Recently, The Lancet published its annual landmark review on lung cancer screening, bringing together the latest medical evidence from around the world to examine in depth the current state of lung cancer screening globally and where it should go next. The review discusses populations at high risk of lung cancer, pulmonary nodules and lung cancer, smoking cessation interventions, cost-effectiveness, and future directions for lung cancer screening, and analyzes those future directions in detail from three angles: personalized screening intervals, artificial intelligence, and biomarkers. On personalized screening intervals, it notes that developing lung cancer risk stratification models to guide individualized screening intervals and greatly improve screening efficiency is one of the important research areas for lung cancer screening in the years ahead. This aligns closely with the direction taken by PT-L1™, ProteinT's product for predicting recurrence risk stratification after surgery in stage I lung cancer. By performing recurrence risk stratification on postoperative tissue from stage I lung cancer, follow-up intervals can be extended for those at low risk while intervention is intensified and screening intervals shortened for those at high risk, making it possible to monitor recurrence risk more effectively and help improve patient survival.
Abstract of the original article:
Lung cancer is one of the malignant tumors with the highest incidence and mortality worldwide. Screening, together with early diagnosis and early treatment, is an effective measure for reducing lung cancer mortality in all populations. The US National Lung Screening Trial (NLST) and the Dutch-Belgian NELSON lung cancer screening trial showed that, compared with chest radiography or no screening, low-dose CT screening reduces lung cancer mortality. These studies have clarified the key questions in lung cancer screening, yet uncertainty remains regarding key aspects of optimizing clinical effectiveness and cost-effectiveness. Recently, The Lancet published its annual landmark review on lung cancer screening, bringing together the latest medical evidence from around the world to provide an integrated analysis of the current state of lung cancer screening globally and its future directions.
Milestones in lung cancer screening
Early clinical trials in the 1970s investigated the potential of chest radiography and sputum analysis for lung cancer screening. Because these trials did not reduce lung cancer mortality, related lung cancer screening research slowed during the 1980s. The first lung cancer screening trial using low-dose CT was conducted in Japan, followed in the 1990s by the Early Lung Cancer Action Project (ELCAP) in North America. Subsequently, two large randomized controlled trials were launched worldwide, the US National Lung Screening Trial (NLST) and the Dutch-Belgian Lung Cancer Screening (NELSON) trial, to determine the effect of low-dose CT screening on mortality.
The NLST randomly assigned 53,454 individuals at high risk of lung cancer across 33 centers in the United States to three annual screenings with low-dose CT or chest radiography between 2002 and 2004. In the low-dose CT group, lung cancer mortality was reduced by a relative 20.0% compared with the radiography group (95% CI 6.8–26.7; P=0.004), and all-cause mortality was reduced by 6.7% (95% CI 1.2–13.6; P=0.02). The number needed to screen to prevent one lung cancer-related death was 320.
The NELSON trial randomized 13,195 men (primary analysis) and 2,594 women (subgroup analysis) between 2003 and 2006 to undergo low-dose CT at baseline, 1 year, 3 years and 5.5 years, or to receive no screening. Unlike the NLST, which defined positive and negative scans using nodule assessment based on two-dimensional (2D) diameter, the NELSON trial relied on semi-automated volumetric nodule measurement to guide management. The study showed that cumulative lung cancer mortality at 10 years fell by 24%. Unlike the NLST, however, the difference in 10-year all-cause mortality between the screening and no-screening groups in the NELSON trial was not statistically significant.
A 2021 population-based prospective cohort study from China (1,016,740 participants) found that one-off low-dose CT screening was associated with a 31% reduction in lung cancer mortality and a 32% reduction in all-cause mortality.
Defining the population for lung cancer screening
Populations at high risk of lung cancer
Lung cancer screening is considered most effective in people at high risk of lung cancer. The criteria defining high risk in the NLST and NELSON studies were relatively simple. In the NLST, this population was defined as individuals aged 55-74 years with a smoking history of ≥30 pack-years and, if they had quit, ≤15 years since quitting. In the NELSON trial, people at high risk of lung cancer were those aged 50-74 years with a smoking history of >25 years, smoking >15 cigarettes per day or >10 cigarettes per day for more than 30 years, and, if they had quit, ≤10 years since quitting. Another approach is to develop multivariable mathematical models that estimate lung cancer risk over a defined period and to select a risk threshold to determine eligibility. The Prostate, Lung, Colorectal and Ovarian (PLCO) model and the Liverpool Lung Project (LLP) model have been well evaluated and used in screening trials.
