AI Tool Enhances Response Assessment and Survival Prediction in Pleural Mesothelioma

Clinicians rely on diameter-based Response Evaluation Criteria in Solid Tumors (RECIST), which struggle with this morphology and can obscure true treatment effects. This uncertainty hinders timely therapy changes and undermines trial readouts. To help address this challenge, the Netherlands Cancer Institute has developed an artificial intelligence (AI) model that replaces diameter estimates with volumetric assessment.

Called ARTIMES, the AI-assisted volumetric response criteria for pleural mesothelioma were created at the Netherlands Cancer Institute. The model processes computed tomography (CT) scans to segment the entire tumor and compute total volume. It then compares volumetric change across timepoints to quantify response.

For pleural mesothelioma, which encases the lung rather than forming discrete masses, diameter measurements are inconsistent. ARTIMES avoids this limitation by replacing line measurements with whole-tumor volume, aligning the metric with disease biology. The approach is described as more accurate than RECIST for predicting how long patients will live.

The team developed and validated ARTIMES in a retrospective, multicohort, multicenter study published in The Lancet Oncology on June 17, 2026. Radiologists and clinicians contributed more than 11,000 CT scans from over 2,000 patients treated at 121 hospitals worldwide. In a separate test using data from eight clinical trials, the volumetric measurements were much more accurate than the criteria used today, strengthening assessments of experimental therapy efficacy.

At the Netherlands Cancer Institute, pulmonologists are already using ARTIMES under an in‑house exemption within current European Union regulations. Every AI output is reviewed by a physician, and the team has produced guidance to support treatment decisions based on volumetric change. The institute is seeking broader approvals so other hospitals can adopt the tool, has made the mesothelioma model publicly available to researchers, and is testing related models for lung cancer and brain metastases.

"We can now assess tumor response to treatments much more accurately. We can discover the lack of response much sooner than before. This allows a patient to stop treatment earlier and, if possible, switch to a different treatment. This not only provides certainty but also spares our patients unnecessary side effects and reduces health care costs," said Sjaak Burgers, pulmonologist at the Netherlands Cancer Institute.

"We obviously want patients worldwide to benefit from this. We are in the process of getting the model approved for use in other hospitals. We are also eagerly awaiting a proposal from the EU to simplify the approval process for this type of medical device," said Kevin Groot Lipman, technical physician and lead author at the Netherlands Cancer Institute.

Related Links
Netherlands Cancer Institute