Challenges and research opportunities for lung cancer screening in China

Benign-malignant classification of pulmonary nodules by low-dose spiral computerized tomography and clinical data with machine learning in opportunistic screening

BACKGROUND

Many people were found with pulmonary nodules during physical examinations. It is of great practical significance to discriminate benign and malignant nodules by using data mining technology.

METHODS

The subjects' demographic data, baseline examination results, and annual follow-up low-dose spiral computerized tomography (LDCT) results were recorded. The findings from annual physical examinations of positive nodules, including highly suspicious nodules and clinically tentative benign nodules, was analyzed. The extreme gradient boosting (XGBoost) model was constructed and the Grid Search CV method was used to select the super parameters. External unit data were used as an external validation set to evaluate the generalization performance of the model.

RESULTS

A total of 135,503 physical examinees were enrolled. Baseline testing found that 27,636 (20.40%) participants had clinically tentative benign nodules and 611 (0.45%) participants had highly suspicious nodules. The proportion of highly suspicious nodules in participants with negative baseline was about 0.12%-0.46%, which was lower than the baseline level except the follow-up of >5 years. In the 27,636 participants with clinically tentative benign nodules, only in the first year of LDCT re-examination was the proportion of highly suspicious nodules (1.40%) significantly greater than that of baseline screening (0.45%) (p < 0.001), and the proportion of highly suspicious nodules was not different between the baseline screening and other follow-up years (p > 0.05). Furthermore, 322 cases with benign nodules and 196 patients with malignant nodules confirmed by surgery and pathology were compared. A model and the top 15 most important clinical variables were determined by XGBoost algorithm. The area under the curve (AUC) of the model was 0.76 [95% CI: 0.67-0.84], and the accuracy was 0.75. The sensitivity and specificity of the model under this threshold were 0.78 and 0.73, respectively. In the validation of model using external data, the AUC was 0.87 and the accuracy was 0.80. The sensitivity and specificity were 0.83 and 0.77, respectively.

CONCLUSIONS

It is important that pulmonary nodules could be more accurately identified at the first LDCT examination. A model with 15 variables which are routinely measured in the clinic could be helpful to distinguish benign and malignant nodules. It could help the radiological team issue a more accurate report; and it may guide the clinical team regarding LDCT follow-up.

A combined non-enhanced CT radiomics and clinical variable machine learning model for differentiating benign and malignant sub-centimeter pulmonary solid nodules

BACKGROUND

Radiomics has been used to predict pulmonary nodule (PN) malignancy. However, most of the studies focused on pulmonary ground-glass nodules. The use of computed tomography (CT) radiomics in pulmonary solid nodules, particularly sub-centimeter solid nodules, is rare.

PURPOSE

This study aims to develop a radiomics model based on non-enhanced CT images that can distinguish between benign and malignant sub-centimeter pulmonary solid nodules (SPSNs, <1 cm).

METHODS

The clinical and CT data of 180 SPSNs confirmed by pathology were analyzed retrospectively. All SPSNs were divided into two groups: training set (n = 144) and testing set (n = 36). From non-enhanced chest CT images, over 1000 radiomics features were extracted. Radiomics feature selection was performed using the analysis of variance and principal component analysis. The selected radiomics features were fed into a support vector machine (SVM) to develop a radiomics model. The clinical and CT characteristics were used to develop a clinical model. Associating non-enhanced CT radiomics features with clinical factors were used to develop a combined model using SVM. The performance was evaluated using the area under the receiver-operating characteristic curve (AUC).

RESULTS

The radiomics model performed well in distinguishing between benign and malignant SPSNs, with an AUC of 0.913 (95% confidence interval [CI], 0.862-0.954) in the training set and an AUC of 0.877 (95% CI, 0.817-0.924) in the testing set. The combined model outperformed the clinical and radiomics models with an AUC of 0.940 (95% CI, 0.906-0.969) in the training set and an AUC of 0.903 (95% CI, 0.857-0.944) in the testing set.

CONCLUSIONS

Radiomics features based on non-enhanced CT images can be used to differentiate SPSNs. The combined model, which included radiomics and clinical factors, had the best discrimination power between benign and malignant SPSNs.

