Abstract Background & aims Idiosyncratic drug‐induced liver injury (DILI) with autoimmune features is a liver condition with laboratory and histological characteristics similar to those of idiopathic autoimmune hepatitis (AIH), which despite being increasingly reported, remains largely undefined. We aimed to describe in‐depth the features of this entity in a large series of patients from two prospective DILI registries. Methods DILI cases with autoimmune features collected in the Spanish DILI Registry and the Latin American DILI Network were compared with DILI patients without autoimmune features and with an independent cohort of patients with AIH. Results Out of 1,426 patients with DILI, 33 cases with autoimmune features were identified. Female sex was more frequent in AIH patients than in the other groups ( p = .001). DILI cases with autoimmune features had significantly longer time to onset ( p < .001) and resolution time ( p = .004) than those without autoimmune features. Interestingly, DILI patients with autoimmune features who relapsed exhibited significantly higher total bilirubin and transaminases at onset and absence of peripheral eosinophilia than those who did not relapse. The likelihood of relapse increased over time, from 17% at 6 months to 50% 4 years after biochemical normalization. Statins, nitrofurantoin and minocycline were the drugs most frequently associated with this phenotype. Conclusions DILI with autoimmune features shows different clinical features than DILI patients lacking characteristics of autoimmunity. Higher transaminases and total bilirubin values with no eosinophilia at presentation increase the likelihood of relapse in DILI with autoimmune features. As the tendency to relapse increases over time, these patients will require long‐term follow‐up.
Summary Background Sofosbuvir, velpatasvir and voxilaprevir (SOF/VEL/VOX) is the recommended rescue therapy for patients with chronic hepatitis C infection who fail direct‐acting antivirals (DAAs). Data are limited on the effectiveness of this treatment after the current first‐line therapies. Our aim was to analyse the effectiveness and safety of SOF/VEL/VOX among patients failing sofosbuvir/velpatasvir (SOF/VEL) or glecaprevir/pibrentasvir (GLE/PIB). Methods Retrospective multicentre study (26 Spanish hospitals), including chronic hepatitis C patients unsuccessfully treated with SOF/VEL or GLE/PIB, and retreated with SOF/VEL/VOX ± ribavirin for 12 weeks between December 2017 and December 2022. Results In total, 142 patients included: 100 (70.4%) had failed SOF/VEL and 42 (29.6%) GLE/PIB. Patients were mainly men (84.5%), White (93.9%), with hepatitis C virus genotype (GT) 3 (49.6%) and 47.2% had liver cirrhosis. Sustained virological response (SVR) was evaluated in 132 patients who completed SOF/VEL/VOX and were followed 12 weeks after end of treatment; 117 (88.6%) achieved SVR. There were no significant differences in SVR rates according to initial DAA treatment (SOF/VEL 87.9% vs. GLE/PIB 90.2%, p = 0.8), cirrhosis (no cirrhosis 90% vs. cirrhosis 87.1%, p = 0.6) or GT3 infection (non‐GT3 91.9% vs. GT3 85.5%, p = 0.3). However, when considering the concurrent presence of SOF/VEL treatment, cirrhosis and GT3 infection, SVR rates dropped to 82.8%. Ribavirin was added in 8 (6%) patients, all achieved SVR. Conclusion SOF/VEL/VOX is an effective rescue therapy for failures to SOF/VEL or GLE/PIB, with an SVR of 88.6%. Factors previously linked to lower SVR rates, such as GT3 infection, cirrhosis and first‐line therapy with SOF/VEL were not associated with lower SVRs.
