National Lung Screening Trial Research Articles

Lung-Reporting and Data System 2.0: Impact of the Updated Approach to Juxtapleural Nodules During Lung Cancer Screening Using the National Lung Cancer Screening Trial Data Set.

To determine the frequency of malignancy of nonperifissural juxtapleural nodules (JPNs) measuring 6 to < 10 mm in a subset of low-dose chest computed tomographies from the National Lung Cancer Screening Trial and the rate of down-classification of such nodules in Lung-Reporting and Data System (RADS) 2.0 compared with Lung-RADS 1.1. A secondary analysis of a subset of the National Lung Screening Trial was performed. An exemption was granted by the Institutional Review Board. The dominant noncalcified nodule measuring 6 to <10 mm was identified on all available prevalence computed tomographies. Nodules were categorized as pleural or nonpleural. Benign or malignant morphology was recorded. Initial and updated categories based on Lung-RADS 1.1 and Lung-RADS 2.0 were assigned, respectively. The impact of the down-classification of JPN was assessed. Both classification schemes were compared using the McNemar test ( P < 0.01). A total of 2813 patients (62 ± 5 y, 1717 men) with 4408 noncalcified nodules were studied. One thousand seventy-three dominant nodules measuring 6 to <10 mm were identified. Three hundred forty-eight (32.4%) were JPN. The updated scheme allowed down-classification of 310 JPN from categories 3 (n = 198) and 4A (n = 112) to category 2. We, therefore, estimate a 4.8% rate of down-classification to category 2 in the entire National Lung Screening Trial screening group. Two/348 (0.57%) JPN were malignant, both nonbenign in morphology. The false-positive rate decreased in the updated classification ( P < 0.01). This study demonstrates the low malignant potential of benign morphology JPN measuring 6 mm to <10 mm. The Lung-RADS 2.0 approach to JPN is estimated to reduce short-term follow-ups and false-positive results.

Overall Survival Among Patients With De Novo Stage IV Metastatic and Distant Metastatic Recurrent Non–Small Cell Lung Cancer

Despite recent breakthroughs in therapy, advanced lung cancer still poses a therapeutic challenge. The survival profile of patients with metastatic lung cancer remains poorly understood by metastatic disease type (ie, de novo stage IV vs distant recurrence). To evaluate the association of metastatic disease type on overall survival (OS) among patients with non-small cell lung cancer (NSCLC) and to identify potential mechanisms underlying any survival difference. Cohort study of a national US population based at a tertiary referral center in the San Francisco Bay Area using participant data from the National Lung Screening Trial (NLST) who were enrolled between 2002 and 2004 and followed up for up to 7 years as the primary cohort and patient data from Stanford Healthcare (SHC) for diagnoses between 2009 and 2019 and followed up for up to 13 years as the validation cohort. Participants from NLST with de novo metastatic or distant recurrent NSCLC diagnoses were included. Data were analyzed from January 2021 to March 2023. De novo stage IV vs distant recurrent metastatic disease. OS after diagnosis of metastatic disease. The NLST and SHC cohort consisted of 660 and 180 participants, respectively (411 men [62.3%] vs 109 men [60.6%], 602 White participants [91.2%] vs 111 White participants [61.7%], and mean [SD] age of 66.8 [5.5] vs 71.4 [7.9] years at metastasis, respectively). Patients with distant recurrence showed significantly better OS than patients with de novo metastasis (adjusted hazard ratio [aHR], 0.72; 95% CI, 0.60-0.87; P < .001) in NLST, which was replicated in SHC (aHR, 0.64; 95% CI, 0.43-0.96; P = .03). In SHC, patients with de novo metastasis more frequently progressed to the bone (63 patients with de novo metastasis [52.5%] vs 19 patients with distant recurrence [31.7%]) or pleura (40 patients with de novo metastasis [33.3%] vs 8 patients with distant recurrence [13.3%]) than patients with distant recurrence and were primarily detected through symptoms (102 patients [85.0%]) as compared with posttreatment surveillance (47 patients [78.3%]) in the latter. The main finding remained consistent after further adjusting for metastasis sites and detection methods. In this cohort study, patients with distant recurrent NSCLC had significantly better OS than those with de novo disease, and the latter group was associated with characteristics that may affect overall survival. This finding can help inform future clinical trial designs to ensure a balance for baseline patient characteristics.

Level of Education Matters in Regard to Participants' Compliance With Screening in the National Lung Screening Trial.

