Radiomics analysis to predict pulmonary nodule malignancy using machine learning approaches.

BACKGROUND: Low-dose CT screening can reduce lung cancer-related mortality. However, most screen-detected pulmonary abnormalities do not develop into cancer and it often remains challenging to identify malignant nodules, particularly among indeterminate nodules. We aimed to develop and assess prediction models based on radiological features to discriminate between benign and malignant pulmonary lesions detected on a baseline screen. METHODS: Using four international lung cancer screening studies, we extracted 2060 radiomic features for each of 16 797 nodules (513 malignant) among 6865 participants. After filtering out low-quality radiomic features, 642 radiomic and 9 epidemiological features remained for model development. We used cross-validation and grid search to assess three machine learning (ML) models (eXtreme Gradient Boosted Trees, random forest, least absolute shrinkage and selection operator (LASSO)) for their ability to accurately predict risk of malignancy for pulmonary nodules. We report model performance based on the area under the curve (AUC) and calibration metrics in the held-out test set. RESULTS: The LASSO model yielded the best predictive performance in cross-validation and was fit in the full training set based on optimised hyperparameters. Our radiomics model had a test-set AUC of 0.93 (95% CI 0.90 to 0.96) and outperformed the established Pan-Canadian Early Detection of Lung Cancer model (AUC 0.87, 95% CI 0.85 to 0.89) for nodule assessment. Our model performed well among both solid (AUC 0.93, 95% CI 0.89 to 0.97) and subsolid nodules (AUC 0.91, 95% CI 0.85 to 0.95). CONCLUSIONS: We developed highly accurate ML models based on radiomic and epidemiological features from four international lung cancer screening studies that may be suitable for assessing indeterminate screen-detected pulmonary nodules for risk of malignancy.

USPSTF2013 versus PLCOm2012 lung cancer screening eligibility criteria (International Lung Screening Trial): interim analysis of a prospective cohort study.

BACKGROUND: Lung cancer is a major health problem. CT lung screening can reduce lung cancer mortality through early diagnosis by at least 20%. Screening high-risk individuals is most effective. Retrospective analyses suggest that identifying individuals for screening by accurate prediction models is more efficient than using categorical age-smoking criteria, such as the US Preventive Services Task Force (USPSTF) criteria. This study prospectively compared the effectiveness of the USPSTF2013 and PLCOm2012 model eligibility criteria. METHODS: In this prospective cohort study, participants from the International Lung Screening Trial (ILST), aged 55-80 years, who were current or former smokers (ie, had ≥30 pack-years smoking history or ≤15 quit-years since last permanently quitting), and who met USPSTF2013 criteria or a PLCOm2012 risk threshold of at least 1·51% within 6 years of screening, were recruited from nine screening sites in Canada, Australia, Hong Kong, and the UK. After enrolment, patients were assessed with the USPSTF2013 criteria and the PLCOm2012 risk model with a threshold of at least 1·70% at 6 years. Data were collected locally and centralised. Main outcomes were the comparison of lung cancer detection rates and cumulative life expectancies in patients with lung cancer between USPSTF2013 criteria and the PLCOm2012 model. In this Article, we present data from an interim analysis. To estimate the incidence of lung cancers in individuals who were USPSTF2013-negative and had PLCOm2012 of less than 1·51% at 6 years, ever-smokers in the Prostate Lung Colorectal and Ovarian Cancer Screening Trial (PLCO) who met these criteria and their lung cancer incidence were applied to the ILST sample size for the mean follow-up occurring in the ILST. This trial is registered at ClinicalTrials.gov, NCT02871856. Study enrolment is almost complete. FINDINGS: Between June 17, 2015, and Dec 29, 2020, 5819 participants from the International Lung Screening Trial (ILST) were enrolled on the basis of meeting USPSTF2013 criteria or the PLCOm2012 risk threshold of at least 1·51% at 6 years. The same number of individuals was selected for the PLCOm2012 model as for the USPSTF2013 criteria (4540 [78%] of 5819). After a mean follow-up of 2·3 years (SD 1·0), 135 lung cancers occurred in 4540 USPSTF2013-positive participants and 162 in 4540 participants included in the PLCOm2012 of at least 1·70% at 6 years group (cancer sensitivity difference 15·8%, 95% CI 10·7-22·1%; absolute odds ratio 4·00, 95% CI 1·89-9·44; p<0·0001). Compared to USPSTF2013-positive individuals, PLCOm2012-selected participants were older (mean age 65·7 years [SD 5·9] vs 63·3 years [5·7]; p<0·0001), had more comorbidities (median 2 [IQR 1-3] vs 1 [1-2]; p<0·0001), and shorter life expectancy (13·9 years [95% CI 12·8-14·9] vs 14·8 [13·6-16·0] years). Model-based difference in cumulative life expectancies for those diagnosed with lung cancer were higher in those who had PLCOm2012 risk of at least 1·70% at 6 years than individuals who were USPSTF2013-positive (2248·6 years [95% CI 2089·6-2425·9] vs 2000·7 years [1841·2-2160·3]; difference 247·9 years, p=0·015). INTERPRETATION: PLCOm2012 appears to be more efficient than the USPSTF2013 criteria for selecting individuals to enrol into lung cancer screening programmes and should be used for identifying high-risk individuals who benefit from the inclusion in these programmes. FUNDING: Terry Fox Research Institute, The UBC-VGH Hospital Foundation and the BC Cancer Foundation, the Alberta Cancer Foundation, the Australian National Health and Medical Research Council, Cancer Research UK and a consortium of funders, and the Roy Castle Lung Cancer Foundation for the UK Lung Screen Uptake Trial.

