Background: Artificial intelligence approaches can integrate complex features and can be used to predict a patient’s risk of developing lung cancer, thereby decreasing the need for unnecessary and expensive diagnostic interventions. Objective: The aim of this study was to use electronic medical records to prescreen patients who are at risk of developing lung cancer. Methods: We randomly selected 2 million participants from the Taiwan National Health Insurance Research Database who received care between 1999 and 2013. We built a predictive lung cancer screening model with neural networks that were trained and validated using pre-2012 data, and we tested the model prospectively on post-2012 data. An age- and gender-matched subgroup that was 10 times larger than the original lung cancer group was used to assess the predictive power of the electronic medical record. Discrimination (area under the receiver operating characteristic curve [AUC]) and calibration analyses were performed. Results: The analysis included 11,617 patients with lung cancer and 1,423,154 control patients. The model achieved AUCs of 0.90 for the overall population and 0.87 in patients ≥55 years of age. The AUC in the matched subgroup was 0.82. The positive predictive value was highest (14.3%) among people aged ≥55 years with a pre-existing history of lung disease. Conclusions: Our model achieved excellent performance in predicting lung cancer within 1 year and has potential to be deployed for digital patient screening. Convolution neural networks facilitate the effective use of EMRs to identify individuals at high risk for developing lung cancer.

Breast and prostate cancer patients may experience physical and psychological distress, and a possible decrease in sleep quality. Subjective and objective methods measure different aspects of sleep quality. Our study attempted to determine differences between objective and subjective measurements of sleep quality using bivariate and Pearson’s correlation data analysis. Forty breast (n = 20) and prostate (n = 20) cancer patients were recruited in this observational study. Participants were given an actigraphy device (ACT) and asked to continuously wear it for seven consecutive days, for objective data collection. Following this period, they filled out the Pittsburgh Sleep Quality Index Questionnaire (PSQI) to collect subjective data on sleep quality. The correlation results showed that, for breast cancer patients, PSQI sleep duration was moderately correlated with ACT total sleeping time (TST) (r = −0.534, p < 0.05), and PSQI daytime dysfunction was related to ACT efficiency (r = 0.521, p < 0.05). For prostate cancer patients, PSQI sleep disturbances were related to ACT TST (r = 0.626, p < 0.05). Both objective and subjective measurements are important in validating and determining details of sleep quality, with combined results being more insightful, and can also help in personalized care to further improve quality of life among cancer patients.

Platinum-induced nephrotoxicity is a severe and unexpected adverse drug reaction that could lead to treatment failure in non-small cell lung cancer patients. Better prediction and management of this nephrotoxicity can increase patient survival. Our study aimed to build up and compare the best machine learning models with clinical and genomic features to predict platinum-induced nephrotoxicity in non-small cell lung cancer patients. Clinical and genomic data of patients undergoing platinum chemotherapy at Wan Fang Hospital were collected after they were recruited. Twelve models were established by artificial neural network, logistic regression, random forest, and support vector machine with integrated, clinical, and genomic modes. Grid search and genetic algorithm were applied to construct the fine-tuned model with the best combination of predictive hyperparameters and features. Accuracy, precision, recall, F1 score, and area under the receiver operating characteristic curve were calculated to compare the performance of the 12 models. In total, 118 patients were recruited for this study, among which 28 (23.73%) were experiencing nephrotoxicity. Machine learning models with clinical and genomic features achieved better prediction performances than clinical or genomic features alone. Artificial neural network with clinical and genomic features demonstrated the best predictive outcomes among all 12 models. The average accuracy, precision, recall, F1 score and area under the receiver operating characteristic curve of the artificial neural network with integrated mode were 0.923, 0.950, 0.713, 0.808 and 0.900, respectively. Machine learning models with clinical and genomic features can be a preliminary tool for oncologists to predict platinum-induced nephrotoxicity and provide preventive strategies in advance.

