New computer program will ‘predict’ who will develop lung cancer in the next decade, enabling life-saving interventions to be done sooner
- In the early stages, there are no obvious signs or symptoms of lung cancer
- When it is detected, the disease has often already spread to other organs
A computer program has been developed that can help identify people who are at risk of developing lung cancer in the next ten years.
Called CanPredict, it is hoped to see public use in the near future and would mean high-risk patients could be screened earlier, potentially saving lives.
Developed by researchers at the University of Oxford and the University of Nottingham, it uses a wide range of measures – from smoking status and age to BMI and socioeconomic status – to calculate the risk of disease.
“We need a way to target those at greatest risk,” said Fergus Gleeson, a professor of radiology at the University of Oxford who worked on the new tool.
Lung cancer is the second most common cancer and the leading cause of cancer death worldwide. In the early stages, there are usually no obvious signs or symptoms and it may go unnoticed for some time.
A computer program has been developed that can help identify people who are at risk of developing lung cancer in the next ten years
Developed by researchers at the University of Oxford and the University of Nottingham, it uses a wide range of measures – from smoking status and age to BMI and socioeconomic status – to calculate risk of disease
About half of patients are diagnosed at a late stage, when the disease has spread and is more difficult to treat. Screening, which aims to catch the disease early, has been shown to improve survival rates.
Screening is performed using computed tomography (CT) scans, but these are expensive and require human resources and time, limiting the number of people who can be screened.
Doctors currently select people for screening using a survey that questions patients of a certain age about smoking status and family history of the disease. It is hoped that CanPredict can provide a much more accurate way of choosing those to be suggested for screening.
Patients deemed to be at high risk will be forwarded for advanced screening, hopefully to detect the disease when it is much easier to treat
Researchers used data from 2.54 million anonymized medical records to see which people were predicted to be at the highest risk of developing lung cancer. They then looked at which of the patients went on to develop the disease. CanPredict correctly identified more people who developed lung cancer than current methods.
Dr. Weiqi Liao, lead author of the study from the University of Oxford, said: ‘It works by examining existing patient records so that it can be managed automatically by GPs.’
Julia Hippisley-Cox, Professor of Clinical Epidemiology and General Practice at the University of Oxford, said: ‘We hope this tool will help better identify patients for screening and detect lung cancer earlier, when treatments are more likely to help.’