Sodium-glucose cotransporter 2 inhibitors (SGLT2-Is) comprise a new class of antidiabetic agents that inhibit glucose reabsorption in the renal proximal tubules. Although a recent report demonstrated the potential ability of SGLT2-Is to attenuate cancer growth of SGLT2-expressing cancer cells, little is known about the effects of SGLT2-Is on hepatocellular carcinoma (HCC). Here, we investigate the anti-cancer properties of a SGLT2-I, canagliflozin, against human liver cancer cells. SGTL2 mRNA and protein expression were detected in Huh7 and HepG2 cells, although not in HLE as well as primary human hepatocytes and hepatic stellate cells. Canagliflozin exerted antiproliferative effects on SGLT2-expressing Huh7 and HepG2 cells in a dose-dependent manner by inhibiting glycolytic metabolism including glucose uptake, lactate and intracellular ATP production. This agent also induced G2/M arrest and apoptosis with inhibited phosphorylation of ERK, p38 and AKT and cleavage of caspase3. Xenograft tumor growth assay showed that oral administration of canagliflozin (10 mg/kg/day) significantly reduced subcutaneous tumor burdens in a glycemic status-independent manner, and attenuated intratumor vascularization in Huh7- and HepG2-derived xenograft tumors in BALB/c nude mice. In vitro, canagliflozin suppressed the increased human umbilical vein endothelial cell (HUVEC) proliferation and tubular formation which were observed in Huh7 or HepG2 co-cultures. By contrast, canagliflozin had no effect on tumor growth and intratumor angiogenesis in SGLT2-null HLE-derived xenograft models. These results indicate that SGLT2-I therapy is a potential new strategy for the treatment of HCC.

AIMTo determine the efficacy of rifaximin for hepatic encephalopathy (HE) with the linkage of gut microbiome in decompensated cirrhotic patients.METHODSTwenty patients (12 men and 8 women; median age, 66.8 years; range, 46-81 years) with decompensated cirrhosis (Child-pugh score > 7) underwent cognitive neuropsychological testing, endotoxin analysis, and fecal microbiome assessment at baseline and after 4 wk of treatment with rifaximin 400 mg thrice a day. HE was determined by serum ammonia level and number connection test (NCT)-A. Changes in whole blood endotoxin activity (EA) was analyzed by endotoxin activity assay. Fecal microbiome was assessed by 16S ribosome RNA (rRNA) gene sequencing.RESULTSTreatment with rifaximin for 4 wk improved hyperammonemia (from 90.6 ± 23.9 μg/dL to 73.1 ± 33.1 μg/dL; P < 0.05) and time required for NCT (from 68.2 ± 17.4 s to 54.9 ± 20.3 s; P < 0.05) in patients who had higher levels at baseline. Endotoxin activity was reduced (from 0.43 ± 0.03 to 0.32 ± 0.09; P < 0.05) in direct correlation with decrease in serum ammonia levels (r = 0.5886, P < 0.05). No statistically significant differences were observed in the diversity estimator (Shannon diversity index) and major components of the gut microbiome between the baseline and after treatment groups (3.948 ± 0.548 at baseline vs 3.980 ± 0.968 after treatment; P = 0.544), but the relative abundances of genus Veillonella and Streptococcus were lowered.CONCLUSIONRifaximin significantly improved cognition and reduced endotoxin activity without significantly affecting the composition of the gut microbiome in patients with decompensated cirrhosis.

AimAlthough some relationships between gut microbiota and liver diseases have been reported, it remains uncertain whether changes in gut microbiota owing to differences in race, food and living environment have similar effects. Response to ursodeoxycholic acid (UDCA) may predict the long‐term prognosis of patients with primary biliary cholangitis (PBC); however, little is known about the significance of the gut microbiome in patients with PBC. We elucidated the relationships among clinical profiles, biochemical response to UDCA and gut microbiome composition in patients with PBC.MethodsFecal samples from 76 patients with PBC treated at our hospital were collected; patients whose UDCA intake period was <1 year were excluded. The microbiome structures of patients were determined using 16S ribosomal RNA gene sequencing and were statistically compared with those of healthy subjects. The structures of patients in the UDCA responder (n = 43) and non‐responder (n = 30) groups were compared according to the Nara criteria (reduction rate of gamma‐glutamyl transpeptidase, ≥69%, after 1 year).ResultsCompared with healthy subjects, bacterial diversity was lower in patients with PBC, with a decreased abundance of the order Clostridiales and increased abundance of Lactobacillales. The UDCA non‐responder group had a significantly lower population of the genus Faecalibacterium, known as butyrate‐producing beneficial bacteria (P < 0.05), although no significant differences in gender, body mass index, medicated drugs or other serological data were indicated between these two groups.ConclusionsGut dysbiosis with loss of beneficial Clostridiales commensals was observed in patients with PBC. Decrease in Faecalibacterium abundance might predict the long‐term prognosis of patients with PBC.

Background and Aim Hypothyroidism might play a crucial role in the pathogenesis of non‐alcoholic fatty liver disease (NAFLD). The association of subclinical hypothyroidism with NAFLD has been inconsistent. The relationship of NAFLD with thyroid function parameters and subclinical hypothyroidism was determined. Methods This cross‐sectional study included 70 patients with subclinical hypothyroidism and 70 controls with euthyroidism matched according to gender, age, and body mass index (BMI). NAFLD was diagnosed via abdominal ultrasonography. The association between NAFLD and subclinical hypothyroidism was analyzed. Results The prevalence of NAFLD was significantly higher in patients with subclinical hypothyroidism than in those with euthyroidism. Multivariate analysis showed that subclinical hypothyroidism was an independent risk factor of NAFLD adjusted by metabolic‐related factors, such as BMI, triglyceride, high‐density lipoprotein‐cholesterol, hypertension, and diabetes. Thyroid‐stimulating hormone (TSH) was an independent risk factor of NAFLD adjusted by the same metabolic‐related factors, but free thyroxine (FT4) was not a risk factor. The FIB‐4 index, a noninvasive marker of liver fibrosis was significantly higher in patients with subclinical hypothyroidism than in those with euthyroidism. Compared with patients with euthyroidism, the proportion of the FIB‐4 index ≥2.67 was significantly higher, and the proportion of the FIB‐4 index <1.30 was lower in patients with subclinical hypothyroidism. Conclusions TSH elevation even within the euthyroid range is an independent risk factor of NAFLD and may influence the progression of liver fibrosis, even with a normal FT4 level.

No abstract