Lung Cancer Risk Stratification Research Articles

Investigating Causal Effects of Hematologic Traits on Lung Cancer: A Mendelian Randomization Study.

Observational studies have suggested blood cell counts may act as predictors of cancer. It is not known whether these hematologic traits are causally associated with lung cancer. Two-sample bidirectional univariable Mendelian randomization (MR) and multivariable MR (MVMR) were performed to investigate the causal association between hematologic traits and the overall risk of lung cancer and three histologic subtypes [lung adenocarcinoma, squamous cell lung cancer, and small cell lung cancer (SCLC)]. The instrumental variables of 23 hematologic traits were strictly selected from large-scale genome-wide association studies. Inverse-variance weighted method and five extra methods were used to obtain robust causal estimates. We found evidence that genetically influenced higher hematocrit [OR, 0.845; 95% confidence interval (CI), 0.783-0.913; P = 1.68 × 10-5] and hemoglobin concentration (OR, 0.868; 95% CI, 0.804-0.938; P = 3.20 × 10-4) and reticulocyte count (OR, 0.923; 95% CI, 0.872-0.976; P = 5.19 × 10-3) decreased lung carcinoma risk, especially in ever smokers. MVMR further identified hematocrit independently of smoking as an independent predictor. Subgroup analysis showed that a higher plateletcrit level increased the risk of small cell lung carcinoma (OR, 1.288; 95% CI, 1.126-1.474; P = 2.25 × 10-4). Genetically driven higher levels of reticulocyte count and hematocrit decreased lung cancer risk. Higher plateletcrit had an adverse effect on SCLC. Hematologic traits may act as low-cost factors for lung cancer risk stratification. Further studies are required to elucidate the potential mechanisms underlying the dysregulation of homeostasis related to hematologic traits, such as subclinical inflammation.

Personalised lung cancer risk stratification and lung cancer screening: do general practice electronic medical records have a role?

BackgroundIn the United Kingdom (UK), cancer screening invitations are based on general practice (GP) registrations. We hypothesize that GP electronic medical records (EMR) can be utilised to calculate a lung cancer risk score with good accuracy/clinical utility.MethodsThe development cohort was Secure Anonymised Information Linkage-SAIL (2.3 million GP EMR) and the validation cohort was UK Biobank-UKB (N = 211,597 with GP-EMR availability). Fast backward method was applied for variable selection and area under the curve (AUC) evaluated discrimination.ResultsAge 55–75 were included (SAIL: N = 574,196; UKB: N = 137,918). Six-year lung cancer incidence was 1.1% (6430) in SAIL and 0.48% (656) in UKB. The final model included 17/56 variables in SAIL for the EMR-derived score: age, sex, socioeconomic status, smoking status, family history, body mass index (BMI), BMI:smoking interaction, alcohol misuse, chronic obstructive pulmonary disease, coronary heart disease, dementia, hypertension, painful condition, stroke, peripheral vascular disease and history of previous cancer and previous pneumonia. The GP-EMR-derived score had AUC of 80.4% in SAIL and 74.4% in UKB and outperformed ever-smoked criteria (currently the first step in UK lung cancer screening pilots).DiscussionA GP-EMR-derived score may have a role in UK lung cancer screening by accurately targeting high-risk individuals without requiring patient contact.

Development and evaluation of a polygenic risk score for lung cancer in never-smoking women: A large-scale prospective Chinese cohort study.

The proportion of lung cancer in never smokers is rising, especially among Asian women, but there is no effective early detection tool. Here, we developed a polygenic risk score (PRS), which may help to identify the population with higher risk of lung cancer in never-smoking women. We first performed a large GWAS meta-analysis (8595 cases and 8275 controls) to systematically identify the susceptibility loci for lung cancer in never-smoking Asian women and then generated a PRS using GWAS datasets. Furthermore, we evaluated the utility and effectiveness of PRS in an independent Chinese prospective cohort comprising 55 266 individuals. The GWAS meta-analysis identified eight known loci and a novel locus (5q11.2) at the genome-wide statistical significance level of P < 5 × 10-8 . Based on the summary statistics of GWAS, we derived a polygenic risk score including 21 variants (PRS-21) for lung cancer in never-smoking women. Furthermore, PRS-21 had a hazard ratio (HR) per SD of 1.29 (95% CI = 1.18-1.41) in the prospective cohort. Compared with participants who had a low genetic risk, those with an intermediate (HR = 1.32, 95% CI: 1.00-1.72) and high (HR = 2.09, 95% CI: 1.56-2.80) genetic risk had a significantly higher risk of incident lung cancer. The addition of PRS-21 to the conventional risk model yielded a modest significant improvement in AUC (0.697 to 0.711) and net reclassification improvement (24.2%). The GWAS-derived PRS-21 significantly improves the risk stratification and prediction accuracy for incident lung cancer in never-smoking Asian women, demonstrating the potential for identification of high-risk individuals and early screening.

