How a new global resolution can advance lung cancer detection

Lung diseases like lung cancer, chronic obstructive pulmonary disease (COPD), and asthma have a profound impact on millions of people, economies and the planet. Multiple determinants of health, such as poverty, air pollution, and geography, contribute to poor outcomes and act as barriers to detection, diagnosis, and implementation of guideline-directed medical therapy (GDMT).

Lung cancer alone claims 1.8 million lives yearly, making it the leading cause of cancer-related death – exceeding breast, colon, and prostate cancers combined. The burden is heaviest in low- and middle-income countries (LMICs), where 70% of cancer deaths occur and survival is less than half that of wealthier nations.

Early lung cancer detection is key. More than 50% of lung cancer cases are diagnosed when they are already stage IV. The five-year survival for stage IV lung cancer is 5%, compared to 65% when diagnosed at stage I.

Tailored, resource-stratified early lung cancer detection strategies are critical. The potential World Health Organization (WHO) adoption of an integrated lung health resolution can help accelerate the implementation of such solutions.

The resolution, for consideration at the World Health Assembly, aims to address shared risk factors and barriers to improve outcomes for people with lung diseases, while strengthening health systems and improving health equity. As part of these objectives, it calls for a renewed focus on early lung disease detection and diagnosis.

A successful resolution would roadmap change, urging member states to prioritise lung health, including lung cancer, within their national health strategies, encompassing the initiation and expansion of access to screening.

Low-dose computed tomography (LDCT) is the gold standard to screen and assess lung cancer risk. However, LDCT’s availability and cost make it unviable for broad-based use in many LMICs. The proposed resolution makes considering AI-enabled chest X-ray analysis, as a triage tool for LDCT assessment, timely.

By using AI to analyse X-rays for cancer signs, healthcare professionals can focus on patients needing further assessment, like an LDCT scan to confirm or rule out lung cancer. This triage approach optimizes the use of available equipment and expertise, ensuring those at the highest risk get the needed attention.

Data recently released at the European Lung Cancer Congress confirms AI-enabled chest X-ray analysis as an effective LDCT triage tool.

The CREATE study – conducted in Egypt, India, Indonesia, Mexico, and Turkey – demonstrated that AI-enabled CXR analysis can accurately predict lung malignancy risk, compared with radiologist assessment of LDCT.

A separate modelling study, assessing the clinical and financial implications of implementing AI-enabled chest X-ray analysis as a triage tool within Vietnam’s healthcare system, predicted that implementation would lead to 3,155 additional lung cancer diagnoses at an early stage and help prevent 4,742 premature deaths over five years. The study concluded that by year five, implementation would be cost-neutral, as earlier cancer detection leads to more efficient treatment and reduced overall healthcare expenses

Studies in other countries have shown that late-stage lung cancer treatment can cost up to 156% more than early-stage treatment, and that the cost of treating late-stage cancer has risen by 55% since 2018.

To improve early lung cancer detection, AI-enabled chest X-ray analysis needs broad implementation within national screening programs, particularly in primary care, where most patients first seek care.

This is crucial for expanding screening beyond smokers, as the incidence of lung cancer among never-smokers has been rising. A large Taiwan study, screening over 12,000 never-smokers with family history, found a 2.6% lung cancer detection rate. While seemingly small, this detection rate is double the lung cancer incidence observed in major European and US screening trials that focused on high-risk smokers.

AI-enabled technology also needs integration across lung specialties. People with tuberculosis (TB) have a 4x higher lung cancer risk, and COPD increases the risk by 2-5x. A Vietnam study used AI to review chest X-rays from TB screenings – of nearly 136,000 patients scanned, 1,733 were identified as high-risk, leading to 500 LDCTs and 133 lung cancer diagnoses.

Integrated care pathways are essential, connecting primary, secondary, and tertiary care, and guiding patients from triage to referral to the best treatment.

The potential WHO resolution offers an opportunity to push the boundaries of early detection and diagnosis, and subsequent implementation of guideline directed medical therapy (GDMT) for all lung diseases. To fully realise this potential, a collaborative implementation effort is essential.

Policy-makers can explore the integration of AI-enabled chest X-ray analysis into screening algorithms, as a triage tool, driving progress towards non-communicable disease reduction targets and unlocking economic benefits.

Healthcare professionals can champion resource-stratified screening approaches within their systems, improving patient outcomes and optimizing resource allocation.

Payers can drive greater value by considering the adoption of innovative technologies and exploration of possible efficiencies across the lung cancer, TB, and COPD settings.

By embracing the potential resolution, prioritising novel solutions aligned with its calls to action, and working together, we can help create a future where lung cancer is no longer a leading cause of death.