Trial marketing in the Pan-Canadian Early Detection of Lung Cancer Study.

BACKGROUND: Recruitment to clinical trials is suboptimal, increasing costs, and delaying the potential implementation of clinical advances. Among other barriers, the lack of marketing experience among trialists may limit recruitment. In this observational study, in the context of the Pan-Canadian Early Detection of Lung Cancer Trial, we assessed the value of a motivational survey of study participants in planning a tailored advertising campaign and analysed the value of individual components of advertising in generating telephone calls to the study and recruited subjects. METHODS: The Pan-Canadian Early Detection of Lung Cancer Trial was a single arm study assessing risk modelling for lung cancer screening by low-dose computed tomography scan and autofluorescence bronchoscopy. Individuals were recruited to eight sites across Canada without a central marketing plan. On contact with the study, individuals reported how they heard about the study according to a predefined list. One site, the Juravinski Cancer Centre, worked with a marketing expert to develop a survey to assess participant motivations, source of study awareness, and personal habits. The survey was used to develop a media campaign for recruitment. Media events were collected from all sites. The primary analysis assessed the number of telephone contacts and recruited subjects associated with various media factors. Individual print media characteristics were assessed for their effect on recruitment. RESULTS: At all sites, 7059 individuals contacted the study, and 2537 were eligible and recruited. Among 52 individuals completing the Juravinski Cancer Centre survey, motivation included concern for personal risk of lung cancer (71%), followed by desire to contribute to a cure (67%), followed by personal knowledge of a person with lung cancer (50%). Most reported hearing of the study from the newspaper (58%) despite no print ad yet being distributed. With survey input, a newsprint campaign was executed. The number of media events varied by site (median: 13, range: 3-28). Among all recruits, 56.4% reported referral by newspaper followed by family/friend (14%). Telephone contacts and recruited subjects per event varied significantly by site, while unpaid media events appeared superior to paid events. Print media characteristics associated with increased telephone contacts and recruitment included use of a rational appeal (vs a mixed rational-emotional), less use of white space, and larger headline font. CONCLUSION: A survey of trial candidates provides useful information regarding personal motivation, media use, and lifestyle. Unpaid media events appear superior in generating recruitment, while print media may be superior to radio and television in selecting eligible recruits. The utility of individual print media characteristics appears to differ from the commercial advertising literature. Further research on marketing in clinical trials is encouraged to improve recruitment ( ClinicalTrials.gov registration: NCT00751660, https://clinicaltrials.gov/ct2/show/NCT00751660 ).

Chronic obstructive pulmonary disease prevalence and prediction in a high-risk lung cancer screening population.

BACKGROUND: Chronic obstructive pulmonary disease (COPD) is an underdiagnosed condition sharing risk factors with lung cancer. Lung cancer screening may provide an opportunity to improve COPD diagnosis. Using Pan-Canadian Early Detection of Lung Cancer (PanCan) study data, the present study sought to determine the following: 1) What is the prevalence of COPD in a lung cancer screening population? 2) Can a model based on clinical and screening low-dose CT scan data predict the likelihood of COPD? METHODS: The single arm PanCan study recruited current or former smokers age 50-75 who had a calculated risk of lung cancer of at least 2% over 6 years. A baseline health questionnaire, spirometry, and low-dose CT scan were performed. CT scans were assessed by a radiologist for extent and distribution of emphysema. With spirometry as the gold standard, logistic regression was used to assess factors associated with COPD. RESULTS: Among 2514 recruited subjects, 1136 (45.2%) met spirometry criteria for COPD, including 833 of 1987 (41.9%) of those with no prior diagnosis, 53.8% of whom had moderate or worse disease. In a multivariate model, age, current smoking status, number of pack-years, presence of dyspnea, wheeze, participation in a high-risk occupation, and emphysema extent on LDCT were all statistically associated with COPD, while the overall model had poor discrimination (c-statistic = 0.627 (95% CI of 0.607 to 0.650). The lowest and the highest risk decile in the model predicted COPD risk of 27.4 and 65.3%. CONCLUSIONS: COPD had a high prevalence in a lung cancer screening population. While a risk model had poor discrimination, all deciles of risk had a high prevalence of COPD, and spirometry could be considered as an additional test in lung cancer screening programs. TRIAL REGISTRATION: (Clinical Trial Registration: ClinicalTrials.gov, number NCT00751660 , registered September 12, 2008).

