Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,112 @@
+---
+license: cc-by-4.0
+tags:
+- healthcare
+- regulatory
+- fda
+- compounding
+- peptides
+- pharmacy
+- drug-policy
+language:
+- en
+size_categories:
+- n<1K
+pretty_name: "Peptide Compounding Status Index — FDA PCAC, July 2026"
+---
+
+# Peptide Compounding Status Index — FDA PCAC, July 2026
+
+A sourced, machine-readable reference of the seven peptides the US FDA's Pharmacy Compounding Advisory Committee (PCAC) evaluated at its July 23–24, 2026 meeting for possible addition to the Section 503A bulk drug substances list.
+
+**None of these peptides is an FDA-approved drug, and none is legal to compound today.**
+
+## The three distinctions this dataset exists to keep straight
+
+Three different things are easy to collapse into one. They are not the same:
+
+1. **Recommended ≠ approved.** PCAC is an advisory committee. Its votes are non-binding recommendations to the FDA, not FDA approval, and not an FDA decision of any kind.
+2. **Recommended ≠ legal to compound.** Adding a substance to the 503A bulks list requires a full FDA rulemaking — a proposed rule, a public comment period, and a final rule the agency can decline to issue. That has not happened for any of these seven. Being added would permit compounding under a valid patient-specific prescription; it would still not make the substance an FDA-approved drug.
+3. **Rejected ≠ an FDA final decision.** The committee voted against one peptide. That is also advice, not an agency determination.
+
+FDA career staff scientists recommended against listing all seven. The committee overrode that advice on the six it recommended, and agreed with staff only in rejecting Emideltide (DSIP).
+
+## Outcomes
+
+| Date | Peptide | Outcome |
+|---|---|---|
+| 2026-07-23 | BPC-157 | Recommended, 8-6-1 |
+| 2026-07-23 | KPV | Recommended, 8-6-1 |
+| 2026-07-23 | TB-500 | Recommended, 8-6-1 |
+| 2026-07-23 | MOTS-c | Recommended, 7-5 |
+| 2026-07-24 | Semax | Recommended, 8-5 |
+| 2026-07-24 | Epitalon | Recommended, 7-5-1 |
+| 2026-07-24 | Emideltide (DSIP) | **Rejected, 6-7-1** — the meeting's only rejection |
+
+Six of seven recommended. All votes advisory and non-binding.
+
+## Columns
+
+The CSV carries the 14 columns below. The JSON carries the same row facts plus `slug` (the URL segment for that peptide's page) and `context` (a prose restatement of that row's own fields, introducing no fact not already in the row or its named sources), and expresses citations as a structured `sources` array rather than `source_titles`/`source_urls`. The JSON rows carry no `url` field — use `slug` and construct `https://www.trycora.io/peptides/<slug>/`. Beyond `rows`, the JSON carries top-level `methodology`, `scope`, `unverifiedNote`, `disclaimer`, `changelog`, a `timeline` of the 503A sequence, and `claimChecks` (sourced Q&A). The CSV carries rows only.
+
+| Column | Description |
+|---|---|
+| `peptide` | Substance name |
+| `aka` | Alternate name, where one is in the sources (Emideltide → DSIP) |
+| `url` | That peptide's page on the publisher's site |
+| `pcac_status` | Outcome, vote count, and date |
+| `legal_to_compound_today` | "No — not on the FDA 503A bulks list; no final rule" on every row. Starts with No, but is not a bare `"No"` like `fda_approved_drug` — match on prefix, not equality |
+| `fda_approved_drug` | Uniformly "No" |
+| `evaluated_use` | Substance description and the use the FDA evaluated |
+| `human_clinical_data` | What human evidence could be located in the named sources |
+| `wada_status` | World Anti-Doping Agency status where the sources address it |
+| `evaluation_track` | Which meeting day evaluated it |
+| `notes` | Per-peptide caveats — on the six recommended rows, the vote margin, the FDA staff position, and a shared advisory clause ending "It is not legal to compound today."; on the rejected row, the FDA staff alignment. Vote counts for every row live in `pcac_status`, and `legal_to_compound_today` is the authoritative legality field — No on all seven. |
+| `source_titles`, `source_urls` | Citations for that row |
+| `last_verified` | Date the row was last checked against its sources |
+
+## Methodology
+
+Row facts trace **only** to the sources named on that row — the FDA PCAC meeting notice (the authoritative agenda and evaluated uses), FDA pages on the Section 503A bulks list, and contemporaneous reporting from Reuters and the Regulatory Affairs Professionals Society (RAPS). The FDA staff position is cited **as reported by Reuters and RAPS** — no FDA staff briefing document is among the named sources.
+
+The JSON additionally carries a `timeline` and a set of `claimChecks`, each with its own sources, including FDA — Compounding Laws and Policies. One statement there — that FDA rulemakings of this kind have historically run 12 to 24 months — is **Cora Health's own characterization** of past rulemakings, is attributable to no cited source, and is not a forecast for this rulemaking. No date is attached to the pending FDA decision.
+
+Where a fact could not be located in those sources, the field says so �� `"Not detailed in the sources reviewed"` or `"None found"` — rather than asserting one. Those placeholders are deliberate data, not gaps to be filled by inference. Several of these peptides have **no located human clinical data at all**, and the dataset does not soften that.
