The Pulse
IQVIA Launches Models That Simulate Clinical Trials
IQVIA introduced Life Science Models at TechIQ 2026, including tools to simulate studies and adjust trial protocols. The company says 85% of predictions in an initial evaluation across 50 trials and 270 endpoints were on target or clinicall
AI.info Team · Written with AI ·
In an initial evaluation spanning 50 trials and 270 endpoints, IQVIA says 85% of its predictions were “on target or clinically close.” The company announced the result on October 7, 2026, alongside the launch of IQVIA Life Science Models, a suite designed to help life-sciences companies simulate clinical studies and test protocol decisions before trials begin.
The figure is an early company-reported result, not a published validation study. IQVIA’s announcement does not detail how it defined a prediction as clinically close, describe the evaluation method, or provide a comparison with other prediction systems. The result offers a first measure of the system’s performance, but leaves outside readers with limited information for judging its reliability across different studies.
Testing protocol choices before enrollment
The models are intended to help sponsors explore decisions that shape a trial, including endpoint selection, patient eligibility criteria, dose selection and sample-size planning. Simulating those choices could help teams spot risks before committing to a study design and enrolling patients. IQVIA also says the models can support digital twins and synthetic controls, and help connect clinical-study predictions with real-world planning and access decisions.
IQVIA says the suite combines its proprietary expert content and domain knowledge with regulatory, compliance and privacy frameworks. The announcement identifies NVIDIA Nemotron among the technologies used and says the models run on Amazon Web Services infrastructure. It does not specify which Nemotron components are involved, how the models are trained, or how customer data will be handled in individual deployments.
“Run the trial before running the trial”
“Our customers can now, essentially, run the trial before running the trial,” said Pete Groves, executive vice president of AI and Technology Solutions at IQVIA. He said the models aim to help customers anticipate outcomes, focus resources and speed the delivery of therapies. The phrase captures the product’s pitch: compare possible study choices through simulation before putting them to the test with participants.
IQVIA describes the launch as an extension of its Predictive Clinical Development approach. The announcement does not name customers using the new models or describe a trial in which a simulation changed a protocol or improved an outcome. Nor does the 85% result establish that a study designed with the models is more likely to succeed; IQVIA presented it as a prediction-performance figure from an initial evaluation.
What the 85% figure leaves unanswered
Clinical-trial outcomes depend on more than protocol choices. Patient recruitment, adherence, site performance and changes in care can all affect whether a study delivers useful results. A model that predicts outcomes well in a retrospective evaluation may not perform as well when sponsors use it to make decisions in new trials, so prospective evidence will matter.
IQVIA’s announcement provides no pricing, access timetable or detail about which customers can use the suite. For now, the clearest numbers are the company’s initial evaluation—50 trials, 270 endpoints and 85% of predictions rated on target or clinically close—and the qualifications that come with it: IQVIA has not provided the methods or independent evidence needed to assess how those predictions translate into better trial decisions.