Evidence for screening

Lung cancer screening trials

Two international randomised controlled trials demonstrated that screening for lung cancer was effective. Both trials showed that low dose computed tomography (LDCT) chest scans among asymptomatic individuals who had a history of heavy smoking, resulted in a 20-24% reduction in the number of deaths from lung cancer. The National Lung Screening Trial (NLST) conducted in the U.S. was published in 2011 and the Dutch-Belgian NELSON trial in 2020. ⁽¹⁾

National Lung Screening Trial (USA)

The National Lung Screening Trial (NLST) was a randomised controlled trial (RCT) conducted in the United States, which compared the effect of low-dose helical computed tomography (LDCT) versus chest radiography (X-ray) on rates of lung cancer mortality. ⁽²⁾ The NLST recruited 53,456 participants (31, 533 men and 21,923 women) ⁽³⁾ between the enrolment period of September 2002 and April 2004, with participants randomised to receive either LDCT or chest radiography, with a baseline screen and two annual follow-up screening examinations completed. ⁽²⁾ Eligibility criteria for the NLST required participants to be aged between 55-74 years, with a 30 or more pack-year history of cigarette smoking, and to either currently smoke, or to have quit within the past 15 years. ⁽²⁾ Participants were initially followed up to Dec 2009.

Throughout the NLST, which took place between 2002 and 2009, there were a total of 1,060 lung cancers (645 per 100,000 person-years) diagnosed in the group screened with LDCT, as compared with 941 (572 per 100,000 person-years) in the chest radiography group. (4) Further to this, there were 1,877 deaths in the LDCT group, compared with 2,000 deaths in the chest radiography, and there was a 20% reduction in mortality from lung cancer observed in the LDCT group. ⁽⁴⁾

NELSON (Dutch-Belgian) trial

In the Nederlands-Leuvens Longkanker Screenings Onderzoek (NELSON) trial, which commenced in 2000, a total of 13,195 men and 2,594 women aged 50-74 years were randomly selected to either undergo CT screening, or to receive no screening, with the aim to demonstrate a reduction in lung cancer mortality by 25% or more when using LDCT. ⁽¹⁾ The trial largely focused on men, as only a small number of women were eligible to participate, due to the lower prevalence of smoking historically among women. Screening participants underwent four rounds of LDCT, at intervals of 1, 2, and 2.5 years, from January 2004 to December 2012. ⁽¹⁾

For the male participants, at the 10-year follow-up, the cumulative incidence of lung cancer was 5.58 per 1,000 person-years for those in the screening group, and 4.91 per 1,000 person-years for those in the control group. Further to this, lung cancers which were detected during screening were more often diagnosed in either stage 1A or 1B, (58.6%), with only 9.4% of screening detected lung cancers diagnosed at stage IV. ⁽¹⁾ At the 10-year follow-up, lung cancer mortality for male participants was 2.50 deaths per 1,000 person-years in the screening group, compared with 3.30 deaths per 1,000 person-years in the control group, with a relative risk reduction of death from lung cancer of 24% . ⁽¹⁾ For the small sample of women included in the trial, the relative risk reduction of death from lung cancer at 10 years of follow-up was 33%.⁽¹⁾ The NELSON trial demonstrated substantial reductions in lung cancer mortality for those who participated in LDCT screening, compared with no screening, with greater benefits shown in women. ⁽¹⁾

Screening accuracy and effectiveness

Numerous studies, both internationally and in Australia have been carried out, to understand the effectiveness and appropriateness of screening for lung cancer with LDCT, and to ensure that it meets the World Health Organization (WHO) principles of screening.

International landmark studies have demonstrated that reduced mortality from lung cancer is achievable when screening for lung cancer with LDCT as opposed to chest radiography, as described above.

Further studies have also set out to demonstrate the increased effectiveness of LDCT screening when using the eligibility standards of the PLCOm2012 criteria.⁽⁵⁾ Analysis from the prospective cohort study of the International Lung Screening Trial (ILST), which aimed to compare the effectiveness of the criteria for lung cancer screening specified by the United States Preventive Services Taskforce (USPSTF) with that of the PLCom2012 model demonstrated that the PLCOm2012 was the more effective model to use when selecting individuals to enrol into a screening program. ⁽⁶⁾ From the complete cohort of 5,819 individuals, 177 cases of lung cancer were diagnosed. ⁽⁶⁾ From this cohort, 135 lung cancers were diagnosed from 4,540 individuals who met the USPSTF criteria, and 162 cases of lung cancer were detected from the 4,540 individuals who had a PLCOm2012 threshold of at least 1.70% at 6 years. ⁽⁶⁾ The PLCOm2012 criteria detected 27 more lung cancers than the USPSTF criteria and provided overall higher cancer detection. ⁽⁶⁾

Predictive modelling has also been applied to a large population-based cohort in Australia to determine the potential effectiveness of the PLCOm2012 criteria for use within the Australian population. ⁽⁷⁾ When applied to the cohort of participants from the Sax Institute’s 45 and Up Study, the PLCOm2012 criteria, with a risk of ≥1.51% was found to be more sensitive in detecting lung cancer cases than the USPSTF criteria and was found to perform best among participants aged 55-74 years. ⁽⁷⁾

Screening eligibility

When developing a screening program, it is important to consider who the target population will be, to ensure that the benefits of screening outweigh the harms. A screening program should also be agile, to incorporate prospective additional risk factors into the screening algorithm as evidence is generated.