Participant selection for lung cancer screening by risk modelling (the Pan-Canadian Early Detection of Lung Cancer [PanCan] study): a single-arm, prospective study.

BACKGROUND: Results from retrospective studies indicate that selecting individuals for low-dose CT lung cancer screening on the basis of a highly predictive risk model is superior to using criteria similar to those used in the National Lung Screening Trial (NLST; age, pack-year, and smoking quit-time). We designed the Pan-Canadian Early Detection of Lung Cancer (PanCan) study to assess the efficacy of a risk prediction model to select candidates for lung cancer screening, with the aim of determining whether this approach could better detect patients with early, potentially curable, lung cancer. METHODS: We did this single-arm, prospective study in eight centres across Canada. We recruited participants aged 50-75 years, who had smoked at some point in their life (ever-smokers), and who did not have a self-reported history of lung cancer. Participants had at least a 2% 6-year risk of lung cancer as estimated by the PanCan model, a precursor to the validated PLCOm2012 model. Risk variables in the model were age, smoking duration, pack-years, family history of lung cancer, education level, body-mass index, chest x-ray in the past 3 years, and history of chronic obstructive pulmonary disease. Individuals were screened with low-dose CT at baseline (T0), and at 1 (T1) and 4 (T4) years post-baseline. The primary outcome of the study was incidence of lung cancer. This study is registered with ClinicalTrials.gov, number NCT00751660. FINDINGS: 7059 queries came into the study coordinating centre and were screened for PanCan risk. 15 were duplicates, so 7044 participants were considered for enrolment. Between Sept 24, 2008, and Dec 17, 2010, we recruited and enrolled 2537 eligible ever-smokers. After a median follow-up of 5·5 years (IQR 3·2-6·1), 172 lung cancers were diagnosed in 164 individuals (cumulative incidence 0·065 [95% CI 0·055-0·075], incidence rate 138·1 per 10 000 person-years [117·8-160·9]). There were ten interval lung cancers (6% of lung cancers and 6% of individuals with cancer): one diagnosed between T0 and T1, and nine between T1 and T4. Cumulative incidence was significantly higher than that observed in NLST (4·0%; p<0·0001). Compared with 593 (57%) of 1040 lung cancers observed in NLST, 133 (77%) of 172 lung cancers in the PanCan Study were early stage (I or II; p<0·0001). INTERPRETATION: The PanCan model was effective in identifying individuals who were subsequently diagnosed with early, potentially curable, lung cancer. The incidence of cancers detected and the proportion of early stage cancers in the screened population was higher than observed in previous studies. This approach should be considered for adoption in lung cancer screening programmes. FUNDING: Terry Fox Research Institute and Canadian Partnership Against Cancer.

Probability of cancer in pulmonary nodules detected on first screening CT.

BACKGROUND: Major issues in the implementation of screening for lung cancer by means of low-dose computed tomography (CT) are the definition of a positive result and the management of lung nodules detected on the scans. We conducted a population-based prospective study to determine factors predicting the probability that lung nodules detected on the first screening low-dose CT scans are malignant or will be found to be malignant on follow-up. METHODS: We analyzed data from two cohorts of participants undergoing low-dose CT screening. The development data set included participants in the Pan-Canadian Early Detection of Lung Cancer Study (PanCan). The validation data set included participants involved in chemoprevention trials at the British Columbia Cancer Agency (BCCA), sponsored by the U.S. National Cancer Institute. The final outcomes of all nodules of any size that were detected on baseline low-dose CT scans were tracked. Parsimonious and fuller multivariable logistic-regression models were prepared to estimate the probability of lung cancer. RESULTS: In the PanCan data set, 1871 persons had 7008 nodules, of which 102 were malignant, and in the BCCA data set, 1090 persons had 5021 nodules, of which 42 were malignant. Among persons with nodules, the rates of cancer in the two data sets were 5.5% and 3.7%, respectively. Predictors of cancer in the model included older age, female sex, family history of lung cancer, emphysema, larger nodule size, location of the nodule in the upper lobe, part-solid nodule type, lower nodule count, and spiculation. Our final parsimonious and full models showed excellent discrimination and calibration, with areas under the receiver-operating-characteristic curve of more than 0.90, even for nodules that were 10 mm or smaller in the validation set. CONCLUSIONS: Predictive tools based on patient and nodule characteristics can be used to accurately estimate the probability that lung nodules detected on baseline screening low-dose CT scans are malignant. (Funded by the Terry Fox Research Institute and others; ClinicalTrials.gov number, NCT00751660.).

The International Association for the Study of Lung Cancer Early Lung Imaging Confederation Open-Source Deep Learning and Quantitative Measurement Initiative.

