Toronto, ON | Barcelona, Spain – Altis Labs, Inc. (“Altis”) is announcing results presented at this year’s European Society of Medical Oncologists conference: IPRO predicted overall survival from baseline CT scans in a phase III aNSCLC clinical trial dataset, and shows potential for enhancing treatment effect quantification beyond RECIST-based SLD.
This study validated the ability of IPRO to predict overall survival in a novel aNSCLC dataset. Altis’ fully-automated AI pipeline was applied to a dataset of 452 stage IIIb and IV NSCLC subjects from Bayer Pharmaceutical’s “NExUS” trial (NCT00449033).
IPRO significantly stratified Overall Survival (OS) across terciles, and had greater separation of curves than SLD, the sum of longest diameters, which underpins many common surrogate endpoints such as Progression Free Survival (PFS). Additionally, IPRO improves estimation of OS compared to using SLD based on c-index and standardized HRs.
This work highlights that IPRO may provide greater utility than SLD as a prognostic factor in future clinical research, and provides increasing power to assessment of clinically relevant outcomes. Ongoing research will evaluate IPRO’s ability to detect treatment effect based on longitudinal imaging.
Dr. Omar F. Khan, a medical oncologist from Tom Baker Cancer Center, University of Calgary, was the lead author of the study.
Read the full abstract here.
About Altis Labs
Altis Labs is the computational imaging company accelerating clinical trials with AI.
Altis has created the industry's largest real-world Imaging, Clinical, and Outcomes database – comprising over 210 million cancer patient images linked to associated demographic, diagnostic, molecular, treatment, and survival information. Deep learning models trained on rwICO can generate profound insights beyond tumor measurements that make up reductionist criteria like RECIST 1.1, yielding greater association with overall survival.
Top 20 biopharma sponsors apply our deep learning models to their clinical trial imaging data to unlock powerful insights. Our models instantaneously generate Imaging-based Prognostication (IPRO) scores and thousands of Spatial Imaging Biomarkers (SIBs) from each scan, which Sponsors can combine for analysis with other data modalities such as ctDNA. These insights are being explored by clinical development teams to quantify treatment effects earlier in order to design more successful clinical trials.
Our multi-disciplinary team is on a mission to help get the most effective novel treatments to patients sooner. Altis is proudly headquartered in Toronto, Canada.
To learn more, email info@altislabs.com.