The success of cancer screening depends on patient adherence to the screening program. The purpose of this study is to assess how the level of education might affect participants' compliance with screening in the National Lung Screening Trial (NLST). Secondary data analyses of the participants in the NLST were performed. A total of 50,104 participants were included in this study. Participants who enrolled in the trial but refused the initial screening were compared with those who completed the screening. A multivariate logistic regression model was used to assess the association between participant noncompliance and education level. A total of 3712 (7.41%) participants refused lung cancer screening in the NLST. Compared with the reference group, participants with an education level of eighth grade or less (odds ratio [OR]: 2.1, CI: 1.68-2.76), ninth-11th grade (OR: 1.9, CI: 1.7-2.34), high school graduates (OR: 1.3, CI: 1.22-1.54), after high school training (OR: 1.1, CI: 1-1.31), or an associate's degree (OR: 1.2, CI: 1.07-1.36) had significantly higher odds of refusing lung cancer screening. Participants with a bachelor's degree showed no significant association with compliance with screening (OR: 0.9, P = 0.86). Multivariate regression analysis also showed that younger, single, male participants with a longer duration of smoking history had significantly higher odds of refusing the screening. A lower level of education was significantly associated with refusing lung cancer screening. A strategic targeted approach for this group might be necessary to promote their compliance rate.

Impact of Diabetes Mellitus on Lung Cancer Screening Efficacy in the National Lung Screening Trial

Impact of Diabetes Mellitus on Lung Cancer Screening Efficacy in the National Lung Screening Trial

Systematic Review of Lung Cancer Screening: Advancements and Strategies for Implementation.

Lung cancer is the leading cause of cancer-related deaths in Europe, with low survival rates primarily due to late-stage diagnosis. Early detection can significantly improve survival rates, but lung cancer screening is not currently implemented in Italy. Many countries have implemented lung cancer screening programs for high-risk populations, with studies showing a reduction in mortality. This review aimed to identify key areas for establishing a lung cancer screening program in Italy. A literature search was conducted in October 2022, using the PubMed and Scopus databases. Items of interest included updated evidence, approaches used in other countries, enrollment and eligibility criteria, models, cost-effectiveness studies, and smoking cessation programs. A literature search yielded 61 scientific papers, highlighting the effectiveness of low-dose computed tomography (LDCT) screening in reducing mortality among high-risk populations. The National Lung Screening Trial (NLST) in the United States demonstrated a 20% reduction in lung cancer mortality with LDCT, and other trials confirmed its potential to reduce mortality by up to 39% and detect early-stage cancers. However, false-positive results and associated harm were concerns. Economic evaluations generally supported the cost-effectiveness of LDCT screening, especially when combined with smoking cessation interventions for individuals aged 55 to 75 with a significant smoking history. Implementing a screening program in Italy requires the careful consideration of optimal strategies, population selection, result management, and the integration of smoking cessation. Resource limitations and tailored interventions for subpopulations with low-risk perception and non-adherence rates should be addressed with multidisciplinary expertise.

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.

Utilization of Diagnostic Procedures After Lung Cancer Screening in the National Lung Screening Trial

ObjectiveTo examine utilization patterns of diagnostic procedures after lung cancer screening among participants enrolled in the National Lung Screening Trial. MethodsUsing a sample of National Lung Screening Trial participants with abstracted medical records, we assessed utilization of imaging, invasive, and surgical procedures after lung cancer screening. Missing data were imputed using multiple imputation by chained equations. For each procedure type, we examined utilization within a year after the screening or until the next screen, whichever came first, across arms (low-dose CT [LDCT] versus chest X-ray [CXR]) and by screening results. We also explored factors associated with having these procedures using multivariable negative binomial regressions. ResultsAfter baseline screening, our sample had 176.5 and 46.7 procedures per 100 person-years for those with a false-positive and negative result, respectively. Invasive and surgical procedures were relatively infrequent. Among those who screened positive, follow-up imaging and invasive procedures were 25% and 34% less frequent in those screened with LDCT, compared with CXR. Postscreening utilization of invasive and surgical procedures was 37% and 34% lower at the first incidence screen compared with baseline. Participants with positive results at baseline were six times more likely to undergo additional imaging than those with normal findings. DiscussionUse of imaging and invasive procedures to evaluate abnormal findings varied by screening modality, with a lower rate for LDCT than CXR. Invasive and surgical workup were less prevalent after subsequent screening examinations compared with baseline screening. Utilization was associated with older age but not gender, race or ethnicity, insurance status, or income.