The Generation of a Lung Cancer Health Factor Distribution Using Patient Graphs Constructed From Electronic Medical Records: Retrospective Study

BACKGROUND

Electronic medical records (EMRs) of patients with lung cancer (LC) capture a variety of health factors. Understanding the distribution of these factors will help identify key factors for risk prediction in preventive screening for LC.

OBJECTIVE

We aimed to generate an integrated biomedical graph from EMR data and Unified Medical Language System (UMLS) ontology for LC, and to generate an LC health factor distribution from a hospital EMR of approximately 1 million patients.

METHODS

The data were collected from 2 sets of 1397 patients with and those without LC. A patient-centered health factor graph was plotted with 108,000 standardized data, and a graph database was generated to integrate the graphs of patient health factors and the UMLS ontology. With the patient graph, we calculated the connection delta ratio (CDR) for each of the health factors to measure the relative strength of the factor's relationship to LC.

RESULTS

The patient graph had 93,000 relations between the 2794 patient nodes and 650 factor nodes. An LC graph with 187 related biomedical concepts and 188 horizontal biomedical relations was plotted and linked to the patient graph. Searching the integrated biomedical graph with any number or category of health factors resulted in graphical representations of relationships between patients and factors, while searches using any patient presented the patient's health factors from the EMR and the LC knowledge graph (KG) from the UMLS in the same graph. Sorting the health factors by CDR in descending order generated a distribution of health factors for LC. The top 70 CDR-ranked factors of disease, symptom, medical history, observation, and laboratory test categories were verified to be concordant with those found in the literature.

CONCLUSIONS

By collecting standardized data of thousands of patients with and those without LC from the EMR, it was possible to generate a hospital-wide patient-centered health factor graph for graph search and presentation. The patient graph could be integrated with the UMLS KG for LC and thus enable hospitals to bring continuously updated international standard biomedical KGs from the UMLS for clinical use in hospitals. CDR analysis of the graph of patients with LC generated a CDR-sorted distribution of health factors, in which the top CDR-ranked health factors were concordant with the literature. The resulting distribution of LC health factors can be used to help personalize risk evaluation and preventive screening recommendations.

Lung cancer incidence and mortality in China: Updated statistics and an overview of temporal trends from 2000 to 2016

Background

Updated systematic statistics on lung cancer are the underpinning cornerstones for formulating prevention and control strategies for the disease. The incidence and mortality data of lung cancer in China in 2016 were estimated, and also used to analyze the temporal trends of lung cancer from 2000 up to 2016 in this study.

Methods

The burden of lung cancer in China in 2016 was estimated using data from 487 cancer registries, which were abstracted from the database of the National Cancer Center of China (NCC). The temporal trends were estimated with the data of 2000 to 2016 from 22 cancer registries.

Results

About 828,100 new lung cancer cases and 657,000 lung cancer deaths were estimated in China in 2016. The crude incidence and mortality rates in the eastern region were the highest nationwide. The incidence and mortality of lung cancer increased with age, and most of the new cases occurred in the age group of over 60 years. The age-standardized incidence increased by about 0.8% per year during 2000 to 2016, especially in woman, whose annual increase rate reached 2.1%. The age-standardized mortality rate decreased by about 0.6% per year, with a decrease of 1.3% per year in urban areas and an increase of 2.3% per year in rural areas. The numbers of new cases and deaths increased by 162.6% and 123.6%, respectively, during 2000 to 2016, which can be explained mainly by the change of the age structure.

Conclusions

The burden of lung cancer is serious in China, and increased especially rapidly for women in rural areas. The disease is expected to threaten the lives of more people in the future due to factors such as aging and population growth. Lung cancer prevention and control strategies and resources should be leveraged toward women and rural areas in the future.

