Peptides vs. small molecules — what makes peptides special, and where their limits lie

The basic distinction

A small molecule typically has a molecular mass below 500 Da, is mostly non-natural (synthetic), often orally active and can be manufactured easily in mass production — aspirin, paracetamol, atorvastatin, metformin are examples. A peptide is a chain of amino acids, typically 5-50 amino acids long (molecular mass 500-5,000 Da), often structurally related to endogenous hormones or neurotransmitters, almost never orally active and industrially produced by solid-phase synthesis (see separate Merrifield article) — insulin, semaglutide, octreotide, GnRH analogs. An antibody is a complete glycoprotein, about 150 kDa, recombinantly produced in mammalian cell culture — trastuzumab, adalimumab, pembrolizumab.

These size classes determine everything else: pharmacokinetics, route of administration, production economics, storage stability, regulatory treatment, price.

First: oral bioavailability

Small molecules can usually be given orally — they survive gastric acid (or are protected in suitable formulations) and cross the intestinal wall by passive diffusion or targeted transporters. Peptides are degraded by proteases in the stomach and small intestine before they can become systemically active. The oral bioavailability of native peptides is typically below 1%. That is why insulin has been injected for 100 years, why semaglutide is injected subcutaneously weekly, why octreotide needs an intramuscular depot formulation.

There are exceptions. A modern solution is the use of permeation enhancers: Rybelsus (oral semaglutide, FDA-approved 2019) contains SNAC (sodium salcaprozate) which transiently permeabilises the stratum-corneum-like gastric-intestinal junction barrier. Oral bioavailability is still only about 1%, but with a sufficiently potent substance that is enough. A second research line is cyclic peptides with conformational protection — they are enzymatically more stable and in individual cases orally tolerable. But as a general principle: peptides are mostly given parenterally.

Second: half-life

Native peptides have short plasma half-lives — typically minutes to at most one hour. This is due to several factors: enzymatic cleavage by endo- and exopeptidases, renal filtration (small peptides pass the glomerular filter), uptake by liver and muscle tissue. Most peptide medicines used clinically are half-life stabilised analogs or depot formulations. The GLP-1 line is a typical example: native GLP-1 has 2 minutes plasma half-life; exenatide (glycine substitution at position 2 against DPP-IV) has 2.4 hours; liraglutide (fatty acid acylation for albumin binding) has 13 hours; semaglutide (additional AIB substitution and longer fatty acid) has about 165 hours. This extension by an order of magnitude per generation is medicinal-chemically the central iteration axis.

Small molecules typically have half-lives of 4-24 hours — which enables once- to thrice-daily oral administration. Antibodies have half-lives of 2-4 weeks — which for IgG antibodies is due to the FcRn receptor-mediated recycling mechanism. Peptides sit between and must be pharmacotechnologically supplemented to reach clinically tolerable application intervals.

Third: selectivity and off-target effects

A clear advantage of peptides over small molecules is receptor selectivity. Peptides are often structural mimicry of endogenous ligands or directly body-identical — insulin, native GLP-1, somatostatin. Their binding specificity to the respective receptor is usually high, with little off-target activity. Small molecules addressing the same receptor often have additional effects at related targets — sulphonylureas bind to SUR1 (desired) but partly also to SUR2A (cardiac), creating clinically relevant off-target effects. Peptides usually have this problem less.

An important exception: short peptides without clear secondary structure (e.g. some cosmetic peptides, older GHRP lines) can bind promiscuously because their conformation is flexible. The selectivity logic applies primarily to structurally stable peptides with formed α-helices, β-turns or cyclic disulphide bridges.

Fourth: immunogenicity

Peptides can — depending on sequence and formulation — trigger antibody responses. With body-identical peptides (human insulin) this is rare; with foreign-species ones (pig insulin in the early insulin era, exenatide from Gila monster saliva) frequent. Anti-drug antibodies can reduce efficacy or trigger allergic reactions. This point is not relevant for small molecules — they are too small to trigger their own antibody responses but can cause secondary reactions as haptens via binding to endogenous proteins (penicillin allergy as classical example). Antibody medicines are highly immunogenic — minimised in chimeric or humanised antibodies by sequence adjustment.

Fifth: production and cost

Small molecules are produced in chemical reactors at tonnage scale. Cost per daily dose is often in the cent range. Peptides are produced via solid-phase peptide synthesis (SPPS) — an iterative process with multiple chemical steps per amino acid, subsequent HPLC purification and demanding characterisation. Production is capital-intensive; cost per daily dose is usually in the single- to double-digit euro/dollar range. Antibodies are produced in mammalian cell cultures; daily-dose costs are often in the triple- to quadruple-digit range.

This cost hierarchy has global supply consequences. Insulin has been available for 100 years but is not affordable in many countries. Making a new peptide medicine available at generic prices usually takes longer than for a small molecule — even after patent expiry, synthesis complexity remains a price floor.

Sixth: storage and use stability

Small molecules in tablet form are stable for years at room temperature. Peptides often need cold storage (2-8 °C), are moisture-sensitive and can aggregate or denature at higher temperatures. That means logistical complexity in the supply chain — a cold chain for insulin in sub-Saharan Africa is a real challenge. Antibodies have similar cold-chain requirements, often even stricter.

What remains

Peptides are not a better or worse substance class than small molecules or antibodies. They are their own tool class with their own strengths and limitations. The strength lies in body-identical mimicry, receptor selectivity and modularity for rational design (see Tirzepatide multi-target article). The limitations lie in oral administrability, half-life and production complexity. An honest pharmacological assessment chooses the substance class by target and application context, not by fashion or marketing attention.

On peptide.journal peptides are therefore not described as 'the new revolutionary substance class' but as an established tool of modern pharmacology with clearly defined properties, the right choice in certain indications and not in others.