Ibrahim Khalil E (1,2), Knöchel J (3), Kechagias S(4,5), Ekstedt M (4,5), Bergenholm L (2)
(1) Department of Pharmaceutical Biosciences, Uppsala university, Sweden, (2) Drug Metabolism and Pharmacokinetics, Cardiovascular, Renal and Metabolism, IMED Biotech Unit, AstraZeneca, Gothenburg, Sweden. (3) Quantitative Clinical Pharmacology, Early Clinical Development, IMED Biotech Unit, AstraZeneca, Gothenburg, Sweden. (4) Department of Gastroenterology and Hepatology & (5) Department of Medical and Health Sciences, Linköping University, Linköping, Sweden.
Objectives:
Nonalcoholic fatty liver disease (NAFLD) is now one of the most common causes of liver-related morbidity and mortality. The best predictor of outcome is liver fibrosis stage (1). NAFLD is defined as the presence of hepatic steatosis in at least 5% of hepatocytes on liver biopsy examination in individuals who consume little or no alcohol, after exclusion of other causes of liver disease. NAFLD is classified into 2 subtypes: nonalcoholic fatty liver (NAFL), which is generally benign with very low risk of progression and nonalcoholic steatohepatitis (NASH), which is characterized by hepatocellular injury, inflammation with a risk of progression to advanced fibrosis (2). Currently there are no approved drug therapies for NASH and lifestyle advice to achieve weight loss is considered the most important part of standard of care. Furthermore, the disease progression has not been fully understood. The aim of this work was to perform a model-based analysis of fibrosis progression from stage 0 to 4 in NAFLD patients including the investigation of body mass index (BMI) impact and other potential covariates on disease progression.
Methods:
A continuous-time Markov model with first order markovian features was used to describe the forward and backward transition probability rates between different fibrosis stages and assess effects of different measured covariates on these transitions. The transition probabilities were fitted against clinical data collected from biopsy-proven NAFLD patients who were followed with repeat biopsy over two decades (3). The analyzed dataset included disease stage observations which were not equally spaced in time. Therefore, the continuous-time Markov modeling approach was chosen as it can also effectively handle the dependency between successive observations of ordered fibrosis stages. The choice of the included covariates was based on their statistical and clinical significance. Missing covariate values were imputed using mean substitution. The software used were NONMEM 7.3.0 and R-studio 3.2.4.
Results:
In total 70 patients underwent biopsy at baseline and at first follow-up after on average 14 (range: 10 – 17) years. 34 of them also underwent a third biopsy after another 10 (range: 8 – 12) years. The developed Markov model allowed the estimation of the transition probability rates between different fibrosis stages. Addition of covariates on specific forward rates for BMI group (obese defined by BMI > 30, and non-obese), age group (cut-off 55 years (4)) and NAFLD activity score (NAS) caused a significant reduction on the objective function value (OFV), while gender and type-2 diabetes did not. The average half-life of the transition probability that correspond to one fibrosis stage of progression from stage 0 to 2 was 8.1 years in non-obese patients and 3.6 years in obese patients both with age below 55 and NAS 3. The transition probability rates increased exponentially with the increase in NAS from fibrosis stage 0 to 3. The average half-life of the backwards transition probability that correspond to one stage of fibrosis regression was 7.9 years. The average half-life of the transition probability that correspond to one fibrosis stage of progression was in line with previous results of a meta-analysis, where the overall fibrosis progression half-life in patients with NAFLD was 5.7 years (95% CI, 4.3-9.9) (5). Moreover, Visual predictive checks using 1000 simulated data sets were made to evaluate the predictive ability of the model.
Conclusions:
The Markov model provided a good description of proportions of patients in the respective fibrosis stages over time. The initial results show that the rate of fibrosis progression is higher in obese patients than in non-obese patients in the early stages of the disease. Finally, the effect of obesity and NAS on fibrosis progression could be successfully quantified.
References:
[1] Hannes Hagström, Patrik Nasr,Mattias Ekstedt, …,Per Stål, Rolf Hultcrantz, Stergios Kechagias. Fibrosis stage but not NASH predicts mortality and time to development of severe liver disease in biopsy-proven NAFLD. J. Hepatol. 2017, 67, 1265–1273
[2] Siddharth Singh, Alina M. Allen, Zhen Wang, Larry J. Prokop, Mohammad H. Murad, and Rohit Loomba. Fibrosis Progression in Nonalcoholic Fatty Liver vs Nonalcoholic Steatohepatitis: A Systematic Review and Meta-analysis of Paired-Biopsy Studies. Clinical Gastroenterology and Hepatology 2015;13:643–654.
[3] Ekstedt M, Franzen LE, Mathiesen UL, et al. Long-term follow-up of patients with NAFLD and elevated liver enzymes. Hepatology 2006;44:865–873.
[4]Nargess Ebrahimi Daryani, Nasser Ebrahimi Daryani, Seyed Moayed Alavian, Ali Zare, Seyed-Mohammad Fereshtehnejad, Mohammad Reza Keramati, Mohammad Reza Pashaei, Peiman Habibollahi. Non-alcoholic steatohepatitis and influence of age and gender on histopathologic findings. 2010 September 7; 16(33): 4169-4175.
[5]Siddharth Singh, Alina M. Allen, Zhen Wang, Larry J. Prokop, Mohammad H. Murad, and Rohit Loomba. Fibrosis Progression in Nonalcoholic Fatty Liver vs Nonalcoholic Steatohepatitis: A Systematic Review and Meta-analysis of Paired-Biopsy Studies. Clinical Gastroenterology and Hepatology 2015;13:643–654.
Reference: PAGE 28 (2019) Abstr 9056 [www.page-meeting.org/?abstract=9056]
Poster: Drug/Disease Modelling - Other Topics