Quantifying Receptor Selectivity in Modern Drug Discovery

Terry's Desk
10 hours ago
4 min read

Diagram illustrating system-independent quantifying receptor selectivity using ratio-of-ratios analysis

Selectivity is one of the most overused—and misunderstood—terms in drug discovery. A compound shows no response in one assay, and we call it “selective.” Another produces a larger shift in EC₅₀ in one system than another, and we assume we’ve found therapeutic separation.

But as Dr. Kenakin demonstrates in this session of Terry’s Corner, what we often measure is not receptor selectivity. It’s a partnership between ligand efficacy and the sensitivity of the cellular system used to detect it.

In this session, you’ll gain:

A framework for cancelling cell effects to isolate true receptor selectivity
Practical methods for calculating system-independent selectivity
Clear distinction between receptor selectivity and signaling bias

Quantifying Receptor Selectivity Requires Canceling the Cell

In early discovery, comparing EC₅₀ values across compounds is a common first-pass strategy for ranking potency. But raw potency differences are not pure reflections of affinity or efficacy.

They also encode:

Receptor expression levels
Coupling efficiency
Signal amplification
Assay sensitivity

Observed agonism is never “just the ligand.” It is always ligand × system.

In the full lecture, Dr. Kenakin reveals how two compounds tested in different systems can appear to differ by thousands-fold in selectivity—until the system contribution is mathematically cancelled. What remains may be a much smaller, but far more meaningful, difference.

True receptor selectivity must transcend the cell line.

Canceling the Cell

If observed potency reflects both drug properties and system sensitivity, then system effects must be neutralized.

The strategy is conceptually simple:

Measure full concentration–response curves
Calculate relative potencies within each system
Perform a ratio-of-ratios comparison

By comparing the relative potency of two agonists across two receptor systems—and then comparing those ratios to each other—system-dependent factors cancel out.

What remains reflects differences in:

Affinity
Efficacy

And importantly, this value becomes portable. It should hold regardless of receptor density or assay format.

This is not just mathematical elegance. It is strategic clarity. It prevents discovery teams from advancing compounds based on artifacts of expression systems rather than intrinsic pharmacology.

Why Full Curves Matter

Selectivity measurements require full concentration–response curves.

Not single concentrations. Not partial windows. Not truncated data.

Why?

Because the absence of observed agonism does not prove absence of efficacy .

A low-efficacy agonist tested in a low-sensitivity system may show no visible response—even at concentrations where 50% receptor occupancy occurs. Move that same ligand into a more sensitive assay, and a curve appears.

Dr. Kenakin uses a lever analogy where:

Efficacy is the weight applied.
System sensitivity determines whether the lever moves enough to be seen.

If the assay threshold is too high, real pharmacology becomes invisible.

This has direct implications:

A “non-selective” compound may simply be under-detected.
A “silent” ligand may be system-limited.
A development decision may hinge on assay sensitivity rather than molecular behavior.

Without full curves, you cannot separate drug properties from detection limitations.

Comparing Full and Partial Agonists

Discovery programs rarely enjoy the simplicity of comparing two full agonists.

More often, one ligand is partial. Now EC₅₀ values alone are insufficient. Maximal response differs. Potency scales distort.

Dr. Kenakin outlines a practical solution: use a potency metric that incorporates both efficacy and EC₅₀ (log maximal response divided by EC₅₀).

This approach:

Corrects distortions introduced by differing maximal responses
Allows comparison across full and partial agonists
Preserves system independence

When handled correctly, partial agonists can be quantified on equal footing with full agonists—without biasing interpretation.

Confidence, Not Just Ratios

Selectivity is not just a number. It is an estimate with uncertainty.

Delta–delta comparisons can be repeated, generating:

Standard errors
95% confidence intervals

And this is where interpretation sharpens.

If confidence intervals include zero, selectivity is not statistically significant. If they exclude zero, the separation is real.

In the full lecture, you will learn how this approach removes subjectivity from interpretation. No more “it looks selective.” Statistics decide.

For teams under pressure to nominate leads, this discipline matters. It prevents overinterpretation of noise as biology.

Receptor Selectivity vs Signaling Bias

Here is where the conversation becomes more nuanced.

Receptor selectivity involves concentration separation. A compound binds receptor A and produces a response at one range of concentrations. Only at much higher concentrations does receptor B become engaged.

Bias is different.

Bias occurs simultaneously with receptor binding. When the ligand engages the receptor, multiple signaling pathways initiate:

G protein activation
β-arrestin recruitment
Calcium signaling
Other downstream cascades

But they do not activate with equal intensity .

Bias reflects differential pathway amplification at the same receptor—not concentration separation across receptors.

This distinction is critical:

Receptor selectivity separates by concentration window.
Bias separates by signaling strength.

You will learn how the same ratio-of-ratios framework can be applied within a receptor to quantify pathway bias.

But caution is required.

If pathway-specific readouts are used to define receptor selectivity, then both agonists must be evaluated using the same pathway. Otherwise, bias contaminates the selectivity calculation.

Whole-Cell Responses: A Historical Complication

Historically, pharmacology relied on whole-cell or tissue responses.

These are integrated outputs. They blend multiple pathways into a single functional readout.

This has advantages:

Physiological relevance
Functional integration

But it obscures pathway-specific behavior.

Modern assays allow isolation of discrete signaling nodes. This precision is powerful—but it introduces complexity. Each pathway can produce a different apparent selectivity profile.

In the end, what matters therapeutically is the integrated response. But during discovery, pathway dissection can clarify mechanism and reveal hidden liabilities.

The key is consistency: define which pathway defines “selectivity,” and stay faithful to it.

Strategic Implications for Drug Hunters

Quantifying receptor selectivity correctly does more than refine pharmacological metrics.

It changes decisions.

It prevents false negatives caused by insensitive assays.
It avoids overestimating subtype separation.
It clarifies whether differentiation is receptor-based or pathway-based.
It creates transportable, system-independent numbers.

In a world of increasingly complex GPCR modulation, these distinctions are not academic. They define risk.

Compounds fail when assumptions about selectivity prove wrong in vivo. Often, the error began at the assay stage.

Quantification—done correctly—protects pipelines.

Watch Course Trailer

Why Terry’s Corner

Terry’s Pharmacology Corner delivers weekly lectures from Dr. Terry Kenakin, monthly live AMAs, and a growing on-demand library built for scientists who need clarity fast.

It is designed for: