Over time evaluation of glycaemic control in direct‐acting antiviral‐treated hepatitis C virus/diabetic individuals with chronic hepatitis or with cirrhosis

Abstract Background Data concerning the impact of hepatitis C virus (HCV) cure on type 2 diabetes mellitus (T2DM) are controversial. The aim of the study was to evaluate the effects of anti‐HCV direct‐acting antiviral (DAA) treatments on long‐term glucose control in HCV/T2DM patients with chronic hepatitis C (CHC) or with cirrhosis. Methods One hundred and eighty‐two consecutive HCV/T2DM patients who achieved a viral clearance by DAA treatment were enrolled. Seventy or 182 of them had CHC, and 112 had cirrhosis. Clinical, biochemical and instrumental parameters were recorded at baseline and at 48, 96 and 120 weeks (48w, 96w and 120w, respectively) after stopping DAA therapy. Results At baseline, the overall study population had a mean of glycated haemoglobin (HbA1c) value of 7.2% (ranging from 5 to 11.2), without any significant differences between CHC and cirrhosis [7.1 and 7.2, respectively]. Evaluation over time of HbA1c variations showed a significant improvement of glucose control at all post‐treatment time points compared with baseline in CHC patients (P = .001). In cirrhotic patients, a significant decrease of HbA1c levels was only found when comparing HbA1c values between baseline and 48w time‐point (P = .001), whereas this improvement disappeared at both 98w and 120w (P = .8 and P = .3, respectively). Multivariate logistic regression analysis showed that patients with chronic hepatitis have a 2.5 (CI 1.066‐5.945) times greater chance of achieving an improvement of glycaemic values than patients with liver cirrhosis (P = .035). Conclusion DAA‐based HCV cure induces a significant and persistent amelioration of glycaemic control in HCV/diabetic patients with chronic hepatitis, whereas cirrhotic HCV/diabetic subjects have only a transient benefit from the virus elimination.


| INTRODUC TI ON
Hepatitis C virus (HCV) is a major cause of liver diseases worldwide. 1 Furthermore, HCV has also been associated with a relatively large spectrum of extra-hepatic clinical disorders, including type 2 diabetes mellitus (T2DM). 2,3 In fact, epidemiological data show that about one third of patients with chronic HCV infection have T2DM. 4,5 Coexistence of chronic hepatitis C (CHC) and T2DM is associated with an increased risk of occurrence of the most severe long-term sequelae of liver including cirrhosis and hepatocellular carcinoma (HCC). 6,7 In addition, the risk of cardiovascular (CVD) events in T2DM patients is increased in HCV-infected patients. 8 The recent availability of direct acting antivirals (DAAs) capable of inducing the permanent clearance of HCV in virtually all treated patients with no or minimal side effects has dramatically changed the evolution of HCV-related liver disease, up to a complete clinical recovery in at least those cases in which cirrhosis has not yet developed. At present, however, data concerning the possible impact of DAAs on the HCV-related extra-hepatic manifestations are still insufficient. [9][10][11] In particular, available studies concerning the effects of DAA treatments on glucose control in T2DM subjects are not only numerically limited but also divergent from each other. [12][13][14] Indeed, the studies performed so far in this field have demonstrated several limitations including short-term follow-up and lack of information on possible differences related to the degree of the liver damage in the DAA-treated patients. [15][16][17] The aims of this study were to evaluate whether the DAA-based HCV cures may have an effect on long-term glucose control in T2DM patients and whether this possible effect might be influenced by the grade of severity of the liver disease.