Prevalence of hospitalized women at high-risk for developing lung cancer

ABSTRACT Background Lung cancer screening with low-dose computer tomography (CT) has been shown to reduce the lung cancer mortality in high-risk individuals by 20%. Despite the proven mortality benefit, the utilization of lung cancer screening among high-risk populations remains low. Objective This study explores the prevalence of high-risk population for developing lung cancer among hospitalized women and evaluates the screening behavior toward other common cancers during a hospital stay. Methods This is a cross-sectional study in which 248 cancer-free hospitalized women aged 50–75 years who reported current or prior smoking were enrolled during hospital admission at an academic center. A bedside survey was conducted to collect socio-demographic, cancer screening behavior, and medical comorbidities for the study patients. Unpaired t-test and Chi-square tests were used to compare characteristics and common cancer screening behavior by lung cancer risk stratification. Results Forty-three percent of the hospitalized women were at intermediate to high-risk for developing lung cancer risk. Intermediate to high-risk women were more likely to be older, Caucasian, retired, or with a disability, and had higher comorbidity burden as compared to the low-risk group. Women at low and intermediate to high risk were equally non-adherent with breast (35% vs 31%, p = 0.59) and colorectal (32% vs 24%, p = 0.20) cancers screening guidelines. Only 38% of women from the intermediate to the high-risk group had a CT chest within the last year. Conclusion The study’s findings suggest that almost half of the hospitalized women who report current or past smoking are at high-risk for developing lung cancer.

High Risk Prognostic Factors Predictive of Outcome Following Stereotactic Body Radiation Therapy (SBRT) for Early-Stage Lung Cancer

High risk prognostic factors for outcomes following SBRT for early-stage lung cancer per SWOG/NRG 1914 include: tumor size > 2cm; moderately/poorly/undifferentiated histology; or high metabolic activity on PET/CT defined as SUV > 6.2. The purpose of this IRB approved retrospective study is to assess outcome for stage I lung cancer and to validate SWOG risk stratification in a single community-based radiation department. A total of 132 patients with 139 tumors treated with SBRT between 2014 and 2019 were stratified by SWOG risk group (high vs. low). To assess differences by risk group in estimated overall survival (OS) at the patient level, as well as Freedom from Local Failure (FFLF), Freedom from Regional Failure (FFRF) and Freedom from Distant Failure (FFDF) at the tumor level, Kaplan-Meier methodology and Cox proportional hazards (PH) modeling for correlated data were used. Statistical significance was concluded at the 0.05 level. Median follow-up for the entire group is 56 months. 77% of patients had high risk tumors. The median patient age was 75 years with 57% female. Patients with high-risk tumors were older (p = 0.023) compared to patients with low-risk tumors. At the tumor level, high risk tumors were more likely to have biopsies performed (91% vs 57% p = 0.002) and more likely to experience regional failure (28% vs 7% p = 0.005). High risk tumors were more likely to experience local failure (8% vs 0% p = NS) and distant failure (27% vs 14% p = NS). Median survival was 49 months for the whole group (46.0 months high risk, 65.1 months low risk). Actuarial OS at 5 years is 38% (35% high risk, 51% low risk). Actuarial FFLF at 5 years is 94% (93% high risk, 100% low risk). Actuarial FFRF at 5 years is 77% (72% high risk, 93% low risk). Actuarial FFDF at 5 years is 74% (70% high risk, 88% low risk). Cox PH models revealed no statistically significant differences in FFRF by risk group (p = 0.08). Excellent local control but higher regional and distant failure was demonstrated for high-risk lung tumors as defined by SWOG/NRG 1914. Clinically important decrement in outcomes were consistently noted for high-risk tumors in this sample which did not demonstrate statistical significance due to lack of events and statistical power. However, our data supports the prognostic importance of tumor size/grade and SUV for identifying patients at high risk. Further validation with larger sample sizes would contribute to our knowledge regarding risk stratification for early-stage lung cancer treated with SBRT.

Quantifying emphysema in lung screening computed tomography with robust automated lobe segmentation.