Screening in people who have never smoked
Although risk assessment in the NLST and NELSON lung cancer screening trials focused on tobacco exposure, at least a quarter of lung cancer patients worldwide have never smoked. The highest prevalence of lung cancer among never-smoking women is in South Asia (83% of women with lung cancer have never smoked), followed by East Asia (61%). Occupational and environmental risk factors also need to be considered for people at high risk of lung cancer, including exposure to radon (present in soil and potentially concentrated in mines and homes), asbestos, chromium, arsenic and air pollution.
Screening in cancer survivors
Compared with the general population, survivors of previous lung cancer and other cancers are at higher risk of developing primary lung cancer. According to an analysis of data from the US Surveillance, Epidemiology, and End Results program, among adult patients diagnosed with one of the ten most common primary malignant tumors, 8.1% developed a second primary tumor, of which lung cancer was the most common.
Management of screening results
Any screening program needs to deliver more benefit than harm. In low-dose CT lung cancer screening, potential physical and psychological harms relate mainly to the detection and workup of non-specific pulmonary nodules and to incidental findings. Lung cancer screening also provides an opportunity to assess COPD and cardiovascular disease. Various guidelines have been introduced to standardize the reporting and management of screening findings, which may reduce both harm and cost.
Pulmonary nodules and lung cancer
Over the past four decades, the management of pulmonary nodules has continued to evolve, based on nodule size, morphology, location and change over time. The main aim of nodule management is to distinguish clinically relevant lung cancer nodules from benign or indolent ones as quickly as possible. Nodule size and growth rate are the most important predictors of malignancy.
The overall goal of workup after referral for suspected cancer is to reach a rapid and safe diagnosis, complete tumor staging, and assess whether the patient is suitable for treatment. Cancers that are likely to shorten life expectancy should be treated, whereas patients with benign or indolent disease should be reassured and continue screening.
Incidental findings
Imaging findings other than pulmonary nodules are commonly detected on low-dose CT scans performed for lung cancer screening.
Incidental findings may lead to increased patient anxiety, radiation exposure from follow-up imaging, complications from investigation or treatment, and costs to the health system. The potential risks associated with investigating incidental findings should be balanced against the potential benefits, and appropriate management protocols for incidental findings are essential.
COPD assessment
Lung cancer screening is an opportunity to diagnose COPD.
Cardiovascular risk assessment
The NLST and NELSON trials showed that the risk of dying from cardiovascular disease is equal to or higher than the risk of dying from cancer.
Smoking cessation interventions
Smoking cessation is the most effective intervention for reducing lung cancer-related mortality, and lung cancer screening offers many opportunities to support individuals in quitting. Combining low-dose CT with an effective smoking cessation program optimizes the morbidity and mortality benefits of a lung cancer screening program by reducing the risk of lung cancer and other tobacco-related diseases, including chronic obstructive pulmonary disease, cardiovascular disease and other cancers, and has been shown to improve the cost-effectiveness of screening.
Cost-effectiveness
The reported cost-effectiveness of lung cancer screening has hindered its implementation in many countries. Cost estimates in the literature vary widely, ranging from US$1,464 to US$2 million per quality-adjusted life year (QALY). The cost-effectiveness of cancer screening also differs between populations.
A key aspect to consider in cost analyses is overdiagnosis, since the treatment of non-progressive or slowly growing cancers is an important driver of increased cost, generally confers no benefit to the patient, and carries potential harm. In the future, personalized screening intervals and refined risk stratification of currently indeterminate pulmonary nodules could improve the cost-effectiveness of cancer screening.
