Lung Cancer Risk Prediction Models Research Articles

Evaluation of risk prediction models to select lung cancer screening participants in Europe: a prospective cohort consortium analysis

Lung cancer risk prediction models might efficiently identify individuals who should be offered lung cancer screening. However, their performance has not been comprehensively evaluated in Europe. We aimed to externally validate and evaluate the performance of several risk prediction models that predict lung cancer incidence or mortality in prospective European cohorts. We analysed 240 137 participants aged 45-80 years with a current or former smoking history from nine European countries in four prospective cohorts from the pooled database of the Lung Cancer Cohort Consortium: the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study (Finland), the Nord-Trøndelag Health Study (Norway), CONSTANCES (France), and the European Prospective Investigation into Cancer and Nutrition (Denmark, Germany, Italy, Spain, Sweden, the Netherlands, and Norway). We evaluated ten lung cancer risk models, which comprised the Bach, the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial 2012 model (PLCOm2012), the Lung Cancer Risk Assessment Tool (LCRAT), the Lung Cancer Death Risk Assessment Tool (LCDRAT), the Nord-Trøndelag Health Study (HUNT), the Optimized Early Warning Model for Lung Cancer Risk (OWL), the University College London-Death (UCLD), the University College London-Incidence (UCLI), the Liverpool Lung Project version 2 (LLP version 2), and the Liverpool Lung Project version 3 (LLP version 3) models. We quantified model calibration as the ratio of expected to observed cases or deaths and discrimination using the area under the receiver operating characteristic curve (AUC). For each model, we also identified risk thresholds that would screen the same number of individuals as each of the US Preventive Services Task Force 2021 (USPSTF-2021), the US Preventive Services Task Force 2013 (USPSTF-2013), and the Nederlands-Leuvens Longkanker Screenings Onderzoek (NELSON) criteria. Among the participants, 1734 lung cancer cases and 1072 lung cancer deaths occurred within five years of enrolment. Most models had reasonable calibration in most countries, although the LLP version 2 overpredicted risk by more than 50% in eight countries (expected to observed ≥1·50). The PLCOm2012, LCDRAT, LCRAT, Bach, HUNT, OWL, UCLD, and UCLI models showed similar discrimination in most countries, with AUCs ranging from 0·68 (95% CI 0·59-0·77) to 0·83 (0·78-0·89), whereas the LLP version 2 and LLP version 3 showed lower discrimination, with AUCs ranging from 0·64 (95% CI 0·57-0·72) to 0·78 (0·74-0·83). When pooling data from all countries (but excluding the HUNT cohort), 33·9% (73 313 of 216 387) of individuals were eligible by USPSTF-2021 criteria, which included 74·8% (1185) of lung cancers and 76·3% (730) of lung cancer deaths occurring over 5 years. Fewer individuals were selected by USPSTF-2013 and NELSON criteria. After applying thresholds to select a population of equal size to USPSTF-2021, the PLCOm2012, LCDRAT, LCRAT, Bach, HUNT, OWL, UCLD, and UCLI, models identified 77·6%-79·1% of future cases, although they selected slightly older individuals compared with USPSTF-2021 criteria. Results were similar for USPSTF-2013 and NELSON. Several lung cancer risk prediction models showed good performance in European countries and might improve the efficiency of lung cancer screening if used in place of categorical eligibility criteria. US National Cancer Institute, l'Institut National du Cancer, Cancer Research UK.

A Lung Cancer Risk Prediction Model from Healthy Korean Adults: A Single Center Cohort Study

A Lung Cancer Risk Prediction Model from Healthy Korean Adults: A Single Center Cohort Study

Enhancing Early Lung Cancer Diagnosis: Predicting Lung Nodule Progression in Follow-Up Low-Dose CT Scan with Deep Generative Model.

