What Tabernanthalog Actually Is
Tabernanthalog, usually shortened to TBG and carrying the development code DLX-007, is a lab-made molecule that shares a skeleton with ibogaine but strips away most of what makes ibogaine dangerous and disorienting. Chemically it is 8-methoxy-3-methyl-2,4,5,6-tetrahydro-1H-azepino[4,5-b]indole, formula C14H18N2O, molecular weight 230.3 g/mol, CAS number 2483829-59-8. It is not found in any plant. It was built in a chemistry lab at the University of California, Davis, by David Olson's team, and first described in a January 2021 Nature paper titled "A non-hallucinogenic psychedelic analogue with therapeutic potential" (Cameron et al.). That title is the whole pitch: keep the parts of ibogaine's pharmacology that seem to drive therapeutic change, cut the parts that make it a 10-to-18-hour hallucinogenic ordeal with real cardiac risk.
The closest natural relative is tabernanthine, one of the minor alkaloids in iboga root alongside ibogaine itself. Olson's lab calls this family of simplified, lab-derived cousins "ibogalogs," and TBG was the first one to get real attention. Simplifying the ibogaine scaffold this aggressively was deliberate function-oriented synthesis: start from a natural lead, delete the parts not doing therapeutic work, keep testing until the reduced molecule still produces the effect researchers care about in animal models.
The Numbers That Matter: Receptor Binding and the hERG Comparison
TBG is not a clean single-target drug. Published binding data (Cameron et al. 2021; later refined by Arias, Rudin, Luethi and colleagues through 2024-2025) show it acting as an agonist at 5-HT1B, 5-HT1F, 5-HT2A, 5-HT2C, and 5-HT6 receptors, a mixed agonist/antagonist at 5-HT2B, an 88% blocker of the serotonin transporter (SERT) at 10 µM, and a 66% inhibitor of MAO-A at the same concentration. Its 5-HT2A affinity (Ki 4,440 nM) is weak compared to classic psychedelics, and its functional efficacy there swings wildly by assay, from 8% to 91%. That swing is part of why researchers still argue about how "non-hallucinogenic" it really is in humans, not just in rodents.
The one number every ibogaine-adjacent article should lead with: TBG is roughly 100-fold less potent at the hERG cardiac ion channel than ibogaine itself. hERG inhibition is the mechanism behind ibogaine's documented QT-interval prolongation, the reason reputable clinics run mandatory EKGs and cardiac screening before every ibogaine session, and the reason ibogaine has a published death toll tied specifically to cardiac arrhythmia. If TBG's hERG safety margin holds up in human trials the way it did in the original rodent and in vitro work, it would represent a real structural fix to ibogaine's single biggest safety problem, not just a marketing claim.
TBG also showed negligible interaction with opioid receptors and the NMDA receptor in the original 2021 characterization, in sharp contrast to ibogaine and its primary metabolite noribogaine, both of which bind opioid receptors directly. Later work (Tae, Ortells, Yousuf et al., 2024) found weak inhibition of some nicotinic acetylcholine receptors and a small GABA-A interaction that the first paper missed, a normal part of how pharmacology estimates get refined as more labs run the assay. Opioid receptor binding is a big part of why ibogaine shows promise against opioid withdrawal in the first place, so TBG's near-absence of that activity means it is working through a different route to get to similar behavioral outcomes, if it works at all in humans.
What the Animal Studies Actually Showed
Four findings carry the therapeutic argument, and all of them come from rodent studies, not human trials:
- No head-twitch response. The head-twitch response (HTR) is the standard rodent proxy for hallucinogenic potency: it correlates with dose in every classic psychedelic ever tested, from LSD to DOI to psilocybin. TBG did not produce it. That is the entire empirical basis for calling TBG "non-hallucinogenic."
- Structural neuroplasticity. TBG promoted dendritic spine growth and increased spine density in cortical neurons in the original study and in several follow-ups (Lu, Tjia, Mullen et al. 2021; Peters & Olson 2021), the same "psychoplastogen" signature first described for ketamine and classic psychedelics. This is the mechanistic story Delix is selling across its whole pipeline, not just TBG.
- Reduced drug-seeking behavior. Heinsbroek, Giannotti, Bonilla et al. (2023) reported TBG reduced motivation for both heroin and alcohol self-administration in rodent models, the addiction-relevant finding that put TBG on NIDA's radar.
- Antidepressant-like effects in standard rodent assays (forced swim test, among others), consistent with the broader neuroplastogen hypothesis.
None of this is nothing. But every single data point above is preclinical. There is no published human efficacy data for TBG specifically, at any dose, for any condition, as of this writing.
Where the Clinical Program Actually Stands (and Where It Doesn't)
This is the part most coverage of TBG gets wrong by omission. Delix Therapeutics licensed TBG from Olson's lab and announced a NIDA partnership to evaluate it for substance use disorders back in December 2021. A Phase 1 trial was publicly targeted for the first half of 2023. As of January 2026, per Wikipedia's most recent sourced update (citing AdisInsight), tabernanthalog has still not entered clinical trials. Three-plus years past the original timeline, TBG remains a preclinical asset.