Deep learning-based age estimation from chest CT scans.

Medical imaging can be used to estimate a patient's biological age, which may provide complementary information to clinicians compared to chronological age. In this study, we aimed to develop a method to estimate a patient's age based on their chest CT scan. Additionally, we investigated whether chest CT estimated age is a more accurate predictor of lung cancer risk compared to chronological age. To develop our age prediction model, we utilized composite CT images and Inception-ResNet-v2. The model was trained, validated, and tested on 13,824 chest CT scans from the National Lung Screening Trial, with 91% for training, 5% for validation, and 4% for testing. Additionally, we independently tested the model on 1849 CT scans collected locally. To assess chest CT estimated age as a risk factor for lung cancer, we computed the relative lung cancer risk between two groups. Group 1 consisted of individuals assigned a CT age older than their chronological age, while Group 2 comprised those assigned a CT age younger than their chronological age. Our analysis revealed a mean absolute error of 1.84years and a Pearson's correlation coefficient of 0.97 for our local data when comparing chronological age with the estimated CT age. The model showed the most activation in the area associated with the lungs during age estimation. The relative risk for lung cancer was 1.82 (95% confidence interval, 1.65-2.02) for individuals assigned a CT age older than their chronological age compared to those assigned a CT age younger than their chronological age. Findings suggest that chest CT age captures some aspects of biological aging and may be a more accurate predictor of lung cancer risk than chronological age. Future studies with larger and more diverse patients are required for the generalization of the interpretations.

Recurrence After Complete Resection for Non-Small Cell Lung Cancer in the National Lung Screening Trial

BackgroundThe objective of this study was to evaluate patterns, predictors, and long-term outcomes of recurrent disease after complete resection for early-stage non-small cell lung cancer (NSCLC) using the National Lung Screening Trial (NLST). MethodsThe frequency of recurrence in patients with pathologic stage I-II NSCLC who underwent complete resection (lobectomy or bilobectomy) in the NLST was evaluated. Predictors of increased risk of recurrence were assessed by Fine-Gray competing risks regression. ResultsOf the 497 patients meeting study inclusion criteria, 94 experienced a recurrence—a rate of 4.9 (95% CI, 4.0-6.0) per 100 person-years. The 5-year cumulative incidence of recurrence was 20.1% (95% CI, 16.5%-23.9%). Most patients experienced recurrences at distant sites alone (n = 47 [50.0%]) or at both locoregional and distant sites (n = 30 [31.9%]). The median time from resection to recurrence was 18.8 (10.6-30.7) months. The incidence rate of recurrence was significantly lower among patients with lung cancer detected by low-dose computed tomography screening during one of the three screening rounds of the NLST when compared with patients with lung cancer detected by chest radiography screening and patients with lung cancer not detected by any form of screening (ie, those diagnosed after a negative or missed screening exam and those diagnosed during follow-up after the three screening rounds of the NLST were completed) (P < .001). Median survival (from the date of recurrence) of patients with pathologic stage I and stage II disease who had recurrences at locoregional, distant, or both sites was 63.0, 23.1, and 9.8 months and 28.9, 8.7, and 10.2 months, respectively. ConclusionsIn this analysis of NLST participants with completely resected stage I-II NSCLC, the 5-year cumulative incidence of recurrence was 20%. Nearly 82% of recurrences were at distant sites and associated with poor survival.

Overall and non-lung cancer incidence in the national lung screening trial (NLST) as indicators of potential for multi-cancer screening.

10633 Background: Single-cancer screening strategies address only one component of the overall population cancer burden. Even high-risk populations in single-cancer screening trials experience a considerable burden of other cancer incidence; therefore, we assessed non-lung cancer incidence in the NLST to evaluate the proportion of unscreened cancers. Methods: The NLST was a randomized controlled trial of screening for lung cancer in adults aged 55–74 years with ≥30 pack-years of smoking history, enrolled in 2002–2004 and actively followed through 2009. Incident cancers were identified via participant questionnaires, death certificates, and direct notifications, and confirmed by medical record abstraction. The present analysis includes first invasive primary cancers diagnosed after study randomization. Non-lung cancer incidence was similar between the two trial arms, which were therefore combined for analysis. Results: Among 53 229 subjects (median follow-up 6.5 years), the incidence of any first primary cancer was 1941 per 100 000 person-years, of which 1327 per 100 000 (68%) was non-lung cancer. Non-lung cancer incidence rates exceeded that for lung cancer in all 5-year age categories (especially at younger ages) and all quintiles of smoking pack-years (especially with fewer pack-years). After lung cancer, the most common cancers were other leading cancer types in the general US population, as well as other smoking-related cancers (Table). No recommended population-based screens exist for 54% of observed cancer cases (3332 of 6142, excluding lung, colon/rectum, female breast, and cervix). Conclusions: In the NLST, only 32% of first primary cancer incidence was lung; non-lung cancer comprised 68%, and most cases were cancer types with no generally recommended screening strategy. Even in a high-risk population, a single-cancer screening test misses most cancers, illustrating the value of multi-cancer screening tests that can detect a broad spectrum of cancers and, therefore, have the potential to address a currently inaccessible portion of cancer morbidity and mortality. [Table: see text]