Early diagnosis of lung cancer can significantly improve patient outcomes. We developed a Growth Predictive model based on the Wasserstein Generative Adversarial Network framework (GP-WGAN) to predict the nodule growth patterns in the follow-up LDCT scans. The GP-WGAN was trained with a training set (N = 776) containing 1121 pairs of nodule images with about 1-year intervals and deployed to an independent test set of 450 nodules on baseline LDCT scans to predict nodule images (GP-nodules) in their 1-year follow-up scans. The 450 GP-nodules were finally classified as malignant or benign by a lung cancer risk prediction (LCRP) model, achieving a test AUC of 0.827 ± 0.028, which was comparable to the AUC of 0.862 ± 0.028 achieved by the same LCRP model classifying real follow-up nodule images (p = 0.071). The net reclassification index yielded consistent outcomes (NRI = 0.04; p = 0.62). Other baseline methods, including Lung-RADS and the Brock model, achieved significantly lower performance (p < 0.05). The results demonstrated that the GP-nodules predicted by our GP-WGAN model achieved comparable performance with the nodules in the real follow-up scans for lung cancer diagnosis, indicating the potential to detect lung cancer earlier when coupled with accelerated clinical management versus the current approach of waiting until the next screening exam.

Effect of novel polygenetic variant of HUNT lung cancer model on lung cancer risk assessment over clinical model among light smokers (average &lt;9 pack-years).

e20066 Background: As lung cancer incidence and mortality increase, methods of early diagnosis are needed to increase survival. The validated HUNT Lung Cancer Model (HUNT LCM) predicts individual 6-year risk of lung cancer in ever smokers with an AUC of 0.87 based on eight clinical variables: sex, age, BMI, pack-years, smoking intensity number of cigarettes per day, quit time, daily cough, and daily indoors smoke exposure. We aimed to explore whether the performance of the HUNT model can be improved by adding genetic information, and in what group of ever smokers these differences in performance emerges. Methods: Based on the eight clinical variables of the HUNT LCM and 22 single nucleotide polymorphisms (SNPs) previously identified in the literature as highly associated with lung cancer risk, the novel lung cancer risk prediction model HUNT Lung-SNP was developed. Clinical and genetical data of ever-smokers from the HUNT2 population study (n = 30746, median follow-up 15.26 years) where 160 individuals were diagnosed with lung cancer within six years were used for the model fitting. External validation was performed in another population-based cohort, the Tromsø Study (n = 2663) where 39 were diagnosed with lung cancer within 6 years. Results: The integrated discrimination improvement index (IDI) between the HUNT Lung-SNP and the HUNT LCM shows that the HUNT Lung-SNP significantly improve the LC risk stratification within six years with an IDI of 0.019 (95% CI 0.015-0.025), p &lt; 0·001 and of 0.013 (95% CI 0.008-0.018), p &lt; 0·001 in the HUNT2 and Tromsø cohorts, respectively. Of the cases, 21/160 (13%) and 10/39 (25%) in the HUNT and Tromsø cohort respectively, were low risk by the HUNT LCM but high risk by the HUNT Lung-SNP model. The main difference in characteristics among cases predicted by the HUNT Lung-SNP model, but missed by the HUNT LCM, was the average number of pack-years: 8.22 vs. 29.8 (p &lt; 0.001) in HUNT2 and 7.7 vs. 33.7 (p &lt; 0.001) in Tromsø cohort. Conclusions: A genetic approach combined with clinical variables of the HUNT LCM can improve the lung cancer risk prediction significantly by increasing the sensitivity among light smokers. Thus, we identified a high-risk population in two independent cohorts that would not be eligible for screening by clinical cutoffs nor with a clinical risk prediction model. This novel HUNT Lung-SNP needs further validation in other populations to evaluate its clinical utility.

Risk-based lung cancer screening performance in a universal healthcare setting.

