Reply

Authors


  • Potential conflict of interest: Nothing to report.

We are glad that our recent article[1] has stimulated discussion on the role of the farnesoid X receptor (FXR) agonist, obeticholic acid (OCA), in experimental portal hypertension and welcome the recent editorial by Gracia-Sancho et al.[2] and letter to the editor by Mookerjee et al.

In brief, we demonstrated reduced FXR expression in two different rat models of cirrhosis and proved that oral administration of OCA, both acute (24 hours) and chronic (10 days), reactivated the FXR downstream pathway, leading to a significant decrease in portal pressure to approximately 15%-20% (which is considered clinically significant) without side effects. Moreover, we substantiated that the beneficial effect of OCA in both models converges through endothelial nitric oxide synthase (eNOS) activation. Interestingly, the signaling pathways to do so were dependent on the etiology of the liver damage: down-regulation of Rho-kinase (an eNOS inhibitor) in the thioacetamide (TAA) model and increased degradation of the endogenous NOS inhibitor, asymmetric dimethylarginine (ADMA), in the bile duct–ligated (BDL) model.

Whereas Gracia-Sancho et al.[2] subscribed to and acknowledged our methodology, study design, and findings, Mookerjee et al. touched on several points to which we would like to reply concisely.

First, the researchers expressed some concerns about the dose, duration, and potential toxicity of OCA treatment in our study. OCA-related toxicity was observed in none of the animals in our study (neither regarding mortality, morbidity, biochemically nor histologically). Preclinical pharmacokinetic studies have shown earlier, in rats, that single doses up to 120 mg/kg are well tolerated without adverse events, whereas chronic toxicity only occurred after daily treatment with OCA at doses of 60 mg/kg or more daily during 6 months.[3] In healthy human volunteers, high doses of OCA up to 500 mg were well tolerated, with peak plasma OCA levels achieved already 4 hours after intake, being highly influenced after by enterohepatic recirculation.[4] We opted for a short-term (24-hour) treatment because we aimed to document short-term hemodynamic effects in both cirrhotic rat models, irrespective of possible long-term FXR-related antifibrotic, anticholeretic, and anti-inflammatory effects. Interestingly enough, the OCA-mediated reduction in portal pressure observed with acute dosing was sustained after 10-day treatment, even with significantly lower doses OCA (5 mg/kg every 2 days in the biliary cirrhosis model).[1]

In vitro contraction experiments with OCA on hepatic stellate cells (HSCs) also showed no toxic effects, because the viability of these cells in response to the different doses was assessed as mentioned in the Materials and Methods. These data demonstrate that the assumption of toxic effects on the HSCs resulting from a high dose does not hold true. Furthermore, the observed marginal loss of contraction at higher doses additionally contradicts this suggestion because, in the case of toxicity, one would not expect almost intact contractility (which is automatically linked to viability).

Mookerjee et al. commented on the possible pharmacokinetic (PK) differences between human data and our rodent experiments, which indeed underscores the species-dependent variations in FXR and OCA metabolism. For the purpose of our study, we performed detailed prestudy dose-finding experiments of acute OCA administration in two rat models of cirrhosis. We agree with Gracia-Sancho et al.[2] that our experimental findings, although reproducible in two types of experimental cirrhosis, remain to be confirmed in patients with cirrhosis in which other PK issues might need to be addressed.

Second, Mookerjee et al. question our finding that dimethylarginine dimethylaminohydrolase-2 (DDAH-2) is the culprit isoform reducing ADMA levels in BDL rats upon FXR agonist treatment. In our study,[1] we nicely showed that expression of both DDAH-1 and -2 were reduced in both TAA and BDL cirrhotic rat livers. Yet, only an increase in hepatic DDAH-2 expression was observed in biliary, but not in TAA, cirrhotic rats after OCA treatment. So, the effect of OCA on DDAH seems to depend on the etiology of cirrhosis, which is not only a novel finding, but again also might form a potential pitfall when going into clinical trial. Whereas we acknowledge that several researchers have demonstrated a DDAH-1-mediated reduction in ADMA upon FXR agonism, which we could not confirm in this study, for the time being, there is at least an equal amount of publications emphasizing either on the importance of DDAH-2 alone or combined with DDAH-1 with regard to ADMA metabolism,[5-11] leaving the solution to this puzzle unavailable at this time. To the point of “unprecedented” FXR agonism leading to DDAH-2 expression, we would like to draw the attention of Mookerjee et al. to the article of Achan et al.,[12] who found that all transretinoic acid, trough nuclear receptor binding, increased nitric oxide synthesis by endothelial cells in vitro through increased DDAH-2 gene expression and promoter activity.

Overall, it can be questioned whether the discussion on the DDAH isoform is really that essential to the main finding that FXR agonists, such as OCA, lower portal pressure and seem to target multiple pathways (such as Rho and DDAH) considered essential in portal hypertension, but which finally all converge through activation of eNOS.

In our view, the issue that really should be addressed at this time is whether OCA is ready for clinical prime time in portal hypertension or if we need more experimental data. We would argue for the former, given the beneficial effects of FXR agonists on liver inflammation, fibrosis, and cholestasis in different animal models of liver disease and our data in two different models of cirrhotic portal hypertension, as well as the promising results on OCA in phase II clinical trials in patients with nonalcoholic fatty liver disease and primary biliary cirrhosis. Considering all of these favorable characteristics, FXR agonists might just prove to be the hepatologist's aspirin for the future.

  • Len Verbeke, M.D.1

  • Jonel Trebicka, M.D., Ph.D.2

  • Wim Laleman, M.D., Ph.D.1

  • 1Department of Liver and Biliopancreatic Disorders

  • University Hospital, Gasthuisberg

  • KU Leuven

  • Leuven, Belgium

  • 2Department of Internal Medicine I

  • University of Bonn

  • Bonn, Germany

Ancillary