Liver disease carries significant healthcare burden and frequently requires a combination of blood tests, imaging, and invasive liver biopsy to diagnose. Distinguishing between inflammatory liver diseases, which may have similar clinical presentations, is particularly challenging. In this study, we implemented a machine learning pipeline for the identification of diagnostic gene expression biomarkers across several alcohol-associated and non-alcohol-associated liver diseases, using either liver tissue or blood-based samples. We collected peripheral blood mononuclear cells (PBMCs) and liver tissue samples from participants with alcohol-associated hepatitis (AH), alcohol-associated cirrhosis (AC), non-alcohol-associated fatty liver disease, chronic HCV infection, and healthy controls. We performed RNA sequencing (RNA-seq) on 137 PBMC samples and 67 liver tissue samples. Using gene expression data, we implemented a machine learning feature selection and classification pipeline to identify diagnostic biomarkers which distinguish between the liver disease groups. The liver tissue results were validated using a public independent RNA-seq dataset. The biomarkers were computationally validated for biological relevance using pathway analysis tools. Utilizing liver tissue RNA-seq data, we distinguished between AH, AC, and healthy conditions with overall accuracies of 90% in our dataset, and 82% in the independent dataset, with 33 genes. Distinguishing 4 liver conditions and healthy controls yielded 91% overall accuracy in our liver tissue dataset with 39 genes, and 75% overall accuracy in our PBMC dataset with 75 genes. Our machine learning pipeline was effective at identifying a small set of diagnostic gene biomarkers and classifying several liver diseases using RNA-seq data from liver tissue and PBMCs. The methodologies implemented and genes identified in this study may facilitate future efforts toward a liquid biopsy diagnostic for liver diseases. Distinguishing between inflammatory liver diseases without multiple tests can be challenging due to their clinically similar characteristics. To lay the groundwork for the development of a non-invasive blood-based diagnostic across a range of liver diseases, we compared samples from participants with alcohol-associated hepatitis, alcohol-associated cirrhosis, chronic hepatitis C infection, and non-alcohol-associated fatty liver disease. We used a machine learning computational approach to demonstrate that gene expression data generated from either liver tissue or blood samples can be used to discover a small set of gene biomarkers for effective diagnosis of these liver diseases.

Introduction: Therapeutic voriconazole concentrations have a narrow window of effectiveness before voriconazole causes cholestatic hepatitis. Case presentation: After one year of voriconazole therapy for pulmonary aspergillosis, a 44-year-old male began 30 mg of lansoprazole for gastroesophegal reflux symptoms. Within five days he presented with fatigue, jaundice and cholestatic hepatitis. The hepatitis promptly resolved after stopping lansoprazole. Sixteen months later, he was given simvastatin as recommended by the American Diabetes Association for diabetic patients over the age of 40 with one additional risk factor to prevent cardiovascular disease. Within 2-weeks of taking simvastatin, the patients redeveloped fatigue, jaundice and cholestatic hepatitis. The patient described that both episodes of fatigue and jaundice were similar in onset and intensity. Conclusion: Voriconazole is metabolized by both cytochrome P450 2C19 and 3A4 isoenzymes. Lansoprazole is an inhibitor of 2C19 isoenzyme. Likely competition between voriconazole and lansoprazole led to increase voriconazole serum concentration and acute cholestatic hepatitis. Simvastatin inhibits the P450 3A4 isoenzyme. After 10 mg of simvastatin a day for two weeks, cholestatic hepatitis occurred. The voriconazole concentration remained elevated (4.1 ug/ml) when measured 15-days after stopping simvastatin. Use of the Naranjo probability scale revealed that the cholestatic hepatitis was probably precipitated by lansoprazole (score 7). Likewise, the Naranjo probability scale revealed that cholestatic hepatitis was definite (score 9) adverse drug reaction precipitated by the addition of the simvastatin to the stable baseline regimen of voriconazole. In a single patient, by two different inhibitors of the p450 pathway, voriconazole induced cholestatic hepatitis occurred twice. While the major P450 pathway for the metabolism and clearance of lansoprazole and simvastatin are different, they both likely contributed to the reduced hepatic clearance of voriconazole. In two independent periods, an increased voriconazole concentration resulted in cholestatic hepatitis within 1-2 weeks of therapy, which resolved after stopping each of the offending medications.

Prandial insulin has been essential for the improved management of the type 1 diabetic patient. Interestingly, many studies have evaluated the addition of prandial insulin to the type 2 diabetic patients with improved control. The greatest drop in A1c with the use of various type of prandial insulins have resulted in the decrease of 1.3% in the A1c measurement. Interestingly, none of the published trials with goal of fasting blood glucose (FBG) have ever obtained the goal A1c. Since a drop in FBG of 28.5mg/dl is equal to a 1% drop in A1c, a simple approach to obtain a target A1c would be to focus on the FBG (per ADA: Average Blood Glucose = A1c (%) x 28.7 - 46.7mg/d). However, average blood glucose requires multiple measurements and may be less accurate then using just a FBG. Since prandial insulin clinical trials have only demonstrated a drop in A1c by 0.3-1.3% the use of only a FBG to help patients get to goal may be easier to teach and to obtain and it might save time and money. Our hypothesis is that if patient obtain a FBG <100mg/dl of a few month period, then 70% will be at an A1c goal <7.0%. After a few months of good fasting glucose control the provider can use this equation (FBG+80)/30 to estimate A1c. For example, a FBG of 130mg/dl would be (130 + 80)/30 = 7.0%; or a FBG of 190 would be (190+80)/30 =eA1c 9% (estimate of A1c). While type 1 diabetes is a very complex daily glucose pattern, the approach to type 2 diabetics on insulin could become simplified.