Anatomy-based quantification of emphysema in a lung screening cohort has the potential to improve lung cancer risk stratification and risk communication. Segmenting lung lobes is an essential step in this analysis, but leading lobe segmentation algorithms have not been validated for lung screening computed tomography (CT). In this work, we develop an automated approach to lobar emphysema quantification and study its association with lung cancer incidence. We combine self-supervised training with level set regularization and finetuning with radiologist annotations on three datasets to develop a lobe segmentation algorithm that is robust for lung screening CT. Using this algorithm, we extract quantitative CT measures for a cohort () from the National Lung Screening Trial and analyze the multivariate association with lung cancer incidence. Our lobe segmentation approach achieved an external validation Dice of 0.93, significantly outperforming a leading algorithm at 0.90 (). The percentage of low attenuation volume in the right upper lobe was associated with increased lung cancer incidence (odds ratio: 1.97; 95% CI: [1.06, 3.66]) independent of risk factors and diagnosis of whole lung emphysema. Quantitative lobar emphysema improved the goodness-of-fit to lung cancer incidence (, ). We are the first to develop and validate an automated lobe segmentation algorithm that is robust to smoking-related pathology. We discover a quantitative risk factor, lending further evidence that regional emphysema is independently associated with increased lung cancer incidence. The algorithm is provided at https://github.com/MASILab/EmphysemaSeg.

Stratification of Lung Cancer Risk with Thoracic Imaging Phenotypes.

In lung cancer screening, estimation of future lung cancer risk is usually guided by demographics and smoking status. The role of constitutional profiles of human body, a.k.a. body habitus, is increasingly understood to be important, but has not been integrated into risk models. Chest low dose computed tomography (LDCT) is the standard imaging study in lung cancer screening, with the capability to discriminate differences in body composition and organ arrangement in the thorax. We hypothesize that the primary phenotypes identified using lung screening chest LDCT can form a representation of body habitus and add predictive power for lung cancer risk stratification. In this pilot study, we evaluated the feasibility of body habitus image-based phenotyping on a large lung screening LDCT dataset. A thoracic imaging manifold was estimated based on an intensity-based pairwise (dis)similarity metric for pairs of spatial normalized chest LDCT images. We applied the hierarchical clustering method on this manifold to identify the primary phenotypes. Body habitus features of each identified phenotype were evaluated and associated with future lung cancer risk using time-to-event analysis. We evaluated the method on the baseline LDCT scans of 1,200 male subjects sampled from National Lung Screening Trial. Five primary phenotypes were identified, which were associated with highly distinguishable clinical and body habitus features. Time-to-event analysis against future lung cancer incidences showed two of the five identified phenotypes were associated with elevated future lung cancer risks (HR=1.61, 95% CI = [1.08, 2.38], p=0.019; HR=1.67, 95% CI = [0.98, 2.86], p=0.057). These results indicated that it is feasible to capture the body habitus by image-base phenotyping using lung screening LDCT and the learned body habitus representation can potentially add value for future lung cancer risk stratification.

Abstract 3312: Comparison of risk factors to molecular risk stratification in Chinese early-stage non-squamous non-small cell lung cancer patients

Abstract Background: Improved risk stratification for non-small cell lung cancer (NSCLC) represents a critical unmet need. A Clinical Laboratory Improvement Amendments (CLIA)-certified, 14-gene, quantitative PCR (qPCR)-based expression assay was found to better stratify mortality risk. The association of molecular risk stratification with other clinicopathological, radiomic, and genetic risk factors have not been fully identified, especially in Chinese NSCLC patients (pts). Methods: We newly recruited 102 early-stage non-squamous NSCLC pts who had undergone complete surgical resection. RNA was extracted from tumor tissue specimens and prospective molecular risk stratification by the 14-gene prognostic assay was performed. Gene mutation status and tumor mutational burden (TMB) were evaluated. Histological sections and clinical records were reviewed for canonical prognostic indicators that have been reported to be associated with lung cancer risk. Results: Of the 102 pts, the mean age was 55 years, 66 (64.7%) were female, and 18 (17.7%) had a smoking history. The majority of pts (84, 82.4%) were stage I, with 1 atypical adenomatous hyperplasia (AAH), 6 adenocarcinomas in situ (AIS), 1 stage II, and 2 stage III. 64 pts were deemed low-risk by molecular testing, 24 were intermediate-risk, and 14 were high-risk. The pts of AAH and AIS were all in the low-risk group. The maximum tumor diameter (MTD) in the high-risk group was the highest. There were no significant differences in age, sex, smoking status, pathologic stage, NCCN risk criteria, micropapillary or solid (MP/S) pattern, or the number of lesions between the three groups. The most common driver mutations were EGFR (65%), TP53 (28%), and RBM10 (13%). The molecular risk stratification had a significant association with TP53 (p = 0.013), TP53 loss of function (LOF, p = 0.01), KRAS (p = 0.011), and APC mutation (p = 0.05), but not with EGFR mutation and co-mutational status. TMB in the low-risk pts was significantly lower than in other pts (p = 0.007). The low-risk pts had lower SUVmax (p = 0.076), significantly lower mean CT value (p = 0.023), and mean enhanced CT value (p = 0.005) compared with others. In the NCCN low-risk pts, 25 were molecular low-risk, 12 were intermediate-risk, and 9 were high-risk. The risk stratification showed a significant association with SUVmax, TP53 mutation, and TMB in the NCCN low-risk pts. Pearson correlation showed the molecular risk score was significantly associated with MTD, TMB, SUVmax, mean CT value, mean enhanced CT value, and max CT value. SUVmax showed the strongest correlation with the risk score. Conclusion: The molecular risk stratification by the 14-gene assay was associated with canonical prognostic indicators, like TP53 LOF, TP53, KRAS, APC mutation, TMB, mean CT value, and mean enhanced CT value. Our study provides more evidence of the clinical utility of the qPCR-based prognostic assay. Citation Format: Naixin Liang, Jianchao Xue, Ming Zhao, Bowen Li, Yadong Wang, Zhicheng Huang, Yankai Xia, Ruirui Li, Zhongxing Bing, Zhibo Zheng, Jianpeng Zhang, Bin Wang, Zhe Feng, Xinyu Liu, Haochen Li, Xiaoqing Yu, Yang Song. Comparison of risk factors to molecular risk stratification in Chinese early-stage non-squamous non-small cell lung cancer patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3312.