+
+This dataset makes no claim that any peptide is safe or effective.
+
+## Changelog
+
+Every correction to a published fact ships with a version bump and an entry here, so anyone holding an older copy can reconcile.
+
+- **v1.4.3 (2026-07-26)** — The "It is not legal to compound today." sentence added in v1.4.1 had reached only the MOTS-c note; all six recommended notes now carry it. Narrowed the methodology statement (the timeline and claim checks carry their own sources, and the 12-to-24-month duration is Cora Health's own characterization attributable to no source). Replaced "the compounded medications Cora does offer" with wording that does not imply Cora is the source. No row, outcome, vote count or date changed.
+- **v1.4.2 (2026-07-26)** — Corrected the methodology statement, which listed "FDA staff materials" among the sources; no FDA staff briefing document is cited on any row, and the FDA staff position traces to the Reuters and RAPS reporting named on each row. No row, outcome, vote count or date changed.
+- **v1.4.1 (2026-07-26)** — Retracted the unsourced descriptor "The only weight-loss-adjacent peptide in the set" from the MOTS-c note, published since v1.0.0 (where it read "the only weight-loss-adjacent peptide in this evaluation set"), replacing it with the sourced "the only peptide in the set whose FDA-evaluated use includes obesity" and stating inline that MOTS-c is not an alternative to a prescribed GLP-1 medication. Removed a duplicated "only rejection" label and vote count from the Emideltide note; corrected singular/plural in the MOTS-c and KPV `evaluated_use` fields. Added a per-peptide `url` column to the CSV (13 → 14 columns) and per-peptide `slug` and `context` fields to the JSON rows. The CSV carries no `slug` or `context` column; the JSON carries no `url` column. (No v1.4.0 was ever published.) Replaced a shared recommended-peptide note with per-peptide notes carrying each peptide's own vote margin and the FDA staff position. No outcome, vote count or date changed.
+- **v1.3.0 (2026-07-26)** — Added a sourced regulatory timeline and a set of straight answers on approval, legality, timing and scope. Also corrected a note stating that some outcomes still read UNVERIFIED; all seven are confirmed.
+- **v1.2.1 (2026-07-25)** — Corrected a statement that every vote went against FDA staff. The Emideltide rejection **agreed** with staff; only the six recommendations went against them. Also removed a comparative characterization of the MOTS-c 7-5 vote as a narrower margin than the 8-6-1 votes — the two-vote gap was the same, only the panel tally differed.
+- **v1.2.0 (2026-07-25)** — Added the confirmed July 24 outcomes.
+- **v1.1.0 (2026-07-24)** — Removed five peptides (LL-37, GHK-Cu, Dihexa, Melanotan-II, PEG-MGF) and the associated evaluation track that described those five as "scheduled for PCAC evaluation approximately February 2027" — a session and date that were **not on the FDA meeting notice and not supported by any source**; the dataset went from 12 rows to the 7 peptides the FDA actually evaluated. Anyone holding a v1.0.0 copy should discard those five rows. The same release resolved the July 23 outcomes: KPV and TB-500 from an unconfirmed "narrow split" to 8-6-1, and MOTS-c from "UNVERIFIED" to 7-5. It also dropped an unsourced comparative characterization of MOTS-c as "the weakest case cited", published in v1.0.0 without a changelog entry at the time.
+- **v1.0.0 (2026-07-23)** — First publication.
+
+## Source and updates
+
+The canonical, continuously-maintained version — including the regulatory timeline and per-peptide pages — is at **https://www.trycora.io/peptide-status-index/**, with machine-readable endpoints at `/peptide-status-index.json` and `/peptide-status-index.csv`. **This Hugging Face copy is a snapshot and may lag the canonical version** — check the canonical endpoints for the current state. The canonical version is updated when the FDA takes another step; the next step is an FDA decision on the recommendations, which carries no date because none exists.
+
+## Citation
+
+```
+Cora Health (2026). Peptide Compounding Status Index — FDA PCAC, July 2026.
+https://www.trycora.io/peptide-status-index/ · CC-BY-4.0
+```
+
+## Corrections
+
+See an error or an update? Email **help@trycora.io** — corrections ship with a version bump and a changelog entry.
+
+## Publisher and disclosures
+
+Published by **Cora Health** (trycora.io), a LegitScript-certified telehealth platform that connects patients with licensed independent providers at Wasef Health, PC for compounded GLP-1 treatment.
+
+**Cora Health does not sell, prescribe, or compound any peptide in this dataset.** Cora does operate a public waitlist at https://www.trycora.io/peptides/ for people who want to be notified if a compliant, provider-guided pathway ever opens — readers should weigh that interest when using this dataset. Cora Health does not prescribe medication at all — licensed providers make every prescribing decision. The compounded medications available through Cora Health are prescribed by licensed providers and prepared by Hallandale Pharmacy (a PCAB-accredited 503A compounding pharmacy) or VialsRx (a US-licensed 503A compounding pharmacy). Compounded medications are not FDA-approved and are not therapeutically equivalent to FDA-approved branded products. Individual results vary.
+
+This dataset is general regulatory reference information. It is not medical or legal advice.