In an Australian-first study, Cancer Council NSW researchers have evaluated the performance of a lung cancer risk tool used to predict current and ex-smokers at the highest risk of developing lung cancer. They found that the tool had excellent predictive performance in Australia. In the future, the tool could help identify people suitable for lung cancer screening.
Background
Lung cancer causes more deaths in Australia than any other cancer. Smoking is the primary factor that puts people at risk, with 80% of lung cancers occurring in people who were smokers.
Unfortunately, the majority of lung cancer cases are diagnosed at a late stage as early diagnosis is difficult as the symptoms can often be vague and non-specific. Lung cancer screening with low-dose computerised tomography (LDCT) is promising, but is only effective at reducing mortality in people at the highest risk of lung cancer. This makes targeting the right group of people for screening essential – screening everyone who has ever smoked a cigarette would be ineffective and would result in unnecessary invasive testing, costs and psychological distress.
The research
The study investigated how reliably a risk tool developed in the USA and Canada would predict lung cancer outcomes in Australians by looking at almost 100,000 Australian current and ex-smokers who are part of the Sax Institute’s 45 and Up Study.
The tool uses a mathematical formula to combine risk factors (including age, smoking intensity, duration, years quit, body mass index and family history) to predict someone’s risk of being diagnosed with lung cancer in the next six years.
The researchers then compared the tool’s predictions with how many smokers in the 45 and Up Study went on to develop lung cancer and found the tool had excellent predictive performance in Australia. It correctly predicted nearly 70% of all subsequent diagnoses.
The impact
The study has major implications for the future of lung cancer screening in Australia as it is the biggest cohort ever analysed with this tool and the tool’s performance is extremely promising.
Applying a risk prediction tool to identify people who are at the highest risk has the potential to reduce the number of people needing to be screened and to make screening more efficient and cost-effective. However, there is also a pressing need for more evidence around whether the tool will work in a clinical setting and how screening should be implemented. There is a joint Australian and Canadian trial underway that will further validate the tool and optimise the trade-off between screening benefits and harms.
In the meantime, it is important to reinforce to people that the best way to reduce lung cancer risk is, and always will be, to quit smoking and stay quit.
Research team
Dr Marianne Weber
Cancer Council NSW
Sarsha Yap
David Goldsbury
Dr Yoon Jung Kang
Michael Caruana
Professor Emily Banks
Dr David Manners
Dr Fraser Brims
Dr Annette McWilliams
Dr Henry Marshall
Professor Kwun Fong
Professor Martin Tammemagi
Professor Karen Canfell