Diagnosis Of Lung Cancer Research Articles

Clinical utility of upfront liquid biopsy in non-small cell lung cancer in the community setting.

316 Background: Molecular diagnosis of advanced non-small cell lung cancer (NSCLC) is critical to guide therapeutic decisions. Blood next generation sequencing (NGS) allows for non-invasive screening for targetable genetic abnormalities. Many studies have validated using the liquid biopsies to screen for resistant mutations at the emergence of treatment failure. However, the clinical utility of liquid biopsy for newly diagnosed NSCLC, as part of a comprehensive molecular profiling, is an area of active research. Methods: Retrospective analysis of the molecular landscape and clinical outcomes in NSCLC patients who had liquid biopsy done between December of 2017 to February of 2022. Patients had EGFR, ALK, ROS-1, and BRAF tested in the tissue, and underwent liquid biopsy for broad molecular profiling. The plasma genotyping was conducted using a 73-gene panel by Guardant 360 test. Actionable mutations were stratified accordingly to the OncoKB database based on the level of preclinical and clinical evidence of drug targetability. Survival analysis was conducted by August of 2023. Results: We enrolled 122 NSCLC patients who 84% were treatment-naïve and 78% had the liquid biopsy performed at the initial diagnosis. We found that 86 patients had oncogenic mutations and 43 (35%) had actionable mutations, with 37 (29%) falling under level 1 of evidence mutations. EGFR and KRAS G12C mutations were the two most common level 1 genetic abnormalities (23% and 11% of all actionable mutations, respectively). We identified two germline BRCA carriers who were referred for genetic counselling. 25 patients (20%) received treatment based on the liquid biopsy results, with 15 (12%) achieving partial response. Six patients had liquid biopsy done at progression of EGFR or MET therapies, which identified a resistance mechanism in three of them. Conclusions: Upfront liquid biopsy identified a significant proportion of patients with targetable mutations. Here, we reported actionable mutations in 35% of our patients who 84% had no prior systemic treatment. Our results complement prior literature and shows that blood next generation sequencing can be a useful tool to guide upfront therapeutical options at the community healthcare setting.

EP.04D.03 Shifting Stages: Implications of a Global Pandemic on Cancer Care and A Multipronged Approach to Early Diagnosis for Non-Small Cell Lung Cancer

EP.04D.03 Shifting Stages: Implications of a Global Pandemic on Cancer Care and A Multipronged Approach to Early Diagnosis for Non-Small Cell Lung Cancer

Robotic One Anesthetic Diagnosis and Treatment: A Novel Program for Lung Cancer Care

BackgroundEarly diagnosis and treatment of lung cancer can improve patient outcomes. To reduce the delay between diagnosis and treatment, we initiated a Robotic One Anesthetic Diagnosis and Treatment (ROADAT) program for lung nodules that combined robotic navigational bronchoscopy, liquid rapid on-site evaluation (ROSE), and surgical resection in the same setting. MethodsA retrospective review was conducted of patients who underwent the ROADAT procedure from December 1, 2021, through January 31, 2024. These patients were compared with sequential controls from the same time window, who would have been ROADAT candidates but underwent diagnosis and treatment using a standard lung nodule workup. ResultsThere were 36 patients in the ROADAT group and 35 controls. Of the 36 ROADAT patients, 31 (86%) underwent lung resections. When final pathology was available, ROSE was concordant in 91% (29 of 32) of cases. Compared with robotic lobectomies alone, ROADAT procedures that resulted in lobectomies added 73 minutes (P = .003). However, total operating room time for when lobectomy and biopsy were done in separate settings was 54 minutes shorter in the ROADAT group (P = .028). Direct (ROADAT, $19,244; controls, $21,737; P = .004) procedure costs and total (ROADAT, $26,668; controls, $29,882; P = .005) procedure costs were lower in the ROADAT group for patients who underwent lobectomy. Time from detection to resection was 35 days shorter in the ROADAT group (ROADAT, 82 days; controls, 117 days; P = .001). ConclusionsThe ROADAT procedure reduced time from nodule detection to surgical resection compared with the standard approach. Liquid-based ROSE had high concordance with final pathology. Total operative time, direct costs, and total costs were lower for the ROADAT procedure.

