- Email: [email protected]
Prevalence and severity of liver injury after pulsed methylprednisolone therapy in multiple sclerosis patients
POSTER PRESENTATIONS obesity is a risk factor for acute liver failure, high cholesterol intake, which often underlies obesity, is suggested to play a role in the mechanism. In the present study, we aimed to elucidate the effect of high-cholesterol (HC) diet on acute APAP-induced liver injury. Methods: C57BL/6 wild-type, toll-like receptor 9 (TLR9) knockout (KO), caspase-1 KO, and apoptosis-associated speck-like protein containing a caspase recruitment domain (ASC) KO mice were fed a HC diet for 4 weeks, followed by treatment with APAP. Liver sinusoidal endothelial cells (LSECs) were isolated from the mice for in vitro analyses. Results: The HC diet significantly increased acute APAP-induced liver injury. In TLR9 KO mice, the exacerbating effects of the HC diet on disease pathology disappeared. Similarly, the HC diet did not exacerbate APAP-induced liver injury in both caspase-1 KO and ASC KO mice. In response to acute APAP-induced liver injury, the expression of liver pro-IL-1β and pro-IL-18 mRNA increased significantly. The HC diet significantly enhanced this effect. In contrast, in the livers of TLR9 KO mice, the enhancement effect with HC intake disappeared. In acute APAP-induced liver injury, cholesterol loading significantly increased the production of mature IL-1β and IL-18 proteins by LSECs, but yielded no change in immune cells. Free cholesterol (FC) accumulation was significantly higher in LSECs isolated from HC mice than in those derived from mice fed a control diet. DNA from dying hepatocytes (dying DNA) significantly enhanced the production of pro-IL-1β and pro-IL-18 mRNA in LSECs from control mice and this enhancement was significantly greater in LSECs from HC mice, in a TLR9 signal-dependent manner. Furthermore, in LSECs with increased cellular FC following treatment with both acetyl low-density lipoprotein and the acyl-CoA:cholesterol acyltransferase inhibitor, dying DNA significantly increased the expression of pro-IL-1β and pro-IL-18 mRNA compared with control LSECs in a TLR9 signal-dependent manner. Conclusions: HC intake exaggerated acute APAP-induced liver injury via the accumulation of FC in LSECs, which enhanced TLR9-mediated signaling, thereby exacerbating the disease pathology through activation of the TLR9/inflammasome pathway.
Methods: We analysed data from 847 DILI episodes that occurred in 816 patients enrolled by the Spanish DILI Registry. Cases were categorized and compared according to age; 65 years or older (N = 279) or younger than 65 years (N = 568). Results: Equal gender distribution was seen among younger DILI patients, while more males (56%) were found in the elderly group. Pattern of liver injury was significantly different between the two groups with the majority of the younger population being hepatocellular (70%), whereas the elderly were equally distributed between hepatocellular and cholestatic/mixed patterns (P < 0.001). Cases with mild severity were more common in the younger group (36% vs. 18%), while cases of moderate severity were more common among the elderly (68% vs. 52%) (P < 0.001). Comorbid conditions such as diabetes, hypertension and dyslipidemia were more frequent in the elderly (P < 0.001) who were characterized by a higher body mass index (BMI). Jaundice and lymphopenia were significantly more common in the elderly (P < 0.001). Percentage of spontaneous recovery was higher in the younger group (64% vs. 57%) whereas the need for hospitalization was significantly higher among the elderly (64% vs. 47%, P < 0.001). Older patients required more time to recover (median of 116 vs. 100 days, P = 0.028). While overall mortality due to any reason was higher among the elderly, liverrelated death/transplantation rates did not differ significantly. Pharmacologic drug classes most commonly implicated in DILI were antiinfectives (35%) and central nervous system agents (14%) for younger patients and antiinfectives (47%) and cardiovascular agents (14%) for the elderly. Conclusions: Male gender, cholestatitc/mixed pattern of injury and higher BMI are the main features that characterize older DILI patients. Increasing age was linked to higher proportions of jaundice and lymphopenia, less spontaneous recovery rates and longer recovery periods. Despite significantly higher proportions of comorbidities, differences in liver-related death and transplantation rates did not reach statistical significance. Funding: AEMPS and ISCIII-FEDER (PI12/00378, PI15/01440,AC-00732013, CIBERehd).