Future directions for lung cancer screening
Personalized screening intervals
For lung cancer screening participants at lower risk, extending the interval between low-dose CT scans could substantially reduce the number of scans performed each year, lower the cost of the screening program, and reduce the individual's radiation dose. Conversely, shortening the CT interval for high-risk individuals could reduce the number of delayed lung cancer diagnoses, and delayed diagnosis greatly shortens patient survival. Developing lung cancer risk stratification models to guide personalized screening intervals could greatly improve the efficiency of lung cancer screening, and this is one of the important research areas for lung cancer screening in the future. The optimal screening interval can be determined in an integrated way from the patient's epidemiological information, information from previous chest CT scans, and biomarkers.
Artificial intelligence
Artificial intelligence, including machine learning and deep learning, holds promise for transforming lung cancer screening. AI can play a role across the entire lung cancer screening workflow, including reducing radiation dose, pulmonary nodule detection, nodule characterization and personalized screening intervals. It can also help with the interpretation of low-dose CT scans in areas where qualified radiologists are unavailable.
To date, the greatest attention has focused on using AI as a first or second reader to improve the sensitivity of pulmonary nodule detection. A drawback of many published AI algorithms is that they have not yet undergone external validation, and many are not yet available commercially for clinical use. An up-to-date overview of AI software products with CE marking and US Food and Drug Administration approval for pulmonary nodule detection and measurement has now been published.
Biomarkers
Blood biomarker tests may play a role in refining the selection criteria for lung cancer screening, helping to manage indeterminate pulmonary nodules, supporting individualized screening intervals, and helping to predict response to adjuvant therapy. Incorporating biomarkers into lung cancer screening may be particularly important in Asian populations, where the incidence of lung cancer in people who have never smoked is high.
· Plasma microRNA signatures
In a retrospective analysis of a subset of samples from the Multicentric Italian Lung Detection (MILD) trial, a plasma microRNA signature reduced the false-positive rate for non-calcified nodules larger than 5 mm in diameter from 19.4% to 3.7%.
A signature of 13 microRNAs was applied to more than 1,000 high-risk participants in the Continuous Observation of Smoking Subjects (COSMOS) study and showed an overall accuracy of 75% (95% CI 72-78) for lung cancer diagnosis. The BIOMILD trial assessed the potential of the microRNA signature to safely determine screening intervals after a baseline low-dose CT scan. Although the high negative predictive value of low-dose CT limits the usefulness of blood miRNA results in people with negative scans, clinical findings indicate that in scan-positive participants the biomarker can effectively support clinical decisions on biopsy or follow-up frequency.
· Lung cancer autoantibodies
Detection of lung cancer autoantibodies in blood is currently the only biomarker that has been evaluated in prospective lung cancer screening. When lung cancer autoantibodies were applied to 1,613 people considered at high risk of lung cancer, blood test results showed a positive predictive value of 9-16% for a lung cancer diagnosis within 6 months, depending on whether a six- or seven-autoantibody panel was used. In a subsequent lung cancer screening program, the Scottish Early Lung Cancer Detection (ECLS) trial randomized more than 12,000 high-risk participants to a seven-autoantibody blood test and low-dose CT (if the autoantibody test was positive) or to standard clinical care, and showed after 2 years that the incidence of stage III or IV lung cancer was lower in the intervention group than in the standard care group.
When evaluated in a subset of participants in the German Lung Cancer Screening Intervention trial, the sensitivity of the test was only 13%, with specificity of 89-91%, indicating that it should not replace low-dose CT in high-risk populations.
· Circulating protein biomarkers
A proof-of-principle study from the Integrative Analysis of Lung Cancer Etiology and Risk (INTEGRAL) consortium showed that a panel of four circulating protein biomarkers can be used to identify individuals at high risk for lung cancer screening. The reported AUC was 0.83, compared with 0.73 for a model based on smoking alone, and this measure may in future help determine eligibility for lung cancer screening and distinguish benign from malignant nodules detected on low-dose CT.
· Other promising blood biomarkers
Other biomarkers with potential for the early detection of lung cancer include complement fragments, circulating tumor DNA, DNA methylation, protein profiling and genetic analysis of endobronchial epithelial cells, single nucleotide polymorphisms, and ELISA-based assays.