Globally, lung cancer is the leading cause of cancer death. Previous trials demonstrated that low-dose computed tomography lung cancer screening of high-risk individuals can reduce lung cancer mortality by 20% or more. Lung cancer screening has been approved by major guidelines in the United States, and over 4,000 sites offer screening. Adoption of lung screening outside the United States has, until recently, been slow. Between June 2017 and May 2019, the Ontario Lung Cancer Screening Pilot successfully recruited 7,768 individuals at high risk identified by using the PLCOm2012noRace lung cancer risk prediction model. In total, 4,451 participants were successfully screened, retained and provided with high-quality follow-up, including appropriate treatment. In the Ontario Lung Cancer Screening Pilot, the lung cancer detection rate and the proportion of early-stage cancers were 2.4% and 79.2%, respectively; serious harms were infrequent; and sensitivity to detect lung cancers was 95.3% or more. With abnormal scans defined as ones leading to diagnostic investigation, specificity was 95.5% (positive predictive value, 35.1%), and adherence to annual recall and early surveillance scans and clinical investigations were high (>85%). The Ontario Lung Cancer Screening Pilot provides insights into how a risk-based organized lung screening program can be implemented in a large, diverse, populous geographic area within a universal healthcare system.

Lung Cancer Risk Prediction Models for Asian Ever-Smokers

IntroductionAlthough lung cancer prediction models are widely used to support risk-based screening, their performance outside Western populations remains uncertain. This study aims to evaluate the performance of 11 existing risk prediction models in multiple Asian populations and to refit prediction models for Asians. MethodsIn a pooled analysis of 186,458 Asian ever-smokers from 19 prospective cohorts, we assessed calibration (expected-to-observed ratio) and discrimination (area under the receiver operating characteristic curve [AUC]) for each model. In addition, we developed the “Shanghai models” to better refine risk models for Asians on the basis of two well-characterized population-based prospective cohorts and externally validated them in other Asian cohorts. ResultsAmong the 11 models, the Lung Cancer Death Risk Assessment Tool yielded the highest AUC (AUC [95% confidence interval (CI)] = 0.71 [0.67–0.74] for lung cancer death and 0.69 [0.67–0.72] for lung cancer incidence) and the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial Model had good calibration overall (expected-to-observed ratio [95% CI] = 1.06 [0.90–1.25]). Nevertheless, these models substantially underestimated lung cancer risk among Asians who reported less than 10 smoking pack-years or stopped smoking more than or equal to 20 years ago. The Shanghai models were found to have marginal improvement overall in discrimination (AUC [95% CI] = 0.72 [0.69–0.74] for lung cancer death and 0.70 [0.67–0.72] for lung cancer incidence) but consistently outperformed the selected Western models among low-intensity smokers and long-term quitters. ConclusionsThe Shanghai models had comparable performance overall to the best existing models, but they improved much in predicting the lung cancer risk of low-intensity smokers and long-term quitters in Asia.

P1.14-02 Validation of Lung Cancer Risk Prediction Models in an Asbestos Exposed Population

P1.14-02 Validation of Lung Cancer Risk Prediction Models in an Asbestos Exposed Population

OA07.06 Development and Validation of a Protein-Based Lung Cancer Risk Prediction Model - Initial Results from the Lung Cancer Cohort Consortium (LC3)

OA07.06 Development and Validation of a Protein-Based Lung Cancer Risk Prediction Model - Initial Results from the Lung Cancer Cohort Consortium (LC3)

Sensitivity of US Preventive Services Task Force and PLCOm2012 lung cancer screening eligibility criteria in individuals with lung cancer in South Dakota self-reporting as Indigenous and non-Indigenous.

