An optical biopsy needle to diagnose cancer

Frédéric Leblond et Joannie Desroches
Performing a biopsy is sometimes like going fishing. It’s not always easy to find the centre of the tumour to remove a small amount of tissue and then analyze it under a microscope to check whether it contains cancer cells. A technological development created by engineers Frédéric Leblond and Joannie Desroches promises to revolutionize biopsies in the future and improve the diagnosis and treatment of cancer.

“We have developed an optical biopsy needle that is able to immediately detect cancerous tissue with a diagnostic accuracy of 84% for brain cancer,” explained Leblond, a researcher at the University of Montreal Hospital Research Centre (CRCHUM) and a professor in the Department of Engineering Physics at Polytechnique Montréal.Aiguille

In a recent article in Scientific Reports, a Nature Publishing Group journal, the scientists present the results of the first trials carried out with this tool designed in collaboration with neurosurgeon Kevin Petrecca at the Montreal Neurological Institute and Hospital and the company Medtronic.

“Using a commercial biopsy needle with a built-in Raman detection system, optical measurements can be taken when the needle is inserted in the brain that will guarantee that the surgeon removes a tissue sample that truly represents the centre of the tumour, not healthy peripheral tissue. Our trials with animals and about ten patients show that we are detecting cancerous tissue containing more than 60% cancer cells with an excellent degree of accuracy,” explained Desroches, a doctoral student working in Frédéric Leblond’s laboratory and the study’s first author.

“We think that this approach will reduce the risks associated with surgical procedures and one day possibly eliminate the need to extract a tissue sample. We could potentially determine the type and grade of the tumour simply by probing the tissue at the tip of the needle,” asserted Leblond.

Currently, a brain biopsy is done using magnetic resonance imaging (MRI). Preoperative images are used to guide the surgeon, who removes a tissue sample and sends it to pathology to validate the diagnosis, while the patient waits on the operating table for about twenty minutes. But to get an exact picture of the cancer in question, its extent and the stage of the disease, it’s important to aim accurately. Sometimes the surgeon misses the mark and removes healthy tissue.

Using the method developed by Leblond’s team, there’s no more guesswork. Their method uses Raman spectroscopy technology, which detects the light scattered by cells. Since cancer cells and normal cells react differently to the light emitted, they display a different molecular signature.

The engineers are now exploring the possibility of using artificial intelligence to perfect their diagnostic tool. “We are building a database to compare the results obtained with the probe to pathology tests. The goal is to be able to guide the needle to the optimal spot in order to accurately diagnose the tumour. It’s an ambitious project, as we need hundreds of pieces of data for each type of tumour,” explained Desroches.

The researchers hope that this new diagnostic tool will be able to go to market within a five- to ten-year time frame. “If the results are conclusive, this approach will allow surgeons to make diagnoses that are faster, safer, more economical and more precise. And, most importantly, we will be better able to predict the behaviour of cancer cells, which will allow us to choose the appropriate follow-up and treatment options,” concluded Leblond.