Preliminary results from the Female Asian Nonsmoker Screening Study (FANSS).

8510 Background: Lung cancer (LC) is the leading cause of cancer death in Asian Americans and unfortunately the majority are diagnosed at advanced or late stages. In Asia, approximately 60-80% of female LC patients are never smokers. Current screening in the U.S. with low-dose CT (LDCT) Chest scans is offered only to current or former smokers based on the National Lung Screening Trial (NLST). The LC detection rate in NLST was 1.1%. The TALENT study, a LC screening study for high-risk nonsmokers in Taiwan reported a LC detection rate of 2.6% which included invasive adenocarcinoma (adeno), adeno in situ, minimally invasive adeno and adenosquamous carcinoma. Invasive adeno detection rate was 1.52%. We are conducting the ongoing FANSS in the U.S. to screen female Asian nonsmokers with LDCT Chest scans to evaluate the feasibility of a LC screening program in this population. We report preliminary results here. Methods: This is an IRB-approved, prospective, multicenter study (NYU, MGH, UCI). Inclusion criteria are women, age between 40-74 years old (yo), never smoked or smoked &lt;100 cigarettes in one's lifetime and identify as from Asian descent (report ancestry or race from the continent of Asia). Participants (pts) with history of LC or treatment of any cancer &lt;5 years ago are excluded. Following informed consent, eligible pts undergo a shared-decision making discussion prior to obtaining a baseline LDCT Chest scan, read according to Lung-RADS 1.1 with plan for annual LDCT for two additional years. A plasma-based assay to analyze cell-free DNA (cfDNA) fragments for early detection of cancer by Delfi Diagnostics (Baltimore, MD) is sent at the time of each scan. A questionnaire regarding ethnicity, family history and environmental exposures is collected at baseline. Results: From 3/1/21 to 1/15/23, 222 pts signed consent and 201 had a baseline LDCT at NYU. Age range was 40-74 yo and median age of 56.8 yo. 83 (41%) reported a family history of LC. Of 201 pts who completed a baseline LDCT, 87 (43%) were Lung-RADS 1, 101 (50%) were Lung-RADS 2, 6 (3%) were Lung-RADS 3 and 7 (3.5%) were Lung-RADS 4. 5 pts with Lung-RADS 3 and 3 pts with Lung-RADS 4 have solid, subsolid or groundglass nodules &gt;6mm that remain in close follow-up. 3 pts were diagnosed with invasive lung adeno for a LC detection rate of 1.5%; 2 are stage IIB and 1 is stage IIIC. All pts were surgically resected, EGFR mutation positive and are receiving adjuvant osimertinib. Analysis of the cfDNA fragmentation profiles is ongoing. Conclusions: Our data shows that LC screening in Asian female nonsmokers is feasible. Preliminary results demonstrate an invasive adeno detection rate comparable with TALENT and higher than in NLST. Early detection brings new meaning with the recent FDA approval for adjuvant targeted therapy in early stage LC. The expansion of LC screening guidelines to other high-risk populations warrants further attention. FANSS is continuing to accrue at additional U.S. sites this year. Clinical trial information: NCT05164757 .

Liquid biopsy in combination with PET/CT versus PET/CT alone in diagnosis of lung cancer in small nodules.