Toward More Effective Lung Cancer Risk Stratification to Empower Screening Programs for the Asian Nonsmoking Population

Lung cancer is currently the leading cause of cancer-related deaths worldwide, although the smoking rate in the global population has gradually declined over time, and tobacco companies with official statements worldwide aimed to prevent smoking initiation among school-aged children in the 1980s [1]. Recently, an estimated microsimulation model demonstrated that the incidence of lung cancers caused by smoking has decreased with policy controls [2,3]. However, the absolute number and proportion of nonsmoking lung cancer in the population is estimated to gradually increase over time in the United States, projected from 2015 to 2065 [1-3].

Association of the interaction between mosaic chromosomal alterations and polygenic risk score with the risk of lung cancer: an array-based case-control association and prospective cohort study.

Mosaic chromosomal alterations (mCAs) detected from blood-derived DNA are large structural alterations of clonal haematopoietic origin and are associated with various diseases, such as haematological malignancies, infections, and solid cancers. We aimed to investigate whether mCAs contribute to the risk of lung cancer and modify the effect of polygenic risk score (PRS) on lung cancer risk prediction. The blood-derived DNA of patients with lung cancer and cancer-free controls with Chinese ancestry from the Nanjing Lung Cancer Cohort (NJLCC) study were genotyped with a Global Screening Array, and mCAs were detected with the Mosaic Chromosomal Alterations (MoChA) pipeline. mCA call sets of individuals with European ancestry were obtained from the prospective cohort UK Biobank (UKB) study, including documented incident lung cancer. All patients with lung cancer from the NJLCC study (aged 15 years or older at diagnosis) were histopathologically confirmed as new lung cancer cases by at least two pathologists and were free of chemotherapy or radiotherapy before diagnosis. Participants with incident lung cancer (aged 37-73 years at assessment) diagnosed after recruitment to the UKB were identified through linkage to national cancer registries. Logistic regression and Cox proportional hazard models were applied to evaluate associations between mCAs and risk of lung cancer in the NJLCC (logistic regression) and UKB (Cox proportional hazard model) studies. The NJLCC study included 10 248 individuals (6445 [62·89%] men and 3803 [37·11%] women; median age 60·0 years [IQR 53·0-66·0]) with lung cancer and 9298 individuals (5871 [63·14%] men and 3427 [36·86%] women; median age 60·0 years [52·0-65·0]) without lung cancer recruited from three sub-regions (north, central, and south) across China between April 15, 2003, and Aug 18, 2017. The UKB included 450 821 individuals recruited from 22 centres across the UK between March 13, 2006, and Nov 1, 2010, including 2088 individuals with lung cancer (1075 [51·48%] men and 1013 [48·52%] women; median age 63·0 years [IQR 59·0-66·0]), and 448 733 participants were free of lung cancer (204 713 [45·62%] men and 244 020 [54·38%] women; median age 58·0 years [IQR 50·0-63·0]). Compared with non-carriers of mosaic losses, carriers had a significantly increased risk of lung cancer in the NJLCC (odds ratio [OR] 1·81, 95% CI 1·43-2·28; p=6·69 × 10^-7) and UKB (hazard ratio [HR] 1·40, 95% CI 1·00-1·95; p=0·048) studies. This increased risk was even higher in patients with expanded cell fractions of mCAs (ie, cell fractions ≥10% vs cell fractions &lt;10%) in the NJLCC (OR 1·61 [95% CI 1·26-2·08] vs 1·03 [0·83-1·26]; p for heterogeneity test=6·41 × 10^-3). A significant multiplicative interaction was observed between PRS and mosaic losses on the risk of lung cancer in both the NJLCC (interaction p value=0·030) and UKB (p=0·043). Compared with non-carriers of mosaic loss abnormalities with low genetic risk, participants with expanded mosaic losses (cell fractions ≥10%) and high genetic risk had around a six-times increased risk of lung cancer in the NJLCC study (OR 6·40 [95% CI 3·22-12·69]), and an almost four-times increased risk of lung cancer (HR 3·75 [95% CI 1·86-7·55]) in the UKB study. The additive interaction also contributed a 3·67 (95% CI 0·49-6·85) relative excess risk of developing lung cancer in the NJLCC study, and a 2·15 (0·12-4·19) relative excess risk in the UKB study. mCAs act as a new endogenous indicator for the risk of lung cancer and might be jointly used with PRS to optimise personalised risk stratification for lung cancer. National Natural Science Foundation of China, Outstanding Youth Foundation of Jiangsu Province, Natural Science Foundation of Jiangsu Province, and Postdoctoral Science Foundation of China. For the Chinese translation of the abstract see Supplementary Materials section.

