Prediction of MET Overexpression in Lung Adenocarcinoma from Hematoxylin and Eosin Images.

Mesenchymal epithelial transition (MET) protein overexpression is a targetable event in non-small cell lung cancer and is the subject of active drug development. Challenges in identifying patients for these therapies include lack of access to validated testing, such as standardized immunohistochemistry assessment, and consumption of valuable tissue for a single gene/protein assay. Development of prescreening algorithms using routinely available digitized hematoxylin and eosin (H&E)-stained slides to predict MET overexpression could promote testing for those who will benefit most. Recent literature reports a positive correlation between MET protein overexpression and RNA expression. In this work, a large database of matched H&E slides and RNA expression data were leveraged to train a weakly supervised model to predict MET RNA overexpression directly from H&E images. This model was evaluated on an independent holdout test set of 300 overexpressed and 289 normal patients, demonstrating a receiver operating characteristic area under curve of 0.70 (95th percentile interval: 0.66 to 0.74) with stable performance characteristics across different patient clinical variables and robust to synthetic noise on the test set. These results suggest that H&E-based predictive models could be useful to prioritize patients for confirmatory testing of MET protein or MET gene expression status.

Independent of Cirrhosis, Hepatocellular Carcinoma Risk Is Increased with Diabetes and Metabolic Syndrome.

BACKGROUND: Hepatocellular carcinoma is the most common primary liver malignancy, commonly a sequelae of hepatitis C infection, but can complicate cirrhosis of any cause. Whether metabolic syndrome and its components, type II diabetes, hypertension, and hyperlipidemia increase the risk of hepatocellular carcinoma independent of cirrhosis is unknown. METHODS: A retrospective cohort study was conducted using the MarketScan insurance claims database from 2008-2012. Individuals with hepatocellular carcinoma aged 19-64 years and age and sex-matched controls were included. Multivariate analysis of hepatocellular carcinoma risk factors was performed. RESULTS: Hepatitis C (odds ratio [OR] 2.102) was the largest risk factor for hepatocellular carcinoma. Other independent risk factors were type II diabetes (OR 1.353) and hypertension (OR 1.229). Hyperlipidemia was protective against hepatocellular carcinoma (OR 0.885). The largest risk increase occurred with hypertension with type II diabetes and hepatitis C (OR 4.580), although hypertension and type II diabetes without hepatitis C still incurred additional risk (OR 3.399). Type II diabetes and hyperlipidemia had a similar risk if hepatitis C was present (OR 2.319) or not (OR 2.395). Metformin (OR 0.706) and cholesterol medications (OR 0.645) were protective in diabetics. Insulin (OR 1.640) increased the risk of hepatocellular carcinoma compared with the general type II diabetes population. CONCLUSION: In the absence of cirrhosis, type II diabetes and hypertension were independent risk factors for hepatocellular carcinoma. Hyperlipidemia and medical management of type II diabetes with metformin and cholesterol medication appeared to reduce the incidence of hepatocellular carcinoma. In contrast, insulin was associated with a higher risk of hepatocellular carcinoma.

Colorectal cancer screening in high-risk groups is increasing, although current smokers fall behind.

BACKGROUND: There is limited information about colorectal cancer (CRC) screening trends in high-risk groups, including the black, obese, diabetic, and smoking populations. For this study, the authors evaluated national CRC screening trends in these high-risk groups to provide insights into whether screening resources are being appropriately used. METHODS: This was a nationally representative, population-based study using the Behavioral Risk Factor Surveillance System from the Centers for Disease Control. Data analysis was performed using bivariate analyses with weighted logistic regression. RESULTS: In the general population, CRC screening increased significantly from 59% to 65% during the years 2006 to 2010. The screening prevalence in non-Hispanic blacks was 58% in 2006 and 65% in 2010. Among obese individuals, the prevalence of up-to-date CRC screening increased significantly from 59% in 2006 to 66% in 2010. Screening prevalence in individuals with diabetes was 63% in 2006 and 69% in 2010. The CRC screening prevalence in current smokers was 45% in 2006 and 50% in 2010. The odds of CRC screening in the non-Hispanic black population, the obese population, and the diabetic population were higher than in non-Hispanic whites, normal weight individuals, and the population without diabetes, respectively. Current smokers had significantly lower odds of CRC screening than never-smokers in the years studied (2006: odds ratio [OR], 0.71; 95% confidence interval [CI], 0.66-0.76; 2008: OR, 0.67; 95% CI, 0.63-0.71; 2010: OR, 0.69; 95% CI, 0.66-0.73). CONCLUSIONS: The prevalence of CRC screening in high-risk groups is trending upward. Despite this, current smokers have significantly lower odds of CRC screening compared with the general population.