INTRODUCTION: With global adoption of computed tomography (CT) lung cancer screening, there is increasing interest to use artificial intelligence (AI) deep learning methods to improve the clinical management process. To enable AI research using an open-source, cloud-based, globally distributed, screening CT imaging data set and computational environment that are compliant with the most stringent international privacy regulations that also protect the intellectual properties of researchers, the International Association for the Study of Lung Cancer sponsored development of the Early Lung Imaging Confederation (ELIC) resource in 2018. The objective of this report is to describe the updated capabilities of ELIC and illustrate how this resource can be used for clinically relevant AI research. METHODS: In this second phase of the initiative, metadata and screening CT scans from two time points were collected from 100 screening participants in seven countries. An automated deep learning AI lung segmentation algorithm, automated quantitative emphysema metrics, and a quantitative lung nodule volume measurement algorithm were run on these scans. RESULTS: A total of 1394 CTs were collected from 697 participants. The LAV950 quantitative emphysema metric was found to be potentially useful in distinguishing lung cancer from benign cases using a combined slice thickness more than or equal to 2.5 mm. Lung nodule volume change measurements had better sensitivity and specificity for classifying malignant from benign lung nodules when applied to solid lung nodules from high-quality CT scans. CONCLUSIONS: These initial experiments revealed that ELIC can support deep learning AI and quantitative imaging analyses on diverse and globally distributed cloud-based data sets.

Breath testing for SARS-CoV-2 infection.

BACKGROUND: From a public health perspective, the identification of individuals with mild respiratory symptoms due to SARS-CoV-2 infection is important to contain the spread of the disease. The objective of this study was to identify volatile organic compounds (VOCs) in exhaled breath common to infection with different variants of the SARS-CoV-2 virus to inform the development of a point-of-care breath test to detect infected individuals with mild symptoms. METHODS: A prospective, real-world, observational study was conducted on mildly symptomatic out-patients presenting to community test-sites for RT-qPCR SARS-CoV-2 testing when the Alpha, Beta, and Delta variants were driving the COVID-19 pandemic. VOCs in exhaled breath were compared between PCR-positive and negative individuals using TD-GC-ToF-MS. Candidate VOCs were tested in an independent set of samples collected during the Omicron phase of the pandemic. FINDINGS: Fifty breath samples from symptomatic RT-qPCR positive and 58 breath samples from test-negative, but symptomatic participants were compared. Of the 50 RT-qPCR-positive participants, 22 had breath sampling repeated 8-12 weeks later. PCA-X model yielded 12 distinct VOCs that discriminated SARS-CoV-2 active infection compared to recovery/convalescence period, with an area under the receiver operator characteristic curve (AUROC), of 0.862 (0.747-0.977), sensitivity, and specificity of 82% and 86%, respectively. PCA-X model from 50 RT-qPCR positive and 58 negative symptomatic participants, yielded 11 VOCs, with AUROC of 0.72 (0.604-0.803) and sensitivity of 72%, specificity 65.5%. The 11 VOCs were validated in a separate group of SARS-CoV-2 Omicron positive patients' vs healthy controls demonstrating an AUROC of 0.96 (95% CI 0.827-0.993) with sensitivity of 80% specificity of 90%. INTERPRETATION: Exhaled breath analysis is a promising non-invasive, point-of-care method to detect mild COVID-19 infection. FUNDING: Funding for this study was a competitive grant awarded from the Vancouver Coastal Research Institute as well as funding from the BC Cancer Foundation.

Economic impact of using risk models for eligibility selection to the International lung screening Trial.

OBJECTIVES: Using risk models as eligibility criteria for lung screening can reduce race and sex-based disparities. We used data from the International Lung Screening Trial(ILST; NCT02871856) to compare the economic impact of using the PLCOm2012 risk model or the US Preventative Services' categorical age-smoking history-based criteria (USPSTF-2013). MATERIALS AND METHODS: The cost-effectiveness of using PLCOm2012 versus USPSTF-2013 was evaluated with a decision analytic model based on the ILST and other screening trials. The primary outcomes were costs in 2020 International Dollars ($), quality-adjusted life-years (QALY) and incremental net benefit (INB, in $ per QALY). Secondary outcomes were selection characteristics and cancer detection rates (CDR). RESULTS: Compared with the USPSTF-2013 criteria, the PLCOm2012 risk model resulted in $355 of cost savings per 0.2 QALYs gained (INB=$4294 at a willingness-to-pay threshold of $20 000/QALY (95 %CI: $4205-$4383). Using the risk model was more cost-effective in females at both a 1.5 % and 1.7 % 6-year risk threshold (INB=$6616 and $6112, respectively), compared with males ($5221 and $695). The PLCOm2012 model selected more females, more individuals with fewer years of formal education, and more people with other respiratory illnesses in the ILST. The CDR with the risk model was higher in females compared with the USPSTF-2013 criteria (Risk Ratio = 7.67, 95 % CI: 1.87-31.38). CONCLUSION: The PLCOm2012 model saved costs, increased QALYs and mitigated socioeconomic and sex-based disparities in access to screening.