Th22 Cells/IL-22 Serves as a Protumor Regulator to Drive Poor Prognosis through the JAK-STAT3/MAPK/AKT Signaling Pathway in Non-Small-Cell Lung Cancer

The interaction of immune cells and cytokines in the tumor microenvironment affects the development and prognosis of tumors with an unclear potential regulatory mechanism. Recent studies have elucidated the protumor role of Th22 cells and its lineage-specific cytokine IL-22 in different human cancers. The present study is aimed at investigating the biological effect of Th22 cells/IL-22 and its molecular mechanism in the pathogenesis process of non-small-cell lung cancer (NSCLC). It was initially found that Th22 cells were enriched in the peripheral blood of NSCLC patients. The level of Th22 cells in peripheral blood mononuclear cells (PBMCs) was positively correlated with the TNM stage, lymph node metastasis, and clinical tumor biomarkers. Furthermore, IL-22 not only antagonized the apoptosis inducing and cell cycle arresting effect by chemotherapy and molecular targeted drugs on NSCLC cell lines but also promoted tumor cell proliferation and tumor tissue growth. Moreover, IL-22 activated the JAK-STAT3/MAPK/AKT signaling pathway, both in vitro and in vivo. Conclusively, the present results confirm that Th22 cells/IL-22 may serve as a negative immune regulator in lung cancer.

The impact of an N1 lymph node examination in patients with early-stage non-small cell lung cancer: a retrospective cohort study

Background

The examination of lymph nodes (LNs) is critical for accurate node staging in patients with non-small cell lung cancer (NSCLC), but a consensus on the examinations of hilar and intrapulmonary (N1 station) LNs has not been reached. This study aimed to evaluate the role of LN dissection and pathological examination of N1 LN stations and their effects on survival in patients with stage IA-IIA NSCLC.

Methods

Data from patients pathologically staged as IA-IIA who underwent radical surgery and confirmed as lacking LN metastases from January 2008 to March 2018 were retrospectively reviewed. The Kaplan-Meier method was used to determine the overall survival (OS) and disease-free survival (DFS). After propensity score matching (PSM), a Cox model was used to determine the prognostic factors.

Results

Of the 1,935 patients investigated, the median number of N1 stations examined was 3. Patients with at least 2 N1 stations examined had apparently better OS (P=0.002) and DFS (P=0.001). All patients were divided into patients with 0–1 N1 station examined and patients with 2–5 N1 stations examined. After PSM, the number of N1 stations examined was an independent prognostic factor for DFS (P=0.004). Patients with 2–5 N1 stations examined experienced prolonged DFS (P=0.010). Patients in group 12 experienced prolonged OS (P=0.021) and DFS (P=0.026). Patients in group 13 or 14 experienced prolonged OS (P=0.028).

Conclusions

A larger extent of N1 station examination was associated with prolonged DFS in patients with stage IA-IIA NSCLC after lobectomy. The dissection and examination of at least 2 N1 stations included LNs from the lobar and segmental drainage fields.

Performance of low-dose computed tomography on lung cancer screening in high-risk populations: The experience over five screening rounds in Sichuan, China

OBJECTIVE

To evaluate the performance of low-dose computed tomography (LDCT) on lung cancer screening in high-risk populations in Sichuan.

METHODS

From April 2014 to July 2018, LDCT was performed annually on 3185 subjects aged 50-74 years who had smoked ≥ 20 pack-years (or subjects having quit smoking within 5 years). Information about all deaths and lung cancer diagnoses were obtained by active investigation, or passive matching to disease surveillance system.

RESULTS

The screening population had a median age of 60 years. 62.4 % of which were current smokers and had smoked 30 pack-years. After participating in the baseline screening, the compliance rates of subjects consecutively completing one round, two rounds, three rounds, and four rounds of annual screening were 67.22 %, 52.84 %, 43.24 %, and 40.04 %, respectively. The positive rates in baseline and annual screening were 6.53 % and 5.79 %, respectively. During the 5 rounds, a total of 9522 person-times were screened by LDCT with a screening sensitivity of 89.13 % (95 % CI: 76.96-95.27), specificity of 94.36 % (95 % CI: 93.88-94.81), positive predictive value of 7.13 % (95 % CI: 5.30-9.53), and negative predictive value of 99.94 % (95 % CI: 99.87-99.98). There were no statistically significant performance differences between baseline and annual screening. The difference in the proportion of screen-detected stage I lung cancer between baseline screening and annual screening was not statistically significant, neither.

