The Lung Cancer Prediction Model "Stress Test": Assessment of Models' Performance in a High-Risk Prospective Pulmonary Nodule Cohort.

BACKGROUND: Pulmonary nodules represent a growing health care burden because of delayed diagnosis of malignant lesions and overtesting for benign processes. Clinical prediction models were developed to inform physician assessment of pretest probability of nodule malignancy but have not been validated in a high-risk cohort of nodules for which biopsy was ultimately performed. RESEARCH QUESTION: Do guideline-recommended prediction models sufficiently discriminate between benign and malignant nodules when applied to cases referred for biopsy by navigational bronchoscopy? STUDY DESIGN AND METHODS: We assembled a prospective cohort of 322 indeterminate pulmonary nodules in 282 patients referred to a tertiary medical center for diagnostic navigational bronchoscopy between 2017 and 2019. We calculated the probability of malignancy for each nodule using the Brock model, Mayo Clinic model, and Veterans Affairs (VA) model. On a subset of 168 patients who also had PET-CT scans before biopsy, we also calculated the probability of malignancy using the Herder model. The performance of the models was evaluated by calculating the area under the receiver operating characteristic curves (AUCs) for each model. RESULTS: The study cohort contained 185 malignant and 137 benign nodules (57% prevalence of malignancy). The malignant and benign cohorts were similar in terms of size, with a median longest diameter for benign and malignant nodules of 15 and 16 mm, respectively. The Brock model, Mayo Clinic model, and VA model showed similar performance in the entire cohort (Brock AUC, 0.70; 95% CI, 0.64-0.76; Mayo Clinic AUC, 0.70; 95% CI, 0.64-0.76; VA AUC, 0.67; 95% CI, 0.62-0.74). For 168 nodules with available PET-CT scans, the Herder model had an AUC of 0.77 (95% CI, 0.68-0.85). INTERPRETATION: Currently available clinical models provide insufficient discrimination between benign and malignant nodules in the common clinical scenario in which a patient is being referred for biopsy, especially when PET-CT scan information is not available.

Explainable machine learning model based on clinical factors for predicting the disappearance of indeterminate pulmonary nodules.

BACKGROUND: During lung cancer screening, indeterminate pulmonary nodules (IPNs) are a frequent finding. We aim to predict whether IPNs are resolving or non-resolving to reduce follow-up examinations, using machine learning (ML) models. We incorporated dedicated techniques to enhance prediction explainability. METHODS: In total, 724 IPNs (size 50-500 mm3, 575 participants) from the Dutch-Belgian Randomized Lung Cancer Screening Trial were used. We implemented six ML models and 14 factors to predict nodule disappearance. Random search was applied to determine the optimal hyperparameters on the training set (579 nodules). ML models were trained using 5-fold cross-validation and tested on the test set (145 nodules). Model predictions were evaluated by utilizing the recall, precision, F1 score, and the area under the receiver operating characteristic curve (AUC). The best-performing model was used for three feature importance techniques: mean decrease in impurity (MDI), permutation feature importance (PFI), and SHAPley Additive exPlanations (SHAP). RESULTS: The random forest model outperformed the other ML models with an AUC of 0.865. This model achieved a recall of 0.646, a precision of 0.816, and an F1 score of 0.721. The evaluation of feature importance achieved consistent ranking across all three methods for the most crucial factors. The MDI, PFI, and SHAP methods highlighted volume, maximum diameter, and minimum diameter as the top three factors. However, the remaining factors revealed discrepant ranking across methods. CONCLUSION: ML models effectively predict IPN disappearance using participant demographics and nodule characteristics. Explainable techniques can assist clinicians in developing understandable preliminary assessments.

Assessing the clinical utility of biomarkers using the intervention probability curve (IPC).

BACKGROUND: Assessing the clinical utility of biomarkers is a critical step before clinical implementation. The reclassification of patients across clinically relevant subgroups is considered one of the best methods to estimate clinical utility. However, there are important limitations with this methodology. We recently proposed the intervention probability curve (IPC) which models the likelihood that a provider will choose an intervention as a continuous function of the probability, or risk, of disease. OBJECTIVE: To assess the potential impact of a new biomarker for lung cancer using the IPC. METHODS: The IPC derived from the National Lung Screening Trial was used to assess the potential clinical utility of a biomarker for suspected lung cancer. The summary statistics of the change in likelihood of intervention over the population can be interpreted as the expected clinical impact of the added biomarker. RESULTS: The IPC analysis of the novel biomarker estimated that 8% of the benign nodules could avoid an invasive procedure while the cancer nodules would largely remain unchanged (0.1%). We showed the benefits of this approach compared to traditional reclassification methods based on thresholds. CONCLUSIONS: The IPC methodology can be a valuable tool for assessing biomarkers prior to clinical implementation.

Identification and evaluation of circulating small extracellular vesicle microRNAs as diagnostic biomarkers for patients with indeterminate pulmonary nodules.