| PATIENTS AND ME THODS
The design of this prospective study was to evaluate over time the glycaemic control in HCV/T2DM patients with severe but still compensated liver disease after successful DAA cure. Prescription of interferon-free, DAA-based anti-HCV therapies has been available in Italy since April 2015, in accordance with the treatment initially limited to some categories of patients. In particular, subjects with cirrhosis in class A of Child-Pugh (C-P) score or with advanced grade of chronic hepatitis (viz., F3 CHC) represented the vast majority of patients that could be treated. 18  Additional criteria for making diagnosis of cirrhosis were a previous liver biopsy showing stage 4 fibrosis, presence of oesophageal and/ or gastric varices at endoscopy (EGS) examination, platelet count lower than 100 × 10 3 /mmc, and typical ultrasound features. [20][21][22] According to the design of the study, the following data were collected from each patient and recorded in a data set at baseline: de-

| S TATIS TIC AL ANALYS IS
The numerical data were expressed as median and range (minimum and maximum) and the categorical variables as number and percentage.
The non-parametric approach was used because most of the numerical variables were not normally distributed, such as verified by test when in addition to considering whether each observation is below or above a value of interest, we want to take into account the dimension of the observations. The same tests were used to perform a paired comparison of clinical characteristics between baseline versus 48, 96 and 120 weeks, respectively.
Finally, univariate and multivariate logistic regression models were estimated in order to identify significant predictive factors of glucose control. The covariates were gender, age, BMI, stage of liver fibrosis (chronic hepatitis vs cirrhosis), HCV genotypes, presence of varices, albumin, gamma globulin, creatinine, AST ALT, GGT, liver decompensation, diabetes duration, diabetes family history, diabetes complications, and insulin treatment. The results were expressed as odds ratio (OR) with relative 95% confidence interval (CI) and P

value.
Statistical analyses were performed using SPSS 22.0 for Windows package.

| RE SULTS
Note: Bold characters identify statistically significant variables.
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; FBG, fasting blood glucose; GGT, gamma glutamyl transpeptidase; HCV, hepatitis C virus. a Elastographic examination was not performed in five CHC and in four cirrhotic patients due to BMI > 33kg/m 2 . All numerical parameters are expressed as median and range except for those otherwise indicated.  Table 2).
Liver stiffness values significantly decreased in all the CHC cases

| D ISCUSS I ON
The relationship between HCV chronic infection and T2DM is an important and largely debated argument, which involves specialists of different areas because of its relevant clinical implications and challenging physio-pathological aspects. [25][26][27][28] After the recent advent of DAA-based therapy which is able to quickly and definitively cure the HCV infection, the interest about interplay between HCV and diabetes is even greater. A major question arising is whether HCV elimination might have any effect on metabolic control in T2DM patients. However, contrasting data are available in this field, because the studies performed so far show substantial differences in the enrolled patient populations and limited durations of the post-DAA-treatment follow up. 29,30 Our study showed that subjects with chronic hepatitis had a significant and persistent amelioration of the glycaemic control and antidiabetic therapy could even be stopped TA B L E 2 Liver biochemistry and liver stiffness in CHC and cirrhotic subjects at baseline and at the end of follow up (120 weeks after stopping DAA treatment) To explain why the HCV recovery has only an initial, transient benefit on glucose metabolism in cases with advanced liver disease, one must consider both the multifactorial pathophysiological basis of diabetes and the large architectural, vascular and functional rearrangements occurring in the liver once cirrhosis is developed. Insulin resistance, muscle sarcopenia, deficit in hormones clearance and a 'toxic' effect on pancreatic islets are all suggested mechanisms determining an impairment of glucose metabolism in patients with liver cirrhosis, thus potentially contributing to the lack of long-term metabolic benefits of DAA treatment observed in these subjects. [33][34][35] Several potential limitations should be taken into account when interpreting our results, including the lack of measurements of insulin, C-peptide, and HOMA-IR as well as of inflammatory markers, which could have added useful information to explain our observations.
The recruitment of all consecutively treated T2DM individuals from a cohort of DAA treated subjects and the length of follow up are among the strengths of this study, rendering of particular relevance the emerged results. We are now continuing to follow the study population, and we are planning to report further data after at least 5 years of follow up, to continue providing information about the impact of DAA-based cure on both diabetes and liver disease of HCV/diabetic subjects.