FRI-444 Clinical characteristics and outcomes of elderly included in the Spanish Drug-Induced Liver Injury (DILI) Registry M. Slim1, J. Sanabria1, M. Robles-Díaz1, I. Medina-Cáliz1, A. González-Jiménez1, A. Ortega1, M. García-Cortés1, B. Garcia-Muñoz1, R. Sanjuán-Jiménez1, M.C. Fernandez2, A. Castiella3, E. Zapata3, H. Hallal4, E. Montane5, G. Soriano6, E. Roman6, J.M. Navarro7, M. Jimenez8, M. Prieto9, I. Conde9, A. Aldea10, M. Hernández-Guerra10, J.M. Moreno11, M. Romero-Gomez12, E. Gomez-Dominguez13, M.I. Lucena1, R.J. Andrade1. 1Instituto de Investigación Biomédica de Málaga (IBIMA), Hospital Universitario Virgen de la Victoria., Universidad de Málaga, Malaga; 2Hospital Torrecardenas, Almeria; 3Hospital de Mendaro, Guipuzcoa; 4Hospital Morales Meseguer, Murcia; 5Hospital Germans Trias i Pujol, Badalona; 6Hospital de la Santa Creu i Sant Pau. CIBERehd, Barcelona; 7Agencia Sanitaria Costa del Sol; 8Instituto de Investigación Biomédica de Málaga (IBIMA), Hospital Regional de Málaga, Malaga; 9 Hospital Universitario y Politécnico La Fe. CIBERehd., Valencia; 10 Hospital Universitario de Canarias, Tenerife; 11Complejo Hospitalario de Albacete, Albacete; 12Hospitales Universitarios Virgen MacarenaVirgen del Rocío. CIBERehd, Sevilla; 13Hospital 12 de Octubre, Madrid, Spain E-mail: [email protected]
FRI-445 Prevalence and severity of liver injury after pulsed methylprednisolone therapy in multiple sclerosis patients M. Biolato1, V. Nociti2, G. Marrone1, A. Liguori1, C. De Fino2, A. Bianco2, M. Mirabella2, A. Grieco1. 1Liver Transplant Medicine Unit, Gastroenterological Area, Gastroenterological and EndocrinoMetabolical Sciences Department; 2Multiple Sclerosis Center, Neuroscience Area, Neuroscience, Aging, Head and neck and Orthopaedics Sciences Department, Fondazione Policlinico Universitario Gemelli, Catholic University of Sacred Heart, Rome, Italy E-mail: [email protected]
Background and Aims: DILI is a rare adverse drug reaction that can be associated with morbidity and mortality. Characterisation and classification of phenotypic sub-groups is crucial for improving diagnosis and causality assessment in DILI. We aimed to compare clinical characteristics and outcomes between elderly and younger patients diagnosed with DILI.
