A tool that can sniff out cancer

SCIENCE

A new tool that sniffs out vapours emanating from blood samples can tell if a cell is cancerous or not with up to 95 percent accuracy.

That’s the claim from researchers at the University of Pennsylvania and Penn’s Perelman School of Medicine, who presented their findings at the 2021 American Society of Clinical Oncology meeting in early June.

The Penn-developed tool — which uses artificial intelligence and machine learning to determine the mixture of volatile organic compounds (VOCs) emitting from cells in blood plasma samples — could serve as a non-invasive way to screen for harder-to-detect cancers, such as pancreatic and ovarian.

Lead research author Dr Charlie Johnson says it’s early days, but “the data shows we can identify these tumours at both advanced and the earliest stages, which is exciting. If developed appropriately for the clinical setting, this could potentially be a test that’s done on a standard blood draw that may be part of your annual physical.”

The tool or ‘e-nose’ is equipped with nanosensors that can detect the make-up of VOCs, which all cells release. It is already known that VOCs released from tissue and plasma from ovarian cancer patients are distinct from those from people with benign tumours.

Among 93 patients tested, the vapour sensors picked up the VOCs from ovarian cancer with 95 percent accuracy and pancreatic cancer with 90 percent accuracy. The tool also correctly identified all patients — a total of eight — with early-stage cancers.

The tool works in a similar way to our own sense of smell, where a distinct mixture of compounds tells the brain what it’s smelling. And it takes no more than 20 minutes to tell the difference between cancer cells and normal cells.

The Penn research team is working with VOC Health to commercialise the tool. So far, the partnership has seen an 20-fold improvement in detection speed. Richard Postrel, VOC Health’s CEO and chief innovation officer, says prototypes “able to detect cancer from liquids and vapours will be ready soon and be provided to the Penn researchers to further their work.”

SOURCE