Estimates of the size of the Australian population that might be eligible for lung cancer screening have varied. A 2016 cross-sectional study of 3,586 individuals aged between 50 and 68 years from the Busselton Shire of Western Australia, aimed to estimate the proportion of an Australian cohort that would be eligible for lung cancer screening based on both the PLCOm2012 and the USPSTF criteria. ⁽⁸⁾ The study revealed that of the entire age-defined cohort, 2.6% would be eligible according to the PLCOm2012 criteria, 4.9% would be eligible according to the USPSTF criteria, and 15.2% would be eligible when merging criteria. ⁽⁸⁾ Extrapolated out to the entire population, these findings suggest that approximately 450,000 individuals in Australia may be eligible for participation in a lung cancer screening program. ⁽⁸⁾ Another study which applied statistical modelling to an existing Australian cohort, estimated that according to the PLCOm2012 risk of >1.51% over six years, 29% of Australians with a history of smoking aged 55-74 years would potentially be eligible for lung cancer screening, which equated to approximately 700,000 people or 1 in 37 people in 2013. ⁽⁷⁾

Cancer Australia has estimated that based on the proposed age and smoking criteria of individuals aged 50-74 years for the Aboriginal and Torres Strait Islander population, and 55-74 years for non-Indigenous Australians, with a history of smoking, 2.9 million people would meet this eligibility criteria. After completing the PLCOm2012 risk tool, it was estimated that approximately 580,000 people would be eligible for LDCT screening in 2020. ⁽⁹⁾

Individuals without a history of smoking

It has been demonstrated that lung screening programs are most clinically impactful, and cost-effective, when targeted only to ‘high-risk’ individuals, which based on data to date includes adults aged 55-74 years, with a history of heavy smoking. ⁽⁹⁾ The Taiwanese TALENT study, in which LDCT screening was offered for never-smokers aged 55-75 years with a family history of lung cancer, demonstrated a T0 lung cancer detection rate of 2.6%. ⁽¹⁰⁾ However, the majority of evidence indicates that screening younger high-risk individuals does not have a favourable balance of harms and benefits, and is not cost-effective. ⁽⁹⁾ Based on current evidence, the harms and costs of screening individuals who have never smoked, outweigh the benefits. A National Lung Cancer Screening Program in Australia will await further practice changing population- based evidence to accrue in this area, before routine LDCT screening is offered to individuals based on alternative screening criteria and risk factors.

Screening costs and cost effectiveness

Another crucial component to consider when developing a screening program, is that of the financial cost-effectiveness, including costs to the healthcare system and potential financial as well as potential for harm as a ‘cost’ to the individual. As stated in the World Health Organization’s principles of screening, the cost of case-finding (which includes diagnosis and treatment), should be economically balanced with respect to possible expenditure that may occur on medical care that results from treating the condition as a whole. ⁽¹¹⁾ Cost-effectiveness also accounts for the balance of harms and benefits of an intervention.

In the Queensland Lung Cancer Screening Study (QLCSS), 256 individuals enrolled aged 60-74 years, with ≥30 pack-year smoking history, who either currently smoked, or had quit smoking within the last 15 years, were recruited to complete a baseline and two annual screening scans using LDCT. ⁽¹²⁾ The average direct medical cost was $3,768 per participant, and the total treatment cost for lung cancer detected by screening was $545, 612. ⁽¹²⁾ In comparison to this, the average total treatment cost for lung cancers detected through symptomatic presentation was $554, 512. ⁽¹²⁾ Estimates of the costs for 14 cancers if screening had not taken place, revealed that whilst the costs for treatment would be similar for screening versus symptomatic presentation, the patient outcomes would be certainly inferior. ⁽¹²⁾

A cost-effectiveness analysis of lung cancer screening in Australia was completed by applying the screening parameters and outcomes from either the NLST, or the NELSON to Australian data. ⁽¹³⁾ The results demonstrated that the incremental cost-effectiveness ratio (ICER) for lung screening compared to usual care was $39,250 per quality adjusted life year (QALY) in the NELSON scenario, compared to an ICER of $76,300 in the NLST estimate. ⁽¹³⁾ Assuming a willingness-to-pay threshold of $30,000 or $50,000/QALY gained, the probability that the NELSON scenario was cost-effective was 15% and 60% respectively, compared to 0.5% and 6.7% in the NLST. ⁽¹³⁾ As such, it is possible that lung cancer screening in Australia could be cost-effective, but this is dependent on the ability to translate the lung cancer mortality benefits demonstrated in trials, into the clinical real world setting. ⁽¹³⁾

