Background
As the adage goes, “time is money,” and the drug development pipeline is no exception. Separately, both “time” and “money” are crucial considerations at every stage of drug development. No one wants to invest years of development and millions of dollars into a compound only for it to enter clinical trials and underperform due to efficacy or toxicity issues. Unfortunately, this is the fate of many compounds.
During the preclinical development stage, in vitro screening and animal models can help identify compounds with potential liabilities. However, compounds with liver safety liabilities often do not emerge until they enter clinical trials, meaning these issues are identified relatively late in the development lifecycle. As a result, this can lead to significant economic and time costs, potentially delaying or halting development altogether.
Historically, sponsors have used DILIsym®, our flagship quantitative systems toxicology (QST) model, in many cases where liver safety signals were observed in clinical trials. In these instances, DILIsym was used to assess observed liver signals and mitigate further hepatotoxic risk. The DILIsym approach requires extensive in vitro assay data and more complete representation of liver exposure based on a wide range of in vitro and in vivo data. This process is time-consuming and requires more data than what is typically available during the early drug discovery phase.
But wouldn’t it be better to assess multiple candidates in the early drug discovery phase to increase our confidence in selecting the best compound to move forward into development, thereby reducing the risk of spending time and money on a compound that may ultimately be shelved?
Simulations Plus has developed new modeling and simulation tools to provide insights into exposure dynamics and hepatotoxic potential during early drug development. These tools, available as the Liver Safety+ package, help focus development efforts and complement existing approaches for selecting the most promising in-class candidates to advance through the pipeline. Applying these new tools not only provides valuable insights but also helps accelerate time-to-market, reduce costs, and avoid wasting resources on undue candidates.
What kind of information can we derive from the tools available in the Liver Safety+ package during early drug development?
Insights into in vivo exposure during early drug
The high-throughput pharmacokinetic (HTPK) module in ADMET Predictor offers mechanistic predictions of in vivo compound exposure based on limited data.
Requiring only the compound structure and basic dose or exposure information, the HTPK module can:
- Estimate exposure based on the provided dose input (IV bolus, IR tablet, IR solution)
- Estimate the dose required to achieve a target plasma concentration
- Calculate and display a simulated plasma concentration versus time curve.
For any of the above exposure predictions, additional estimates are provided:
- Hepatic, or hepatic + renal clearance
- Lipid-adjusted percent unbound to plasma proteins (fup_adj%)
- Tissue-specific partitioning coefficients (Kp_{tissue})
- Fraction absorbed in intestinal tract (%Fa) and relative oral bioavailability (%Fb)
During early drug development, absorption or bioavailability may pose challenges, e.g., absorption may be limited by low solubility or permeability, while bioavailability may be restricted by high first-pass metabolism, ultimately impacting the apparent potency of a compound. The HTPK module accounts for these cases, although it cannot predict all pharmacological effects, such as nonlinear transporter or metabolism kinetics, which are difficult to predict in early drug development. Nevertheless, the early predictions provided by the HTPK module offer valuable insights into a compound’s potential exposure, helping to distinguish promising candidates from less-than-ideal ones.
As with most modeling and simulation software, as more data are collected for a compound, parameters within the HTPK module can be adjusted to increase confidence and improve exposure predictions.
Examples of exposure-related questions in early drug development that are addressed with Simulations Plus software:
- How readily will a given compound be absorbed and enter the circulation?
- What is the predicted dose needed to achieve desired plasma exposure (for efficacy purposes)?
- What is the estimated partitioning into a tissue of interest?
Insights into compound-specific mechanisms of hepatotoxicity during early drug development
The ADMET Predictor DILIsym (APD) module provides both qualitative and quantitative predictions of a compound’s potential to induce in vitro signals in assays designed to assess hepatotoxic risk.