Diagnosis and management of multiple primary lung cancer.

Multiple primary lung cancer (MPLC), can be categorized as synchronous multiple primary lung cancer (sMPLC) and metachronous multiple primary lung cancer (mMPLC), which are becoming increasingly common in clinical practice. A precise differential diagnosis between MPLC and intrapulmonary metastases (IPM) is essential for determining the appropriate management strategy. MPLC is primarily diagnosed through histology, imaging, and molecular methods. Imaging serves as an essential foundation for preoperative diagnosis, while histology is a critical tool for establishing a definitive diagnosis. As molecular biology advances, the diagnosis of MPLC has stepped into the era of molecular precision. Surgery is the preferred treatment approach, with stereotactic radiotherapy and ablation being viable options for unresectable lesions. Targeted therapy and immunotherapy can be considered for specific patients. A multidisciplinary team approach to evaluation and the application of combination therapy can benefit more patients. Looking ahead, the development of more authoritative guidelines will be instrumental in streamlining the diagnosis and management of MPLC.

Association between Medicaid expansion and stage at diagnosis and treatment rates of non-small cell lung cancer.

87 Background: Lung cancer is a leading cause of cancer mortality in the United States, with Non-small cell lung cancer (NSCLC) being the most common subtype. Under the Affordable Care Act, 39 states adopted Medicaid Expansion (ME) in 2014, while 12 did not. ME improved insurance coverage and lung cancer screening rates. This study examined the association between ME and NSCLC stage at diagnosis and treatment rates. Methods: We performed a retrospective observational study using Surveillance, Epidemiology, and End Results (SEER) database including patients aged 55-64 diagnosed with NSCLC from 2007 to 2021. Multivariate logistic regression analysis was performed to compare the odds of Early NSCLC diagnosis (Stages 1 and 2) and odds of treatment for early NSCLC between ME states and non-ME states before and after 2014. Results: During the study period, 191,785 patients were diagnosed with NSCLC, of which 42,338 were early NSCLC. The odds of early NSCLC diagnosis and treatment in early NSCLC were similar in ME and non-ME states before 2014. After 2014, the odds of early NSCLC diagnosis were significantly higher in ME states [AOR-1.53 (1.45-1.63)] compared to non-ME states [AOR-1.34 (1.25-1.44)]. However, there was no significant difference in odds of treatment for early NSCLC in ME states [1.36 (1.01 -1.81) and non-ME states 1.25 (0.93-1.68) after 2014. Conclusions: We conclude that ME is associated with increased odds of early-stage diagnosis of NSCLC. However, the odds of receipt of treatment were not affected. These findings highlight the benefits of ME as an effective strategy to improve access to care, but also underscore that other barriers to receiving treatment exist despite early diagnosis. Further studies are warranted to identify the barriers and potential limitations of policy change to improve outcomes. Logistic analysis for odds of early-stage NSCLC diagnosis and odds of treatment in early NSCLC. Early NSCLC Diagnosis Odds Ratio p-value 95% Confidence Interval Non – ME States (Pre- Expansion) 1.00 (Reference) Non – ME States (Post- Expansion) 1.34 0.00 1.25 - 1.44 ME States (Pre- Expansion) 1.04 0.16 0.98 - 1.11 ME States (Post- Expansion) 1.53 0.00 1.45 - 1.63 Treatment of Early NSCLC Non - ME States (Pre- Expansion) 1.00 (Reference) Non-ME States (Post- Expansion) 1.25 0.13 0.93-1.68 ME States (Pre- Expansion) 1.40 0.05 1.00 - 1.95 ME States (Post- Expansion) 1.36 0.04 1.01- 1.81 NSCLC- Non-Small Cell Lung Cancer, ME- Medicaid Expansion.