An economic impact study compared the cost-effectiveness of using the PLCOm2012 risk tool with the USPSTF age and smoking criteria. ⁽¹⁴⁾ The base-case scenario demonstrated that utilising the risk-model resulted in $355 cost-savings per 0.20 QALYs compared with the USPSTF-2013 criteria. ⁽¹⁴⁾ At a willingness-to-pay threshold of $20,000/QALY, the mean incremental net benefit was $4,294, and probabilistic sensitivity analyses revealed that 90% of the simulations were cost-effective at $20,000/QALY. (14) As such, using a risk tool such as the PLCOm2012 model can save costs and increase QALYs. ⁽¹⁴⁾

In the Medical Services Advisory Committee’s (MSAC) advice to the Minister of Health and Aged Care on the feasibility of implementing a National Lung Cancer Screening Program, it was determined that screening individuals aged 50 to 70, with a 30 pack-year history, who currently smoke or have quit within the last 10 years, will produce an ICER of approximately $65,000/QALY gained. ⁽¹⁵⁾ This was considered a robust estimate and acceptable by MSAC. ⁽¹⁵⁾

In addition, it is important to acknowledge that primary prevention of lung cancer through tobacco control is still the most cost-effective long-term strategy for reducing the burden of disease associated with lung cancer and would provide an additional cost benefit gain across cancer control and health optimisation beyond robust lung cancer specific estimates described above. ⁽¹⁶⁾ For more information on Cancer Council’s tobacco control policy priorities, please visit the tobacco chapter of the National Cancer Prevention Policy.

Current use, behaviours, attitudes, and knowledge of lung cancer screening

Several studies have been completed in recent years in order to understand current attitudes towards lung cancer screening in Australia. These have included perceptions of lung cancer risk and the harms and benefits of lung cancer screening amongst both the general population, and health professionals. In 2013, a targeted questionnaire was developed by Flynn, Peters, and Morgan to understand attitudes to lung cancer screening in an Australian high-risk population. ⁽¹⁷⁾ The results revealed that 95% of participants would participate in a screening program, and 91% of participants would consider surgical options if lung cancer was detected during screening. ⁽¹⁷⁾ However, despite the participants being at an increased risk for lung cancer, only 43% of those who completed the questionnaire, including those with a history of smoking, considered themselves to be at risk for lung cancer. ⁽¹⁷⁾

Preferences for lung cancer screening in the general population have been found to be influenced by a variety of factors. A 2020 survey designed to understand public attitudes on lung cancer screening, revealed that factors such as previous cancer diagnoses, and duration of smoking, indicated that respondents would be less likely to opt out of a lung cancer screening program. ⁽¹⁸⁾ A further study revealed that individuals who worry about lung cancer or have perceived seriousness of lung cancer have a preference for participating in lung cancer screening. ⁽¹⁹⁾ Additionally, 72% of participants either agreed or strongly agreed that they would partake in a lung cancer screening program if it was offered to them, despite only 45.6% of participants being eligible for screening. ⁽¹⁹⁾ A study that aimed to identify the potential drivers of screening participation among those eligible for the International Lung Screening Trial at two Australian sites identified self-efficacy and accessibility as two key factors. The report recommended that a screening program would benefit from assisted, pre-screening decision making, including funding for clinicians to support decision making, and mobile screening to improve accessibility.⁽²⁰⁾

Therefore, while there may be a strong willingness and interest from the general public to participate in a lung cancer screening program, it is clear that education to inform participants both of their risk of developing lung cancer, and of their likelihood to be eligible to partake in a screening program is required.

A 2018 study investigating current lung cancer screening practices by general practitioners (GPs) in the absence of a National Program, found that 69% of GPs requested either a chest X-ray or LDCT within the last 12 months, for one or more of their patients who were asymptomatic but deemed to be at ‘high risk’ for lung cancer based on their personal assessment. ⁽²¹⁾ Further to this, the majority of GPs stated that they would recommend lung cancer screening, with either LDCT or chest X-ray, to asymptomatic individuals with a history of smoking. ⁽²¹⁾ An additional survey which was distributed to a nationally representative survey of GPs in Australia revealed that 71% of participants were aware that lung cancer screening is not currently Medicare reimbursed in Australia. ⁽²²⁾ Despite this, 46% of participants requested a chest X-ray and almost 21% requested a LDCT screen during the survey period. ⁽²²⁾

These studies suggest that lung cancer screening is taking place opportunistically in some places in Australia, and there is confusion as to the current lung cancer screening recommendations and funding, with 15% of GPs believing that screening is funded by Medicare, and 40% unsure of the status of funding in Australia. ⁽²²⁾ As such, this research demonstrates the need for a clear framework for GPs around lung cancer screening, and as importantly, the procedure to communicate and refer on abnormal findings for timely investigation and management.