Circulating proteome for pulmonary nodule malignancy.

BACKGROUND: Although lung cancer screening with low-dose computed tomography is rolling out in many areas of the world, differentiating indeterminate pulmonary nodules remains a major challenge. We conducted one of the first systematic investigations of circulating protein markers to differentiate malignant from benign screen-detected pulmonary nodules. METHODS: Based on 4 international low-dose computed tomography screening studies, we assayed 1078 protein markers using prediagnostic blood samples from 1253 participants based on a nested case-control design. Protein markers were measured using proximity extension assays, and data were analyzed using multivariable logistic regression, random forest, and penalized regressions. Protein burden scores (PBSs) for overall nodule malignancy and imminent tumors were estimated. RESULTS: We identified 36 potentially informative circulating protein markers differentiating malignant from benign nodules, representing a tightly connected biological network. Ten markers were found to be particularly relevant for imminent lung cancer diagnoses within 1 year. Increases in PBSs for overall nodule malignancy and imminent tumors by 1 standard deviation were associated with odds ratios of 2.29 (95% confidence interval: 1.95 to 2.72) and 2.81 (95% confidence interval: 2.27 to 3.54) for nodule malignancy overall and within 1 year of diagnosis, respectively. Both PBSs for overall nodule malignancy and for imminent tumors were substantially higher for those with malignant nodules than for those with benign nodules, even when limited to Lung Computed Tomography Screening Reporting and Data System (LungRADS) category 4 (P < .001). CONCLUSIONS: Circulating protein markers can help differentiate malignant from benign pulmonary nodules. Validation with an independent computed tomographic screening study will be required before clinical implementation.

Omega 3 supplementation reduces C-reactive protein, prostaglandin E2 and the granulocyte/lymphocyte ratio in heavy smokers: An open-label randomized crossover trial.

Objectives: Given the current controversy concerning the efficacy of omega 3 supplements at reducing inflammation, we evaluated the safety and efficacy of omega 3 on reducing inflammation in people with a 6-year lung cancer risk >1.5% and a C reactive protein (CRP) level >2 mg/L in a phase IIa cross-over study. Materials and methods: Forty-nine healthy participants ages 55 to 80, who were still smoking or had smoked in the past with ≥30 pack-years smoking history, living in British Columbia, Canada, were randomized in an open-label trial to receive 2.4 g eicosapentaenoic acid (EPA) + 1.2 g docosahexaenoic acid (DHA)/day for 6 months followed by observation for 6 months or observation for 6 months first and then active treatment for the next 6 months. Blood samples were collected over 1 year for measurement of plasma CRP, plasma and red blood cell (RBC) membrane levels of EPA, DHA and other fatty acids, Prostaglandin E2 (PGE2), Leukotriene B4 (LTB4) and an inflammatory marker panel. Results: Twenty one participants who began the trial within the active arm completed the trial while 20 participants who started in the control arm completed the study. Taking omega 3 resulted in a significant decrease in plasma CRP and PGE2 but not LTB4 levels. Importantly, the effect size for the primary outcome, CRP values, at the end of the intervention relative to baseline was medium (Cohen's d = 0.56). DHA, but not EPA levels in RBC membranes inversely correlated with PGE2 levels. Omega 3 also led to a significant reduction in granulocytes and an increase in lymphocytes. These high-dose omega 3 supplements were well tolerated, with only minor gastrointestinal symptoms in a subset of participants. Conclusion: Omega 3 fatty acids taken at 3.6 g/day significantly reduce systemic inflammation with negligible adverse health effects in people who smoke or have smoked and are at high risk of lung cancer.ClinicalTrials.gov, NCT number: NCT03936621.

High-Ambient Air Pollution Exposure Among Never Smokers Versus Ever Smokers With Lung Cancer.

INTRODUCTION: Air pollution may play an important role in the development of lung cancer in people who have never smoked, especially among East Asian women. The aim of this study was to compare cumulative ambient air pollution exposure between ever and never smokers with lung cancer. METHODS: A consecutive case series of never and ever smokers with newly diagnosed lung cancer were compared regarding their sex, race, and outdoor and household air pollution exposure. Using individual residential history, cumulative exposure to outdoor particulate matter (PM2.5) in a period of 20 years was quantified with a high-spatial resolution global exposure model. RESULTS: Of the 1005 patients with lung cancer, 56% were females and 33% were never smokers. Compared with ever smokers with lung cancer, never smokers with lung cancer were significantly younger, more frequently Asian, less likely to have chronic obstructive pulmonary disease or a family history of lung cancer, and had higher exposure to outdoor PM2.5 but lower exposure to secondhand smoke. Multivariable logistic regression analysis revealed a significant association with never-smoking patients with lung cancer and being female (OR = 4.01, 95% confidence interval [CI]: 2.76-5.82, p < 0.001), being Asian (ORAsian versus non-Asian = 6.48, 95% CI: 4.42-9.50, p < 0.001), and having greater exposure to air pollution (ORln_PM2.5 = 1.79, 95% CI: 1.10-7.2.90, p = 0.019). CONCLUSIONS: Compared with ever-smoking patients with lung cancer, never-smoking patients had strong associations with being female, being Asian, and having air pollution exposures. Our results suggest that incorporation of cumulative exposure to ambient air pollutants be considered when assessing lung cancer risk in combination with traditional risk factors.