FocalNeXt: A ConvNeXt augmented FocalNet architecture for lung cancer classification from CT-scan images

Early and accurate diagnosis of lung cancer, a life-threatening disease, is critical to the successful treatment of patients with the disease. On the other hand, it is well known that the integration of computer-aided diagnosis (CAD) systems into the diagnostic workflow facilitates the process and improves the diagnostic results of automated systems by reducing the difficulties associated with human observation. In this paper, we present FocalNeXt, a new ConvNeXt-augmented FocalNet architecture specifically designed for automated lung cancer detection from computed tomography (CT) scan images. By combining the strong attention mechanism of FocalNet and the feature extraction mechanism of ConvNeXt within the vision transformer paradigm, FocalNeXt aims to improve diagnostic accuracy. Our comprehensive evaluation of a publicly available Iraq-Oncology Teaching Hospital/National Center for Cancer Diseases (IQ-OTH/NCCD) CT-scan dataset includes rigorous comparisons and an ablation study. FocalNeXt achieved an accuracy of 99.81%, outperforming the state-of-the-art methods. The model also excelled in sensitivity (99.78%), recall (99.36%), and F1-score (99.56%), positioning FocalNeXt as a leading model for lung cancer detection. The ablation study further demonstrated its efficacy and underscored the robustness of FocalNeXt in different configurations. The results underline its potential to contribute to advances in medical imaging and personalized healthcare.

P4.17F.02 Sociodemographic and Clinical Predictors of Diagnosis of Lung Cancer in the Emergency Department in a Large, Integrated Health System

P4.17F.02 Sociodemographic and Clinical Predictors of Diagnosis of Lung Cancer in the Emergency Department in a Large, Integrated Health System

Using a Polygenic Risk Score to Improve the Risk Prediction of Non-Small Cell Lung Cancer in Taiwan.

Low-dose computed tomography (LDCT) can help reducing lung cancer mortality. In Taiwan, the existing screening criteria revolve around smoking habits and family history of lung cancer. The role of genetic variation in non-small cell lung cancer (NSCLC) development is increasingly recognized. In this study, we aimed to investigate the potential benefits of polygenic risk scores (PRSs) in predicting NSCLC and enhancing the effectiveness of screening programs. We conducted a retrospective cohort study that included participants without prior diagnosis of lung cancer and later received LDCT for lung cancer screening. Genetic data for these participants were obtained from the project of Taiwan Precision Medicine Initiative. We adopted the model of genome-wide association study-derived PRS calculation using 19 susceptibility loci associated with the risk of NSCLC as reported by Dai et al. We studied a total of 2,287 participants (1,197 male, 1,090 female). More female participants developed NSCLC during the follow-up period (4.4% v 2.5%, P = .015). The only risk factor of NSCLC diagnosis among male participants was age. Among female participants, independent risk factors of NSCLC diagnosis were age (adjusted hazard ratio [aHR], 1.08 [95% CI, 1.04 to 1.11]), a family history of lung cancer (aHR, 3.21 [95% CI, 1.78 to 5.77]), and PRS fourth quartile (aHR, 2.97 [95% CI, 1.25 to 7.07]). We used the receiver operating characteristics to show an AUC value of 0.741 for the conventional model. With the further incorporation of PRS, the AUC rose to 0.778. The evaluation of PRS for NSCLC prediction holds promise for enhancing the effectiveness of lung cancer screening in Taiwan especially in women. By incorporating genetic information, screening criteria can be tailored to identify individuals at higher risks of NSCLC.

Rapid Detection of Explicit Volatile Organic Compounds for Early Diagnosis of Lung Cancer Using MoSi2N4 Monolayer.