Lung cancer is the leading cause of cancer deaths. Screening individuals who are at elevated risk using low-dose computed tomography reduces lung cancer mortality by ≥20%. Individuals who have community-based factors that contribute to an increased risk of developing lung cancer have high lung cancer rates and are diagnosed at younger ages. In this study of lung cancer in South Dakota, the authors compared the sensitivity of screening eligibility criteria for self-reported Indigenous race and evaluated the need for screening at younger ages. US Preventive Services Task Force (USPSTF) 2013 and 2021 (USPSTF2013 and USPSTF2021) criteria and two versions of the PLCOm2012 risk-prediction model (based on the 2012 Prostate, Lung, Colorectal, and Ovarian [PLCO] Cancer Screening Trial), one with a predictor for race and one without, were applied at USPSTF-equivalent thresholds of ≥1.7% in 6 years and ≥1.0% in 6 years to 1565 individuals who were sequentially diagnosed with lung cancer (of whom 12.7% self-reported as Indigenous) at the Monument Health Cancer Care Institute in South Dakota (2010-2019). Eligibility sensitivities of USPSTF criteria did not differ significantly between individuals who self-reported their race as Indigenous and those who did not (p>.05). Sensitivities of both PLCOm2012 models were significantly higher than comparable USPSTF criteria. The sensitivity of USPSTF2021 criteria was 66.1% and, for comparable PLCOm2012 models with and without race, sensitivity was 90.7% and 89.6%, respectively (both p<.001); 1.4% of individuals were younger than 50years, and proportions did not differ by Indigenous classification (p=.518). Disparities in screening eligibility were not observed for individuals who self-reported their race as Indigenous. USPSTF criteria had lower sensitivities for lung cancer eligibility. Both PLCOm2012 models had high sensitivities, with higher sensitivity for the model that included race. The PLCOm2012noRace model selected effectively in this population, and screening individuals younger than 50years did not appear to be justified. Lung cancer is the leading cause of cancer deaths. Studies show that using low-dose computed tomography scans to screen people who smoke or who used to smoke and are at elevated risk for lung cancer reduces lung cancer deaths. This study of 1565 individuals with lung cancer in South Dakota compared screening eligibility using US Preventive Services Task Force (USPSTF) criteria and a lung cancer risk-prediction model (PLCOm2012; from the 2012 Prostate, Lung, Colorectal, and Ovarian [PLCO] Cancer Screening Trial). The model had higher sensitivity and picked more people with lung cancer to screen compared with USPSTF criteria. Eligibility sensitivities were similar for individuals who self-reported as Indigenous versus those who did not between USPSTF criteria and the model.

Construction of a Risk Prediction Model for Lung Cancer Based on Lifestyle Behaviors in the UK Biobank Large-Scale Population Cohort

To identify the risk factors related to lifestyle behaviors that affect the incidence of lung cancer, to build a lung cancer risk prediction model to identify, in the population, individuals who are at high risk, and to facilitate the early detection of lung cancer. The data used in the study were obtained from the UK Biobank, a database that contains information collected from 502 389 participants between March 2006 and October 2010. Based on domestic and international guidelines for lung cancer screening and high-quality research literature on lung cancer risk factors, high-risk population identification criteria were determined. Univariate Cox regression was performed to screen for risk factors of lung cancer and a multifactor lung cancer risk prediction model was constructed using Cox proportional hazards regression. Based on the comparison of Akaike information criterion and Schoenfeld residual test results, the optimal fitted model assuming proportional hazards was selected. The multiple factor Cox proportional hazards regression was performed to consider the survival time and the population was randomly divided into a training set and a validation set by a ratio of 7:3. The model was built using the training set and the performance of the model was internally validated using the validation set. The area under the receiver operating characteristic (ROC) curve ( AUC) was used to evaluate the efficacy of the model. The population was categorized into low-risk, moderate-risk, and high-risk groups based on the probability of occurrence of 0% to <25%, 25% to <75%, and 75% to 100%. The respective proportions of affected individuals in each risk group were calculated. The study eventually covered 453 558 individuals, and out of the cumulative follow-up of 5 505 402 person-years, a total of 2 330 cases of lung cancer were diagnosed. Cox proportional hazards regression was performed to identify 10 independent variables as predictors of lung cancer, including age, body mass index (BMI), education, income, physical activity, smoking status, alcohol consumption frequency, fresh fruit intake, family history of cancer, and tobacco exposure, and a model was established accordingly. Internal validation results showed that 8 independent variables (all the 10 independent variables screened out except for BMI and fresh fruit intake) were significant influencing factors of lung cancer ( P<0.05). The AUC of the training set for predicting lung cancer occurrence at one year, five years, and ten years were 0.825, 0.785, and 0.777, respectively. The AUC of the validation set for predicting lung cancer occurrence at one year, five years, and ten years were 0.857, 0.782, and 0.765, respectively. 68.38% of the individuals who might develop lung cancer in the future could be identified by screening the high-risk population. We established, in this study, a model for predicting lung cancer risks associated with lifestyle behaviors of a large population. Showing good performance in discriminatory ability, the model can be used as a tool for developing standardized screening strategies for lung cancer.