Cystic Primary Lung Cancer: Evolution of Computed Tomography Imaging Morphology Over Time.

PURPOSE: Primary lung cancers associated with cystic airspaces are increasingly being recognized; however, there is a paucity of data on their natural history. We aimed to evaluate the prevalence, pathologic, and imaging characteristics of cystic lung cancer in a regional thoracic surgery center with a focus on the evolution of computed tomography morphology over time. MATERIALS AND METHODS: Consecutive patients referred for potential surgical management of primary lung cancer between January 2016 and December 2018 were included. Clinical, imaging, and pathologic data were collected at the time of diagnosis and at the time of the oldest computed tomography showing the target lesion. Descriptive analysis was carried out. RESULTS: A total of 441 cancers in 431 patients (185 males, 246 females), median age 69.6 years (interquartile range: 62.6 to 75.3 y), were assessed. Overall, 41/441 (9.3%) primary lung cancers were cystic at the time of diagnosis. The remaining showed solid (67%), part-solid (22%), and ground-glass (2%) morphologies. Histopathology of the cystic lung cancers at diagnosis included 31/41 (76%) adenocarcinomas, 8/41 (20%) squamous cell carcinomas, 1/41 (2%) adenosquamous carcinoma, and 1/41 (2%) unspecified non-small cell lung carcinoma. Overall, 8/34 (24%) cystic cancers at the time of diagnosis developed from different morphologic subtype precursor lesions, while 8/34 (24%) cystic precursor lesions also transitioned into part-solid or solid cancers at the time of diagnosis. CONCLUSIONS: This study demonstrates that cystic airspaces within lung cancers are not uncommon, and may be seen transiently as cancers evolve. Increased awareness of the spectrum of cystic lung cancer morphology is important to improve diagnostic accuracy and lung cancer management.

Risk Perception Among a Lung Cancer Screening Population.

BACKGROUND: A successful lung cancer screening program requires a patient cohort at sufficient risk of developing cancer who are willing to participate. Among other factors, a patient's lung cancer risk perception may inform their attitudes toward screening and smoking cessation programs. RESEARCH QUESTION: This study analyzed data from the Pan-Canadian Early Detection of Lung Cancer (PanCan) Study to address the following questions: Which factors are associated with the perception of lung cancer risk? Is there an association between risk perception for lung cancer and actual calculated risk? Is there an association between risk perception for lung cancer and the intent to quit smoking? Are there potential targets for lung cancer screening awareness? STUDY DESIGN AND METHODS: The PanCan study recruited current or former smokers aged 50 to 75 years who had at least a 2% risk of developing lung cancer over 6 years to undergo low-dose screening CT. Risk perception and worry about lung cancer were captured on a baseline questionnaire. Cumulative logistic regression analysis was used to assess the relationship between baseline risk variables and both lung cancer risk perception and worry. RESULTS: Among the 2,514 individuals analyzed, a higher perceived risk of lung cancer was positively associated with calculated risk (P = .032). Younger age, being a former smoker, respiratory symptoms, lower FEV1, COPD, and a family history of lung cancer were associated with higher perceived risk. Conversely, a consistent relationship between calculated risk and worry was not identified. There was a positive association between risk perception and lung cancer worry and reported intent to quit smoking. INTERPRETATION: Individuals' lung cancer risk perception correlated positively with calculated risk in a screening population. Promotion of screening programs may benefit from focusing on factors associated with higher risk perception; conversely, harnessing worry seemingly holds less value.

Perifissural nodules seen at CT screening for lung cancer.