In this study, we investigate the adsorption of MoSi2N4) and MoSi2N4-VNtowards five potential lung cancer volatile organic compounds (VOCs).Density functional theory calculations reveal that MoSi2N4 weakly adsorb the mentioned VOCs, whereas introduction of nitrogen vacancies significantly enhances the adsorption energies ([[EQUATION]]), both in gas phase and aqueous medium. The MoSi2N4-VN monolayers exhibit a reduced bandgap and facilitate charge transfer upon VOCs adsorption, resulting in enhanced [[EQUATION]] values of -0.83, -0.76, -0.49, -0.61, and -0.50 eV for 2,3,4-trimethyl hexane, 4-methyl octane, o-toluidine, Aniline, and Ethylbenzene, respectively. Bader charge analysis and spin-polarized density of states (SPDOS) elucidate the charge redistribution and hybridization between MoSi2N4-VN and the adsorbed VOCs. The work function of MoSi2N4-VN is significantly reduced upon VOCs adsorption due to induced dipole moments, enabling smooth charge transfer and selective VOCs sensing. Notably, MoSi2N4-VN monolayers exhibit sensor responses ranging from 16.2% to 26.6% towards the VOCs, with discernible selectivity. Importantly, the recovery times of the VOCs desorption is minimal, reinforcing the suitability of MoSi2N4-VN as a rapid, and reusable biosensor platform for efficient detection of lung cancer biomarkers. Thermodynamic analysis based on Langmuir adsorption model shows improved adsorption and detection capabilities MoSi2N4-VN under diverse operating conditions of temperatures and pressures.

EP.05A.08 The Accuracy of Bronchoscopy and Transthoracic Needle Biopsy for Diagnosing Lung Cancer - A Systematic Review

EP.05A.08 The Accuracy of Bronchoscopy and Transthoracic Needle Biopsy for Diagnosing Lung Cancer - A Systematic Review

Brain metastases in newly diagnosed lung cancer: epidemiology and conditional survival.

The brain serves as the primary site for metastasis in patients with both non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC). The presence of lung cancer with brain metastasis (LCBM) is a debilitating condition associated with considerable morbidity and mortality. The objective of this study was to assess the incidence and survival rates of LCBM in the United States population. We analyzed a total of 9,212 patients diagnosed with LCBM between 2010 and 2015, extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Our analysis assessed the incidence, relative survival, and conditional survival (CS) of LCBM. We utilized the Kaplan-Meier method to estimate overall survival and determine CS at year y+x after x years of survival, following the formula CS(y|x) = CS(y+x)/CS(x). Prognostic factor selection was performed using the least absolute shrinkage and selection operator (LASSO) regression approach, and multivariate Cox regression was employed to demonstrate the impact of these predictors on outcomes and construct a CS-based nomogram. The overall age-adjusted incidence rate of LCBM was 5.82 cases per 100,000, with a slight decline observed during our study period. Patient relative survival showed a continuous decline with increasing age. CS analysis revealed that the 5-year CS rate for patients initially diagnosed with LCBM adjusted from 3% to 13%, 28%, 52%, and 73% over successive years of survival (1-4 years). Identified predictors included age at diagnosis, sex, race, tumor size, tumor grade, surgery, radiotherapy, and chemotherapy. These predictors, along with the CS formula, were employed to develop a CS-based nomogram for real-time prognosis prediction. Calibration curve, area under the time-dependent receiver operating characteristic (ROC) curve, concordance index (c-index), and decision curve analysis (DCA) demonstrated the model's strong predictive capabilities. This study deepened our understanding of LCBM patients, summarizing their epidemiological characteristics and CS patterns. We successfully developed a novel CS-based nomogram model for dynamic survival estimation, offering real-time and personalized prognostic information that is clinically valuable.

Association of area-level broadband and non-small cell lung cancer diagnosis and guideline-concordant care in the US.

