A study led by researchers from Weill Cornell Medicine, the New York Genome Center (NYGC), NewYork-Presbyterian and Memorial Sloan Kettering Cancer Center has revealed that a new AI-powered liquid biopsy technology to detect tumour DNA in blood demonstrated sensitivity in prediction of cancer recurrence.

Utilising ML, the technology could identify circulating tumour DNA (ctDNA) with high accuracy.

This breakthrough could significantly enhance cancer care by detecting recurrence earlier and monitoring tumour response during treatment.

The study involved patients with various cancer types including lung, melanoma, breast and colorectal cancers, as well as precancerous colorectal polyps.

Weill Cornell Medicine hematology and medical oncology division medicine professor Dan Landau said: “We were able to achieve a remarkable signal-to-noise enhancement, and this enabled us, for example, to detect cancer recurrence months or even years before standard clinical methods did so.”

The study introduced a ML strategy, similar to that used by ChatGPT and other AI applications, to discern cancer patterns from sequencing data.

A system, named MRD-EDGE, was trained to recognise patient-specific tumour mutations in 15 colorectal cancer patients.

It successfully predicted residual cancer in nine patients post-surgery and chemotherapy, with subsequent confirmations of recurrence months later using less sensitive methods.

MRD-EDGE could also detect mutant DNA from precancerous colorectal adenomas, indicating its potential in identifying early-stage cancer development.

Landau further added: “It had not been clear that these polyps shed detectable ctDNA, so this is a significant advance that could guide future strategies aimed at detecting premalignant lesions.”

The researchers further showed that the system could monitor responses to immunotherapy in melanoma and lung cancer patients without prior training on patient tumour data.