CONCLUSION

The application of LDCT on lung cancer screening in high-risk populations shows favorable compliance and a high screening performance in the project area of Sichuan,China.

Preoperative diagnosis of malignant pulmonary nodules in lung cancer screening with a radiomics nomogram

Background

Lung cancer is the most commonly diagnosed cancer worldwide. Its survival rate can be significantly improved by early screening. Biomarkers based on radiomics features have been found to provide important physiological information on tumors and considered as having the potential to be used in the early screening of lung cancer. In this study, we aim to establish a radiomics model and develop a tool to improve the discrimination between benign and malignant pulmonary nodules.

Methods

A retrospective study was conducted on 875 patients with benign or malignant pulmonary nodules who underwent computed tomography (CT) examinations between June 2013 and June 2018. We assigned 612 patients to a training cohort and 263 patients to a validation cohort. Radiomics features were extracted from the CT images of each patient. Least absolute shrinkage and selection operator (LASSO) was used for radiomics feature selection and radiomics score calculation. Multivariate logistic regression analysis was used to develop a classification model and radiomics nomogram. Radiomics score and clinical variables were used to distinguish benign and malignant pulmonary nodules in logistic model. The performance of the radiomics nomogram was evaluated by the area under the curve (AUC), calibration curve and Hosmer‐Lemeshow test in both the training and validation cohorts.

Results

A radiomics score was built and consisted of 20 features selected by LASSO from 1288 radiomics features in the training cohort. The multivariate logistic model and radiomics nomogram were constructed using the radiomics score and patients’ age. Good discrimination of benign and malignant pulmonary nodules was obtained from the training cohort (AUC, 0.836; 95% confidence interval [CI]: 0.793‐0.879) and validation cohort (AUC, 0.809; 95% CI: 0.745‐0.872). The Hosmer‐Lemeshow test also showed good performance for the logistic regression model in the training cohort (P = 0.765) and validation cohort (P = 0.064). Good alignment with the calibration curve indicated the good performance of the nomogram.

Conclusions

The established radiomics nomogram is a noninvasive preoperative prediction tool for malignant pulmonary nodule diagnosis. Validation revealed that this nomogram exhibited excellent discrimination and calibration capacities, suggesting its clinical utility in the early screening of lung cancer.

Causes of death in long-term bladder cancer survivors: A population-based study

BACKGROUND

Long-term (> 5 years) bladder cancer survivors represent a distinct subgroup of bladder cancer patients and information about the causes of death in this subgroup is limited. The aim of this study was to review the causes of death in long-term bladder cancer survivors.

METHOD

The Surveillance, epidemiology and end results (SEER) database was used to analyze the causes of death of long-term bladder cancer survivors. Patients' characteristics and survival outcomes were reported for the entire cohort in our study. Kaplan-Meier analysis and Cox proportional hazard model for survival analysis.

RESULTS

A total of 147 781 bladder cancer patients with >5 years survival were identified. This cohort included 81 843 patients surviving 5-10 years and 65 938 patients surviving >10 years. Among the patients who survived 5-10 years, 6.9% died because of primary bladder cancer, 11.0% due to cardiac disease and 7.7% due to nonmalignant pulmonary disease. Among patients surviving >10 years, 3.1% died because of primary bladder cancer, 8.6% due to cardiac disease and 5.8% due to nonmalignant pulmonary disease. On multivariate analysis, factors associated with longer cardiac disease-specific survival among long-term bladder cancer survivors include younger age at diagnosis（<40 years; vs. 40-69 years, P = 0.030 or >69 years, P < 0.001), married status (vs. single status, P < 0.001), white race (vs. African American race, P = 0.002), male (vs. female, P < 0.001), grade I (vs. grade III, P = 0.003 or grade IV, P < 0.001).

CONCLUSION

The probability of death from primary bladder cancer is still important among various causes of death even 20 years after being diagnosed with bladder cancer. Furthermore, cardiopulmonary causes contributed to a considerable proportion of deaths in long-term bladder cancer survivors.