Development of lung cancer risk prediction models based on F-18 FDG PET images.

We aimed to evaluate whether the degree of F-18 fluorodeoxyglucose (FDG) uptake in the lungs is associated with an increased risk of lung cancer and to develop lung cancer risk prediction models using metabolic parameters on F-18 FDG positron emission tomography (PET). We retrospectively included 795 healthy individuals who underwent F-18 FDG PET/CT scans for a health check-up. Individuals who developed lung cancer within 5years of the PET/CT scan were classified into the lung cancer group (n = 136); those who did not were classified into the control group (n = 659). The healthy individuals were then randomly assigned to either the training (n = 585) or validation sets (n = 210). Clinical factors including age, sex, body mass index (BMI), and smoking history were collected. The standardized uptake value ratio (SUVR) and metabolic heterogeneity (MH) index were obtained for the bilateral lungs. Logistic regression models including clinical factors, SUVR, and MH index were generated to quantify the probability of lung cancer development using a training set. The prediction models were validated using a validation set. The lung SUVR and lung MH index in the lung cancer group were significantly higher than in the control group (p < 0.001 and p < 0.001, respectively). In the combined prediction model 1, age, sex, BMI, smoking history, and lung SUVR were significantly associated with lung cancer development (age: OR 1.07, p < 0.001; male: OR 2.08, p = 0.015; BMI: OR 0.93, p = 0.057; current or past smoker: OR 5.60, p < 0.001; lung SUVR: OR 1.13, p < 0.001). In the combined prediction model 2, age, sex, BMI, smoking history, and lung MH index showed a significant association with lung cancer development (age: OR 1.06, p < 0.001; male: OR 1.87, p = 0.045; BMI: OR 0.93, p = 0.010; current or past smoker: OR 4.78, p < 0.001; lung MH index: OR 1.33, p < 0.001). In the validation data, combined prediction models 1 and 2 exhibited very good discrimination [area under the receiver operator curve (AUC): 0.867 and 0.901, respectively]. The metabolic parameters on F-18 FDG PET are related to an increased risk of lung cancer. Metabolic parameters can be used as biomarkers to provide information independent of the clinical parameters, related to lung cancer risk.

Lung cancer risk score for ever and never smokers in China.

Most lung cancer risk prediction models were developed in European and North-American cohorts of smokers aged ≥ 55 years, while less is known about risk profiles in Asia, especially for never smokers or individuals aged < 50 years. Hence, we aimed to develop and validate a lung cancer risk estimate tool for ever and never smokers across a wide age range. Based on the China Kadoorie Biobank cohort, we first systematically selected the predictors and explored the nonlinear association of predictors with lung cancer risk using restricted cubic splines. Then, we separately developed risk prediction models to construct a lung cancer risk score (LCRS) in 159,715 ever smokers and 336,526 never smokers. The LCRS was further validated in an independent cohort over a median follow-up of 13.6 years, consisting of 14,153 never smokers and 5,890 ever smokers. A total of 13 and 9 routinely available predictors were identified for ever and never smokers, respectively. Of these predictors, cigarettes per day and quit years showed nonlinear associations with lung cancer risk (Pnon-linear < 0.001). The curve of lung cancer incidence increased rapidly above 20 cigarettes per day and then was relatively flat until approximately 30 cigarettes per day. We also observed that lung cancer risk declined sharply within the first 5 years of quitting, and then continued to decrease but at a slower rate in the subsequent years. The 6-year area under the receiver operating curve for the ever and never smokers' models were respectively 0.778 and 0.733 in the derivation cohort, and 0.774 and 0.759 in the validation cohort. In the validation cohort, the 10-year cumulative incidence of lung cancer was 0.39% and 2.57% for ever smokers with low (< 166.2) and intermediate-high LCRS (≥ 166.2), respectively. Never smokers with a high LCRS (≥ 21.2) had a higher 10-year cumulative incidence rate than those with a low LCRS (< 21.2; 1.05% vs. 0.22%). An online risk evaluation tool (LCKEY; http://ccra.njmu.edu.cn/lckey/web) was developed to facilitate the use of LCRS. The LCRS can be an effective risk assessment tool designed for ever and never smokers aged 30 to 80 years.

