Here's how a 'little saliva' could detect Stage 1 cancer

Researchers believe that analyzing the sugar in people's saliva could — in the future — help them detect Stage 1 cancers.

Scientists at the University of Gothenburg in Sweden have developed a method for detecting the changes in sugar molecules using artificial intelligence (AI) that could indicate cancer.

• Your health headlines all here

The molecules are called glycans, and they are linked to the protein in human cells, a summary of a small-scale study published on Wednesday in Cell Report Methods reads. These molecules can be collected through a simple saliva swab.

"We have analyzed data from about 220 patients with 11 differently diagnosed cancers and have identified differences in the substructure of the glycan depending on the type of cancer," Daniel Bojar, associate senior lecturer in bioinformatics at the University of Gothenburg and lead author of the study, said in the press release.

Glycans have been used to indicate inflammation or disease, the press release says.

Researchers say AI can find patterns in what glycans look like when a patient has cancer. When the "structure" of the glycan is linked to a form of cancer it could provide a "precise answer" on the health of the person.

"By letting our newly developed method, enhanced by AI, work through large amounts of data, we were able to find these connections," Bojar said.

Analyzing the changes of the glycan's structure and seeing the patterns associated with different conditions is a new method of research into the detection of cancer, the press release said.

Researchers said they were able to ensure the type of cancer and glycan structure were consistent with each test by including AI.

Previous research using glycans looked at whether the level of sugar was higher or lower, but this is "not reliable," according to the study.

"We can rely on our results; they are statistically significant. If we know what we are looking for, it is easier to find the correct result," Bojar said.

With a 4 million Swedish krona (C$519,304) grant from the Lundberg Foundation, the team hopes it can develop a faster method to detect cancer itself and the type through saliva or a blood sample.

Bojar said the team might be able to perform clinical trials on human samples in four to five years.