PURPOSE: To describe and characterize the potential for malignancy of noncalcified lung nodules adjacent to fissures that are often found in current or former heavy smokers who undergo computed tomography (CT) for lung cancer screening. MATERIALS AND METHODS: Institutional review board approval and informed consent were obtained. Baseline and follow-up thin-section multidetector CT scans obtained in 146 consecutive subjects at high risk for lung cancer (age range, 50-75 years; > 30 pack-year smoking history) were retrospectively reviewed. Noncalcified nodules (NCNs) were categorized according to location (parenchymal, perifissural), shape, septal connection, manually measured diameter, diameter change, and lung cancer outcome at 7(1/2) years. RESULTS: Retrospective review of images from 146 baseline and 311 follow-up CT examinations revealed 837 NCNs in 128 subjects. Of those 837 nodules, 234 (28%), in 98 subjects, were adjacent to a fissure and thus classified as perifissural nodules (PFNs). Multiple (range, 2-14) PFNs were seen in 47 subjects. Most PFNs were triangular (102/234, 44%) or oval (98/234, 42%), were located inferior to the carina (196/234, 84%), and had a septal connection (171/234, 73%). The mean maximal length was 3.2 mm (range, 1-13 mm). During 2-year follow-up in 71 subjects, seven of 159 PFNs increased in size on one scan but were then stable. The authors searched a lung cancer registry 7(1/2) years after study entry and found 10 lung cancers in 139 of 146 study subjects who underwent complete follow-up; none of these cancers had originated from a PFN. CONCLUSION: PFNs are frequently seen on screening CT scans obtained in high-risk subjects. Although PFNs may show increased size at follow-up CT, the authors in this study found none that had developed into lung cancer; this suggests that the malignancy potential of PFNs is low. (c) RSNA, 2010.

Protocol and Rationale for the International Lung Screening Trial.

Rationale: The NLST (National Lung Screening Trial) reported a 20% reduction in lung cancer mortality with low-dose computed tomography screening; however, important questions on how to optimize screening remain, including which selection criteria are most accurate at detecting lung cancers and what nodule management protocol is most efficient. The PLCOm2012 (Prostate, Lung, Colorectal and Ovarian) Cancer Screening Trial 6-year and PanCan (Pan-Canadian Early Detection of Lung Cancer) nodule malignancy risk models are two of the better validated risk prediction models for screenee selection and nodule management, respectively. Combined use of these models for participant selection and nodule management could significantly improve screening efficiency.Objectives: The ILST (International Lung Screening Trial) is a prospective cohort study with two primary aims: 1) Compare the accuracy of the PLCOm2012 model against U.S. Preventive Services Task Force (USPSTF) criteria for detecting lung cancers and 2) evaluate nodule management efficiency using the PanCan nodule probability calculator-based protocol versus Lung-RADS.Methods: ILST will recruit 4,500 participants who meet USPSTF and/or PLCOm2012 risk ≥1.51%/6-year selection criteria. Participants will undergo baseline and 2-year low-dose computed tomography screening. Baseline nodules are managed according to PanCan probability score. Participants will be followed up for a minimum of 5 years. Primary outcomes for aim 1 are the proportion of individuals selected for screening, proportion of lung cancers detected, and positive predictive values of either selection criteria, and outcomes for aim 2 include comparing distributions of individuals and the proportion of lung cancers in each of three management groups: next surveillance scan, early recall scan, or diagnostic evaluation recommended. Statistical powers to detect differences in the four components of primary study aims were ≥82%.Conclusions: ILST will prospectively evaluate the comparative accuracy and effectiveness of two promising multivariable risk models for screenee selection and nodule management in lung cancer screening.Clinical trial registered with www.clinicaltrials.gov (NCT02871856).

Prediction of lung cancer risk at follow-up screening with low-dose CT: a training and validation study of a deep learning method.

Background: Current lung cancer screening guidelines use mean diameter, volume or density of the largest lung nodule in the prior computed tomography (CT) or appearance of new nodule to determine the timing of the next CT. We aimed at developing a more accurate screening protocol by estimating the 3-year lung cancer risk after two screening CTs using deep machine learning (ML) of radiologist CT reading and other universally available clinical information. Methods: A deep machine learning (ML) algorithm was developed from 25,097 participants who had received at least two CT screenings up to two years apart in the National Lung Screening Trial. Double-blinded validation was performed using 2,294 participants from the Pan-Canadian Early Detection of Lung Cancer Study (PanCan). Performance of ML score to inform lung cancer incidence was compared with Lung-RADS and volume doubling time using time-dependent ROC analysis. Exploratory analysis was performed to identify individuals with aggressive cancers and higher mortality rates. Findings: In the PanCan validation cohort, ML showed excellent discrimination with a 1-, 2- and 3-year time-dependent AUC values for cancer diagnosis of 0·968±0·013, 0·946±0·013 and 0·899±0·017. Although high ML score cohort included only 10% of the PanCan sample, it identified 94%, 85%, and 71% of incident and interval lung cancers diagnosed within 1, 2, and 3 years, respectively, after the second screening CT. Furthermore, individuals with high ML score had significantly higher mortality rates (HR=16·07, p<0·001) compared to those with lower risk. Interpretation: ML tool that recognizes patterns in both temporal and spatial changes as well as synergy among changes in nodule and non-nodule features may be used to accurately guide clinical management after the next scheduled repeat screening CT.

Health-related quality of life and anxiety in the PAN-CAN lung cancer screening cohort.

