STUDresearch · Non-peptide
Seltorexant
Also known as
JNJ-42847922 · MIN-202 · 2-SORA · selective OX2 antagonist · seltorexant 20 mg
Community talk. May be wrong. Not medical advice. Not a protocol. Not for human or animal use.
Non-peptide
Lots of talk
Systemic
Oral
Sleep / mood
Systemic oral selective OX2 (orexin-2) blocker.
What people say
Seltorexant is an investigational oral selective orexin-2 receptor antagonist studied for insomnia and depression. It is neither a peptide nor an approved insomnia drug.
Doses people talk about
Rows are separate insomnia, depression and community contexts, not a progression. Trial material cannot validate a gray-market product.
How it may feel
Trials show sleep-promoting and somnolence effects, while community reports split between faster sleep onset, little effect and insufficient sleep maintenance. Short plasma half-life does not guarantee a clear morning.
Recent community reports20–40 mg oral across separate authors
Some authors described better sleep onset at 20 mg; others reported underwhelming effects, short maintenance or little benefit up to 40 mg. Accounts included other sleep medicines in some cases.
Insomnia phase 2b comparison arms5, 10 or 20 mg oral nightly for 14 days
Randomized against placebo and zolpidem in adults and older adults with insomnia; 10 and 20 mg improved selected sleep measures in the reported dataset.
MDD adjunct phase 2b arm20 mg oral once daily for 6 weeks
Studied on top of an existing SSRI or SNRI; 10 and 40 mg arms also appeared during the adaptive study. This is not a monotherapy or insomnia label.
Half-life & effect duration
- Half-life in the body
- OralAbout 2–3 hours
- Felt duration people report
- Positive onset accountsAsleep within about 20–40 minutes
- Other accountsNot enough sleep maintenance, underwhelming effects or next-day sleepiness
Cycles people discuss
The insomnia study used nightly dosing for 14 days; MDD programs used longer daily blocks. Community week-long trials are reports, not evidence that intermittent or indefinite use is equivalent.
Timing
Insomnia trials used oral nighttime dosing; community notes describe roughly 20–40 minutes before bed, with late caffeine and next-morning grogginess complicating the routine. Those reports do not validate a bedtime interval or guarantee morning alertness; depression-study dosing is a separate adjunctive setting.
Stacks
Posts compare or combine seltorexant with magnesium, glycine, DSIP, L-theanine and other sleep agents. These mixed stacks confound both benefit and morning effects; no combination is established here.
Access talk
Gray tabs vs Dayvigo: Lemborexant is the pharmacy DORA. Seltorexant is the selective OX2 research tab. The two products are not interchangeable.
Watch for
Next-day grogginess at 20+ mg: Recurring individual reports. 5–10 mg is the “clear morning” pitch, not a guarantee.
Tap a line to jump into the full notes. Research only — may be wrong.
Timing context & sources
Half-life in the body
Human oral studies report a mean parent half-life of about 2–3 hours.
Healthy-subject multiple-dose work reported median Tmax of 0.5–1.5 hours and mean half-life of 2–3 hours. Earlier single-dose 10–80 mg cohorts reported roughly 0.33–0.5-hour Tmax and around 2-hour half-life.
Study formulations and controlled healthy adults do not validate community products or guarantee subjective sleep timing or morning clarity.
- Multiple daytime administration of JNJ-42847922 in healthy subjects (opens in a new tab)Abstract and results reviewed: healthy-subject oral 5–60 mg once-daily cohorts over 10 days; median Tmax 0.5–1.5 hours, mean half-life 2–3 hours, somnolence at 20 mg and above, and central-function assessments from 4 hours onward.Small healthy-volunteer daytime study, not insomnia efficacy, community-product validation or a direct subjective bedtime-duration study. The study was Janssen-sponsored and most listed authors were Janssen employees.
- Characteristics of seltorexant for depression and anxiety (opens in a new tab)Full review pharmacokinetic section and Table 1 reviewed: healthy-male single oral 10, 20, 40 and 80 mg data with mean Tmax 0.33–0.50 hours and mean half-life 2.02–2.44 hours, plus multiple-dose study summary.Secondary review of early sponsored studies; not approval status or community product evidence.