BACKGROUND: The identification of indeterminate pulmonary nodules (IPNs) following a low-dose computed tomography (LDCT) is a major challenge for early diagnosis of lung cancer. The inadequate assessment of IPNs' malignancy risk results in a large number of unnecessary surgeries or an increased risk of cancer metastases. However, limited studies on non-invasive diagnosis of IPNs have been reported. METHODS: In this study, we identified and evaluated the diagnostic value of circulating small extracellular vesicle (sEV) microRNAs (miRNAs) in patients with IPNs that had been newly detected using LDCT scanning and were scheduled for surgery. Out of 459 recruited patients, 109 eligible patients with IPNs were enrolled in the training cohort (n = 47) and the test cohort (n = 62). An external cohort (n = 99) was used for validation. MiRNAs were extracted from plasma sEVs, and assessed using Small RNA sequencing. 490 lung adenocarcinoma samples and follow-up data were used to investigate the role of miRNAs in overall survival. RESULTS: A circulating sEV miRNA (CirsEV-miR) model was constructed from five differentially expressed miRNAs (DEMs), showing 0.920 AUC in the training cohort (n = 47), and further identified in the test cohort (n = 62) and in an external validation cohort (n = 99). Among five DEMs of the CirsEV-miR model, miR-101-3p and miR-150-5p were significantly associated with better overall survival (p = 0.0001 and p = 0.0069). The CirsEV-miR scores were calculated, which significantly correlated with IPNs diameters (p < 0.05), and were able to discriminate between benign and malignant PNs (diameter ≤ 1 cm). The expression patterns of sEV miRNAs in the benign, adenocarcinoma in situ/minimally invasive adenocarcinoma, and invasive adenocarcinoma subgroups were found to gradually change with the increase in aggressiveness for the first time. Among all DEMs of the three subgroups, five miRNAs (miR-30c-5p, miR-30e-5p, miR-500a-3p, miR-125a-5p, and miR-99a-5p) were also significantly associated with overall survival of lung adenocarcinoma patients. CONCLUSIONS: Our results indicate that the CirsEV-miR model could help distinguish between benign and malignant PNs, providing insights into the feasibility of circulating sEV miRNAs in diagnostic biomarker development. TRIAL REGISTRATION: Chinese Clinical Trials: ChiCTR1800019877. Registered 05 December 2018, https://www.chictr.org.cn/showproj.aspx?proj=31346 .

Surgery or Non-surgical Treatment of ≤8 mm Non-small Cell Lung Cancer: A Population-Based Study.

Background: Timing for intervention of small indeterminate pulmonary nodules has long been a topic of debate given the low incidence of malignancy and difficulty in obtaining a definite preoperative diagnosis. We sought to determine survival outcomes of surgical and non-surgical managements in non-small cell lung cancer (NSCLC) ≤8 mm, which may provide a reference for prospective decision-making for patients with suspected NSCLC. Method: A total of 1,652 patients with Stage IA NSCLC ≤8 mm were identified from the Surveillance, Epidemiology, and End Results (SEER) database and categorized into surgery and non-surgery groups. Chi-square test, t-test and Mann-Whitney U test were used to compare the baseline characteristics between groups. Survival curves were depicted using Kaplan-Meier method and compared by log-rank test. Cox proportional hazard model was used for univariate and multivariate analyses. Adjustment of confounding factors between groups was performed by propensity score matching. Results: The surgery and non-surgery groups included 1,438 and 208 patients, respectively. Patients in surgery group demonstrated superior survival outcome than patients in non-surgery group both before [overall survival (OS): HR, 16.22; 95% CI, 11.48-22.91, p < 0.001; cancer-specific survival (CSS): HR, 49.6; 95% CI, 31.09-79.11, p < 0.001] and after (OS: HR, 3.12; 95% CI, 2.40-4.05, p < 0.001; CSS: HR, 3.85; 95% CI, 2.74-5.40, p < 0.001) propensity score matching. The 30-day mortality rates were 3.1 and 12.0% in surgery and non-surgery groups, respectively. Multivariate analysis suggested age, sex, race, tumor size, grade, pathological stage were all independent prognostic factors in patients with ≤8 mm NSCLC. A comparison of surgical resections revealed a survival superiority of lobectomy over sub-lobectomy. In terms of CSS, no statistically significant difference was found between segmentectomy and wedge resection. Conclusion: The current SEER database showed better prognosis of surgical resection than non-surgical treatment in patients with ≤8 mm NSCLC. However, the factors that should be essentially included in the proper propensity-matched analysis, such as comorbidity, cardiopulmonary function and performance status were unavailable and the true superiority or inferiority should be examined further by ongoing randomized trial, especially comparing surgery and stereotactic body irradiation.

Perioperative identifications of non-palpable pulmonary nodules: a narrative review.