OBJECTIVES: The impact of lung cancer screening with low-dose chest CT (LDCT) on participants' anxiety levels and health-related quality of life (HRQoL) is an important consideration in the implementation of such programmes. We aimed to describe changes in anxiety and HRQoL in a high-risk Canadian cohort undergoing LDCT lung cancer screening. METHODS: 2537 subjects who had 2% or greater lung cancer risk over 6 years using a risk prediction tool were recruited from eight centres across Canada in the Pan-Canadian Early Detection of Lung Cancer Study (2008-2010). We compared HRQoL and anxiety levels before and after screening of 1237 participants with LDCT (excluding a subset of 1300 participants who also underwent autofluorescence bronchoscopy screening), as well as after investigations performed because of a positive screening examination. The 12-item short-form Physical and Mental Component Scales (SF-12), EQ-5D-3L scores and State Trait Anxiety Inventory-State anxiety were used at each assessment. RESULTS: Overall, there were no clinically significant differences in HRQoL outcomes between baseline and each of the survey time points following initial screening. No mean change in anxiety in the overall cohort was noted following baseline LDCT, but more participants had clinically significant increase in anxiety versus decrease after baseline screening (increase >minimal clinically important difference (MCID) (n=180) vs decrease >MCID (n=50), p<0.001). This finding persisted but to a lesser degree at the 12 month time point (increase >MCID (n=146) vs decrease >MCID (n=87), p<0.001). CONCLUSIONS: CT screening for lung cancer has no major overall impact on HRQoL among participants, although a minority of participants (number-needed-to-harm=7 after baseline screening and 18 at 1 year) demonstrated clinically significant increased anxiety levels. TRIALREGISTRATION NUMBER: NCT00751660; Results.

Predicting Malignancy Risk of Screen-Detected Lung Nodules-Mean Diameter or Volume.

OBJECTIVE: In lung cancer screening practice low-dose computed tomography, diameter, and volumetric measurement have been used in the management of screen-detected lung nodules. The aim of this study was to compare the performance of nodule malignancy risk prediction tools using diameter or volume and between computer-aided detection (CAD) and radiologist measurements. METHODS: Multivariable logistic regression models were prepared by using data from two multicenter lung cancer screening trials. For model development and validation, baseline low-dose computed tomography scans from the Pan-Canadian Early Detection of Lung Cancer Study and a subset of National Lung Screening Trial (NLST) scans with lung nodules 3 mm or more in mean diameter were analyzed by using the CIRRUS Lung Screening Workstation (Radboud University Medical Center, Nijmegen, the Netherlands). In the NLST sample, nodules with cancer had been matched on the basis of size to nodules without cancer. RESULTS: Both CAD-based mean diameter and volume models showed excellent discrimination and calibration, with similar areas under the receiver operating characteristic curves of 0.947. The two CAD models had predictive performance similar to that of the radiologist-based model. In the NLST validation data, the CAD mean diameter and volume models also demonstrated excellent discrimination: areas under the curve of 0.810 and 0.821, respectively. These performance statistics are similar to those of the Pan-Canadian Early Detection of Lung Cancer Study malignancy probability model with use of these data and radiologist-measured maximum diameter. CONCLUSION: Either CAD-based nodule diameter or volume can be used to assist in predicting a nodule's malignancy risk.

The Cost-Effectiveness of High-Risk Lung Cancer Screening and Drivers of Program Efficiency.

INTRODUCTION: Lung cancer risk prediction models have the potential to make programs more affordable; however, the economic evidence is limited. METHODS: Participants in the National Lung Cancer Screening Trial (NLST) were retrospectively identified with the risk prediction tool developed from the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial. The high-risk subgroup was assessed for lung cancer incidence and demographic characteristics compared with those in the low-risk subgroup and the Pan-Canadian Early Detection of Lung Cancer Study (PanCan), which is an observational study that was high-risk-selected in Canada. A comparison of high-risk screening versus standard care was made with a decision-analytic model using data from the NLST with Canadian cost data from screening and treatment in the PanCan study. Probabilistic and deterministic sensitivity analyses were undertaken to assess uncertainty and identify drivers of program efficiency. RESULTS: Use of the risk prediction tool developed from the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial with a threshold set at 2% over 6 years would have reduced the number of individuals who needed to be screened in the NLST by 81%. High-risk screening participants in the NLST had more adverse demographic characteristics than their counterparts in the PanCan study. High-risk screening would cost $20,724 (in 2015 Canadian dollars) per quality-adjusted life-year gained and would be considered cost-effective at a willingness-to-pay threshold of $100,000 in Canadian dollars per quality-adjusted life-year gained with a probability of 0.62. Cost-effectiveness was driven primarily by non-lung cancer outcomes. Higher noncurative drug costs or current costs for immunotherapy and targeted therapies in the United States would render lung cancer screening a cost-saving intervention. CONCLUSIONS: Non-lung cancer outcomes drive screening efficiency in diverse, tobacco-exposed populations. Use of risk selection can reduce the budget impact, and screening may even offer cost savings if noncurative treatment costs continue to rise.