Mature neurosteroid precursor with modest clinical signal and downstream-conversion interpretability issues
Look, pregnenolone has been sitting in the research conversation since at least the 1940s, and that long tenure cuts against it pretty hard on novelty. It’s an OTC supplement anyone can buy, synthesized endogenously from cholesterol by CYP11A1, and it’s the upstream ancestor of basically every steroid hormone your body makes. That pedigree sounds impressive until you realize it mostly means that any experiment you run with pregnenolone is immediately complicated by a cascade of metabolic daughters, DHEA, cortisol, progesterone, sex steroids, that you didn’t necessarily intend to modulate. Clean mechanistic work is genuinely hard to do with this compound.
Where pregnenolone earns some respect is its neurosteroid activity. The sigma-1 receptor agonism is real and reasonably well-characterized, and the positive allosteric modulation of NMDA receptors at micromolar concentrations gives it a plausible CNS mechanism that isn’t just a simple single-receptor binding story. These aren’t new findings, but they’re substantive. The problem is that if you’re actually trying to probe sigma-1 biology, there are far more selective ligands available, and if NMDA allosteric modulation is the target, ketamine and memantine give you cleaner pharmacology without the hormonal noise. Pregnenolone doesn’t really define a research question the way a best-in-class probe should.
The clinical evidence is modest and mixed. Phase 2 work in schizophrenia has produced some positive signals, the UCSD group has published on augmentation strategies and there’s been multi-site work, but effect sizes are unimpressive and replication has been inconsistent. Cognitive enhancement signals in older adults are plausible given the age-related CNS decline in neurosteroid levels, but this hasn’t translated into anything that would justify prioritizing pregnenolone over more tractable interventions. The evidence base is better than pure anecdote but falls well short of anything convincing.
What we dont like most is the interpretability problem. Any time you administer pregnenolone and observe an effect, you’re looking at a compound that has simultaneously altered sigma-1 tone, NMDA receptor function, GABA-A receptor modulation through allopregnanolone conversion, and potentially cortisol and sex steroid levels. Teasing apart which of those pathways drove the outcome is genuinely difficult without heroic experimental design. That makes pregnenolone a somewhat frustrating research tool even for investigators who find the neurosteroid biology interesting.
Honestly, if you’re building a research program around neurosteroid biology, the more interesting entry points right now are the 3-alpha-reduced metabolites like allopregnanolone (brexanolone has FDA approval and a well-validated GABA-A mechanism) or the specific sigma-1 literature. Pregnenolone sits upstream of all of that, which sounds like a feature but mostly functions as a limitation. It’s worth understanding, especially as a reference standard in steroidogenesis assays, but we think researchers in 2025 have better options for almost every question pregnenolone was once the only tool to answer.