The state of the art for artificial intelligence in lung digital pathology.

Lung diseases carry a significant burden of morbidity and mortality worldwide. The advent of digital pathology (DP) and an increase in computational power have led to the development of artificial intelligence (AI)‐based tools that can assist pathologists and pulmonologists in improving clinical workflow and patient management. While previous works have explored the advances in computational approaches for breast, prostate, and head and neck cancers, there has been a growing interest in applying these technologies to lung diseases as well. The application of AI tools on radiology images for better characterization of indeterminate lung nodules, fibrotic lung disease, and lung cancer risk stratification has been well documented. In this article, we discuss methodologies used to build AI tools in lung DP, describing the various hand‐crafted and deep learning‐based unsupervised feature approaches. Next, we review AI tools across a wide spectrum of lung diseases including cancer, tuberculosis, idiopathic pulmonary fibrosis, and COVID‐19. We discuss the utility of novel imaging biomarkers for different types of clinical problems including quantification of biomarkers like PD‐L1, lung disease diagnosis, risk stratification, and prediction of response to treatments such as immune checkpoint inhibitors. We also look briefly at some emerging applications of AI tools in lung DP such as multimodal data analysis, 3D pathology, and transplant rejection. Lastly, we discuss the future of DP‐based AI tools, describing the challenges with regulatory approval, developing reimbursement models, planning clinical deployment, and addressing AI biases. © 2022 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Abstract 2227: Development and validation of a blood-based protein-only predictor of 5-year lung cancer risk in ever smokers

Abstract Background: Lung cancer is the second most common cancer type and is the leading cause of cancer death globally, with smoking and advancing age as the leading causal risk factors. The USPSTF guidelines for lung cancer screening recommends annual screening for select current or former smokers over 50 years of age. While annual screening via low dose CT has been demonstrated to decrease lung cancer mortality, compliance with screening guidelines remains low. Additional prognostic tools for future lung cancer risk stratification, particularly those without immutable demographic and health history, may be beneficial in increasing screening compliance and monitoring changing risk across time. Methods: Using modified-aptamer proteomics technology, SomaScan® assay v4.0, we scanned ~5000 proteins in 6085 EDTA plasma samples from “Ever Smokers” (current or former smokers, aged 50-73) with no known prevalent cancer at visit 3 of the Atherosclerosis Risk in Communities (ARIC) study, for a total of ~30 million protein measurements. A total of 348 incident lung cancer diagnoses occurred in this sample set, with 75 occurring within 5 years of visit 3 blood-draw. Time to lung cancer diagnosis events were modeled with protein measurements using machine learning methods in 70% of ARIC visit 3 ever smokers. A model was selected based on performance in a 15% holdout sample subset and validated in the remaining 15% ARIC visit 3 samples not used for model training or selection. Results: A 7-feature protein-only accelerated failure time (AFT) Weibull model was successfully developed to predict the probability of a lung cancer diagnosis within 5 years of blood draw. Model performance in training, model selection, and validation datasets was AUC equal to 0.76, 0.72, and 0.83, respectively. Based on predicted probabilities from the model, individuals were stratified into 3 risk bins (low, medium, and high) with a 5-year event rate of 0.49% vs 2.74% in low vs high risk bins. Model performance was additionally assessed in an independent Japanese cohort. Conclusion: We successfully developed a blood-based protein-only model that predicts risk of developing lung cancer in ever smokers. Performance of the protein model out-performs traditional risk factors for lung cancer and given the lack of immutable factors it has the potential to provide real-time risk which can be repeatedly assessed over time. Proteomics-driven risk stratification may have ability to increase adherence to lung cancer screening guidelines and/or influence a positive behavior change in modifiable risk-related behaviors. Citation Format: Clare Paterson, Leigh Alexander, Rachel Ostroff, Joseph Gogain, Yolanda Hagar, Hannah Biegel, Stephen Williams. Development and validation of a blood-based protein-only predictor of 5-year lung cancer risk in ever smokers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2227.

