The most important ketamine study running right now is not a commercial trial. It is a federal one, and it tests the assumption the entire commercial field is built on. The National Institute of Mental Health has opened a Phase 2 trial of (2R,6R)-hydroxynorketamine, the principal metabolite of ketamine, as a standalone antidepressant in treatment-resistant depression, alongside a companion Phase 1 trial that probes the AMPA-throughput theory of ketamine’s mechanism directly in patients. Together they ask the question underneath every rapid-acting antidepressant program in development: can ketamine’s antidepressant benefit be separated from the dissociation and abuse liability that limit it?
The hypothesis the field is built on
In 2016, a team led by researchers at NIMH, the National Institute on Aging, and the University of Maryland reported in Nature that hydroxynorketamine, abbreviated HNK, reproduced ketamine’s antidepressant effects in mice without ketamine’s side effects, and, more provocatively, that the effect was independent of NMDA-receptor inhibition, working instead through activation of AMPA receptors. The same work showed that when ketamine’s metabolism into HNK was blocked, ketamine lost its antidepressant effect in mice, which implied the metabolite, not the parent drug, was doing the therapeutic work.
If that holds in humans, the implication is large. Ketamine’s dissociation and abuse potential are tied to NMDA-receptor blockade, the action of the parent compound. If the antidepressant benefit instead comes from a metabolite acting through a different receptor system, then the benefit and the baggage are separable, and a drug can in principle deliver one without the other. This is the mechanistic foundation of the entire clean-ketamine enterprise, the premise behind oral and deuterated reformulations, the R-enantiomer programs, and metabolite-based candidates alike.
Why it is contested
The premise is not settled science. It has been disputed for a decade on three separate fronts, and a careful reader should hold all three in view.
The first is whether HNK is actually NMDAR-independent. A group led by researchers including the Monteggia lab challenged the original claim in 2017, showing that HNK does inhibit NMDA receptors at relevant concentrations and is therefore not NMDAR-independent in its action. The original investigators replied that antidepressant-relevant concentrations of HNK do not block NMDARs. The mechanistic question remains open.
The second is whether HNK reliably works at all. Replication has been mixed. At least one study found no antidepressant effect of HNK in a rat learned-helplessness model and argued that R-ketamine is more potent and longer-lasting, and other groups have favored R-ketamine over the metabolite. The preclinical case is contested even before reaching humans.
The third, and most consequential, is that the human data so far is discouraging. A 2020 NIMH analysis found that higher HNK plasma levels predicted poorer antidepressant response in a treatment-resistant depression trial, the opposite of what the hypothesis would predict, prompting a published paper that asked in its title whether that was the end of the HNK pipeline.
Why the trial matters anyway
The discouraging human signal came from correlational data, plasma levels measured in trials of the parent drug, not from HNK administered directly. Correlational findings can mislead, and the only way to resolve whether HNK itself is an antidepressant is to give it to patients in a controlled trial and measure the result. That is exactly what the NIMH Phase 2 does. It is the most substantive public-sector investment in ketamine mechanism science currently active, run by the NIMH group that has been central to ketamine and esketamine development for two decades. One disclosure worth noting: the senior investigator is a co-inventor on the HNK patents, with the rights assigned to the US government and a share of any royalties, which is a standard government-research arrangement and also a reason to read the program’s framing with the usual care.
There is a clarifying point about what the trial can and cannot settle. The decade-long argument over whether HNK blocks NMDA receptors is a mechanistic question. The clinical question, whether HNK relieves depression without producing dissociation, is answerable directly regardless of how the mechanism debate resolves. The Phase 2 is built to answer the clinical question. The companion Phase 1, testing AMPA throughput in patients, goes at the mechanism.
The stakes for the commercial field
The commercial pipeline has largely moved on to engineering, oral formulations, deuterated metabolites, abuse-deterrent versions, as though the separability of benefit from baggage were already established. It is not. ACADIA’s ACP-211, an oral deuterated form of the related metabolite R-norketamine, is a different bet on the same premise. The R-ketamine programs and the broader pharmacokinetic-engineering trend the desk has covered all assume that the antidepressant effect can be retained while the liabilities are designed out. HNK is the purest and most direct test of whether that assumption is biologically true.
If the NIMH Phase 2 shows HNK works as a standalone antidepressant without dissociation, it validates the foundation under all of those programs. If it fails, and the human signal so far leans that way, it suggests the dissociation and the NMDA engagement may be harder to separate from the benefit than the field has assumed, which would be a caution for every candidate built on the clean-ketamine premise. The result will not pick a winner among the commercial programs. It will tell the field whether it is optimizing the delivery of an effect that can be cleanly separated from ketamine’s liabilities, or chasing a separation the biology may not allow.