Guiding Lead Optimization for Solubility Improvement with Physics-Based Modeling.

Area of Science:

• Computational chemistry
• Drug discovery
• Physical chemistry

Background:

• Existing aqueous solubility models often require training data or neglect solid-state contributions.
• Accurate prediction of aqueous solubility is vital for lead optimization in drug design.

Purpose of the Study:

• To validate a physics-based aqueous solubility approach that accounts for solid-state contributions without prior data.
• To guide the improvement of poor solubility during lead optimization.

Main Methods:

• Application of a quantum mechanical (QM)-based thermodynamic cycle approach.
• Comparison of QM-based method with a molecular mechanical (MM)-based approach.
• Analysis of two pharmaceutical series to assess performance.

Main Results:

• The QM-based thermodynamic cycle approach demonstrated superior performance compared to the MM-based method.
• The QM model identified sources of poor solubility in lead compounds.
• Effective strategies for chemical modification and formulation were suggested.

Conclusions:

• The validated physics-based QM approach is highly attractive for guiding solubility improvements in late drug discovery and early development.
• This method offers insights into solid-state contributions, aiding rational drug design.