A Large-Scale Genome-Wide Gene-Gene Interaction Study of Lung Cancer Susceptibility in Europeans With a Trans-Ethnic Validation in Asians

IntroductionAlthough genome-wide association studies have been conducted to investigate genetic variation of lung tumorigenesis, little is known about gene-gene (G × G) interactions that may influence the risk of non-small cell lung cancer (NSCLC). MethodsLeveraging a total of 445,221 European-descent participants from the International Lung Cancer Consortium OncoArray project, Transdisciplinary Research in Cancer of the Lung and UK Biobank, we performed a large-scale genome-wide G × G interaction study on European NSCLC risk by a series of analyses. First, we used BiForce to evaluate and rank more than 58 billion G × G interactions from 340,958 single-nucleotide polymorphisms (SNPs). Then, the top interactions were further tested by demographically adjusted logistic regression models. Finally, we used the selected interactions to build lung cancer screening models of NSCLC, separately, for never and ever smokers. ResultsWith the Bonferroni correction, we identified eight statistically significant pairs of SNPs, which predominantly appeared in the 6p21.32 and 5p15.33 regions (e.g., rs521828C6orf10 and rs204999PRRT1, ORinteraction = 1.17, p = 6.57 × 10−13; rs3135369BTNL2 and rs2858859HLA-DQA1, ORinteraction = 1.17, p = 2.43 × 10−13; rs2858859HLA-DQA1 and rs9275572HLA-DQA2, ORinteraction = 1.15, p = 2.84 × 10−13; rs2853668TERT and rs62329694CLPTM1L, ORinteraction = 0.73, p = 2.70 × 10−13). Notably, even with much genetic heterogeneity across ethnicities, three pairs of SNPs in the 6p21.32 region identified from the European-ancestry population remained significant among an Asian population from the Nanjing Medical University Global Screening Array project (rs521828C6orf10 and rs204999PRRT1, ORinteraction = 1.13, p = 0.008; rs3135369BTNL2 and rs2858859HLA-DQA1, ORinteraction = 1.11, p = 5.23 × 10−4; rs3135369BTNL2 and rs9271300HLA-DQA1, ORinteraction = 0.89, p = 0.006). The interaction-empowered polygenetic risk score that integrated classical polygenetic risk score and G × G information score was remarkable in lung cancer risk stratification. ConclusionsImportant G × G interactions were identified and enriched in the 5p15.33 and 6p21.32 regions, which may enhance lung cancer screening models.

Potential of Inflammatory Protein Signatures for Enhanced Selection of People for Lung Cancer Screening.

Simple SummarySelection of appropriate high-risk smokers is one of the major challenges for implementing low-dose computed tomography screening for lung cancer. Many lung cancer risk prediction models have been proposed to supplement lung cancer screening. This study evaluated the potential of inflammatory protein markers to enhance lung cancer risk stratification beyond lung cancer risk models. The addition of inflammatory protein markers to existing lung cancer risk models improved risk prediction. The inflammatory protein markers may enhance current risk stratification and may be useful to identify high-risk populations for lung cancer screening.Randomized trials have demonstrated a substantial reduction in lung cancer (LC) mortality by screening heavy smokers with low-dose computed tomography (LDCT). The aim of this study was to assess if and to what extent blood-based inflammatory protein biomarkers might enhance selection of those at highest risk for LC screening. Ever smoking participants were chosen from 9940 participants, aged 50–75 years, who were followed up with respect to LC incidence for 17 years in a prospective population-based cohort study conducted in Saarland, Germany. Using proximity extension assay, 92 inflammation protein biomarkers were measured in baseline plasma samples of ever smoking participants, including 172 incident LC cases and 285 randomly selected participants free of LC. Smoothly clipped absolute deviation (SCAD) penalized regression with 0.632+ bootstrap for correction of overoptimism was applied to derive an inflammation protein biomarker score (INS) and a combined INS-pack-years score in a training set, and algorithms were further evaluated in an independent validation set. Furthermore, the performances of nine LC risk prediction models individually and in combination with inflammatory plasma protein biomarkers for predicting LC incidence were comparatively evaluated. The combined INS-pack-years score predicted LC incidence with area under the curves (AUCs) of 0.811 and 0.782 in the training and the validation sets, respectively. The addition of inflammatory plasma protein biomarkers to established nine LC risk models increased the AUCs up to 0.121 and 0.070 among ever smoking participants from training and validation sets, respectively. Our results suggest that inflammatory protein biomarkers may have potential to improve the selection of people for LC screening and thereby enhance screening efficiency.