Prediction of 6-year risk of bladder cancer using the HUNT Lung Cancer Risk Model in ever-smokers: A HUNT study.

e16581 Background: Bladder cancer is the 6th most common cancer in men and 17th most common cancer in women and the mortality is high when detected in advanced stage. Previous or current cigarette smoke is the main risk factor. There is no screening program for bladder cancer. Given the pathophysiological similarity with lung cancer, we postulated that a lung cancer risk model could work for bladder cancer. The validated HUNT Lung Cancer Risk Model (HUNT LCM) predicts lung cancer in 6 years with an area under the ROC curve (AUC) of 0.87. Here we aimed to test the performance of the HUNT LCM in the prospective HUNT2 cohort for a 6-year prediction of bladder cancer. Methods: The HUNT2 cohort from Norway, a prospective population study of 65 200 individuals 19-100 years of age, was followed up for a median of 15 years. This population was linked with the Norwegian Cancer Registry and all cases diagnosed with bladder cancer were registered. The HUNT Lung Cancer Model was used to predict risk of bladder cancer within 6 years. Results: In the HUNT2 cohort 38 846 ever-smokers were identified. 82.6% of those developing bladder cancer were ever-smokers. 31 269 had all the variables needed for the model and of those, 191 developed bladder cancer in six years. The HUNT Lung Cancer Model prediction of bladder cancer in 6 years AUC was 0.751. Conclusions: To our knowledge, this is the first time a lung cancer risk prediction model has been applied to predict bladder cancer. The performance of the model is lower than for lung cancer but still has a good performance. This may open a path for screening and early detection of bladder cancer.

A strategy to reduce the false-positive rate after low-dose computed tomography in lung cancer screening: A multicenter prospective cohort study.

The ability of lung cancer screening to manage pulmonary nodules was limited because of the high false-positive rate in the current mainstream screening method, low-dose computed tomography (LDCT). We aimed to reduce overdiagnosis in Chinese population. Lung cancer risk prediction models were constructed using data from a population-based cohort in China. Independent clinical data from two programs performed in Beijing and Shandong, respectively, were used as the external validation set. Multivariable logistic regression models were used to estimate the probability of lung cancer incidence in the whole population and in smokers and nonsmokers. In our cohort, 1,016,740 participants were enrolled between 2013 and 2018. Of 79,581 who received LDCT screening, 5165 participants with suspected pulmonary nodules were allocated into the training set, of which, 149 lung cancer cases were diagnosed. In the validation set, 1815 patients were included, and 800 developed lung cancer. The ages of patients and radiologic factors of nodules (calcification, density, mean diameter, edge, and pleural involvement) were included in our model. The area under the curve (AUC) values of the model were 0.868 (95% CI: 0.839-0.894) in the training set and 0.751 (95% CI: 0.727-0.774) in the validation set. The sensitivity and specificity were 70.5% and 70.9%, respectively, which could reduce the 68.8% false-positive rate in simulated LDCT screening. There was no substantial difference between smokers' and nonsmokers' prediction models. Our models could facilitate the diagnosis of suspected pulmonary nodules, effectively reducing the false-positive rate of LDCT for lung cancer screening.

Evaluation of the accuracy of the PLCOm2012 6-year lung cancer risk prediction model among smokers in the CARTaGENE population-based cohort.