3072 Background: In the National Lung Screening Trial, almost two-thirds of patients that underwent an invasive procedure for diagnosis had benign findings, suggesting a significant number of patients were exposed to risk that could be avoided. Currently, to help distinguish malignant from benign nodules, Positron Emission Tomography/Computed Tomography (PET/CT) is used. However, PET/CT can be limited in its use for lesions less than 2 cm in size. The detection of circulating tumor DNA (ctDNA) and cell-free DNA (cfDNA) in the blood stream via liquid biopsy has demonstrated potential as a non-invasive approach in both screening and diagnostic tests. The purpose of this study is to assess whether combining liquid biopsy with PET/CT findings can better detect malignancy in small lung nodules 6-20 mm in size compared to PET/CT alone, possibly allowing earlier cancer detection and reduction in unneeded invasive diagnostic testing. Methods: This was a single-center prospective observational study for which the primary objective was to compare liquid biopsy plus PET/CT to PET/CT alone for detection of lung cancer in patients with newly discovered or newly growing solid lung nodules. Patients with nodules 6-20 mm in size who were scheduled to undergo 18F-Fluorodeoxyglucose PET/CT also underwent blood collection for plasma ctDNA/cfDNA testing using an amplicon-based next-generation sequencing assay incorporating molecular barcodes for error suppression. Tumor diagnosis was based on resolution of the nodule or tissue biopsy. For liquid biopsy, a multiomic model previously built using (1) cfDNA methylation frequencies, (2) ctDNA abundance, (3) somatic ctDNA mutations, and (4) cfDNA fragment size ratio was used to predict malignancy. For imaging, lesions with a maximum standardized uptake value of 2 or greater were considered malignant. Results: Between December 2021 and November 2022, 24 patients were enrolled. Of these, 13 had a definitive diagnosis and were included in the analysis with the remainder still undergoing serial imaging follow-up. Of the 13 patients, 9 were biopsy or surgically proven to be malignant, and 4 were benign based on biopsy or resolution. The median (IQR) nodule size was 14 mm (8-15). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for PET/CT were 0.67, 1, 1, and 0.57, respectively. For multiomic liquid biopsy alone, these values were the same. However, combining multiomic liquid biopsy and PET/CT, the sensitivity, specificity, PPV, and NPV were 0.89, 1, 1, and 0.8, respectively. Conclusions: Combining multiomic liquid biopsy with PET/CT results in improved sensitivity for the detection of early-stage malignant solid lung nodules by 33% compared to PET/CT alone with no reduction in specificity. This could result in improved selection of patients requiring biopsy. Further evaluation with a larger cohort is needed to confirm the findings.

Preparing CT imaging datasets for deep learning in lung nodule analysis: Insights from four well-known datasets

Background: Deep learning is an important means to realize the automatic detection, segmentation, and classification of pulmonary nodules in computed tomography (CT) images. An entire CT scan cannot directly be used by deep learning models due to image size, image format, image dimensionality, and other factors. Between the acquisition of the CT scan and feeding the data into the deep learning model, there are several steps including data use permission, data access and download, data annotation, and data preprocessing. This paper aims to recommend a complete and detailed guide for researchers who want to engage in interdisciplinary lung nodule research of CT images and Artificial Intelligence (AI) engineering.Methods: The data preparation pipeline used the following four popular large-scale datasets: LIDC-IDRI (Lung Image Database Consortium image collection), LUNA16 (Lung Nodule Analysis 2016), NLST (National Lung Screening Trial) and NELSON (The Dutch-Belgian Randomized Lung Cancer Screening Trial). The dataset preparation is presented in chronological order.Findings: The different data preparation steps before deep learning were identified. These include both more generic steps and steps dedicated to lung nodule research. For each of these steps, the required process, necessity, and example code or tools for actual implementation are provided.Discussion and conclusion: Depending on the specific research question, researchers should be aware of the various preparation steps required and carefully select datasets, data annotation methods, and image preprocessing methods. Moreover, it is vital to acknowledge that each auxiliary tool or code has its specific scope of use and limitations. This paper proposes a standardized data preparation process while clearly demonstrating the principles and sequence of different steps. A data preparation pipeline can be quickly realized by following these proposed steps and implementing the suggested example codes and tools.

The probability of lung cancer in patients with incidentally detected pulmonary nodules detected via low-dose computed tomography screening in China.