Multiple well-established hepatotoxicity mechanisms are considered when predicting liver liabilities in DILIsym. These mechanisms include:
- Mitochondrial dysfunction
- Reactive oxygen species production
- Bile acid transporter inhibition
- Phospholipid transporter inhibition
While the standard DILIsym approach necessitates performing and analyzing in vitro assays to assess a compound’s effect on each of these mechanisms, the APD module only needs the compound structure and provides:
- Qualitative (yes/no) predictions for active toxicity mechanisms relating to hepatotoxicity
- Quantitative predictions – such as MEC, AC50, IC50 (defined in table below) – for toxicity mechanisms related to hepatotoxicity
Qualitative predictions (i.e., yes/no classifications for active hepatotoxicity mechanisms) can help prioritize in vitro assays and provide insight into potential liver liabilities. While these classifications are not directly linked to hepatotoxic risk, they indicate a compound’s potential to induce liver injury via that mechanism.
Quantitative predictions offer greater insight into liver liabilities, but their interpretation requires considering liver exposure, underlying liver biochemistry, and interactions with other toxicity mechanisms to assess overall hepatotoxicity risk. This will be discussed in the next section.
Examples of hepatotoxicity mechanism-related questions in early drug development that are addressed with Simulations Plus software:
- Are hepatotoxicity mechanisms a cause for concern for my compound?
- Which in vitro assays should I prioritize for my compound?
Insights into hepatotoxic risk rankings of in-class compounds during early drug development
Hepatotoxic predictions with minimal information? Yes, it is possible!
The HTPK and DILIsym module in ADMET Predictor offers insights into compound-specific exposure and active hepatotoxicity mechanisms. While these are valuable insights, it may be prudent to use them to compare the relative hepatotoxic risk of all in-class compounds during early drug development.
In the earliest stages of drug development, there may be little or no information about the desired dose or target plasma concentration, making it difficult to gain insights into exposure, even with the HTPK module. Although this may seem discouraging at first glance, we have developed an approach that still provides insights into hepatotoxic risk at this early stage of discovery. No exposure information? No problem!
Two approaches can be taken based on the insights provided above:
Approach #1: If exposure information (either in vivo or predicted using the HTPK module) is available for all in-class compounds, the following steps can be applied to each compound:
- Combine plasma exposure data with predicted Kp_liver to estimate liver exposure
- Use qualitative and quantitative hepatotoxicity predictions to parameterize relevant drug-induced liver injury (DILI) mechanisms
- Run DILIsym simulations using predicted dynamic liver concentrations to assess liver injury biomarker elevations
- Compare liver injury biomarker elevations across all in-class compounds to generate a ranked hepatotoxicity risk prediction
Approach #2: If exposure information is unavailable for at least one in-class compound, the following steps can be applied to each compound:
- Use qualitative and quantitative hepatotoxicity predictions to parameterize relevant DILI mechanisms
- Run DILIsym simulations using constant liver concentrations to assess liver injury biomarker elevations
- Simulate multiple constant liver concentrations to cover a wide range that exceeds the hypothesized effective concentration range
- Compare liver injury biomarker elevations across all in-class compounds to generate a ranked hepatotoxic risk prediction
While these early drug development hepatotoxicity predictions do not indicate the frequency of liver injury, the relative hepatotoxicity ranking of in-class compounds can help guide decisions on which compounds to advance in the development pipeline.
Examples of hepatotoxicity risk-related questions in early drug development that are addressed with Simulations Plus software:
- Which in-class compounds have the lowest and highest hepatotoxic risk when compared to similar compounds?
- Is there potential for a compound to induce elevations in liver injury biomarkers?
- Given equal exposure, efficacy, etc., which compound is the safest to advance based on the hepatotoxic ranking of in-class compounds?
One-stop package for early drug discovery needs!
The Liver Safety+ package includes all the software and licenses required to predict exposure and hepatotoxic risks during early drug development.
The package contains:
• ADMET Predictor HTPK module
• ADMET Predictor DILIsym module
• DILIsym
If you’re interested in learning more about the Liver Safety+ package or in evaluating DILI risk using DILIsym, let us know.