The non-small cell lung cancer (NSCLC) is the leading cause of cancer death worldwide. In NSCLC, the oncogenic AKT-mTOR, ERK and STAT3 pathways are commonly dysregulated and have emerged as attractive targets for therapeutic developments. In a relatively limited subset of NSCLC, these pathways driven by mutant EGFR can be treated by the tyrosine kinase inhibitors (TKIs)-mediated targeted therapy. However, for the most NSCLC, more novel targeted agents are imperatively needed. Therefore, we investigated the inhibitory effects of the active fraction HS7 from Taiwanofungus camphoratus, a unique medicinal fungus in Taiwan, on these pathways in CL1-0 EGFR wild-type human NSCLC cells. The active fraction HS7 was prepared by n-hexane extraction of T. camphoratus followed by silica gel chromatography. Its effects on the cell viabilities were determined by sulforhodamine B colorimetric assay. Flow cytometry was used to analyze cell-cycle regulation and apoptosis induction. The changes in cellular protein levels were examined by Western blot. The active fraction HS7 vigorously inhibits AKT-mTOR, ERK and STAT3 signaling pathways in CL1-0 cells. At dose of 25 μg/mL, these signaling pathways were almost completely inhibited by HS7, accompanied with induction of cyclin-dependent kinase inhibitors such as p15, p21 and p27. Accordingly, the AKT-mTOR downstream targets p-p70S6K and HIF-1α were also suppressed as well. At this dose, the cell proliferation was profoundly suppressed to 23.4% of control and apoptosis induction was observed. The active fraction HS7 from n-hexane extract of T. camphoratus exerts multi-targeting activity on the suppression of AKT-mTOR, ERK and STAT3 pathways and induction of p15, p21 and p27 in EGFR wild-type NSCLC cells. This multi-targeting activity of HS7 suggests its potential as an alternative medicine for the treatment of EGFR TKIs resistant NSCLC.

Lung cancer is the leading cause of cancer death worldwide; the most common pathologic type is lung adenocarcinoma (LADC). In spite of the recent progress in targeted therapy, most LADC patients eventually expired due to the inevitable recurrence and drug resistance. New complementary agent with evidence‐based molecular mechanism is urgently needed. MiR‐34a is an important p53 downstream tumor suppressor, which regulates apoptosis, cell‐cycle, EMT (epithelial mesenchymal transition), and so forth. Its expression is deficient in many types of cancers including LADC. Here, we show that a Chinese herbal formula JP‐1 activates p53/miR‐34a axis in A549 human LADC cells (p53 wild‐type). Treatment with JP‐1 induces p53 and its downstream p21 and BAX proteins as well as the miR‐34a, resulting in growth inhibition, colony formation reduction, migration repression, and apoptosis induction. Accordingly, the decreases of miR‐34a downstream targets such as CDK6, SIRT1, c‐Myc, survivin, Snail, and AXL were observed. Moreover, JP‐1 activates AMPKα and reduces mTOR activity, implying its inhibitory effect on the energy‐sensitive protein synthesis and cell proliferation signaling. Our results show that JP‐1 activates p53/miR‐34a tumor suppressor axis and decreases proteins related to proliferation, apoptosis resistance, and metastasis, suggesting its potential as a complementary medicine for LADC treatment.

Air pollution has been recognized to be a risk factor for lung cancer. The objective of this study was to investigate the effects of air pollution on heavy metal alterations in the pleural effusion of lung cancer patients. Pleural effusion was collected from patients with lung cancer and congestive heart failure (CHF). One-year average levels of particulate matter with an aerodynamic diameter of < 10 µm (PM₁₀), PM_2.5, NO₂, and SO₂ were linked to the exposure of these subjects. Traffic-related metals, included Al, Fe, Cu, Zn, and Pb, were determined in the pleural effusion. Logistic regression models were used to examine their associations. There were 63 lung cancer patients and 31 CHF patients enrolled in the current study. We found that PM₁₀, PM_2.5, and NO₂ were negatively correlated with Al in the pleural effusion, whereas PM_2.5 was positively correlated with Zn in the pleural effusion. Increases in 1 μg/m³ of PM_2.5 and 1 ng/mL of Zn were associated with lung cancer (adjusted OR=2.394, 95% CI= 1.446-3.964 for PM_2.5; adjusted OR=1.003, 95% CI=1.000-1.005 for Zn). Increases in PM_2.5 and Zn in the pleural effusion increased the risk of malignant pleural effusion in lung cancer patients (adjusted OR=1.517; 95% CI=1.082-2.127 for PM_2.5; adjusted OR=1.002, 95% CI=1.000-1.005 for Zn). Furthermore, we observed that adenocarcinomas increased in association with a 1-μg/m³ increase in PM_2.5 (crude OR=1.683; 95% CI=1.006-2.817) in lung cancer patients. In conclusion, PM_2.5 exposure and the possible resultant Zn in the pleural effusion associated with the development of malignant pleural effusion in lung cancer.