10569 Background: Low dose computed tomography (LDCT) screening has been proven to be effective in reducing lung cancer mortality, but the ensuing high false-positive and overdiagnosis rates shackle the effectiveness of lung cancer screening (LCS) in China. Nodule malignancy prediction models may be an applicable solution. Methods: We developed and internally validated the model using data from the ongoing Henan province Cancer Screening Program in Urban China (CanSPUC). From 2013 to 2021, 23031 heavy smokers underwent baseline screening with LDCT; 2553 participants were diagnosed with pulmonary nodules. Detailed questionnaire, physical assessment and follow-up were completed for all participants. We then externally validated the model using the National Lung Screening Trial (NLST) dataset consisting of 10485 lung nodules in 4660 participants. Multivariable Cox proportional risk regression models were used. Data analysis was performed from July 1, 2022, through December 31, 2022. Results: A total of 111 and 1089 lung cancer cases with a median follow-up duration of 3.7 years occurred in the Henan CanSPUC and NLST study, respectively. Age, gender, physical activity, consumption of pickled food, history of silicosis or pneumoconiosis, nodule type, size, calcification, and pleural retraction sign were included into the model. The AUC (95% CI) of the model for 5-year lung cancer risk was 0.857 (0.801, 0.914) and 0.847 (0.829, 0.881) in the training set and validation set, respectively. Compared with Mayo model, VA model, PKU model, and Brock model, the Henan CanSPUC model yield statistically better discriminatory performance (all P values &lt; 0.05). The model calibrated well across the deciles of predicted risk in both the overall population and all subgroups. Conclusions: The model developed and validated in this study may be used to estimate the probability of lung cancer in nodules detected at baseline LDCT, allowing more efficient risk-adapted follow-up in population-based LCS programs.

A multivariate analysis of low-dose CT lung cancer screening in a Michigan community hospital network.

10576 Background: Lung cancer is the leading cause of cancer deaths in the United States. The USPSTF recommended screening for lung cancer with low-dose chest computer tomography (LDCT) in 2013. This was based on the National Lung Screening Trial (NLST) which showed a 20% relative reduction in mortality from LDCT compared with chest radiography in high-risk individuals. The major NLST eligibility criteria were age 55-74, a 30 + pack year smoking history and current smoking status or having quit in the last 15 years. The impact of other risk factors on screening results besides inclusion criteria is unknown. We sought to identify other factors for the risk of lung cancer detection in screened individuals. Methods: This was a retrospective cohort study employing a Michigan based McLaren Hospital Network Lung Cancer Screening Program (LCSP) database from 2015-2022. Participant eligibility in screening was the USPSTF criteria except for an age range of 50-80. Analyses were stratified by BMI, smoking pack-years, insurance status, smoking status, sex, other CT findings, pre- vs. during COVID pandemic (before and after January 2020). Multivariate logistic regression analysis was used for odds ratios (OR) with 95% confidence intervals (CI) as estimates of the relative risk of an elevated Lung-RADS category ≥3 (high risk of detecting lung cancer). Results: A total of 2638 subjects, 49.91% females, 50.09% males, mean BMI 29.50 ± 7.17 (SD), pack-years 44.52 ±18.29, Medicaid 6.66%, Medicare 32.01%, Private Insurance 57.62%, Self-pay /unknown 3.71%. Current smoker 65.49%, Former smoker 34.51%, other CT findings 51.99%, pre-COVID 46.08%, during COVID pandemic 53.92%. Lung-RADS category 0 (0.04%), 1 (35.33%), 2, (50.61%), 3 (6.56%), 4a (4.85%), 4b (1.55%), 4x (1.06%). A statistically significant association with a Lung-RADS category ≥3 was found with lower BMI, OR=0.980 (95% CI, 0.969-0.991), p=0.0005, and female gender, OR=1.232 (95% CI, 1.039-1.462), p=0.0166 (Table). Conclusions: In our study cohort a lower-than-average BMI and also female gender were statistically correlated with an elevated Lung-RADS category ≥3. These findings might be incorporated into a risk stratification model, however further larger studies are needed to validate our findings. [Table: see text]

Leveraging social media to increase lung cancer screening awareness, knowledge and uptake among high-risk populations (The INSPIRE-Lung Study): study protocol of design and methods of a community-based randomized controlled trial

BackgroundLung cancer is the leading cause of cancer death for both men and women in the United States. The National Lung Screening Trial (NLST) demonstrated that low-dose computed tomography (LDCT) screening can reduce lung cancer mortality among high-risk individuals, but uptake of lung screening remains low. Social media platforms have the potential to reach a large number of people, including those who are at high risk for lung cancer but who may not be aware of or have access to lung screening.MethodsThis paper discusses the protocol for a randomized controlled trial (RCT) that leverages FBTA to reach screening-eligible individuals in the community at large and intervene with a public-facing, tailored health communication intervention (LungTalk) to increase awareness of, and knowledge about, lung screening.DiscussionThis study will provide important information to inform the ability to refine implementation processes for national population efforts to scale a public-facing health communication focused intervention using social media to increase screening uptake of appropriate, high-risk individuals.Trial registrationThe trial is registered at clinicaltrials.gov (#NCT05824273).