Early detection of lung cancer is the key to improving treatment and prognosis of this disease, and the advent of advances in computed tomography (CT) imaging and national screening programs have improved the detection rate of very small pulmonary lesions. As such, the management of this sub-centimetric and often sub-solid lesions has become quite challenging for clinicians, especially for choosing the most suitable diagnostic method. In clinical practice, to fulfill this diagnostic yield, transthoracic needle biopsy (TTNB) is often the first choice especially for peripheral nodules. For lesions for which TTNB could present technical difficulties or failed, other diagnostic strategies are needed. In this case, video-assisted thoracic surgery (VATS) is the gold standard to reach the diagnosis of lung nodules suspect of being malignant. Nonetheless it's often not easy the identification of such lesions during VATS because of their little dimensions, non-firm consistency, deep localization. In literature various marking techniques have been described, in order to improve intraoperative nodules detection and to reduce conversion rate to thoracotomy: CT-guided hookwire positioning, methylene blue staining, intra-operative ultrasound and electromagnetic navigation bronchoscopy are the most used. The scientific evidence on this matter is weak because there are no randomized clinical trials but only case series on single techniques with no comparison on efficacy, so there are no guidelines to refer. From this standing, in this article we conducted a narrative review of the existing literature on the subject, with the aim of outlining a framework as complete as possible. We analyzed strengths and weaknesses of the main techniques reported, so as to allow the clinician to orient himself with greater ease.

Identification of Proteomic Features To Distinguish Benign Pulmonary Nodules from Lung Adenocarcinoma.

We hypothesized that distinct protein expression features of benign and malignant pulmonary nodules may reveal novel candidate biomarkers for the early detection of lung cancer. We performed proteome profiling by liquid chromatography-tandem mass spectrometry to characterize 34 resected benign lung nodules, 24 untreated lung adenocarcinomas (ADCs), and biopsies of bronchial epithelium. Group comparisons identified 65 proteins that differentiate nodules from ADCs and normal bronchial epithelium and 66 proteins that differentiate ADCs from nodules and normal bronchial epithelium. We developed a multiplexed parallel reaction monitoring (PRM) assay to quantify a subset of 43 of these candidate biomarkers in an independent cohort of 20 benign nodules, 21 ADCs, and 20 normal bronchial biopsies. PRM analyses confirmed significant nodule-specific abundance of 10 proteins including ALOX5, ALOX5AP, CCL19, CILP1, COL5A2, ITGB2, ITGAX, PTPRE, S100A12, and SLC2A3 and significant ADC-specific abundance of CEACAM6, CRABP2, LAD1, PLOD2, and TMEM110-MUSTN1. Immunohistochemistry analyses for seven selected proteins performed on an independent set of tissue microarrays confirmed nodule-specific expression of ALOX5, ALOX5AP, ITGAX, and SLC2A3 and cancer-specific expression of CEACAM6. These studies illustrate the value of global and targeted proteomics in a systematic process to identify and qualify candidate biomarkers for noninvasive molecular diagnosis of lung cancer.

Indeterminate pulmonary nodules in colorectal cancer.

AIM: To investigate the clinicopathologic parameters of pulmonary metastasis in colorectal cancer (CRC) patients after lung operation of indeterminate pulmonary nodules (IPNs). METHODS: From a prospective database of CRC patients, 40 cases that underwent lung operation between November 2008 and December 2012 for suspicious metastatic pulmonary nodules on chest computed tomography (CT) were enrolled. The decision to perform a lung operation was made if the patient met the following criteria: (1) completely resected or resectable primary CRC; (2) completely resectable IPNs; (3) controlled or controllable extrapulmonary metastasis; and (4) adequate general condition and pulmonary function to tolerate pulmonary operation. Lung operation was performed by a thoracic surgeon without CT-guided biopsy for pathologic confirmation. RESULTS: A total of 40 cases of lung resection was performed in 29 patients. Five patients underwent repeated lung resection. The final pathology result showed metastasis from the CRC in 30 cases (75%) and benign pathology in 10 cases (25%). The primary tumor site was the rectum in 26/30 (86.6%) cases with pulmonary metastasis, but only 3/10 (30%) cases in the benign group had a primary rectal cancer (P = 0.001). Positron emission tomography (PET)-CT was performed for 22/30 (73.4%) patients in the lung metastasis group and for 6/10 (60.0%) patients in the benign group. PET-CT revealed hot uptake of (18)fluorine 2-fluoro-2-deoxy-D-glucose with all IPNs in both groups. The group with pulmonary metastasis had a higher incidence of primary rectal cancer (P = 0.001), a more advanced tumor stage (P = 0.011), and more frequent lymphatic invasion of tumor cells (P = 0.005). Six cases with previous liver metastasectomy were present in the lung metastasis group. Serum carcinoembryonic antigen levels before lung operation were not elevated in any of the patients. CONCLUSION: The stage and location of the primary tumor and tumor cell infiltration of lymphatics provide useful indicators for deciding on lung resection of IPNs in CRC.