96 Background: Access to broadband is a social determinant of health facilitating connections with healthcare services and other resources and alleviating burdens of travel distance, time, and cost. We used nationwide data to examine the association of zip code-level broadband usage and stage at diagnosis, as well as the receipt of quality care among individuals newly diagnosed with non-small cell lung cancer (NSCLC). Methods: Individuals ≥18 years newly diagnosed with first primary NSCLC between April - December of 2019-2021, were identified from the National Cancer Database (January - March were excluded to match the COVID-19 timeframe in 2020). Broadband usage percentage (the percentage of individuals per zip code with access to fixed broadband at speeds of 25 Mbps/3 Mbps in 2020) were combined with the cohort based on individual’s residential zip code. Multivariable logistic regression models examined associations between broadband usage quartiles and stage at diagnosis, as well as the receipt of guideline-concordant care based on National Comprehensive Cancer Network guidelines, including surgery and radiation, ≥10 lymph nodes sampling, chemotherapy, and neoadjuvant chemoradiation (NACR). Analyses were conducted overall and stratified by metropolitan status and year, adjusting for age group, sex, race and ethnicity, comorbidity, distance to facility, state, and zip code-level social deprivation index (SDI). Results: Cohort included 305,198 individuals newly diagnosed with NSCLC. Individuals living in areas with the highest broadband usage were more likely to be aged 65-74 years, female, non-Hispanic White, with Medicare coverage, living in metropolitan and low SDI areas. In adjusted models, individuals living in areas with the highest broadband usage were more likely to be diagnosed with stage I NSCLC (adjusted odds ratio [AOR] = 1.10, 95% confidence interval [95%CI] = 1.01-1.20) and to receive guideline-concordant NACR (AOR = 1.38, 95%CI = 1.09-1.73) compared with those living in areas with the lowest broadband usage in 2020. When stratified by metropolitan status, associations of broadband and diagnosis and receipt of NACR were only observed among individuals living in non-metropolitan areas with medium-low and medium-high broadband usage, respectively (AOR = 1.15, 95%CI = 1.01-1.32; AOR = 1.80, 95%CI = 1.25-2.58). No significant associations were identified for other care modalities or during 2019 and 2021. Conclusions: Increased access to broadband usage is associated with earlier stage NSCLC diagnosis and receiving guideline-concordant NACR in 2020. Our results suggest that increased broadband access, combined with the expansion of telehealth services during the COVID-19 pandemic, may have improved telehealth utilization throughout 2020, leading to enhanced quality of cancer care, particularly benefiting individuals residing in non-metropolitan areas.

Evaluation of completeness of commonly used data elements for clinical trial eligibility criteria using a registry-enhanced data collection process: Results from patients with non-small cell lung cancer (NSCLC) and colorectal cancer (CRC) at three community oncology practices.

346 Background: Data elements required for clinical trial eligibility assessments are often unstandardized and incomplete in EHR making clinical research participant identification a challenging and biased process. Moreover, the absence of new patient records and disease codes results in monitoring large patient populations until records are completed. These issues can lead to missed opportunities for timely patient identification. N-Power Medicine (NPM) developed a registry-enhanced oncology network platform for unbiased, and real-time quality data collection. We report from this platform on completeness rates of key variables for eligibility assessment in NSCLC and CRC clinical trials. Methods: NPM’s platform consists of 3 parts: patient registry (Kaleido), onsite and remote staff to consent patients and ensure data completeness, quality and standardization, and technology to structure and process the data. Patients with NSCLC and CRC from 3 community oncology practices across different US regions were enrolled in Kaleido and visited the clinic between February and May 2024. Data for 11 key variables were abstracted. To minimize the need for extensive monitoring until a diagnosis was confirmed, we developed a machine learning (ML) model for lung cancer using non-C34 ICD10 codes to construct multiple classes of tunable predictive models. Results: A total of 728 (NSCLC=281, CRC=447) patients were evaluated. The male-to-female ratio was 1 for NSCLC and 1.3 for CRC and 90% of NSCLC and 68% of CRC patients were over 60 years of age. The platform achieved 99% completeness for critical data points such as diagnosis date, histology, metastatic status, recent systemic anti-cancer treatments, ECOG performance status and other cancer diagnoses with diagnosis date. 47% of NSCLC and 43% of CRC patients had metastatic disease at the time of evaluation. Over 90% of patients with metastasis had their current therapy line and relevant genomic test results successfully documented. Imaging-based cancer status completeness was 95% for NSCLC and 89% for CRC. The ML model would identify over 50% of all patients that would eventually receive a lung cancer diagnosis, with over 80% specificity up to 3 weeks before a diagnosis. Conclusions: NPM’s platform demonstrated high completeness rates for the majority of essential data elements, improving the inclusion of all eligible patients in clinical trials. The predictive ML model shows promise in identifying suitable patients before an ICD10 code is assigned, reducing the need for extensive monitoring and ensuring timely completeness of necessary assessments for eligibility. Importantly, these advancements were achieved without adding financial or resource burdens to clinical workflows.

