Peptide nasal sprays: what actually absorbs

The one rule that decides most of it

The nasal mucosa is a membrane with a job, and that job is not letting things through. What crosses it easily is small, and peptides are not small. Molecular weight — measured in daltons — predicts most of what happens next:

• Under about 1,000 Da — crosses reasonably well. This is where the peptides with real intranasal evidence live.
• 1,000 to 3,000 Da — crosses poorly. Something gets in; how much is rarely known.
• Above 3,000 Da — effectively nothing crosses intact. A spray at this size is a spray of something that will be swallowed and digested.

Those boundaries are soft, and the exceptions matter — see below. But if you only remember one thing, remember that a 4,800-dalton molecule is not going to behave like an 800-dalton one because it was put in a nicer bottle.

There are two routes, and people conflate them

This is where most nasal-spray arguments go wrong. A drug sprayed into the nose has two possible destinations, and they are answered by different evidence:

Intranasal insulin is the cleanest example. Insulin is 5,808 Da — far too large for meaningful systemic nasal absorption, which is why nasal insulin never worked as a diabetes treatment. It is nonetheless studied for cognition, because enough reaches the brain along the olfactory route to change something there. "Doesn't absorb" and "does nothing" are not the same claim.

The practical consequence: when somebody tells you a nasal peptide works, the useful question is works how, and measured against what.

What a real nasal peptide looks like

Nasal peptide drugs exist and are approved. They are the benchmark, and they are worth knowing because they show that the size rule has real exceptions when a formulation is engineered properly rather than dissolved in bacteriostatic water:

• Desmopressin (~1,069 Da) — nasal spray, approved for decades.
• Nafarelin (~1,322 Da) — nasal, approved.
• Calcitonin, salmon (~3,432 Da) — nasal, approved, and comfortably above the "nothing gets through" line.
• Oxytocin (~1,007 Da) — nasal formulations exist; systemic bioavailability is low single digits, and most of the interesting research is about central effects.

What those four have in common is a formulation designed for the nose — pH, tonicity, viscosity, sometimes a permeation enhancer — and clinical work establishing what dose does what. A research peptide reconstituted at home and decanted into a spray bottle has none of that. It can still work. It is simply a different thing from the products above, and it should not borrow their credibility.

Every compound people spray, by size

Molecular weights are approximate — salt form and acetate content shift them slightly. The verdict column is about human evidence for the nasal route specifically, not about whether the compound does anything when injected.

The short list that holds up

Semax (~814 Da) and Selank (~752 Da) are the two with the strongest case. Both were developed in Russia specifically as intranasal drugs, both are approved there, and both are small enough that nasal delivery is plausible rather than aspirational. The caveat that belongs with them: the clinical literature is largely Russian-language and has not been replicated to Western regulatory standards. That is not a reason to dismiss it, and it is not the same as an FDA approval either.

DSIP (~849 Da) is small enough. The human evidence for what it does is thin regardless of route — the size argument is the strong part, not the efficacy argument.

Oxytocin (~1,007 Da) works nasally in the sense that matters for the research it appears in: central effects, at low systemic bioavailability.

Small molecules that are not peptides at all — NAD⁺ (~663 Da), glutathione (~307 Da) — clear the size bar easily. Clearing the size bar is not the same as having evidence of benefit, and for both of those the route is the least uncertain thing about them.

The list that does not, whatever the label says

Semaglutide (~4,114 Da) and tirzepatide (~4,814 Da). Both are far above the threshold. There is a good reason the oral semaglutide product needed a dedicated absorption enhancer and an enormous dose increase to work at all through the gut — getting a molecule this size across a mucous membrane is a formulation science problem, not a bottle problem. A nasal GLP-1 sold as an alternative to injecting is not one.

TB-500 / thymosin β4 (~4,963 Da), tesamorelin (~5,136 Da), CJC-1295 (~3,368 Da), sermorelin (~3,358 Da), thymosin α1 (~3,108 Da). All above the line. All sold nasally by somebody.

BPC-157 (~1,419 Da) sits in the awkward middle. It is not hopeless the way a 5,000-dalton peptide is hopeless, and it is not efficient either. There is no human nasal PK study. If you see a specific percentage attached to nasal BPC-157, it was estimated.

Ipamorelin (~712 Da) is the instructive one. It is comfortably small enough — and there is essentially no clinical evidence for intranasal ipamorelin doing anything. Small enough to absorb is a necessary condition, not a sufficient one. This is the trap in every simple size table, including the one above.

If you are going to use one anyway

• Volume is capped by anatomy. About 0.1–0.2 mL per nostril. Anything more runs down the back of the throat and gets digested. Splitting a dose between nostrils genuinely helps; doubling the volume into one does not.
• Aim outward, not up. Point the nozzle toward the outer wall of the nostril, not at the septum. Sniff gently — hard sniffing drives it straight past the mucosa and down the throat.
• Congestion changes everything. A blocked nose is a different absorption surface. So is a nose that has just had a decongestant through it.
• Peptides in solution are less stable than in powder. A spray bottle sitting at room temperature on a bathroom shelf is the worst storage condition in the protocol. Refrigerate, and treat a made-up bottle as short-dated.
• Dose is not comparable to injection. Whatever the true nasal fraction is, it is not 1.0 — so an injectable dose sprayed nasally is a smaller effective dose, by an unknown multiple. This is the reason nasal dosing guidance in the community is so inconsistent.

The short version

Look any of these up

Every compound named here has a full guide in the DoseIQ library — dosing, route, half-life, what the evidence actually shows, and which of them interact with anything else you are taking.