Post-Frey recovery · live window 7/22 → 8/1 · canonical IF returns ~Aug 1
Kunitz and Bowman-Birk are serine-protease inhibitors — trypsin and chymotrypsin only. They do not inhibit lipase. Creon is dosed in USP lipase units and titrated on fat absorption, so the ceiling of any soy effect is the protease fraction. "Soy stops Creon working" is retired. No dose amount changed. No slot moved. Nothing added or reinstated.
DAO (diamine oxidase) is the histamine-degrading enzyme. Its cofactors are copper · calcium · topaquinone (TPQ) — not B6. Ranked for this phenotype, the real levers are:
1. Substrate competition — Agmatine 1 g/day is a confirmed DAO substrate (moved to 4 PM to clear the NaturDAO collision). · 2. Mucosal integrity — DAO is made by gut-lining cells; post-Frey + ulcers + EPI is the likely suppressor. Intervention already live: L-Glutamine 15 g/day. · 3. Copper — the actual metal cofactor, now added to the 8/1 draw.
⚠ B6 is not a DAO lever — it's upstream on histamine synthesis (histidine decarboxylase). Do not switch to NaturDAO Plus: it bundles P5P on the same retired rationale.
Acetaminophen — documented adverse, VM July 2026 · NSAIDs — ulcer / GI-bleed history · Potassium supplements — on the ARB
⚠ Soy has moved out of this tier at rev 7.21 — it is now activity-gated, not categorical. See below.
Exclude: raw / unheated / high-TIA soy — raw soy flour, raw soybeans, unheated soy protein powders. · Cleared: soy lecithin · soybean oil · fermented soy · hydrolysed soy protein / soy peptone · a declared trace from a fermentation medium. SerinAid PS, NaturDAO and Erinamax all sit here — not as exceptions, as the general case.
Ceiling of the effect. Kunitz and Bowman-Birk are serine-protease inhibitors — trypsin and chymotrypsin only. Not lipase. Not amylase. Creon is dosed in USP lipase units and titrated on fat absorption. "Soy stops Creon working" overstates the ceiling by the whole therapy and is retired.
Why a constraint survives at all — and it is not enzyme inactivation: a large active STI load removes luminal-protease feedback and raises CCK → pancreatic stimulation. Wrong direction for a post-Frey pancreas. Requires raw-flour-scale exposure.
Yardstick — use this, not the word "soy": NaturDAO ~96 mg/day legume protein = cleared. Erinamax ≈5 µg/day upper bound = ~19,000× below that.
🏷 Provenance ⟨?⟩ — untagged in every rev, no source anywhere in the document, no soy interaction on the Creon FDA label, and none of the cited literature touches PERT, EPI or fat absorption. It was the only Standing Constraint that explained itself instead of citing something. Origin unrecalled — not assumed in either direction.
Dark or tarry stool · visible blood · new dizziness · resting HR climbing · fever or rigors · any BP reading ≥180/120 → urgent care, not a message
The regimen produces atypically suppressed inflammatory markers (WBC, CRP). Clinical suspicion outweighs standard lab thresholds when assessing acute infection, cholangitis, or pancreatitis flare. A normal white count does not exclude infection in this patient.
Adversarial passes have lodged zero findings against the chronic core. Not one of the ~40 compounds was faulted on evidence quality, dose, timing, interaction, or sourcing — outside the post-surgical window. What gets corrected are the overlays, the instruments, and the reasoning — all of which sit around the regimen, not in it. This rev is one more of those: a stated rationale was wrong; the compound stayed.
The record: GMI 5.7–5.8% across 180 days · zero readings above 250 · eGFR 112 · normal hepatic function despite an obstruction history · HbA1c 6.1% in T3c diabetes, without insulin.