What has moved is Delix's other lead candidate. Zalsupindole (DLX-001, also called AAZ-A-154), a 5-MeO-DMT-derived non-hallucinogenic neuroplastogen, reached a Phase 1b readout: an 18-patient study reporting an average 12-point reduction on the Montgomery-Åsberg Depression Rating Scale (MADRS) through day 36, per reporting on Delix's pipeline progress. That is encouraging for the company's platform thesis, but it is a different molecule with a different chemical backbone, and the sample size is too small to draw firm conclusions from on its own. In September 2026, Psychedelic Alpha reported Johnson & Johnson was set to lead an $85 million Series C round for Delix, a serious pharma-grade vote of confidence in the platform, following a $25 million Series B in September 2025 and roughly $100-133 million raised across the company's history depending on which funding tracker you check. None of that capital has, as of now, been publicly tied to restarting a TBG-specific clinical timeline. The DoD also awarded Delix an $825,000 grant, but for a hearing-loss neuroplastogen program, not TBG.
Translation for anyone searching "tabernanthalog clinical trial status 2026": the honest answer is that the company's institutional momentum and investor interest are real and growing, but it is currently flowing toward zalsupindole, not toward TBG. If you see claims that TBG is "in trials" or has a trial start date, ask for the actual ClinicalTrials.gov registration number before taking it at face value. We checked and found none as of this writing.
Tabernanthalog vs. Ibogaine: The Direct Comparison
| Property | Ibogaine | Tabernanthalog (TBG) |
|---|---|---|
| Source | Natural alkaloid, Tabernanthe iboga root | Fully synthetic, lab-designed analog |
| hERG cardiac risk | Significant: documented QTc prolongation, cardiac deaths reported | ~100x lower potency at hERG in preclinical testing |
| Subjective experience | Intense, often 10-18+ hours, visionary/dissociative | No head-twitch response in rodents; anecdotal human reports describe subtle or minimal effects, contested |
| Opioid receptor activity | Direct binding (ibogaine + noribogaine) | Negligible in original characterization |
| Legal status (US) | Schedule I controlled substance | Not an explicitly controlled substance as of 2026 |
| Human efficacy data | Decades of clinic-level and some published outcome data | None published; still preclinical |
| Availability | Clinics worldwide, decades of treatment experience | Sold online by research-chemical vendors as an unregulated "nootropic," no quality control |
The table makes the trade-off obvious. TBG's whole value proposition is removing ibogaine's cardiac risk and multi-hour subjective ordeal while chasing the same addiction-and-depression mechanism. What it has not yet done, and may never do, is prove in a controlled human trial that it produces anything close to ibogaine's reported outcomes. Ibogaine's real-world track record, for all its legitimate safety concerns, is built on actual patients in actual clinics. TBG's case rests entirely on mouse and cell data, more than five years after the original discovery paper.
The Legal Status Problem and the "Non-Hallucinogenic" Critique
Because TBG is a novel molecule rather than a scheduled compound, it is not explicitly controlled in the United States or Canada as of 2025-2026. That legal gray zone is exactly why it shows up for sale through research-chemical vendors marketing it as a nootropic, a channel with zero quality control, no dosing standards, and no clinical oversight, closer in practice to the unregulated research-chemical market than to anything resembling medicine. Anecdotal reports collected by Psychedelic Alpha, The Atlantic ("Tripping on Nothing," October 2024), and a widely circulated Medium account from AI researcher Arthur Juliani describe genuinely divergent experiences: some report essentially no perceptual effects at all, others describe subtle mood or perspective shifts that are hard to separate from expectation.
That gap between "no head-twitch in mice" and "no hallucinogenic effects in humans" is also where the field's harshest scientific critique lands. A January 2026 analysis published by the OPEN Foundation argued that HTR, the single behavioral test underpinning most "non-hallucinogenic" claims across this entire drug class, is a poor proxy once you move beyond classic 5-HT2A-selective tryptamines into polypharmacological compounds like TBG that also hit SERT, MAO-A, and multiple other serotonin receptor subtypes. The criticism isn't that TBG is secretly hallucinogenic. It's that the rodent test used to support the marketing claim wasn't built for a molecule with TBG's receptor profile, and nobody has run the human studies that would settle the question properly.
The Bottom Line
Tabernanthalog is a legitimate, well-published piece of neuroscience: a rationally designed ibogaine analog that appears, in preclinical models, to decouple the therapeutic signal from the cardiac risk and the hallucinogenic burden. That is a real scientific achievement regardless of what happens next. What it is not, five years after the Nature paper and more than three years past its originally announced Phase 1 target, is a clinically validated treatment, or even a compound currently in human trials. Delix Therapeutics is a well-funded, increasingly pharma-backed company (the reported Johnson & Johnson-led Series C is a serious signal), but its clinical momentum right now is running through zalsupindole, not TBG. If you're researching TBG as an alternative to ibogaine treatment, the honest 2026 answer is: not yet, maybe eventually, and nobody currently selling it to you as a "legal ibogaine alternative" has human data to back that framing.
For readers dealing with the conditions TBG is aiming at, including opioid addiction, depression, and alcohol use disorder, ibogaine itself, used at a screened, cardiac-monitored clinic, remains the option with an actual track record today. TBG is worth watching. It isn't yet something to choose instead.