P1.01B.04 A Highly Unmet Need: Patterns of Lung Cancer Screening in Head and Neck Cancer Survivors Preceding a Lung Cancer Diagnosis

P1.01B.04 A Highly Unmet Need: Patterns of Lung Cancer Screening in Head and Neck Cancer Survivors Preceding a Lung Cancer Diagnosis

P2.04B.11 Factors Associated with Early-Stage Lung Cancer Diagnosis (ESLCD) In Brazil - 2013 - 2019: A Multivariable Analysis

P2.04B.11 Factors Associated with Early-Stage Lung Cancer Diagnosis (ESLCD) In Brazil - 2013 - 2019: A Multivariable Analysis

EP.04C.08 The Potential of Chest X-Ray Artificial Intelligence Detecting Missed Lung Cancer Diagnosis in a Community Based Cancer in Thailand

EP.04C.08 The Potential of Chest X-Ray Artificial Intelligence Detecting Missed Lung Cancer Diagnosis in a Community Based Cancer in Thailand

Bronchial washing fluid sequencing is useful in the diagnosis of lung cancer with necrotic tumor

BackgroundEarly-stage lung cancers detected by low-dose computed tomography (CT) often require confirmation through invasive procedures due to the absence of endobronchial lesions. This study assesses the diagnostic utility of bronchial washing fluid (BW) sequencing, a less invasive alternative, aiming to identify patient characteristics most suited for this approach. MethodsFrom June 2017 to March 2018, we conducted a prospective cohort study by enrolling patients with incidental lung lesions suspected of early-stage lung cancer at two independent hospitals, and 114 were diagnosed with lung cancer while 50 were diagnosed with benign lesions. BW sequencing was performed using a targeted gene panel, and the clinical characteristics of patients detected with cancer through sequencing were identified. ResultsMalignant cells were detected in 33 patients (28.9 %) through BW cytology. By applying specificity-focused mutation criteria, BW sequencing classified 42 patients (36.8 %) as having cancer. Among the cancer patients who were BW sequencing positive and BW cytology negative, 15 patients (75.0 %) showed necrosis on CT. The sensitivity of BW sequencing was particularly enhanced in patients with necrotic tumors, reaching 75 %. ConclusionsBW sequencing presents a viable, non-invasive diagnostic option for early-stage lung cancer, especially valuable in patients with necrotic lesions. By potentially reducing the reliance on more invasive diagnostic procedures, this method could streamline clinical workflows, decrease patient burden, and improve overall diagnostic efficiency.

Gold Nanocluster-Based Capture-Detection Platform for Clinical Diagnosis of Early Lung Cancer

Gold Nanocluster-Based Capture-Detection Platform for Clinical Diagnosis of Early Lung Cancer

The American Cancer Society National Lung Cancer Roundtable strategic plan: Optimizing strategies for lung nodule evaluation and management.

