Non-risk-based lung cancer screening with low-dose computed tomography: JAMA

In 2015, the First Affiliated Hospital of Guangzhou Medical University, in collaboration with the Guangdong Hetaoxiang Medical Charity Foundation and the Guangzhou Charity Association, launched a community-based lung cancer screening initiative targeting low-income and underserved populations. Subsequently, with support from the Guangzhou Municipal Government, the Health and Family Planning Commission, the Civil Affairs Bureau, and KingMed Diagnostics, the hospital and the National Clinical Research Center for Respiratory Disease undertook the “LUNG-CARE Project”, a large-scale lung cancer screening initiative in Yuexiu District, Guangzhou. By leveraging community networks, the project provided eligible residents with free low-dose computed tomography (LDCT) scans to promote early diagnosis and treatment of lung cancer.

While randomized controlled trials have shown that heavy smokers benefit from LDCT screening, recent studies indicate a rising incidence of lung cancer among younger, non-smoking populations in Asia. The LUNG-CARE Project incorporated detailed epidemiologic questionnaires and follow-up protocols to identify potential risk factors beyond smoking and to refine screening strategies for the Chinese population. A central aim was to compare lung cancer detection rates between individuals with and without conventional high-risk factors (e.g., smoking or family history) and to uncover novel population-specific indicators of risk.

Between 2015 and 2021, residents aged 40–74 years without a history of lung cancer or related symptoms were recruited from four communities in Guangzhou to undergo LDCT-based screening. Lung cancer detection rates, the proportion of early-stage (stage I) diagnoses, and the diagnostic performance of LDCT were assessed. Participants were stratified into “high-risk” categories according to the NCCN guidelines and the Chinese Medical Association’s expert consensus. A total of 11,708 participants underwent LDCT screening [male: 5,452 (46.6%), female: 6,256 (53.4%), median age 59 years (IQR: 51-65)). Positive findings - defined as solid or part-solid nodules ≥5 mm or pure ground-glass nodules ≥8 mm - were observed in 2,245 participants (19.2%). Of these, 231 underwent invasive diagnostic procedures (including surgery), leading to 200 confirmed cases of lung cancer (1.7%), of which 165 (82.5%) were stage 0-I. Most cases were lung adenocarcinomas.

The diagnostic performance of a single LDCT scan showed sensitivity 96.6% (200/207), specificity 82.2% (9,456/11,501), positive predictive value (PPV) 8.9% (200/2,245), negative predictive value (NPV) 99.9% (9,456/9,463). Based on NCCN and Chinese expert consensus criteria, only 1,883 (16.1%) and 4,902 (41.9%) participants, respectively, were classified as “high-risk”. However, only 38 (19.0%) and 112 (56.0%) of the confirmed lung cancer cases met the respective high-risk definitions-corresponding to missed detection rates of 81.0% (162/200) and 44.0% (88/200). No significant difference in lung cancer detection rates was observed between the NCCN-defined high-risk and non-high-risk groups (2.0% vs. 1.6%, P = 0.245). In contrast, the Chinese consensus definition yielded a significantly higher detection rate in the high-risk group (2.3% vs. 1.3%, P < 0.001). These trends remained consistent in sensitivity analyses, excluding adenocarcinoma in situ (AIS) and minimally invasive adenocarcinoma (MIA) cases.

Incorporating non-high-risk individuals into lung cancer screening programs, based on the unique epidemiological characteristics of lung cancer in Chinese populations, demonstrates for the first time the value of a non-risk-based strategy in China. This approach may offer a more rational alternative to traditional risk-based screening by enabling the detection of a greater number of early-stage, potentially curable lung cancers. Future research should prioritize the identification of clinically significant lung cancer subtypes warranting intervention to optimize patient outcomes, as well as the development of high-risk biomarkers or pre-screening enrichment strategies tailored to populations not conventionally classified as high-risk.

At the 2022 European Society for Medical Oncology (ESMO) Congress, Professors Jianxing He and Wenhua Liang’s team presented the initial findings from the LUNG-CARE Project as a Late-Breaking Abstract (LBA48). Key results included the following: among individuals undergoing LDCT screening, 19.2% had clinically significant solitary nodules, and 1.7% were pathologically confirmed to have lung cancer. Notably, 86% of diagnosed lung cancers were stage 0–I, stages generally associated with curative potential. Lung cancer prevalence increased progressively with age, peaking in the 60-64 -year age group; however, the proportion of stage I cancers declined with advancing age, underscoring the importance of balancing detection rates and curability when designing screening programs.

Using an unscreened cohort of approximately 110,000 residents from the same community as a comparator, the LDCT screening cohort was associated with a 63% reduction in lung cancer mortality (hazard ratio [HR]: 0.37), markedly lower than the HR of 0.69 reported in China’s national one-off screening initiative. This enhanced survival benefit may be attributable to the inclusion of non-high-risk individuals in the LUNG-CARE Project. Overall, the screened cohort consistently exhibited superior prognostic outcomes compared with the unscreened cohort, likely due to a higher proportion of early-stage cancer detection.

Reference:

Li C, Cheng B, Li J, et al. Non–Risk-Based Lung Cancer Screening With Low-Dose Computed Tomography. JAMA. Published online April 21, 2025. doi:10.1001/jama.2025.4017