Felt duration people report
Reports conflict on sleep onset, maintenance and next-day effects despite the short parent half-life.
Controlled studies found sleep-promoting effects and daytime somnolence at higher doses. Community authors separately described faster sleep onset, insufficient maintenance or underwhelming effect; one combined lemborexant.
Different indications, products, doses and co-medications; self-reports cannot establish prevalence or equate parent elimination with perceived sleep benefit.
- Multiple daytime administration of JNJ-42847922 in healthy subjects (opens in a new tab)Abstract and results reviewed: healthy-subject oral 5–60 mg once-daily cohorts over 10 days; median Tmax 0.5–1.5 hours, mean half-life 2–3 hours, somnolence at 20 mg and above, and central-function assessments from 4 hours onward.Small healthy-volunteer daytime study, not insomnia efficacy, community-product validation or a direct subjective bedtime-duration study. The study was Janssen-sponsored and most listed authors were Janssen employees.
- NCT03375203 phase 2b insomnia dose-response study (opens in a new tab)ClinicalTrials.gov record and linked protocol reviewed: 5, 10 and 20 mg seltorexant versus placebo and zolpidem, with repeated nightly administration for 14 days and sleep-onset/maintenance endpoints.Registry/protocol context; does not authorize use, validate nontrial products or replace full statistical review of every endpoint. The registry names Janssen Research & Development as sponsor and responsible party.
- Have you tried the seltorexant yet? (opens in a new tab)Multiple distinct authors and visible follow-ups reviewed: reports ranged from strong sleep benefit to easier sleep onset without maintenance, no useful effect, or little benefit for maintenance insomnia; product-purchase remarks were not transferred.Multiple unrelated authors, unverified products/doses and incomplete co-medications; cannot establish prevalence, product identity or comparative efficacy.
- Users of seltorexant: community comparison discussion (opens in a new tab)Multiple distinct authors reviewed: underwhelming or null reports, one report of faster but brief effects up to 40 mg, separate 20 mg sleep-onset benefit reports, and one disclosed lemborexant combination. Authors remain separate.Very recent self-selected thread, unverified products, several missing exact amounts and one combination; cannot establish safety, tolerance or comparative effectiveness.
Other context in this card
- Seltorexant as adjunctive therapy in major depressive disorder phase 2b study (opens in a new tab)PubMed abstract reviewed: adaptive placebo-controlled 10, 20 and 40 mg once-daily arms on top of existing SSRI/SNRI therapy over six weeks, with reported 20 mg findings and somnolence, headache and nausea adverse events.Adjunctive MDD study, not an insomnia label, monotherapy instruction or evidence for community products. Most listed authors were Janssen employees and the trial was industry-sponsored.
What people say
- n=1 continuity: Wearable logs (Oura / Whoop-style) describe fewer micro-awakenings and higher sleep efficiency more than extra total hours. One August 2026 log: 94.6% efficiency vs 88.5% baseline, same total sleep. anecdote
- “I can finally shut off”: Peptide-Twitter wording — exhausted-but-wired insomnia, not sedation. Plausible if you buy the OX2 story. forum
- Next-day vs 20–25 mg: 5–10 mg logs are the “clear morning” band. 20 mg and 25 mg get groggy-next-day complaints in the same week’s posts. forum
- Vs DSIP: Multiple 2026 posts: DSIP did nothing, seltorexant put them out. Selection bias is loud. forum
- JAMA Psychiatry insomnia trial (people quote this): 10 mg and 20 mg beat placebo on falling asleep and staying asleep over 14 nights; 20 mg also beat zolpidem on latency in that read. 5 mg was the weak arm. trial
- MDD + insomnia adjunct: J&J Phase 3 story is 20 mg once daily on top of an SSRI/SNRI for depression scores *and* sleep disturbance — not a party-sleep pill. trial
Doses people talk about
- n=1 sweet-spot talk: 5–10 mg 30 minutes pre-bed is the copied biohack band. forum