· Exhaled breath
Exhaled breath, including exhaled breath condensate, has also been investigated for lung cancer detection. Exhaled breath condensate contains cells and DNA fragments, while volatile organic compounds in exhaled breath can be analyzed by gas chromatography or mass spectrometry, nanosensors and colorimetric sensors.
Considerations for implementing lung cancer screening
Because of the high resource requirements of lung cancer screening programs and the evolving screening evidence, relatively few organized national lung cancer screening programs have been implemented. Countries with national or nationally funded screening programs, or where lung cancer screening is a national insurance benefit, include the United States, Poland, Croatia and South Korea. Canada and many European countries have launched pilot projects or smaller-scale lung cancer screening programs. In 2022, the UK National Screening Committee recommended that all four nations of the United Kingdom implement targeted lung cancer screening for people aged 55-74 at high risk of lung cancer, together with comprehensive smoking cessation support. Also in 2022, the Australian Government's Medical Services Advisory Committee recommended a national lung cancer screening program for people aged 50-70 whose smoking puts them at higher risk of lung cancer.
Even in well-resourced settings, implementation of lung cancer screening faces many challenges and requires a large and dedicated team, multidisciplinary expertise, and strong clinical and organizational leadership. Common challenges along the screening pathway must be addressed, including participant selection, recruitment methods, acquisition and interpretation, nodule management and referral, and communication of results. Maximizing benefit and minimizing harm is essential for any lung cancer screening program. Yet participation in lung cancer screening remains low, both in pilot settings and in established programs.
Conclusions
Randomized controlled trials have shown that low-dose CT screening for lung cancer helps reduce lung cancer mortality, and over the past three decades a broad body of evidence has supported lung cancer screening in many health systems with differing characteristics. Further research on personalized screening intervals, integrated assessment of COPD and cardiovascular risk, artificial intelligence and novel biomarkers represents a key opportunity to improve the efficiency of lung cancer screening. Geographic differences in lung cancer incidence, such as the large proportion of lung cancer patients who have never smoked in Asian countries, will require particular attention to refining selection criteria and risk models so that high-risk groups can be engaged effectively and health equity promoted. Continued research and broad adoption of evidence-based strategies could extend lung cancer screening to more people and improve health outcomes for populations worldwide.
ProteinT | PT-L1™
Recurrence Risk Prediction for Patients After Surgery for Stage I Lung Cancer
ProteinT-Lung1™ (PT-L1™) uses liquid chromatography-tandem mass spectrometry (LC-MS/MS) to perform proteomic testing on the tumor tissue of the individual being tested, and calculates a recurrence score based on a company-developed recurrence risk model for stage I lung cancer to further assess recurrence risk, providing a basis for clinicians and patients to develop more personalized screening plans and interventions. Its sensitivity and specificity are 96% and 86% respectively, and its negative predictive value (NPV) is 98%. Increasing the frequency of follow-up and screening for high-risk patients can markedly reduce the loss of survival time caused by delayed diagnosis, while reducing the frequency of screening and follow-up for low-risk patients can lower screening costs and improve quality of life.
ProteinT | n-LAPE/MS™
A Next-Generation Nano-Enrichment Liquid Biopsy Platform for Novel Biomarker Discovery
n-LAPE/MS™ is ProteinT's proprietary next-generation nano-scale low-abundance protein enrichment mass spectrometry platform for liquid biopsy-based novel biomarker discovery. Using self-synthesized and modified nanomaterials, it achieves specific and efficient enrichment of the low-abundance blood proteome, increasing the number of proteins detected in blood by mass spectrometry more than tenfold. Detection on a high-resolution LC-MS/MS platform enables identification of more than 4,000 proteins within an hour, with stable data. The platform now offers end-to-end service capabilities covering automated sample preparation, a library of dozens of self-synthesized and modified nanomaterials, and proprietary experimental methods. It can be widely applied in clinical research and exploration across cancer, cardiovascular and cerebrovascular disease, drug target discovery and other fields, providing a powerful tool for the discovery of liquid biomarkers in lung cancer and beyond.
References:
1.https://www.163.com/dy/article/HS5RBR040514C42P.html
2.Lung cancer screening - The Lancet DOI:https://doi.org/10.1016/S0140-6736(22)01694-4