The PLCOm2012 prediction tool for risk of lung cancer has been proposed for a pilot program for lung cancer screening in Quebec, but has not been validated in this population. We sought to validate PLCOm2012 in a cohort of Quebec residents, and to determine the hypothetical performance of different screening strategies. We included smokers without a history of lung cancer from the population-based CARTaGENE cohort. To assess PLCOm2012 calibration and discrimination, we determined the ratio of expected to observed number of cases, as well as the sensitivity, specificity and positive predictive values of different risk thresholds. To assess the performance of screening strategies if applied between Jan. 1, 1998, and Dec. 31, 2015, we tested different thresholds of the PLCOm2012 detection of lung cancer over 6 years (1.51%, 1.70% and 2.00%), the criteria of Quebec's pilot program (for people aged 55-74 yr and 50-74 yr) and recommendations from 2021 United States and 2016 Canada guidelines. We assessed shift and serial scenarios of screening, whereby eligibility was assessed annually or every 6 years, respectively. Among 11 652 participants, 176 (1.51%) lung cancers were diagnosed in 6 years. The PLCOm2012 tool underestimated the number of cases (expected-to-observed ratio 0.68, 95% confidence interval [CI] 0.59-0.79), but the discrimination was good (C-statistic 0.727, 95% CI 0.679-0.770). From a threshold of 1.51% to 2.00%, sensitivities ranged from 52.3% (95% CI 44.6%-59.8%) to 44.9% (95% CI 37.4%-52.6%), specificities ranged from 81.6% (95% CI 80.8%-82.3%) to 87.7% (95% CI 87.0%-88.3%) and positive predictive values ranged from 4.2% (95% CI 3.4%-5.1%) to 5.3% (95% CI 4.2%-6.5%). Overall, 8938 participants had sufficient data to test performance of screening strategies. If eligibility was estimated annually, Quebec pilot criteria would have detected fewer cancers than PLCOm2012 at a 2.00% threshold (48.3% v. 50.2%) for a similar number of scans per detected cancer. If eligibility was estimated every 6 years, up to 26 fewer lung cancers would have been detected; however, this scenario led to higher positive predictive values (highest for PLCOm2012 with a 2.00% threshold at 6.0%, 95% CI 4.8%-7.3%). In a cohort of Quebec smokers, the PLCOm2012 risk prediction tool had good discrimination in detecting lung cancer, but it may be helpful to adjust the intercept to improve calibration. The implementation of risk prediction models in some of the provinces of Canada should be done with caution.

Developing and Validating a Lung Cancer Risk Prediction Model: A Nationwide Population-Based Study.

Lung cancer can be challenging to diagnose in the early stages, where treatment options are optimal. We aimed to develop 1-year prediction models for the individual risk of incident lung cancer for all individuals aged 40 or above living in Denmark on 1 January 2017. The study was conducted using population-based registers on health and sociodemographics from 2007-2016. We applied backward selection on all variables by logistic regression to develop a risk model for lung cancer and applied the models to the validation cohort, calculated receiver-operating characteristic curves, and estimated the corresponding areas under the curve (AUC). In the populations without and with previously confirmed cancer, 4274/2,826,249 (0.15%) and 482/172,513 (0.3%) individuals received a lung cancer diagnosis in 2017, respectively. For both populations, older age was a relevant predictor, and the most complex models, containing variables related to diagnoses, medication, general practitioner, and specialist contacts, as well as baseline sociodemographic characteristics, had the highest AUC. These models achieved a positive predictive value (PPV) of 0.0127 (0.006) and a negative predictive value (NPV) of 0.989 (0.997) with a 1% cut-off in the population without (with) previous cancer. This corresponds to 1.2% of the screened population experiencing a positive prediction, of which 1.3% would be incident with lung cancer. We have developed and tested a prediction model with a reasonable potential to support clinicians and healthcare planners in identifying patients at risk of lung cancer.

Machine Learning and Real-World Data to Predict Lung Cancer Risk in Routine Care.

