JCO Article Insights: Improving Lung Cancer Screening Using Blood-Based Biomarkers

In this JCO Article Insights episode, Davide Soldato summarized finding from the original article published in the September JCO issue: “Mortality Benefit of a Blood-Based Biomarker Panel for Lung Cancer on the Basis of the Prostate, Lung, Colorectal, and Ovarian Cohort”. The summary provides information regarding the ability of a blood-based panel of 4 biomarkers in improving the identification of individuals at risk of developing lethal lung cancer and potential of combined screening strategies to improve trade-off between potential harms and benefit of the screening process. TRANSCRIPT The guest on this podcast episode has no disclosures to declare. Davide Soldato: Welcome to the JCO Article Insights episode for the September issue of the Journal of Clinical Oncology. This is Davide Soldato, your host, and today I will be providing a summary on one article focused on the refinement of screening strategies for lung cancer. The article, titled "Mortality Benefit of a Blood-Based Biomarker Panel for Lung Cancer on the Basis of the Prostate, Lung, Colorectal, and Ovarian Cohort" by Dr. Irajizad and colleagues, investigated the ability of a panel of circulating blood biomarkers in improving the identification of individuals at risk of developing lethal lung cancer. We already know that lung cancer screening based on the use of low dose CT is associated with a reduction in mortality, as already demonstrated by the National Lung Cancer Screening Trial and the NELSON Trial. Furthermore, the US Preventive Task Force has recently recommended an expansion of screening criteria for lung cancer. Currently, based on this recommendation, screening strategies are recommended for individuals 50 years of age and older with a smoking history of at least 20 pack-years and who are current smokers at the moment of the screening time or have quit within the past 15 years. Despite this positive data and this recommendation, the uptake of lung cancer screening in the US is still low, with reported uptake rates below 15%. The risk of false positive results, the unnecessary follow-up procedures, uneven access to lung cancer screening programs, and fear of cancer diagnosis and treatment have all been identified as potential barriers to optimal implementation and uptake of lung cancer screening. And so, in order to overcome some of these barriers, several efforts have been made in the last years to develop lung cancer screening prediction models with the aim of selecting a higher risk population who would derive higher benefit from lung cancer screening.  In the present manuscript, the author builds on their previous work where they developed and tested a clinical prediction model and a blood-based prediction model in the context of the PLCO cohort. The Prostate, Lung, Colon and Ovarian Cancer Screening Trial was a randomized, multicenter trial in the US which aimed to evaluate the impact of early detection procedures on disease-specific mortality for the aforementioned cancers. Two lung cancer screening prediction model had already been developed and tested in the cohort. The PLCOm2012 model is based on several clinical and demographic characteristics, including age, race and ethnicity group, education, BMI, chronic obstructive pulmonary disease, personal history of cancer, family history of lung cancer, smoking status and intensity, duration and quit time. In a previous study, this model demonstrated a higher sensitivity and positive predictive value with no loss in specificity for lung cancer diagnosis compared to the National Lung Screening Trial criteria. Additionally, in the same cohort, the 4MP was a blood-based panel that included the precursor form of surfactant protein B, cancer antigen 125, carcinoembryonic antigen, and cytokeratin-19 fragment. In a previous study, a combination of this blood-based panel and the PLCOm2012 model was associated with a better identification of patients at high risk of developing lung cancer that would consequently benefit from lung cancer screening. In the manuscript that was published in the current issue of the JCO, the authors aim to expand on these previous results and test the ability of the combined 4MP and PLCO model to identify individuals at high risk of developing lung cancer death. The study used prediagnostic sera of 552 individuals that were diagnosed with lung cancer within one year from the blood draw and 2000  non-cases. In the study, the authors assessed the performance of this combined four 4MP and PLCO model at a risk threshold of 1% and 1.7% of developing lung cancer at six years. Among the more than 500 individuals who were diagnosed with lung cancer, 70% died from it and 18% died of other causes, and the median survival times for lung cancer cases was 2.7 years. The combined 4MP and PLCO model had an area under the curve (AUC) of 0.88 for the prediction of lung cancer-specific mortality. The performance of this combined model using both clinical demographic and also a blood-based panel was higher than the ones of the two models considered alone. Furthermore, the model had similar predictive performance for both non small cell lung cancer and small cell lung cancer-related deaths.  Subsequently, the authors compared the performance of the combined 4MP and PLCO model with the 2013 and 2021 US Preventive Task Force criteria and observed that the combined model had improved sensitivity, specificity, and positive predictive value for the prediction of lung cancer-specific mortality compared to both types of criteria. Finally, the authors assessed whether the combined 4MP and PLCO model were able to determine the survival probability among individuals who had a smoking history of at least 10 pack-years. Cases and non-cases were classified as either test positive or negative according to model scores at the 1.7 and 1% risk threshold at six years. For both thresholds, the combined 4MP and PLCO model identified a significantly higher number of lung cancer deaths in test positive individuals compared with test negative ones. So, in conclusion, these studies identify patients who are at higher risk of developing lung cancer-specific mortality using a combination of blood-based biomarkers and clinical demographic characteristics. The combined models showed higher sensitivity and specificity and positive predictive value compared to the standard US Preventive Task Force criteria. The results of this study are important because the identification of individuals at higher risk of lung cancer diagnosis and death offers the opportunity for a more favorable tradeoff between potential harms and benefits of the screening process. And so these results could assist in the design of future screening and intervention studies as well as facilitate the uptake of lung cancer screening, especially for those test-positive patients that have a higher risk of lung cancer death. Application of this model could potentially lead to a higher number of patients diagnosed in earlier stages and thus eligible for curative intent treatment.  That concludes this episode of JCO Article Insights regarding a summary of the article "Mortality Benefit of a Blood-Based Biomarker Panel for Lung Cancer on the Basis of the Prostate, Lung, Colorectal, and Ovarian Cohort" by Dr. Irajizad and colleagues.  This is Davide Soldato. Thank you for your attention and stay tuned for the next episode of JCO Article Insights.  The purpose of this podcast is to educate and to inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions. Guests on this podcast express their own opinions, experiences, and conclusions. Guest statements on the podcast do not express the opinions of ASCO. The mention of any product, service, organization, activity, or therapy should not be construed as an ASCO endorsement.