DXM has one of the most complicated pharmacokinetics and I have been fascinated by it since the first time I tried a third plateau dose.

It is an over-the-counter cough medicine, originally supposed to replace codeine due to codeine’s addiction liability and risk of fatal respiratory depression.

It belongs to the morphinan chemical class, meaning it looks very similar to morphine and codeine chemically, but they have completely different effects and mechanisms of action in the brain.

At low doses, it’s very effective for suppressing coughs, but at high doses, it’s a portal to another world. Let me explain what it does.

When it’s consumed, it enters the bloodstream and goes to the liver for first-pass metabolism, like most drugs. Then the liver starts metabolizing it rapidly using the enzyme CYP2D6, turning dextromethorphan into dextrorphan (DXO).

DXO is more lipophilic, meaning it’s attracted to fats and lipids. The brain is all fats and lipids, so DXO crosses the blood-brain barrier very quickly and gets into the brain.

At high doses, some DXM is left behind due to enzyme inhibition, because the enzyme is busy turning DXM into DXO. DXM is active and affects the brain in several ways, but it’s less lipophilic, so it creeps in more slowly.

Now we have two drugs in the brain, and they’re not very similar. The metabolite DXO is a strong NMDA antagonist, like ketamine. It causes strong dissociation at high doses, but the effect is not exactly like ketamine. Why?

Well, we have DXM in the brain too. DXM is strongly serotonergic and activates the sigma-1 receptor, causing neuronal excitability and sharpening the effects of DXO.

DXM also inhibits serotonin reuptake, similar to what SSRIs do, and this is the reason why you should never, ever take it with SSRIs, because it can greatly increase the risk of serotonin syndrome, which can be fatal or at best extremely traumatizing.

DXM is also a very weak NMDA antagonist, nothing like DXO, but when combined, it synergistically enhances the effects.

Now we have two different drugs in the system: one releases serotonin, norepinephrine, and dopamine, stimulating the brain and providing some neuroprotective effects through sigma-1 activation. The other acts like ketamine, blocking NMDA receptors aggressively and causing strong dissociation.

But the thing is that DXM itself has very therapeutic effects, especially for depression. It can reduce depression symptoms very quickly, unlike SSRIs that can take weeks to months to show effects and not very reliably.

DXM itself can be a reliable fast-acting antidepressant, and this got me wondering: why don’t we have good analogues that prevent conversion to DXO?

There are drugs that use combinations of strong CYP2D6 inhibitors like Wellbutrin and DXM to reduce the conversion to DXO, like the drug Auvelity, but the issue is that it doesn’t completely prevent DXO conversion, and it’s not very ideal to always have CYP2D6 inhibition since it can affect other medications’ metabolism.

This got me wondering if there was another way to solve the problem. Instead of inhibiting CYP2D6, what if the DXM molecule itself could be modified so that CYP2D6 could no longer metabolize it into DXO?

My first thought was replacing the methoxy group with 3-fluoromethoxy. On paper, this seemed like it could be an effective way to interfere with CYP2D6-mediated demethylation, but synthesizing a fluorinated DXM analogue would be difficult and costly.

Fluorine has a much stronger bond than hydrogen, and this strong bond would be very difficult for CYP2D6 to metabolize and demethylate. At the same time, fluorine can sometimes act as a functional replacement for hydrogen without dramatically changing the pharmacokinetic properties of a drug.

Since fluorination wasn’t a good idea, I started thinking about a simpler approach: replacing the methoxy group with isopropoxy.

The idea was to make the molecule physically bulkier, so that it would no longer fit properly into the CYP2D6 active site and would be much harder for the enzyme to metabolize.

So now I had a chemical compound on paper that, at least theoretically, could prevent DXO conversion entirely while preserving as much of DXM’s pharmacokinetics as possible.

This led to 3-isopropoxy-dextromethorphan, or DPO as I like to call it.

The basic idea behind DPO is simple: replace DXM’s 3-methoxy group with a bulkier 3-isopropoxy group and, in doing so, potentially prevent or greatly reduce the conversion into DXO because it would be too bulky to fit properly into CYP2D6.

If this worked, DPO could theoretically retain the pharmacological properties of the parent DXM while removing a major source of its complicated pharmacology.

About DPO

I wanted to separate the effects of DXM from the effects of its metabolite because it genuinely has great therapeutic effects for depression and possibly many more conditions.

The structure of DPO is (+)-3-(isopropoxy)-N-methylmorphinan, with the 3-methoxy group of DXM replaced by a 3-isopropoxy group.

My hypothesis is that the bulk around the ether could make the molecule a much poorer substrate for CYP2D6 O-dealkylation. If DPO were resistant to this metabolic pathway, substantially less DXO would be produced.

That would theoretically leave us with a molecule that has the parent drug’s serotonin reuptake inhibition, sigma-1 activity, and relatively weak NMDA antagonism, while reducing the contribution of the much stronger NMDA antagonism of DXO.

In other words, instead of having DXM turn into a second, pharmacologically very different drug inside the body, DPO would ideally remain DPO.

Hypothesis of pharmacological profile

I expect DPO to retain much of the parent DXM pharmacology, although this cannot be assumed without experimental receptor-binding and lab tests. But I tried using prediction tools such as BioTransformer and SwissTargetPrediction to map out its pharmacokinetic and pharmacological properties.

I calculated the SMILES for the compound:

CN1CC[C@@]23CCCC[C@@H]2[C@@H]1CC4=C3C=C(C=C4)OC(C)C

My hypothesis is that when DPO enters the body, since it wouldn’t fit into CYP2D6 as easily, CYP3A4 could become one of the main metabolic pathways.

One possible pathway would be CYP3A4 converting DPO into 3-isopropoxy-morphinan, which could then go through glucuronidation pathways and finally be excreted.

The SwissTargetPrediction results support my hypothesis to some extent since it predicts affinity for several receptors.

Also, SwissADME predicts that the molecule would penetrate the BBB, so it would likely be psychoactive.

If this prediction is correct, CYP3A4 could become one of the most important enzymes determining DPO’s pharmacokinetics, rather than CYP2D6 as it is with DXM.

Also, based on research from BenchChem, the half-life of DPO could be over 24 hours.

My hypothesis on half-life is that since DXM in CYP2D6 poor metabolizers can take around 24 hours to eliminate half of the drug, and DPO doesn’t metabolize with CYP2D6, it could be similar to that of a poor metabolizer or even possibly much longer. So dosing would be different than DXM.

This is all hypotheses, and the actual metabolic pathway would need to be confirmed experimentally, including which CYP enzymes metabolize DPO, what metabolites are produced, and how long DPO remains in the body.

Synthesizing process

I am not a chemist, I could be entirely wrong here!

The basic idea is relatively simple. DXO already has the same morphinan structure as DPO, but instead of having an isopropoxy group, it has a hydroxyl group at the 3-position.

The main problem is that DXO also contains an amine, and that amine can react during the process. So the first step would be to temporarily protect the amine so that it doesn’t interfere.

Then the isopropoxy group could be added, and finally the protection would be removed.

In simple terms:

DXO → protect the amine → add isopropoxy → remove the protection → DPO

Afterward, the compound would need to be tested to make sure the correct molecule was actually produced. NMR and mass spectrometry could be used to confirm its structure and purity.

But making the molecule is only the first step. The real question is whether DPO actually behaves the way I predicted.