Altis Labs’ artificial intelligence (AI)-powered imaging endpoint has shown its ability to detect early, notable treatment effects more effectively than standard endpoints.
In Johnson & Johnson’s Phase III MAIRPOSA study (NCT04487080), which investigated Rybrevant (amivantamab) and Lazcluze (Lazertinib) in patients with epidermal growth factor receptor (EGFR)-mutated advanced non-small-cell lung cancer (NSCLC), Altis’ AI imaging software analysed approximately 10,000 radiology scans, generating prognostic outcome measures for each patient at baseline and on-treatment assessments.
Go deeper with GlobalData
Discover B2B Marketing That Performs
Combine business intelligence and editorial excellence to reach engaged professionals across 36 leading media platforms.
The software, called IPRO, defined response as a ≥50% improvement in the IPRO-α score relative to baseline, and the treatment effect was estimated as the IPRO Response Rate ratio for the investigational treatment versus control arm treatment at landmark early imaging assessments.
IPRO was deemed to favour the investigational treatment beginning at week 16 and throughout subsequent time points, while Response Evaluation Criteria in Solid Tumours (RECIST)-based objective response rate (ORR) failed to anticipate the significant overall survival (OS) benefit throughout.
When evaluating pooled, patient-level IPRO trajectories over time, IPRO deterioration was consistently associated with OS detriment, while IPRO improvement was consistently associated with OS benefit.
Founder and CEO of Altis Labs, Felix Baldauf-Lenschen, said: “This readout proves that AI can anticipate meaningful clinical benefit that traditional imaging endpoints like ORR may fail to detect.”
Altis Labs presented the data at the World Conference on Lung Cancer (WCLC) 2026 which took place in Seoul, Republic of Korea.
IPRO is a fully automated AI system that predicts patient survival directly from the CT scans already acquired in clinical trials. Rather than automating RECIST, IPRO goes beyond measuring target lesion size by analysing three-dimensional scans. This allows the technology to consider prognostic imaging biomarkers spanning tumour burden, body composition, and organ health that contribute to multifactorial survival outcomes.
By detecting potential treatment failure at earlier periods, a physician will be able to alter the patient’s treatment earlier, potentially impacting survival benefit.
In oncology trials, OS has always been the gold standard endpoint, with the US Food and Drug Administration (FDA) having debuted industry draft guidance last year recommending it as a key pre-specified endpoint in all cancer clinical trials.
