yet about 90% of drug candidates entering clinical trials fail, usually due to inadequate efficacy or safety. Although researchers can now draw on vast amounts of genetic, genomic, molecular, and clinical evidence to improve these efforts, these data are fragmented across disciplines, making them difficult to integrate and interpret at the scale required for modern drug development. To help with this, Harrison Zhang and colleagues introduce the Virtual Biotech, a multi-agent AI platform designed to address bottlenecks by coordinating specialized AI “scientist” agents under the direction of a virtual chief scientific officer. According to the authors, unlike earlier AI systems that typically focus on individual biological analyses or rely on a single agent, the Virtual Biotech is designed to retrieve and analyze diverse biomedical and clinical evidence to support early-stage drug development decisions.

Zhang et al. tested the platform at three key points in drug development. Across 55,984 clinical trials, more than 37,000 AI agents found that drugs targeting genes specific to particular cell types were 48% more likely to reach the market and associated with 32% fewer adverse events. The system also combined diverse biological evidence to suggest a therapeutic strategy for lung cancer and analyzed a failed ulcerative colitis trial to identify possible reasons for its failure, demonstrating how human-guided AI teams could provide transparent, large-scale analysis to inform drug-development decisions. However, the authors note that Virtual Biotech’s conclusions are limited by the quality and breadth of available data, are not yet suitable for poorly studied diseases or targets, and still must be experimentally validated. “The Virtual Biotech illustrates a shift from isolated AI tools toward coordinated systems that reason across biological scales and stages of translation,” write the authors. “Rather than replacing scientists, agentic systems could expand the scope and speed of therapeutic hypothesis exploration while making the reasoning process more transparent and reproducible.”

An inherited genetic mutation may increase lung cancer risk among people who have never smoked, with carriers of the rare EGFR T790M variant facing roughly 25 times the risk of lung cancer overall and 62 times the risk among never-smokers, according to a new study. What’s more, the authors traced the occurrence of the mutation to Southern Appalachian populations in the United States, where it likely arrived with British and Irish settlers during the colonial era. Although most lung cancer cases are tobacco-use related, lung cancer in individuals who have never smoked is an increasingly important global health concern – one that will persist even as smoking-related lung cancers continue to decline. Yet the inherited genetic factors that contribute to these cancers remain poorly understood. One important example is the EGFR T790M mutation, which can be inherited and significantly increase the risk of lung adenocarcinoma, particularly when paired with a second, cancer-driving EGFR mutation. However, because this mutation is extremely rare, previous studies have not been able to precisely quantify its prevalence nor the cancer risk it confers.

To obtain more reliable estimates, Jaclyn LoPiccolo and colleagues analyzed genetic data from 3.37 million people of European ancestry to evaluate the association between EGFR T790M and lung cancer. LoPiccolo et al. found that carrying the inherited EGFR T790M mutation was associated with about a 25-fold higher risk of lung cancer, with the association particularly strong among never-smokers, who had roughly a 62-fold higher risk. By contrast, smoking was associated with a roughly 4-fold increase in risk, indicating that T790M was a powerful risk factor for lung cancer, especially in people who have never smoked. Moreover, the authors found that the mutation had no significant links to other types of cancer or respiratory conditions. LoPiccolo et al. also found that EGFR T790M carriers were disproportionately concentrated in the United States’ Southern Appalachian regions, especially in Tennessee and Alabama. This suggests the mutation arose in Europe and was brought to the Southern Appalachian region by British and Irish settlers during the colonial era, where it became more common among the region’s relatively isolated populations. In a related Perspective, Stephen Chanock discusses the study and its findings in greater detail.

An artificial intelligence (AI) model called RADAR, designed to provide broad diagnostic interpretation of abdominal CT scans, substantially outperformed existing AI systems across a wide range of diseases and clinical settings, researchers report. The findings demonstrate the potential of vision-language models to advance expert-level generalist AI tools for radiology. The goal of AI in radiology is to create an expert-level generalist system capable of providing accurate, comprehensive diagnostic support across a wide range of real-world clinical situations. This, however, remains particularly challenging for contrast-enhanced abdominal computed tomography (CT), which often involves the evaluation of dozens of organs and the identification of hundreds of potential diseases, a task that requires physicians with specialized expertise and training. Although vision-language models offer a promising alternative by learning from associations between medical images and their accompanying text, abdominal CT presents unique challenges because relevant diagnostic information is sparse and highly dependent on complex anatomical context.

Here, Qi Zhang and colleagues present Rapid Abdominal Diagnosis with AI and Radiology (RADAR) – a vision-language AI model designed to provide broad diagnostic interpretation of contrast-enhanced abdominal CT scans. Trained on a dataset of 424,911 examinations, containing 1.5 million image-text pairs and more than 15 million anatomy-specific pairs, RADAR breaks CT volumes into individual anatomical structures and links those structures to corresponding descriptions in radiology reports, using a large-scale contrastive-learning approach to improve the connection between imaging findings and diagnoses. Throughout internal and external evaluations across multiple medical centers and varied clinical scenarios, Zhang et al. found that RADAR substantially outperformed existing specialist AI models, achieving a mean AUC (the average probability that a model can correctly separate positive and negative cases across evaluations) of 0.913 across 146 abdominal CT findings, compared with 0.776 for the best competing vision-language model. The model also performed well in challenging emergency settings, despite not being specifically trained on emergency data, achieving an AUC of 0.904 across more than 27,000 emergency CT cases. According to the authors, its performance remained strong across organs, diseases, and emergency cases, including uncommon findings. Moreover, in testing in cohorts at eight external centers, RADAR maintained high accuracy (AUC 0.895), demonstrating robust generalization across diverse patients, clinical settings, and imaging protocols.

Fasting plasma glucose (FPG) and hemoglobin A1c (HbA1c) are the clinical gold standards for screening for blood sugar levels yet they provide only static snapshots that fail to reflect the highly dynamic and heterogeneous nature of metabolic impairment. However, continuous glucose monitoring (CGM) is reshaping our understanding of these levels. While the medical community already has begun capitalizing on CGM for diabetes management, a new study has the potential to change current clinical practices for cardiovascular disease risk stratification in individuals without diabetes.

Researchers at Boston University Chobanian & Avedisian School of Medicine have found that glucose patterns captured by CGM devices among adults without diabetes are related to markers of cardiometabolic risk. The findings, published in Diabetes Care, could ultimately inform future approaches to identifying individuals at greater risk for poor cardiometabolic health.