- Higher RC experiments: 15, 20, 25 mg — groggy reports climb. forum
- One reported 5-to-25 mg comparison: A public log described its best wearable continuity at 5 mg and more grogginess without better deep-sleep numbers at 15–25 mg. This unverified account is not a dose-escalation or dose-selection rule. anecdote
- Not 80 mg: Early development explored up to 80 mg. That is not the Twitter protocol. trial
- MDD 20 mg is adjunct on an SSRI/SNRI — not a DIY antidepressant milligram. trial
- Framing: Trial milligrams and RC n=1s — not a prescription, not FDA insomnia directions. forum
- Insomnia RCT arms: 5 mg, 10 mg, 20 mg oral at night vs placebo vs zolpidem. 10 and 20 moved the needle. trial
- MDD adjunct program: 20 mg once daily is the Phase 3 number. trial
How it may feel
- Nights 1–7: Sleep-efficiency gain shows on wearables before “I slept 8 hours” does. Fragmentation down, time-in-bed sometimes up. anecdote
- Dose curve in n=1s: One public log peaked continuity at 5 mg; 10 and 15 mg added nothing; 15 mg was a worse night. Another ran 20 mg × 7 nights and called deep+REM better. anecdote
- Groggy band: 20–25 mg morning-after fog is the recurring complaint. Short 2–3 h half-life is supposed to prevent that — not always. forum
- Not a cure for short sleepers: Several n=1s: consolidation yes, extra hours no. forum
- Split n=1s in Aug 2026: One public log (10 mg × 3 nights) posted 95+ wearable scores and extra total hours; another on the same-week Kimera tabs called 10–25 mg “no score change, groggy as hell, trash.” Hit-or-miss is the honest map. anecdote
- Fragmentation vs extra hours: The useful reports are fewer micro-awakenings / higher efficiency, not a benzo knockout. forum
- 20 mg trial band vs 5–10 mg Twitter band: Insomnia RCT 10 and 20 moved latency; 5 mg was the weak arm. Biohack logs still treat 5–10 as the “clear morning” band because 20 mg somnolence showed up in both the RCT table and the n=1s. trialforum
- Night 1: Fast Tmax (about 0.3–1.5 h). People who respond describe lights-out in 20–40 minutes, not a benzo knockout. trial
- Week 2+: Insomnia trial still showed 10/20 mg working on night 13. Not a one-night trick in that dataset. trial
Around the dose
- Next-morning reports: Existing notes contrast clearer mornings at 5–10 mg with groggier 20–25 mg logs. These unverified reports do not establish a clear-morning band, a dose-response rule or driving safety. forum
- Solo first: Isolate it from DSIP / pinealon / GABA hammers for a week so the wearable change (or lack of one) can be blamed on one tab. forum
- Caffeine cutoff: Late coffee plus an OX2 block is a common “why am I still wired” confound. forum
- After-dose discussion: Existing notes compare next-night carryover with longer DORAs, but a 2–3-hour parent half-life does not establish return to baseline or a rebound profile. forum
- Dose-comparison report: An existing public log says continuity peaked at 5 mg, 15 mg was a worse night, and higher amounts did not reliably improve the wearable score. This unverified account does not establish a dose-selection rule. anecdote
- Clock: Healthy-adult PK studies place median Tmax around 0.5–1.5 hours after oral dosing. That daytime absorption window does not prove sleep onset or a universal bedtime interval. trial
Cycles people discuss
- Nightly while testing: Most n=1s run it every night for a week+, then keep or drop. forum
- Not a weekend benzo: People treating it as Ambien-as-needed sometimes miss that the trial design was nightly blocks. forum
- Stop: Short half-life — next night is closer to baseline than a long DORA tail. Rebound talk is thinner than with zolpidem in these threads. forum
- Trial blocks: 14-day insomnia RCT; 6-week MDD adjunct. trial
Timing
- Residual 4 h later: Daytime single-dose work cited no residual at 4 hours — that is why n=1s contrast it with lemborexant’s longer linger. forum
- Half-life: About 2–3 hours. Pitched as “asleep, then gone.” trial
- Tmax: 0.3–1.5 hours — dose 20–40 minutes pre-bed in the n=1s. trial