Significant Incidental Findings in the National Lung Screening Trial

Low-dose computed tomography (LDCT) lung screening has been shown to reduce lung cancer mortality. Significant incidental findings (SIFs) have been widely reported in patients undergoing LDCT lung screening. However, the exact nature of these SIF findings has not been described. To describe SIFs reported in the LDCT arm of the National Lung Screening Trial and classify SIFs as reportable or not reportable to the referring clinician (RC) using the American College of Radiology's white papers on incidental findings. This was a retrospective case series study of 26 455 participants in the National Lung Screening Trial who underwent at least 1 screening examination with LDCT. The trial was conducted from 2002 to 2009, and data were collected at 33 US academic medical centers. Significant incident findings were defined as a final diagnosis of a negative screen result with significant abnormalities that were not suspicious for lung cancer or a positive screen result with emphysema, significant cardiovascular abnormality, or significant abnormality above or below the diaphragm. Of 26 455 participants, 10 833 (41.0%) were women, the mean (SD) age was 61.4 (5.0) years, and there were 1179 (4.5%) Black, 470 (1.8%) Hispanic/Latino, and 24 123 (91.2%) White individuals. Participants were scheduled to undergo 3 screenings during the course of the trial; the present study included 75 126 LDCT screening examinations performed for 26 455 participants. A SIF was reported for 8954 (33.8%) of 26 455 participants who were screened with LDCT. Of screening tests with a SIF detected, 12 228 (89.1%) had a SIF considered reportable to the RC, with a higher proportion of reportable SIFs among those with a positive screen result for lung cancer (7632 [94.1%]) compared with those with a negative screen result (4596 [81.8%]). The most common SIFs reported included emphysema (8677 [43.0%] of 20 156 SIFs reported), coronary artery calcium (2432 [12.1%]), and masses or suspicious lesions (1493 [7.4%]). Masses included kidney (647 [3.2%]), liver (420 [2.1%]), adrenal (265 [1.3%]), and breast (161 [0.8%]) abnormalities. Classification was based on free-text comments; 2205 of 13 299 comments (16.6%) could not be classified. The hierarchical reporting of final diagnosis in NLST may have been associated with an overestimate of severe emphysema in participants with a positive screen result for lung cancer. This case series study found that SIFs were commonly reported in the LDCT arm of the National Lung Screening Trial, and most of these SIFs were considered reportable to the RC and likely to require follow-up. Future screening trials should standardize SIF reporting.

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 867: Identifying plasma exosomal small non-coding RNAs as potential biomarkers for lung cancer early detection

Abstract Introduction: Lung cancer is the second most common cancer and the leading cause of cancer death in the US and worldwide. The National Lung Screening Trial (NLST) through thoracic low-dose computed tomography (LDCT) scans has shown that screening decreases mortality by 20%, but nearly 40% of patients in the NLST had a positive screen at one point, with 96.4% of these being false positive. It is urgent to develop new non-invasive methods with high efficacy, such as blood molecular biomarkers as companion diagnostic tools to complement LDCT screening. Recent studies have shown that small non-coding RNAs (ncRNAs) play important roles in cancer development, and exosomes are actively secreted extracellular vesicles that contain proteins, lipids, and nucleic acids, including miRNAs and other classes of small ncRNAs. Our study aimed to screen and identify noval plasma exosomal small ncRNAs for lung cancer early detection. Methods: A supersensitive human plasma exosome isolation and purification as well as exosome small RNA sequencing were conducted on 233 human subjects including 117 lung cancer early-stage patients and 116 non-cancer subjects (54 benign and 62 normal samples). Sequencing reads were first processed by trimming off the adapters and low-quality bases using cutadapt. Following trimming, the sequences were aligned to the human genome (Genome Reference Consortium GRCh38) and mapped to different small ncRNA databases using STAR. Read counts for each RNA category (miRNA, piRNA, snRNA, snoRNA, tRNA, tsRNA) were calculated from the mapping results. Differential gene expression analysis was performed by the DESeq2 bioconductor package. The genes with Benjamini and Hochberg q-value &amp;lt; 0.05 were called differentially expressed. Based on the differential gene lists, diagnostic models were built by 10-fold cross validated Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression. Results: A total of 5470 small ncRNAs were initially identified, including 1094 miRNA, 2172 piRNA, 1185 snRNA, 376 snoRNA, 582 tRNA, 61 tsRNA. Using DESeq2 with an FDR q-value &amp;lt; 0.05 with a cut off of 1.5-fold change, we found 508 small ncRNAs significantly differentially expressed between cancer and non-cancer groups, consisting of 231 small ncRNAs up-regulated and 277 down-regulated in tumor. By diagnostic modeling, a set of 16 small ncRNAs were identified as promising biomarkers, which could differentiate patients with cancer from non-cancer with an area under the receiver operating (ROC) curve (AUC) of 96.71%. Using the same sixteen-smallRNA signature to distinguish cancer from benign, the AUC achieved 97.23%; while the same panel could distinguish cancer from normal with an AUC of 96.26%. Conclusion: This preliminary study demonstrated the potential diagnostic value of the exosomal small ncRNA biomarkers in lung cancer early detection. The results will be evaluated by more cases in the future study. Citation Format: Yuanyuan Fu, Masaki Nasu, Mayumi Jijiwa, Yu Chen, Zitong Gao, Isam Ibrahim, Youping Deng. Identifying plasma exosomal small non-coding RNAs as potential biomarkers for lung cancer early detection [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 867.