Background and Aims: Few cases of high dose pulsed methylprednisolone-related liver injury in patients with Multiple Sclerosis are reported in literature but a prospective investigation was never been performed. The aim of the study is to evaluate the prevalence and severity of liver injury in patients affected by Multiple Sclerosis and treated with pulsed methylprednisolone because of clinical relapses. Methods: Prospective observational single-center study on patients with multiple sclerosis, in relapsing phase of the disease, treated with i.v. methylprednisolone 1,000 mg/day for 5 days. We tested liver function tests (aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, total bilirubin) before treatment and after two weeks. In case of severe liver injury, defined according to “Hy’s law”, a comprehensive hepatologic work-up was performed. Results: During 12 month observation period, we collected 251 cycles of i.v. high dose steroid treatment from 175 patients (65.1% females, mean age of 40.8 ± 12.2 years; mean Expanded disability status scale 2.8 ± 2.1). After excluding 8 cycles presenting basal alteration of liver function tests, serum aminotransferase elevation (any grade) was observed on twenty-one cycles two weeks after methylprednisolone
Journal of Hepatology 2017 vol. 66 | S333–S542
S399
POSTER PRESENTATIONS therapy ( prevalence 8.6%). Six of them ( prevalence 2.5%) presented a severe liver injury according to Hy’s law; after comprehensive hepatologic work-up, three out of them received a final diagnosis of drug-induced liver injury and other three a final diagnosis of autoimmune hepatitis. Conclusions: Liver injury should be considered a possible adverse event in patients with multiple sclerosis treated with pulsed methylprednisolone therapy for clinical relapse. Aminotransferase monitoring two weeks after pulsed methylprednisolone treatment is useful for clinical management. FRI-446 Jnk2 is indispensable in murine and human Ibuprofen-induced acute liver failure M.E. Zoubek1, M.M. Woitok1, S. Sydor2, L.J. Nelson3, L.P. Bechmann2, M.I. Lucena4,5, R.J. Andrade4,5, A. Bast6, G.H. Koek7, C. Trautwein1, F.J. Cubero1. 1Department of Internal Medicine III, University Hospital RWTH Aachen, Aachen; 2Department of Gastroenterology and Hepatology, University Hospital, University Duisburg-Essen, Essen, Germany; 3 Laboratory Department of Hepatology, Royal Infirmary of Edinburgh, University of Edinburgh, Edinburgh, United Kingdom; 4 IBIMA, H. Virgen de la Victoria, Universidad de Málaga, Malaga; 5 CIBERehd, Barcelona, Spain; 6Department of Pharmacology and Toxicology, Maastricht University; 7Division of Gastroenterology and Hepatology, Department of Internal Medicine, Maastricht University Medical Centre, Maastricht, The Netherlands E-mail: [email protected] Background and Aims: Ibuprofen is a frequently used non-steroidal anti-inflammatory drug (NSAID). Although it is generally welltolerated it can, nevertheless, cause serious liver injury. Our earlier investigations clearly demonstrated that the JNK pathway plays a relevant role in the pathophysiology of Ibuprofen-induced ALF, thus, we aimed to analyze the role of Jnk1 and Jnk2 which, at present, remains unknown. Methods: First we investigated activation of essential pathways in liver sections of Ibuprofen-induced ALF patients. Next we developed a novel murine model of Ibuprofen intoxication. Overnight fasted male C57BL/6 mice (6–8 weeks of age) were i.p injected with either vehicle or Ibuprofen 600 mg/kg for 8 h. To assess the role of JNK, we used animals carrying constitutive deletion of Jnk1 (Jnk1-/-) or Jnk2 (Jnk2-/-) expression and, later, expanded our investigations to animals with either genetic deletion of Jnk1 (Jnk1Δhepa) or siRNA-mediated deletion of Jnk2 (siJnk2Δhepa) exclusively in hepatocytes. Results: JNK phosphorylation was increased in the cytoplasm of hepatocytes and other non-LPCs derived from Ibuprofen-induced ALF liver samples of human and murine origin, compared with healthy tissue. Ibuprofen overdose produced a significantly stronger degree of liver injury compared with vehicle-treated controls as evidenced by transaminases, macroscopic as well as microscopic analysis of liver tissues. Next, we sought to investigate the molecular pathways associated with Ibuprofen overdose in mice. Interestingly, we found increased activation of PKCα, AKT and JNK, and decreased p38 protein levels, 8 h after Ibuprofen intoxication. Constitutive Jnk1-/- and Jnk2-/deficient mice exhibited a greater liver dysfunction than wildtype (WT) animals. In contrast, siJnk2Δhepa animals showed a remarkable decline of liver injury during Ibuprofen-induced ALF, which correlated with significantly higher serum liver enzymes and worsened liver histology, compared to Jnk1Δhepa or WT animals. Conclusions: In our study, cytoplasmic JNK activation in hepatocytes and other non-LPCs is a hallmark of human and murine Ibuprofeninduced ALF. Ibuprofen overdose activates PKCa/AKT/JNK- and attenuates p38-signaling in mice. Functional analysis in vivo demonstrates that JNK2 in hepatocytes is essential in mediating protection during Ibuprofen-dependent ALF and thus is a molecular target for treatment.
