Lunit has announced findings from a study suggesting that its artificial intelligence (AI) technology may help forecast treatment outcomes for patients with rare cancers receiving pembrolizumab, an immune checkpoint inhibitor.
The study was conducted in partnership with The University of Texas MD Anderson Cancer Center, said Lunit, which provides AI solutions for cancer diagnostics and treatment.
According to Lunit, immunotherapy, particularly immune checkpoint inhibitors such as pembrolizumab, has introduced new approaches to cancer treatment. However, not all patients respond equally, especially in cases involving rare tumour types where both treatment options and data are limited.
Accurately predicting which patients will respond has posed a significant challenge, said the company.
The study, led by Dr. Aung Naing, professor of Investigational Cancer Therapeutics at MD Anderson, utilised Lunit’s AI-powered whole-slide image analysis tool, Lunit SCOPE IO, to assess the tumour microenvironment in biopsy samples taken both before and during treatment from rare tumour patients on pembrolizumab.
The study included more than 500 slides across over 10 rare tumour types.
The findings indicate that Lunit SCOPE IO can identify patterns in tumour samples that are associated with improved responses to immunotherapy.
According to the study, patients whose tumour samples exhibited AI-detected changes in both the density of immune cells within the tumour, referred to as intratumoral tumour-infiltrating lymphocytes (iTIL), and tumour content, were more likely to have favourable treatment outcomes.
The study found that, for certain tumour types, patients with higher pre-treatment iTIL density had a 51% lower risk of disease progression or death, aligning with improved progression-free survival (PFS) with a hazard ratio (HR) of 0.49.
Additionally, patients who exhibited an increase in iTIL density in their on-treatment biopsies had a 35% reduced risk of disease progression or death and a 41% reduction in mortality risk, corresponding with improved overall survival (OS) and an HR of 0.59.
A reduction in tumour content during treatment was also associated with better outcomes, said the firm. Patients with decreased tumour content showed a 49% lower risk of disease progression or death and a 46% reduction in mortality risk, represented by HRs of 0.51 and 0.54, respectively.
Furthermore, patients who exhibited both an increase in iTIL density and a decrease in tumour content showed the most substantial improvement in outcomes. They experienced a 68% reduction in the risk of disease progression or death and a 72% reduction in mortality risk.
Lunit CEO Brandon Suh said: “These findings highlight how our AI technology can provide deep insights into the unique and challenging tumor microenvironment seen in rare cancers and represent a critical advancement in our understanding of rare tumor biology.
“This study has demonstrated the value of Lunit SCOPE IO in an important clinical setting, showcasing its potential to personalise treatment for patients who have limited therapeutic options. We believe these advancements are a testament to the transformative impact AI can have on oncology and patient outcomes.”