Abstract Drug-induced liver injury (DILI) diagnosis and classification (hepatocellular, cholestatic, and mixed) relies on traditional clinical biomarkers (eg ALT and ALP), despite limitations such as extrahepatic interferences, narrow dynamic ranges, and low mechanistic value. microRNAs may be very useful for complementing traditional DILI biomarkers but most studies in this direction have considered only paracetamol poisoning. Thus the value of microRNAs (miRNAs) as biomarkers for idiosyncratic DILI has not yet been demonstrated. In this study, we first examined the effect of model cholestatic drugs on the human hepatocyte miRNome by RNAseq and RT-qPCR. Results demonstrated that chlorpromazine, cyclosporin A, and ANIT induced (miR-21-3p, -21-5p, -22-3p, -27a-5p, -1260b, -34a-5p, and -98-5p) and repressed (-122-5p, -192-5p, -30c-5p, -424-5p, and -16-5p) specific miRNAs in sandwich-cultured upcyte hepatocytes. However, no common signature was found for cholestatic drugs. Next we investigated the levels of these miRNA in human serum and found that most were also significantly altered in cholestatic/mixed DILI patients upon hospital/ambulatory admission. However, miR-122-5p, -192-5p, -34a-5p, and -22-3p demonstrated a much more significant induction in patients with hepatocellular DILI, thus revealing better specificity for hepatocellular damage. Time-course analyses demonstrated that -1260b and -146 had a very similar profile to ALP, but with wider dynamic ranges, while -16-5p and -451a showed a negative correlation. Conversely, -122-5p and -192-5p correlated with ALT but with wider dynamic ranges and faster recoveries. Finally, the 122/451a and 122/16 ratios showed excellent prediction performances in both the study [area under the receiver operating characteristic curve (AUROC) >0.93] and the validation cohort (AUROC > 0.82), and can, therefore, be postulated for the first time as circulating miRNA biomarkers for idiosyncratic DILI.
Understanding central mechanisms underlying drug-induced toxicity plays a crucial role in drug development and drug safety. However, a translation of cellular in vitro findings to an actual in vivo context remains challenging. Here, physiologically based pharmacokinetic (PBPK) modeling was used for in vivo contextualization of in vitro toxicity data (PICD) to quantitatively predict in vivo drug response over time by integrating multiple levels of biological organization. Explicitly, in vitro toxicity data at the cellular level were integrated into whole-body PBPK models at the organism level by coupling in vitro drug exposure with in vivo drug concentration-time profiles simulated in the extracellular environment within the organ. PICD was exemplarily applied on the hepatotoxicant azathioprine to quantitatively predict in vivo drug response of perturbed biological pathways and cellular processes in rats and humans. The predictive accuracy of PICD was assessed by comparing in vivo drug response predicted for rats with observed in vivo measurements. To demonstrate clinical applicability of PICD, in vivo drug responses of a critical toxicity-related pathway were predicted for eight patients following acute azathioprine overdoses. Moreover, acute liver failure after multiple dosing of azathioprine was investigated in a patient case study by use of own clinical data. Simulated pharmacokinetic profiles were therefore related to in vivo drug response predicted for genes associated with observed clinical symptoms and to clinical biomarkers measured in vivo. PICD provides a generic platform to investigate drug-induced toxicity at a patient level and thus may facilitate individualized risk assessment during drug development.
Background and Aims: Drug-induced steatosis is a major reason for drug failure in clinical trials and post-marketing withdrawal; and therefore, predictive biomarkers are essential. These could be particularly relevant in non-alcoholic fatty liver disease (NAFLD), where most patients show features of the metabolic syndrome and are prescribed with combined chronic therapies, which can contribute to fatty liver. However, specific biomarkers to assess the contribution of drugs to NAFLD are lacking. We aimed to find microRNAs (miRNAs) responsive to steatotic drugs and to investigate if they could become circulating biomarkers for drug-induced steatosis. Methods: Human HepG2 cells were treated with drugs and changes in miRNA levels were measured by microarray and qRT-PCR. Drug-induced fat accumulation in HepG2 was analyzed by high-content screening and enzymatic methods. miRNA biomarkers were also analyzed in the sera of 44 biopsy-proven NAFLD patients and in 10 controls. Results: We found a set of 10 miRNAs [miR-22-5p, -3929, -24-2-5p, -663a, -29a-3p, -21 (5p and 3p), -27a-5p, -1260 and -202-3p] that were induced in human HepG2 cells and secreted to the culture medium upon incubation with model steatotic drugs (valproate, doxycycline, cyclosporin A and tamoxifen). Moreover, cell exposure to 17 common drugs for NAFLD patients showed that some of them (e.g., irbesartan, fenofibrate, and omeprazole) also induced these miRNAs and increased intracellular triglycerides, particularly in combinations. Finally, we found that most of these miRNAs (60%) were detected in human serum, and that NAFLD patients under fibrates showed both induction of these miRNAs and a more severe steatosis grade. Conclusion: Steatotic drugs induce a common set of hepatic miRNAs that could be used in drug screening during preclinical development. Moreover, most of these miRNAs are serum circulating biomarkers that could become useful in the diagnosis of iatrogenic steatosis.