Abstract 5433: Representation learning for histological profiling of lung squamous premalignant lesions and tumors

Abstract Lung squamous cell carcinoma (LSCC) is preceded by the development of bronchial premalignant lesions (PMLs). PMLs progress through a series of histologic grades characterized by molecular and morphologic alterations. To enhance our understanding of PML progression to lung cancer, we are developing deep learning methods to quantitate PML histologic features and localize PML severity along the histologic disease spectrum. We leveraged whole slide images (WSIs) of lung tumors (lung adenocarcinoma, LUAD and LSCC) and non-involved adjacent (referred to as ‘normal’) tissue from the Clinical Proteomic Tumor Analysis Consortium (CPTAC), the National Lung Screening Trial (NLST) and the Cancer Genome Atlas (TCGA). The NLST WSIs were used for training a self-supervised contrastive learning (simCLR) model. A graph isomorphism network (GIN) trained on CPTAC WSIs was developed using the features from simCLR and perform WSI-level predictions on tissue subtype (LUAD, LSCC, normal). Model performance was assessed using area under the receiver operating curves (AUC). The model was used to learn WSI-level features of endobronchial biopsies of PMLs from two datasets (PCGA, n=365 samples spanning all histologic grades, and CIS, n=112, only high-grade samples). The features from the final layer of the GIN were used to compute principal components (PCs) and visualize the relationships between samples using t-SNE and UMAP. To interpret how the GIN processes WSI data, we performed gradient-based class activation mapping (Grad-CAMs) on the graphs. High model performance was observed on the CPTAC test dataset (Accuracy = 87%, AUC &amp;gt;= 0.89) but dropped slightly on the TCGA dataset (Accuracy = 72%, AUC &amp;gt;= 0.75). The GIN-CAMs identified WSI regions that were highly associated with the output label based on expert pathologist annotation on TCGA cases. Clustering revealed distinct groupings of normal, LUAD and LSCC subtypes (PC1 &amp; PC2, p&amp;lt;0.01). Most WSIs from the PCGA data clustered with normal tissues (PC1 &amp; PC2: p&amp;lt;0.01 for PCGA v LUAD, PCGA v LSCC). However, a portion of the CIS samples were closely grouped with LSCC tumor cases (74% of CIS WSIs were classified as LSCC tumors. PC1 &amp; PC2: p&amp;lt;0.01 for CIS v LUAD; PC1: p&amp;lt;0.01 for CIS v LSCC). Additionally, the feature distribution of CIS samples that progressed towards LUSC tumors were significantly different than those that regressed (PC1 &amp; PC3: p &amp;lt; 0.05 for Progressive v Regressive). The GIN recognizes and generates robust features that are distinctly and consistently observed in the different histologic groups across multiple cohorts. These features are also able to stratify progressive and regressive high-grade lesions. The stratification of PMLs is an important step in designing novel interception strategies to prevent the development of lung cancer, and our results suggest that pathology data may be efficacious to include in future biomarkers. Citation Format: Rushin Gindra, Yi Zheng, Regan Conrad, Emily Green, Sarah Mazzilli, Ehab Billatos, Mary Reid, Eric Burks, Vijaya Kolachalama, Jennifer E. Beane. Representation learning for histological profiling of lung squamous premalignant lesions and tumors. [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 5433.