Where optimization starts
Screening gives you a hit: a molecule that does something measurable to your target. After you confirm it is real, build a little SAR around it, and show it is tractable to chemistry, it earns the name lead. A lead is a promising starting point — not a drug. It is usually too weak, too unselective, or too poorly behaved in the body to ever reach a patient.
Lead optimization is the phase where chemists, biologists, and pharmacologists work that lead up to a candidate: a molecule whose whole profile — potency, selectivity, ADMET, safety, and synthesizability — is good enough to commit to expensive preclinical and clinical testing. The job is rarely to make one number great; it is to make every important number acceptable at once.
Why it is a multi-front campaign
The hard truth of optimization is that properties are coupled. Adding a greasy group to fill a pocket often boosts potency but worsens solubility and metabolic stability. Tightening a molecule to lock in the binding conformation can improve both potency and selectivity, but may introduce a new off-target. This is why we call the goal multiparameter optimization (MPO): you are steering many gauges at once, and pushing one usually nudges the others.
How the work actually flows
Optimization runs as repeated design–make–test cycles. Each loop is small and concrete, and over dozens of loops the series climbs toward the goal.
- Design. Propose a few new analogues based on the current SAR, a structure if you have one, and the properties you most need to fix.
- Make. Synthesize them — ideally by a short, robust route so you can turn cycles quickly.
- Test. Run the panel: a potency assay, key selectivity counter-screens, and core ADMET measures like solubility, permeability, and metabolic stability.
- Learn and repeat. Read the new data into your SAR, update your hypothesis, and design the next set. The molecule that survives every gauge becomes the candidate.