S400
FRI-447 Development of a novel 3D organotypic human HepaRG-based non-invasive imaging platform for pre-clinical drug screening N.J. Martucci1, K. Morgan1, G. Anderson1, P.C. Hayes1, J.N. Plevris1, P.O. Bagnaninchi2, L. Nelson1. 1Hepatology; 2MRC Centre for Regenerative Medicine, University of Edinburgh, Edinburgh, United Kingdom E-mail: [email protected] Background and Aims: Drug-induced liver injury (DILI) is the leading cause of attrition in pre-clinical drug development and withdrawals, hampered by both the lack of physiologically-relevant human models and reliance on animal data. Therefore, strong incentives exist to create new human-based, in vitro liver models, which would offer better pre-clinical prediction of DILI. Human hepatic HepaRG (hepatocyte: cholangiocyte) coculture maintain in vivo-like liver-specific functions and functional polarity and are considered a sustainable surrogate to primary human hepatocytes. We aimed to develop organotypic 3D liver spheroids, coupled with a novel imaging platform based on optical coherence tomography (OCT), as a high throughput amenable approach for drug toxicity screening using paracetamol as a model hepatotoxin. Methods: HepaRGs were cultured at different densities (0–50,000/ well) to optimize 3D spheroid size using ultra low-adherence plates [InSpheroTM]. OCT was used to monitor 3D spheroids in real time and to quantify cell viability, proliferation; and correlated with biochemical assays. H&E staining was performed (d7; d14) to verify viability (presence/absence of necrotic core) of spheroids. The optimal spheroid size was defined as having a round spheroidal shape with no necrotic core. To further validate the 3D model, viability (OCT; ATP), metabolic competency (Prestoblue) and cell proliferation (wst8) was quantified following 24-hour dose-response to paracetamol. Results: Enhanced viability, proliferation, and metabolic competency was observed with an optimal spheroid size of 2,500 HepaRG cells (also vs 2D HepaRG model). OCT was able to image 200 μm Ø spheroids non-destructively with a 5 μm resolution (depth/ penetration); whilst time-lapse OCT image-acquisition with optical fluctuation -maps were found to correlate directly to cell viability and calculated at each dose; giving unique insight into both drug penetration and toxicity. Finally, OCT cell viability values were integrated all over the samples and correlated to biochemical assays.
Conclusions: The 3D-HepaRG-OCT platform has the potential to assess noninvasively and label-free drug toxicity/penetration in organotypic 3D-models providing an easily reproducible model system for improved predictive drug toxicity screening and potentially to reduce the 3Rs. Future work involves further validation of the model’s similarity to liver function in vivo, and testing other modes of DILI, such as cholestasis.
