New ‘Google maps’ approach to revolutionise lung cancer treatment
Non-small cell lung cancer tissue mapped cell-by-cell through multiplex spatial imaging.
(Photo credit: Dr James Monkman and A/Prof Arutha Kulasinghe )
Key points
- Researchers have developed a way to predict how the most common form of lung cancer will respond to different therapies.
- Non-small cell lung cancer was mapped cell-by-cell using spatial biology and AI, taking the guess work out of drug treatment.
- The same approach could be used to inform treatments for other malignancies like melanoma, head and neck, and bladder cancer.
Associate Professor Arutha Kulasinghe from UQ’s Frazer Institute in collaboration with researchers at Yale University studied the tumours of 234 patients with non-small cell lung cancer (NSCLC) across 3 cohorts in Australia, the United States and Europe.
“Using AI and spatial biology, we mapped NSCLC, cell by cell, to understand and predict its response to drug treatment.” Dr Kulasinghe said.
“This ‘Google maps’ approach can pinpoint areas within tumours that are both responsive and resistant to therapies, which will be a gamechanger for lung cancer treatment.
(L to R) Mr Rafael Tubelleza, Dr James Monkman and Associate Professor Arutha Kulasinghe.
(Photo credit: The University of Queensland)
“Rather than having to use a trial-and-error approach, oncologists will now know which treatments are most likely to work with new precision medicine tools.”
The Anthony N. Brady Professor of Pathology at Yale, Professor David Rimm said the “work provides a road map for a new diagnostic test that could optimise treatment choice in lung cancer”.
Lung cancer is the leading cause of cancer death in the world, with an estimated 1.8 million deaths annually, and non-small cell lung cancer makes up 85 per cent of all cases.
Immunotherapy treatments cost between $400,000 and $500,000 per patient per year and are effective in only 20-30 per cent of patients.
“These therapies also carry significant risks for patients receiving them, including severe immune-related toxicity that can be fatal,” Dr Kulasinghe said.
“These challenges highlight the critical need to classify patients according to their likelihood to benefit from treatment.
“By integrating data on the molecular geography of cancer and machine learning techniques, we can improve treatment decision-making and improve patient outcomes for lung cancer patients.
“This same approach can also be used to inform treatments for other malignancies where immunotherapies are used, for example melanoma, head and neck, and bladder cancer.’’
The research has been published in Nature Genetics.
Collaboration and acknowledgements
The study was carried out in collaboration with Yale School of Medicine, WEHI and supported by NIH Yale SPORE in Lung Cancer, Cure Cancer, the Princess Alexandra Research Foundation and Wesley Research Institute. UQ’s Frazer Institute is based at the Translational Research Institute (TRI).Topics
Related articles
Vaping cannabis linked to teen depression and anxiety
