Off-Target Activity

No molecule binds one protein only. Off-target activity is quantified as a selectivity ratio, comparing the concentration needed at the intended target with the concentration needed at each other one. It arises mostly from homology: receptor subtypes descended from a common ancestor share binding pockets, so a ligand shaped for one frequently fits its relatives, and peptide analogues drift towards their family neighbours as residues are substituted for stability.

Drug development addresses this with breadth. Safety pharmacology panels screen candidates against dozens to hundreds of receptors, transporters, ion channels and enzymes, with hERG channel activity tested specifically because of its link to QT prolongation. The melanocortin family shows the cost of poor selectivity: bremelanotide, approved in 2019 for hypoactive sexual desire disorder, is not selective across melanocortin receptor subtypes, and MC1R activation is what produces skin darkening. Unregulated tanning peptides in the same family have generated case reports of new and changing melanocytic lesions.

The finding changes a decision only in combination with exposure. Micromolar activity at a secondary target is irrelevant to a drug whose free plasma concentration peaks in the low nanomolar range, and a long list of weak interactions can look alarming while meaning nothing.

The assumption that peptides are inherently selective because they resemble natural signals is the error worth naming. Modification for half-life and stability routinely changes the selectivity profile of the parent hormone, and the profile of the parent was never a guarantee to begin with. A clean panel is also incomplete on its own, because it usually tests the parent molecule rather than its metabolites, and in unregulated products the effect may belong to a synthesis impurity rather than the peptide at all.