Journal of Hepatology 2017 vol. 66 | S333–S542
Methods: We analysed data from 847 DILI episodes that occurred in 816 patients enrolled by the Spanish DILI Registry. Cases were categorized and compared according to age; 65 years or older (N = 279) or younger than 65 years (N = 568). Results: Equal gender distribution was seen among younger DILI patients, while more males (56%) were found in the elderly group. Pattern of liver injury was significantly different between the two groups with the majority of the younger population being hepatocellular (70%), whereas the elderly were equally distributed between hepatocellular and cholestatic/mixed patterns (P < 0.001). Cases with mild severity were more common in the younger group (36% vs. 18%), while cases of moderate severity were more common among the elderly (68% vs. 52%) (P < 0.001). Comorbid conditions such as diabetes, hypertension and dyslipidemia were more frequent in the elderly (P < 0.001) who were characterized by a higher body mass index (BMI). Jaundice and lymphopenia were significantly more common in the elderly (P < 0.001). Percentage of spontaneous recovery was higher in the younger group (64% vs. 57%) whereas the need for hospitalization was significantly higher among the elderly (64% vs. 47%, P < 0.001). Older patients required more time to recover (median of 116 vs. 100 days, P = 0.028). While overall mortality due to any reason was higher among the elderly, liverrelated death/transplantation rates did not differ significantly. Pharmacologic drug classes most commonly implicated in DILI were antiinfectives (35%) and central nervous system agents (14%) for younger patients and antiinfectives (47%) and cardiovascular agents (14%) for the elderly. Conclusions: Male gender, cholestatitc/mixed pattern of injury and higher BMI are the main features that characterize older DILI patients. Increasing age was linked to higher proportions of jaundice and lymphopenia, less spontaneous recovery rates and longer recovery periods. Despite significantly higher proportions of comorbidities, differences in liver-related death and transplantation rates did not reach statistical significance. Funding: AEMPS and ISCIII-FEDER (PI12/00378, PI15/01440,AC-00732013, CIBERehd).
FRI-444 Clinical characteristics and outcomes of elderly included in the Spanish Drug-Induced Liver Injury (DILI) Registry M. Slim1, J. Sanabria1, M. Robles-Díaz1, I. Medina-Cáliz1, A. González-Jiménez1, A. Ortega1, M. García-Cortés1, B. Garcia-Muñoz1, R. Sanjuán-Jiménez1, M.C. Fernandez2, A. Castiella3, E. Zapata3, H. Hallal4, E. Montane5, G. Soriano6, E. Roman6, J.M. Navarro7, M. Jimenez8, M. Prieto9, I. Conde9, A. Aldea10, M. Hernández-Guerra10, J.M. Moreno11, M. Romero-Gomez12, E. Gomez-Dominguez13, M.I. Lucena1, R.J. Andrade1. 1Instituto de Investigación Biomédica de Málaga (IBIMA), Hospital Universitario Virgen de la Victoria., Universidad de Málaga, Malaga; 2Hospital Torrecardenas, Almeria; 3Hospital de Mendaro, Guipuzcoa; 4Hospital Morales Meseguer, Murcia; 5Hospital Germans Trias i Pujol, Badalona; 6Hospital de la Santa Creu i Sant Pau. CIBERehd, Barcelona; 7Agencia Sanitaria Costa del Sol; 8Instituto de Investigación Biomédica de Málaga (IBIMA), Hospital Regional de Málaga, Malaga; 9 Hospital Universitario y Politécnico La Fe. CIBERehd., Valencia; 10 Hospital Universitario de Canarias, Tenerife; 11Complejo Hospitalario de Albacete, Albacete; 12Hospitales Universitarios Virgen MacarenaVirgen del Rocío. CIBERehd, Sevilla; 13Hospital 12 de Octubre, Madrid, Spain E-mail: [email protected]
FRI-445 Prevalence and severity of liver injury after pulsed methylprednisolone therapy in multiple sclerosis patients M. Biolato1, V. Nociti2, G. Marrone1, A. Liguori1, C. De Fino2, A. Bianco2, M. Mirabella2, A. Grieco1. 1Liver Transplant Medicine Unit, Gastroenterological Area, Gastroenterological and EndocrinoMetabolical Sciences Department; 2Multiple Sclerosis Center, Neuroscience Area, Neuroscience, Aging, Head and neck and Orthopaedics Sciences Department, Fondazione Policlinico Universitario Gemelli, Catholic University of Sacred Heart, Rome, Italy E-mail: [email protected]
Background and Aims: DILI is a rare adverse drug reaction that can be associated with morbidity and mortality. Characterisation and classification of phenotypic sub-groups is crucial for improving diagnosis and causality assessment in DILI. We aimed to compare clinical characteristics and outcomes between elderly and younger patients diagnosed with DILI.