Lung nodules are frequently detected on low-dose computed tomography scans performed for lung cancer screening and incidentally detected on imaging performed for other reasons. There is wide variability in how lung nodules are managed by general practitioners and subspecialists, with high rates of guideline-discordant care. This may be due in part to the level of evidence underlying current practice guideline recommendations (primarily based on findings from uncontrolled studies of diagnostic accuracy). The primary aims of lung nodule management are to minimize harms of diagnostic evaluations while expediting the evaluation, diagnosis, and treatment of lung cancer. Potentially useful tools such as lung cancer probability calculators, automated methods to identify patients with nodules in the electronic health record, and multidisciplinary team evaluation are often underused due to limited availability, accessibility, and/or provider knowledge. Finally, relatively little attention has been paid to identifying and reducing disparities among individuals with screening-detected or incidentally detected lung nodules. This contribution to the American Cancer Society National Lung Cancer Roundtable Strategic Plan aims to identify and describe these knowledge gaps in lung nodule management and propose recommendations to advance clinical practice and research. Major themes that are addressed include improving the quality of evidence supporting lung nodule evaluation guidelines, strategically leveraging information technology, and placing emphasis on equitable approaches to nodule management. The recommendations outlined in this strategic plan, when carried out through interdisciplinary efforts with a focus on health equity, ultimately aim to improve early detection and reduce the morbidity and mortality of lung cancer. PLAIN LANGUAGE SUMMARY: Lung nodules may be identified on chest scans of individuals who undergo lung cancer screening (screening-detected nodules) or among patients for whom a scan was performed for another reason (incidental nodules). Although the vast majority of lung nodules are not lung cancer, it is important to have evidence-based, standardized approaches to the evaluation and management of a lung nodule. The primary aims of lung nodule management are to diagnose lung cancer while it is still in an early stage and to avoid unnecessary procedures and other harms.

Baseline computed tomography imaging findings could assist in early diagnosis of visceral pleural invasion for newly discovered early subpleural non-small cell lung cancer: T1 or T2.

Preoperative accurate visceral pleural infiltration (VPI) diagnosis for T1-size non-small cell lung cancer (NSCLC) is significant for clinical decision-making. The study aimed to explore the diagnostic efficacy of computed tomography (CT) imaging features and serum biomarkers in diagnosing VPI in newly discovered subpleural NSCLC ≤3 cm. There were 447 patients with NSCLC ≤3 cm retrospectively enrolled and assigned to the VPI group (n=81) and the non-VPI group (n=366) based on elastic fiber staining results. The serum biomarkers and CT imaging features were obtained for each subject. Univariate and multivariate analyses were used to identify the independent predictors for VPI. Area under the receiver operating characteristic (ROC) curve (AUC) was used to evaluate the diagnostic performance of each independent predictor and combined predictors in predicting VPI, with performance compared using the DeLong test. For tumor biomarkers, the VPI group had a significantly higher percentage of cases with abnormal carcino-embryonic antigen (CEA) level, cytokeratin 19 fragment (CYFRA21-1) level, and pro-gastrin-releasing peptide (ProGRP) level than that of the non-VPI group (P<0.001, P=0.003, P=0.004). However, in multivariate analysis, only the lesion-pleura relationship patterns type Ia [odds ratio (OR) =20.689; 95% confidence interval (CI): 5.058-84.622; P<0.001], type Ib (OR =5.155; 95% CI: 1.178-22.552; P=0.03), type II (OR =7.154; 95% CI: 1.733-29.53; P=0.007) with type III as reference, solid lesion density (OR =9.954; 95% CI: 4.976-19.911; P<0.001) with part-solid density as reference were identified as the independent predictors for VPI. In predicting VPI, the combined model (AUC =0.885) significantly outperformed models based on lesion density (AUC =0.833) and lesion-pleura relationship patterns (AUC =0.655) (all P<0.001). The CT predictors for VPI in patients with subpleural NSCLC (≤3 cm) were lesion density and lesion-pleura relationship patterns (pleural attachment and indentation), but not serum tumor biomarkers.