A study published in Nature Cancer illustrates how advances in artificial intelligence (AI) have paved the way for precision oncology.
AI software developed by Pangea Biomed can predict the likelihood of a patient's response to cancer treatment by analysing tumour histopathology images.
The study demonstrated the capability of Pangea Biomed's ENLIGHT-DP to predict patient responses based on previously unseen tumour samples. The software uses a deep-learning framework called DeepPT to predict gene expression in a tumour from haematoxylin and eosin (H&E) stained slides, eliminating the need for costly RNA sequencing.
Pangea Bio’s ENLIGHT AI uses these predictions to recommend targeted or immune-based treatments. Researchers found that patients predicted by ENLIGHT DEEP-PT to respond to a therapy were two to five times more likely to do so compared to those predicted not to respond. This pattern was consistent across four different treatments and six cancer types within a cohort of 234 patients.
Higher response rate
The study results indicated that the response rate among patients matched to their treatment by ENLIGHT-DP was 39.5% higher than the baseline response rate.
Pangea Bio noted that current methods for predicting treatment responses directly from histopathology slides are often limited by the need for large datasets of matched imaging and response data for each specific treatment, which are frequently unavailable.
This limitation raises concerns about the applicability of these methods for different patient populations and treatment regimens, according to the Tel Aviv, Israel-based start-up. By bypassing the need for extensive datasets, ENLIGHT-DP could enable quicker therapy recommendations, potentially improving patient outcomes.
“These findings highlight the potential of using AI and digital pathology to enhance precision oncology, making advanced cancer treatment predictions more accessible and accurate,” said Pangea Bio chief executive Tuvik Beker. He also noted that this method could offer rapid and precise treatment recommendations, particularly in low- and middle-income countries (LMICs) where access to cancer diagnostics can be limited.
The study was conducted by Pangea in collaboration with researchers from the Australian National University and the US National Cancer Institute.
“Sometimes doctors and patients don’t have that much time – there’s a need to start treatment right away,” Eytan Ruppin, chief of the NCI’s Cancer Data Science Laboratory said.
“ENLIGHT DEEP-PT analysis can be done in any place. Just imagine the impact if indeed it is proven beneficial in prospective studies, as we hope.”