Half-Maximal Inhibitory Concentration (IC50)

IC50 answers a narrow question: at what concentration does this compound cut the observed response in half? What counts as the response varies enormously, from the turnover of a purified enzyme to the displacement of a radiolabelled ligand to the fraction of cells alive after three days. Unlike a binding constant, IC50 is not a fixed property of the molecule and target pair, because it depends on how much substrate or tracer it has to compete against. For competitive inhibitors, the Cheng-Prusoff relationship converts an IC50 into an inhibition constant given the substrate concentration and its own affinity.

The value is the standard currency of screening. DPP-4 inhibitors such as sitagliptin and linagliptin inhibit the enzyme at low nanomolar concentrations, which is what lets them prolong the survival of endogenous GLP-1. Kinase inhibitors are routinely profiled as an IC50 across hundreds of kinases at once, producing selectivity maps rather than a single number. Cytotoxicity IC50 values in tumour cell lines drive most early oncology triage.

What makes it useful is the ratio, not the absolute figure. An inhibitor with a nanomolar IC50 at its intended target and micromolar values everywhere else has a thousand-fold selectivity window, and comparing the on-target value with the free plasma concentration a dose achieves tells you whether inhibition is plausible in a patient at all.

Where it bites hardest is comparison. An IC50 measured at low substrate flatters a competitive inhibitor and can shift several-fold with a change in assay conditions, so numbers from different laboratories are not interchangeable and should be converted to inhibition constants before being ranked. Cytotoxicity values are worse: peptides advertised as anticancer on the strength of a cell-line IC50 in the tens or hundreds of micromolar are quoting concentrations no plasma level approaches.