Background and Aims: Few cases of high dose pulsed methylprednisolone-related liver injury in patients with Multiple Sclerosis are reported in literature but a prospective investigation was never been performed. The aim of the study is to evaluate the prevalence and severity of liver injury in patients affected by Multiple Sclerosis and treated with pulsed methylprednisolone because of clinical relapses. Methods: Prospective observational single-center study on patients with multiple sclerosis, in relapsing phase of the disease, treated with i.v. methylprednisolone 1,000 mg/day for 5 days. We tested liver function tests (aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, total bilirubin) before treatment and after two weeks. In case of severe liver injury, defined according to “Hy’s law”, a comprehensive hepatologic work-up was performed. Results: During 12 month observation period, we collected 251 cycles of i.v. high dose steroid treatment from 175 patients (65.1% females, mean age of 40.8 ± 12.2 years; mean Expanded disability status scale 2.8 ± 2.1). After excluding 8 cycles presenting basal alteration of liver function tests, serum aminotransferase elevation (any grade) was observed on twenty-one cycles two weeks after methylprednisolone
Journal of Hepatology 2017 vol. 66 | S333–S542
S399
POSTER PRESENTATIONS therapy ( prevalence 8.6%). Six of them ( prevalence 2.5%) presented a severe liver injury according to Hy’s law; after comprehensive hepatologic work-up, three out of them received a final diagnosis of drug-induced liver injury and other three a final diagnosis of autoimmune hepatitis. Conclusions: Liver injury should be considered a possible adverse event in patients with multiple sclerosis treated with pulsed methylprednisolone therapy for clinical relapse. Aminotransferase monitoring two weeks after pulsed methylprednisolone treatment is useful for clinical management. FRI-446 Jnk2 is indispensable in murine and human Ibuprofen-induced acute liver failure M.E. Zoubek1, M.M. Woitok1, S. Sydor2, L.J. Nelson3, L.P. Bechmann2, M.I. Lucena4,5, R.J. Andrade4,5, A. Bast6, G.H. Koek7, C. Trautwein1, F.J. Cubero1. 1Department of Internal Medicine III, University Hospital RWTH Aachen, Aachen; 2Department of Gastroenterology and Hepatology, University Hospital, University Duisburg-Essen, Essen, Germany; 3 Laboratory Department of Hepatology, Royal Infirmary of Edinburgh, University of Edinburgh, Edinburgh, United Kingdom; 4 IBIMA, H. Virgen de la Victoria, Universidad de Málaga, Malaga; 5 CIBERehd, Barcelona, Spain; 6Department of Pharmacology and Toxicology, Maastricht University; 7Division of Gastroenterology and Hepatology, Department of Internal Medicine, Maastricht University Medical Centre, Maastricht, The Netherlands E-mail: [email protected] Background and Aims: Ibuprofen is a frequently used non-steroidal anti-inflammatory drug (NSAID). Although it is generally welltolerated it can, nevertheless, cause serious liver injury. Our earlier investigations clearly demonstrated that the JNK pathway plays a relevant role in the pathophysiology of Ibuprofen-induced ALF, thus, we aimed to analyze the role of Jnk1 and Jnk2 which, at present, remains unknown. Methods: First we investigated activation of essential pathways in liver sections of Ibuprofen-induced ALF patients. Next we developed a novel murine model of Ibuprofen intoxication. Overnight fasted male C57BL/6 mice (6–8 weeks of age) were i.p injected with either vehicle or Ibuprofen 600 mg/kg for 8 h. To assess the role of JNK, we used animals carrying constitutive deletion of Jnk1 (Jnk1-/-) or Jnk2 (Jnk2-/-) expression and, later, expanded our investigations to animals with either genetic deletion of Jnk1 (Jnk1Δhepa) or siRNA-mediated deletion of Jnk2 (siJnk2Δhepa) exclusively in hepatocytes. Results: JNK phosphorylation was increased in the cytoplasm of hepatocytes and other non-LPCs derived from Ibuprofen-induced ALF liver samples of human and murine origin, compared with healthy tissue. Ibuprofen overdose produced a significantly