Emideltide — delta sleep-inducing peptide, universally called DSIP — was the one compound the committee declined to recommend. It lost 6 to 7, with one abstention.
A single vote the other way would have produced a clean seven-for-seven outcome. What separated it is worth understanding, because it was not obviously the weakest science on the docket.
The basics
| Field | Detail |
|---|---|
| Structure | Nonapeptide (nine amino acids) |
| Origin | Identified in the 1970s in research on sleep-related factors in cerebral blood |
| Reviewed for | Opioid withdrawal, chronic insomnia, narcolepsy |
| PCAC vote | 6 yes, 7 no, 1 abstention — not recommended |
| FDA staff position | Recommended against inclusion |
| Current status | Not eligible for compounding |
The name is a hypothesis, not a finding
DSIP was named for an observed association with delta-wave sleep in early research. That name has done an enormous amount of unearned work in the decades since, because it states a conclusion in the label.
The subsequent literature has not been consistent. Reproducing a clear, reliable sleep effect has proven difficult, and the compound's endogenous role remains incompletely characterized. A molecule named for an effect it may not dependably produce is a recurring hazard in this field — the name persists long after the evidence stops supporting it.
Why it failed: the scope problem
Look at what each compound was nominated for:
| Compound | Indications reviewed | Count | Outcome |
|---|---|---|---|
| TB-500 | Wound healing | 1 | Passed 8–6 |
| Epitalon | Insomnia | 1 | Passed 7–4 |
| BPC-157 | Ulcerative colitis | 1 | Passed 8–6 |
| MOTS-c | Obesity, osteoporosis | 2 | Passed 7–5 |
| Semax | Cerebral ischemia, migraine, trigeminal neuralgia | 3 | Passed 8–5 |
| Emideltide | Opioid withdrawal, chronic insomnia, narcolepsy | 3 | Failed 6–7 |
Semax also carried three indications and passed, so scope alone does not explain the outcome. What distinguishes DSIP's three is how far apart they are. Cerebral ischemia, migraine and trigeminal neuralgia are all neurological conditions with related mechanistic territory. Opioid withdrawal, chronic insomnia and narcolepsy span addiction medicine and two distinct sleep disorders with different underlying pathology.
Each of those would require its own evidence base. A committee already working at the edge of sufficiency had three separate literatures to be satisfied about, and the material did not extend that far.
The transferable lesson
Committee members who voted against emideltide pointed to evidence quality. But the pattern across both days suggests something more specific: nominations succeeded when they asked one coherent question and struggled when they asked several unrelated ones. How a nomination is framed appears to matter alongside what supports it.
Narcolepsy was the hardest ask
Of the three, narcolepsy is the most demanding. It is a specific neurological disorder with defined diagnostic criteria and, in the most common form, a well-characterized underlying mechanism involving loss of orexin-producing neurons. It also has approved treatments with established evidence.
Asking a committee to accept a compound with an inconsistent sleep literature for a condition that is both mechanistically specific and already treatable is a considerably heavier lift than asking about general insomnia.
What the rejection does not mean
It does not mean DSIP is dangerous, and the committee made no such finding. It does not mean it is pharmacologically inert. And it does not mean the other six were shown to work — they were recommended against FDA staff advice, on a determination about characterization and safe handling rather than effectiveness.
The practical status of DSIP is identical to the other six: not approved, not eligible for compounding, sold only as research material outside the medical system.
Can it come back?
Nothing prevents a future nomination, and a narrower one — a single indication, framed around the strongest available literature — would plausibly fare better. There is no announced plan to revisit it, and it is not on the agenda for the meeting scheduled before the end of February 2027.
Common questions
It lost 6 to 7 with one abstention. Members who voted against it cited evidence quality. The pattern across both days also suggests its unusually broad indication set — spanning addiction medicine and two distinct sleep disorders — made it a heavier lift than single-indication nominations.
No. The committee made no safety finding against it. A vote not to recommend inclusion on a compounding bulks list is not a determination of harm.
The literature is inconsistent. It was named for an association observed in early research, but reproducing a clear and reliable sleep effect has proven difficult, and its endogenous role remains incompletely characterized.
Its practical status is the same as the six recommended compounds: not approved, not eligible for 503A compounding, and available only as research-use-only material outside the medical system.
Nothing prevents a future nomination, and a narrower one would plausibly do better. None has been announced, and it is not on the February 2027 agenda.
Sources
- US Food and Drug Administration. PCAC meeting, July 23–24, 2026. Meeting docket FDA-2026-N-2979; bulk substances docket FDA-2025-N-6895. fda.gov
- McDermott Will & Schulte. “Bulk-list bound? PCAC backs majority of peptides in two-day public meeting.” July 27, 2026.
- Orrick. “FDA Peptide Compounding Vote: What to Watch at the July PCAC Meeting.” July 2026.
- Sheppard Mullin. “Compounded Peptides on the Loose: What the Recent PCAC Meeting Means for Industry.” August 2026.
- Fierce Pharma. “Peptide adcomm Day 2: Emideltide voted down in panel’s 1st pushback.” July 2026.