Brainomix AI shows its worth in AZ pulmonary fibrosis trial

Artificial intelligence-powered software developed by Brainomix has been shown to identify patients with idiopathic pulmonary fibrosis (IPF) at risk of decline in a clinical trial of AstraZeneca’s experimental therapy tralokinumab.

The e-Lung software was able to stratify patients at risk of IPF progression more effectively than standard techniques, with implications for the way subjects are enrolled into clinical trials of new therapies, according to a paper on the study published in the American Journal of Respiratory and Critical Care Medicine (AJRCCM).

The technology, which provides automated processing and interpretation of CT scans, was used in a 52-week trial of tralokinumab, an anti-IL-13 antibody that is already on the market as Adtralza for atopic dermatitis.

AZ licensed rights to the antibody in dermatology applications to LEO Pharma, retaining rights to the drug in other indications. It has been tested in asthma, chronic obstructive pulmonary disease (COPD), and ulcerative colitis, as well as IPF - although, those programmes have been discontinued, according to Adis Insight data.

Despite that, the 52-week trial in IPF showed the value of assessing patients using e-Lung, which makes use of a novel biomarker – the weighted reticulovascular score (WRVS) – that assesses reticular opacities and vascular structures of the lung.

“In this analysis […] we have shown that, from a single baseline CT scan, WRVS is able to identify patients at risk of decline in forced vital capacity (FVC) over the next 52 weeks, outperforming standard measures,” commented Dr Peter George, the lead author on the publication, who is a lung disease specialist at the Royal Brompton Hospital and senior medical director of Brainomix.

Enrolling IPF patients who remain stable over the course of a clinical trial is a risk, as progression is necessary to be able to show a treatment effect, and could be one reason why so many drug candidates for IPF fail in pivotal studies, despite early promise.

The data suggests "that the e-Lung WRVS tool may allow for enrichment of clinical trials with progressive patients, could identify patients at low risk of IPF progression, might facilitate well-matched treatment arms, and could reduce the size of future clinical trials,” said Dr George.

The paper acknowledges that the study was relatively small, with 62 patients analysed, and further work is needed to validate the findings.

The study builds on previously published data showing that WRVS was more strongly associated with transplant-free survival of IPF patients than FVC alone, and was also able to predict both short- and long-term outcomes of the patients, including lung function decline and survival.

In 2022, the UK government funded a 12-month study of e-ACT – another Brainomix technology used to interpret CT scans – to look at its potential in diagnosing and tracking lung cancer. The University of Oxford spin-out has also developed e-STROKE, an AI used to improve the diagnosis and treatment of people with suspected stroke based on analysis of CT images.