This study used machine learning to develop a 3-year lung cancer risk prediction model with large real-world data in a mostly younger population. Over 4.7 million individuals, aged 45 to 65 years with no history of any cancer or lung cancer screening, diagnostic, or treatment procedures, with an outpatient visit in 2013 were identified in Optum's de-identified Electronic Health Record (EHR) dataset. A least absolute shrinkage and selection operator model was fit using all available data in the 365 days prior. Temporal validation was assessed with recent data. External validation was assessed with data from Mercy Health Systems EHR and Optum's de-identified Clinformatics Data Mart Database. Racial inequities in model discrimination were assessed with xAUCs. The model AUC was 0.76. Top predictors included age, smoking, race, ethnicity, and diagnosis of chronic obstructive pulmonary disease. The model identified a high-risk group with lung cancer incidence 9 times the average cohort incidence, representing 10% of patients with lung cancer. Model performed well temporally and externally, while performance was reduced for Asians and Hispanics. A high-dimensional model trained using big data identified a subset of patients with high lung cancer risk. The model demonstrated transportability to EHR and claims data, while underscoring the need to assess racial disparities when using machine learning methods. This internally and externally validated real-world data-based lung cancer prediction model is available on an open-source platform for broad sharing and application. Model integration into an EHR system could minimize physician burden by automating identification of high-risk patients.

Estimating the area under the ROC curve when transporting a prediction model to a target population.

We propose methods for estimating the area under the receiver operating characteristic (ROC) curve (AUC) of a prediction model in a target population that differs from the source population that provided the data used for original model development. If covariates that are associated with model performance, as measured by the AUC, have a different distribution in the source and target populations, then AUC estimators that only use data from the source population will not reflect model performance in the target population. Here, we provide identification results for the AUC in the target population when outcome and covariate data are available from the sample of the source population, but only covariate data are available from the sample of the target population. In this setting, we propose three estimators for the AUC in the target population and show that they are consistent and asymptotically normal. We evaluate the finite-sample performance of the estimators using simulations and use them to estimate the AUC in a nationally representative target population from the National Health and Nutrition Examination Survey for a lung cancer risk prediction model developed using source population data from the National Lung ScreeningTrial.

A lung cancer risk prediction model for nonsmokers: A retrospective analysis of lung nodule cohorts in China.

The risk of lung cancer in nonsmokers is increasing; however, there are relatively few studies on the risks of lung cancer in nonsmokers. We collected epidemiological and clinical data from 429 nonsmoking patients with lung nodules from the Affiliated Li Huili Hospital as a training cohort and 123 nonsmoking patients with lung nodules as a testing cohort. We identified variables that might be related to malignant lung nodules from 27 variables by performing least absolute shrinkage and selection operator analysis. Univariate and multivariate analyses of these variables were conducted using binary logistic regression. Significant variables were used to generate a lung cancer risk prediction model for nodules in nonsmokers. We successfully constructed a predictive nomogram incorporating density, ground-glass opacities, pulmonary nodule size, hypertension, plasma fibrinogen levels, and blood urea nitrogen. This model exhibited good discriminative ability, with a C-index value of 0.788 (95% confidence interval [CI]: 0.742-0.833) in the training cohort and 0.888 (95% CI: 0.835-0.941) in the testing cohort; it was well-calibrated in both cohorts. Decision curve analyses supported the clinical value of this predictive nomogram when used at a lung cancer possibility threshold of 18%. Ten-fold cross-validation indicated good stability and accuracy of the model (kappa=0.416 ± 0.128; accuracy=0.751 ± 0.056; area under the curve=0.768 ± 0.049). Our risk model can reasonably predict the risks of lung cancer in nonsmoking Chinese patients with lung nodules.

The Journey for Lung Cancer Screening where we Stand Today.

Lung cancer remains a leading cause of cancer mortality worldwide with many patients presenting with advanced disease. We reviewed the available literature for lung cancer screening using low dose computed tomography (LDCT). We reviewed the National Lung Screening Trial (NLST), Early Lung Cancer Action Program (ELCAP) and the (Nederlands-Leuvens Longkanker Screenings Onderzoek (NELSON) trials. We also look at different lung cancer risk prediction models that may aid in identifying target populations and also discuss the cost-effectiveness of LDCT screening in different groups of smokers and ex-smokers. Lastly, we discuss recent guideline changes that have occurred in line with new and emerging evidence on lung cancer screening. LDCT has been shown reduce lung cancer mortality in certain groups of current and former smokers and should be considered to help in the early diagnosis of lung cancer.