- CYP3A4: Metabolized by CYP3A4. Grapefruit / inhibitor talk is thinner than with tadalafil but the same enzyme family. trial
More on what it is
- Why people talk about it: 2026 biohacking discussions compare this short-half-life investigational OX2 antagonist with longer DORAs and peptide sleep ideas. Actual forum reports range from strong sleep-onset benefit to little effect or weak sleep maintenance. forum
- Vs DORAs: Lemborexant and daridorexant block OX1 and OX2. Seltorexant is the selective one — that is the whole “cleaner NREM / less next-day” argument. forum
- Vs DSIP / pinealon: Peptide sleep cards have mixed human evidence, while seltorexant has randomized insomnia and MDD programs. Unverified community products are not Janssen trial material. forumtrial
- Not Ambien: Mechanism is not GABA. Some clinic quotes still say “why not just use zolpidem” — that is insurance talk, not the n=1 crowd. forum
- What it is: Janssen/Minerva’s selective orexin-2 antagonist (JNJ-42847922 / MIN-202). Oral. Still investigational — not an FDA insomnia label as of the 2026 threads. Not a peptide. trial
- How it works (plain): Orexin is a stay-awake peptide in the brain. Seltorexant blocks mainly OX2 (~100× over OX1 in the pitch), so the “keep running” signal quiets without the classic Ambien GABA hammer. trial
Stacks
- Solo first: The n=1s that look useful isolated it from pinealon/epitalon for a week. forum
- + Magnesium glycinate / glycine: House-of-biohacking-style sleep stack posts. forum
- + DSIP nasal: Some still run both; others say DSIP became pointless. forum
- + L-theanine: Vendor threads pair OX2 block with GABA-ish calm. forum
- Not with a GABA hammer on night one: People comparing to Ambien are usually switching, not stacking. forum
Access talk
- Gray tabs vs Dayvigo: Lemborexant is the pharmacy DORA. Seltorexant is the selective OX2 research tab. The two products are not interchangeable. forum
- Kimera / RC bottles are the 2026 biohack Twitter source people actually name. Identity and milligram are trust-the-label problems. forum
- Vendor heat ≠ approval: August 2026 research-chem leaderboards had seltorexant as a top-ordered SKU. Sales rank is not an NDA. forum
- Not an FDA insomnia pill (as of these threads). Investigational 2-SORA (JNJ-42847922 / MIN-202). Phase 3 is MDD-adjunct, not a Dayvigo substitute you can fill at CVS. trial
- MDD3005 (Psych Congress 2025): 20 mg adjunct vs quetiapine XR missed statistical superiority on MADRS response (57.4% vs 53.4%) with less weight gain and less somnolence. Comparable, not a win — and still not a sleep-label. trial
Storage notes
- No BAC water. Finished oral solid in almost every log. forum
- RC bottles: Light, dry, original pack — treat gray tabs as identity-unverified. forum
Watch for
- Next-day grogginess at 20+ mg: Recurring individual reports. 5–10 mg is the “clear morning” pitch, not a guarantee. forum
- Does not fix sleep apnea / blocked nose: One OX2 n=1 said so explicitly. forum
- More ≠ better: Public dose-finding n=1 got worse metrics at 15 mg than 5 mg. anecdote
- Not a peptide. Do not mix mg charts with DSIP mcg charts. forum
- Gray identity: Research tabs are not Janssen trial material. Dose roulette is the product. forum
- Not FDA-approved for insomnia (as of these threads): Investigational. Gray-market tabs are identity + dose roulette. trial
- Depression program is adjunct, not a DIY antidepressant. 20 mg sat on top of SSRI/SNRI in Phase 3. trial
- CYP3A4 interactions: Same enzyme family as many other night pills. trial
- Phase 3 is messy: One MDD-insomnia arm missed its primary vs placebo in a 2025 public posting; another (MDD3001) is cited as a win; MDD3005 missed statistical superiority vs quetiapine XR (57.4% vs 53.4% MADRS response) with less weight gain and less somnolence. Do not read “sleep RC of the month” as an approved depression drug. trial
- CYP3A4: Same enzyme family as many night pills. Grapefruit / inhibitor talk is thinner than tadalafil but the same family. trial
Evidence mix
Mostly community / anecdote tags
Full: every bullet (trial + community). Use Scan for a faster bro-science read.