stronger degree of liver injury compared with vehicle-treated controls as evidenced by transaminases, macroscopic as well as microscopic analysis of liver tissues. Next, we sought to investigate the molecular pathways associated with Ibuprofen overdose in mice. Interestingly, we found increased activation of PKCα, AKT and JNK, and decreased p38 protein levels, 8 h after Ibuprofen intoxication. Constitutive Jnk1-/- and Jnk2-/deficient mice exhibited a greater liver dysfunction than wildtype (WT) animals. In contrast, siJnk2Δhepa animals showed a remarkable decline of liver injury during Ibuprofen-induced ALF, which correlated with significantly higher serum liver enzymes and worsened liver histology, compared to Jnk1Δhepa or WT animals. Conclusions: In our study, cytoplasmic JNK activation in hepatocytes and other non-LPCs is a hallmark of human and murine Ibuprofeninduced ALF. Ibuprofen overdose activates PKCa/AKT/JNK- and attenuates p38-signaling in mice. Functional analysis in vivo demonstrates that JNK2 in hepatocytes is essential in mediating protection during Ibuprofen-dependent ALF and thus is a molecular target for treatment.
S400
FRI-447 Development of a novel 3D organotypic human HepaRG-based non-invasive imaging platform for pre-clinical drug screening N.J. Martucci1, K. Morgan1, G. Anderson1, P.C. Hayes1, J.N. Plevris1, P.O. Bagnaninchi2, L. Nelson1. 1Hepatology; 2MRC Centre for Regenerative Medicine, University of Edinburgh, Edinburgh, United Kingdom E-mail: [email protected] Background and Aims: Drug-induced liver injury (DILI) is the leading cause of attrition in pre-clinical drug development and withdrawals, hampered by both the lack of physiologically-relevant human models and reliance on animal data. Therefore, strong incentives exist to create new human-based, in vitro liver models, which would offer better pre-clinical prediction of DILI. Human hepatic HepaRG (hepatocyte: cholangiocyte) coculture maintain in vivo-like liver-specific functions and functional polarity and are considered a sustainable surrogate to primary human hepatocytes. We aimed to develop organotypic 3D liver spheroids, coupled with a novel imaging platform based on optical coherence tomography (OCT), as a high throughput amenable approach for drug toxicity screening using paracetamol as a model hepatotoxin. Methods: HepaRGs were cultured at different densities (0–50,000/ well) to optimize 3D spheroid size using ultra low-adherence plates [InSpheroTM]. OCT was used to monitor 3D spheroids in real time and to quantify cell viability, proliferation; and correlated with biochemical assays. H&E staining was performed (d7; d14) to verify viability (presence/absence of necrotic core) of spheroids. The optimal spheroid size was defined as having a round spheroidal shape with no necrotic core. To further validate the 3D model, viability (OCT; ATP), metabolic competency (Prestoblue) and cell proliferation (wst8) was quantified following 24-hour dose-response to paracetamol. Results: Enhanced viability, proliferation, and metabolic competency was observed with an optimal spheroid size of 2,500 HepaRG cells (also vs 2D HepaRG model). OCT was able to image 200 μm Ø spheroids non-destructively with a 5 μm resolution (depth/ penetration); whilst time-lapse OCT image-acquisition with optical fluctuation -maps were found to correlate directly to cell viability and calculated at each dose; giving unique insight into both drug penetration and toxicity. Finally, OCT cell viability values were integrated all over the samples and correlated to biochemical assays.
Conclusions: The 3D-HepaRG-OCT platform has the potential to assess noninvasively and label-free drug toxicity/penetration in organotypic 3D-models providing an easily reproducible model system for improved predictive drug toxicity screening and potentially to reduce the 3Rs. Future work involves further validation of the model’s similarity to liver function in vivo, and testing other modes of DILI, such as cholestasis.
Journal of Hepatology 2017 vol. 66 | S333–S542