Epigenome-wide scan identifies differentially methylated regions for lung cancer using pre-diagnostic peripheral blood

ABSTRACT Background: DNA methylation markers have been associated with lung cancer risk and may identify aetiologically relevant genomic regions, or alternatively, be markers of disease risk factors or biological processes associated with disease development. Methods: In a nested case–control study, we measured blood leukocyte DNA methylation levels in pre-diagnostic samples collected from 430 participants (208 cases; 222 controls) in the 1989 CLUE II cohort. We compared DNA methylation levels with case/control status to identify novel genomic regions, both single CpG sites and differentially methylated regions (DMRs), while controlling for known DNA methylation changes associated with smoking using a previously described pack-years-based smoking methylation score. Stratification analyses were conducted over time from blood draw to diagnosis, histology, and smoking status. Results: We identified 16 single CpG sites and 40 DMRs significantly associated with lung cancer risk (q < 0.05). The identified genomic regions were associated with genes including H19, HOXA3/HOXA4, RUNX3, BRICD5, PLXNB2, and RP13. For the single CpG sites, the strongest association was noted for cg09736286 in the DIABLO gene (OR [for 1 SD] = 2.99, 95% CI: 1.95–4.59, P-value = 4.81 × 10–7). We found that CpG sites in the HOXA3/HOXA4 region were hypermethylated in cases compared to controls. Conclusion: The single CpG sites and DMRs that we identified represented significant measurable differences in lung cancer risk, providing potential biomarkers for lung cancer risk stratification. Future studies will need to examine whether these regions are causally related to lung cancer.

Early candidate nasal swab classifiers developed using machine learning and whole transcriptome sequencing may improve early lung cancer detection.

8551 Background: The goal of lung nodule management is to make an early diagnosis in patients with lung cancer while avoiding unnecessary, costly and potentially harmful procedures in patients with benign lesions. Increased implementation of low dose CT screening will lead to increased numbers of both benign and malignant nodules that will require effective management. We have previously described the feasibility of detecting gene expression changes associated with lung cancer (“field of injury”) in nasal epithelium utilizing whole transcriptome RNA sequencing from non-invasive nasal brush samples. Using this approach, we now report the performance of candidate nasal classifiers that combine both genomic and clinical features to risk stratify lung nodules from ever-smokers to aid in the diagnosis of lung cancer. Methods: Patients from several clinical cohorts who were ever-smokers with a lung nodule &lt; 30 mm and without a history of prior cancer underwent nasal epithelium sampling. All patients had at least one year of follow up or until a final diagnosis of benign or malignant nodule was made. Candidate classifiers were developed using whole-transcriptome RNA sequencing and machine learning. Training of these classifiers included genomic and clinical information (age, sex, pack-years, years-since-quit, nodule size and nodule spiculation). Two decision boundaries were chosen to maximize sensitivity and specificity for low and high-risk nodules, respectively. The performance of these classifiers was evaluated using cross-validation (CV). Results: All candidate nasal classifiers underwent CV assessment on over 700 patients with lung nodules. All candidate classifiers achieved CV performance of &gt; 40% specificity at 95% sensitivity in low- risk nodules and all candidate classifiers achieved CV performance of &gt; 60% sensitivity at 90% specificity in high-risk nodules. The classifiers stratified benign nodules as low- risk with &gt; 95% negative predictive value (NPV), intermediate risk nodules were stratified as 5-65% risk and malignant nodules were stratified as high- risk with &gt; 65% positive predictive value (PPV) in a population with estimated 25% prevalence. This performance was robust across subgroups of age ( &lt; 65 year vs. &gt;65 year), sex, and current versus former smokers. Conclusions: The nasal genomic-clinical candidate classifiers have a high NPV effectively identifying benign nodules when calling them low- risk, thereby, informing decisions to more safely avoid a diagnostic workup. Additionally, the high PPV of these classifiers identifies malignant nodules when calling them high- risk and informs decisions regarding the urgency of a further diagnostic work-up. A nasal genomic-clinical classifier has the potential to serve as a non-invasive tool for lung cancer risk-stratification to help inform decision making in patients with lung nodules.

Screen-detected solid nodules: from detection of nodule to structured reporting.

