Introduction: Paracetamol toxicity is the leading cause of acute liver failure in many countries. The most commonly discussed mechanism for paracetamol poisoning is the production of the highly reactive toxic metabolite N-acetyl-p-benzoquinone imine (NAPQI). NAPQI can deplete liver glutathione (GSH) and bind with liver protein and cause liver injury with higher doses of paracetamol. Among the three dominant metabolism pathways (i.e. glucuronidation, sulfation and oxidation), sulfation is saturated within therapeutic doses which has led sulfation to be overlooked as an important pathway in determining toxicity. Current practice to treat paracetamol poisoning is N-acetylcysteine (NAC), which replenishes cysteine. Although paracetamol poisoning has been widely studied in the literature, there is a lack of a comprehensive understanding of the optimal treatment of paracetamol overdose, particularly in relation to the diagnostic biomarkers and toxicity assessment.
Objectives: The overall aim of this work is to optimise the management of paracetamol overdose. In this work we aim to develop and calibrate a QSP model for paracetamol catabolism that can be further developed to describe known and putative toxicity mechanisms.
Methods: A system model for paracetamol metabolism and toxicity model was developed in a stepwise manner. A published model with paracetamol metabolism component  was extended and calibrated under normal dose of paracetamol (1.5g). The model then incorporated parameters estimated from a population pharmacokinetic (PK) model for supratherapeutic paracetamol and its metabolites (glucuronidate and sulfate conjugation). Finally the model was re-calibrated under both normal and supratherapeutic dose (6g). Both first-order and Michaelis-Menten processes were used to describe substrate transportation and enzyme catalysed biochemical reactions. First-order rate constants (k), volume parameters and maximum reaction rates (Vmax) were calibrated, while the potency parameter (KM) and initial conditions for endogenous substrates were sourced from the literature. Urine recovery from each metabolism pathway (glucuronidation, sulfation and oxidation metabolites) and plasma concentration profile for parent drug and its metabolites (glucuronidation and sulfation metabolites) were calibrated to match with published data under different dose levels. The system model was written in a series of ordinary differential equations (ODEs) in MATLAB (R2017a).
Results: The system model consisted of 47 states and 109 parameters. Five different body locations are included, i.e. gut, liver, blood, tissue and urine. The system model integrates five components, including (1) input of paracetamol, (2) paracetamol metabolism and inter-organ transportation of parent drug and metabolites, (3) physiological turnover of endogenous sulfur-containing substrates, including the metabolism of cysteine to produce GSH and inorganic sulfate, etc., (4) a published general model for biomarker release from injured tissue to blood , (5) a published PK model for NAC and three proposed mechanisms of NAC effect. After the final calibration, urine recovery for each metabolism pathway is comparable with the literature value for different dose levels (1.5g, 10.3g and 11.8g). Plasma concentration profiles for paracetamol and its major metabolites simulated from the system model have a good agreement with the clinical data under 1.5g and 6g of paracetamol.
Discussion: A system pharmacology model that includes known mechanisms of paracetamol induced liver toxicity and NAC treatment was developed. The model was able to describe paracetamol metabolism from therapeutic dose to overdose. With the interaction between paracetamol and endogenous sulfur-containing substrate and the release of biomarker after liver injury (e.g. hepatic transaminases, protein adduct), the system model has the potential to quantitatively describe the whole causal pathway for paracetamol induced liver toxicity. Incorporation of NAC and its mechanism of action will allow the system model to investigate the ‘what if’ scenarios in the rescue of paracetamol poisoning.
- Ben-Shachar et al, Theor Biol Med Model 2012; 9(1): 1-22; 2. Li et al, JPKPD 2020; 1-14