Lung cancer screening (LCS) is gaining some interest worldwide after positive results from International trials. Unlike other screening practices, LCS is performed by an extremely sensitive test, namely low-dose computed tomography (LDCT) that can detect the smallest nodules in lung parenchyma. Up-to-date detection approaches, such as computer aided detection systems, have been increasingly employed for lung nodule automatic identification and are largely used in most LCS programs as a complementary tool to visual reading. Solid nodules of any size are represented in the vast majority of subjects undergoing LDCT. However, less than 1% of solid nodules will be diagnosed lung cancer. This fact calls for specific characterization of nodules to avoid false positives, overinvestigation, and reduce the risks associated with nodule work up. Recent research has been exploring the potential of artificial intelligence, including deep learning techniques, to enhance the accuracy of both detection and characterisation of lung nodule. Computer aided detection and diagnosis algorithms based on artificial intelligence approaches have demonstrated the ability to accurately detect and characterize parenchymal nodules, reducing the number of false positives, and to outperform some of the currently used risk models for prediction of lung cancer risk, potentially reducing the proportion of surveillance CT scans. These forthcoming approaches will eventually integrate a new reasoning for development of future guidelines, which are expected to evolve into precision and personalized stratification of lung cancer risk stratification by continuous fashion, as opposed to the current format with a limited number of risk classes within fixed thresholds of nodule size. This review aims to detail the standard of reference for optimal management of solid nodules by low-dose computed and its projection into the fine selection of candidates for work up.

FP09.02 Calibration of the LLP Lung Cancer Risk Stratification Model: Prospective Validation in the UKLS Cohort of 76,000 People

FP09.02 Calibration of the LLP Lung Cancer Risk Stratification Model: Prospective Validation in the UKLS Cohort of 76,000 People

Abstract IA16: Lung Cancer Risk Assessment in Diverse Populations: One Size Cannot Fit All

Abstract Lung cancer is the leading cause of cancer related deaths in the United States with an estimate of over 200,000 new cases and 135,720 deaths in 2020. Despite the development of precision therapies, immunotherapy and improved surgical techniques, the overall 5-year survival has remained below that of many other malignancies and significant disparities continue to persist. Black American males tend to present with later-stage disease, are less likely to undergo appropriate staging and thus, have a higher lung cancer mortality. The National Lung Screening Trial I (NLST) determined that low-dose CT (LDCT) screening could reduce lung cancer specific mortality by nearly 20% and overall mortality by 6.7%. However, studies suggest that clinically useful models for improved lung cancer risk stratification will be essential to the identification of those most likely to benefit from screening while minimizing unnecessary screening and associated risks. In NLST, only 4% of enrolled participants were Black Americans. Recent investigation demonstrates that current screening eligibility criteria are insufficient in identifying Black Americans at risk for lung cancer a large percentage of whom are diagnosed at a younger age than White Americans and with less tobacco consumption. As a result, current draft recommendations include reducing both age and smoking screening eligibility to 50 years of age and 20 pack years respectively with a goal of narrowing gaps in disparities in early lung cancer diagnosis and outcome. Here, we will review current models for lung cancer risk stratification that incorporate sociodemographic factors, comorbid illnesses and non-invasive biomarkers and discuss their clinical application to diverse populations. Citation Format: Patrick Nana-Sinkam. Lung Cancer Risk Assessment in Diverse Populations: One Size Cannot Fit All [abstract]. In: Proceedings of the AACR Virtual Conference: Thirteenth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2020 Oct 2-4. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(12 Suppl):Abstract nr IA16.

Artificial neural networks improve LDCT lung cancer screening: a comparative validation study

BackgroundThis study proposes a prediction model for the automatic assessment of lung cancer risk based on an artificial neural network (ANN) with a data-driven approach to the low-dose computed tomography (LDCT) standardized structure report.MethodsThis comparative validation study analysed a prospective cohort from Chiayi Chang Gung Memorial Hospital, Taiwan. In total, 836 asymptomatic patients who had undergone LDCT scans between February 2017 and August 2018 were included, comprising 27 lung cancer cases and 809 controls. A derivation cohort of 602 participants (19 lung cancer cases and 583 controls) was collected to construct the ANN prediction model. A comparative validation of the ANN and Lung-RADS was conducted with a prospective cohort of 234 participants (8 lung cancer cases and 226 controls). The areas under the curves (AUCs) of the receiver operating characteristic (ROC) curves were used to compare the prediction models.ResultsAt the cut-off of category 3, the Lung-RADS had a sensitivity of 12.5%, specificity of 96.0%, positive predictive value of 10.0%, and negative predictive value of 96.9%. At its optimal cut-off value, the ANN had a sensitivity of 75.0%, specificity of 85.0%, positive predictive value of 15.0%, and negative predictive value of 99.0%. The area under the ROC curve was 0.764 for the Lung-RADS and 0.873 for the ANN (P = 0.01). The two most important predictors used by the ANN for predicting lung cancer were the documented sizes of partially solid nodules and ground-glass nodules.ConclusionsCompared to the Lung-RADS, the ANN provided better sensitivity for the detection of lung cancer in an Asian population. In addition, the ANN provided a more refined discriminative ability than the Lung-RADS for lung cancer risk stratification with population-specific demographic characteristics. When lung nodules are detected and documented in a standardized structured report, ANNs may better provide important insights for lung cancer prediction than conventional rule-based criteria.