# Tolera Bio — Regulatory Strategy
## An antigen-specific tolerance vaccine for eosinophilic esophagitis, delivered N-of-1

*Confidential — for VC / VP discussion. Strategy brief, v1. Not legal or regulatory advice; a formal path requires FDA interaction (INTERACT/pre-IND) and regulatory counsel.*

---

## 1. Regulatory identity of the product

Tolera's product (working name **TOL-EoE**) is a **per-patient-manufactured biologic**: a single-chain pMHC-II construct on a shared, qualified backbone, conjugated to an iron-oxide nanoparticle, in which the patient's own HLA-II allele domain and identified food-epitope cassette are the individualized elements. Regulatorily this places TOL-EoE in the **individualized-therapeutic** class alongside individualized neoantigen cancer vaccines — a class where the FDA has shifted from a "one molecule, one label" review to treating **the manufacturing/selection *process* as the regulated entity**, with per-batch release criteria, sterility testing, and chain-of-identity tracking. The company's core regulatory task is therefore to qualify the **platform and its CMC control strategy** once, and file individual products against it.

**CBER** (Center for Biologics Evaluation and Research), likely the Office of Therapeutic Products, is the expected review division (therapeutic biologic / cell-and-gene-adjacent individualized product).

---

## 2. The two-layer regulatory architecture (platform + product)

| Layer | What is qualified | How | Precedent |
|---|---|---|---|
| **Platform layer** | Shared MHC-II backbone, NP conjugation chemistry, the *epitope-selection algorithm*, and the individualized CMC control strategy (parametric/real-time release, chain-of-identity) | One IND built on platform data; FDA may wish to **inspect the selection algorithm** (as with AI-driven neoantigen selection) | Individualized neoantigen vaccines; FDA's process-as-product approach |
| **Product layer** | Each patient's specific {HLA × cassette} construct | Manufactured against the qualified platform; released on pre-validated criteria rather than de-novo review each time | Autologous cell therapy per-lot release |

**Strategic implication for the pitch:** we do *not* file a new IND per patient. We file **one platform IND**, and the personalization tiers (A/B/C from the platform spec) map to how much incremental characterization a given patient's components need — not to separate regulatory submissions.

---

## 3. Expedited-pathway eligibility

EoE with no approved *tolerance-inducing* therapy is a strong fit for all four FDA expedited tools:

1. **Orphan Drug Designation (ODD).** US EoE prevalence (~167,500; ~100,500 diagnosed) is well under the 200,000-patient threshold. ODD → 7-year market exclusivity, fee waivers, tax credits. *Highest-priority near-term filing — cheap, fast, de-risks the exclusivity story for investors.* (Teplizumab carried Orphan designation into its T1D approval.)
2. **Fast Track.** Serious disease + unmet need (no antigen-specific/curative option) → rolling review, more frequent FDA interaction.
3. **Breakthrough Therapy Designation (BTD).** Available once early clinical data show substantial improvement over available therapy on a meaningful endpoint. Teplizumab and mRNA-4157 both hold BTD — direct precedent for both the tolerance mechanism and the individualized modality. Target after Phase 1/early Phase 2 tolerance signal.
4. **RMAT (Regenerative Medicine Advanced Therapy).** Worth testing eligibility: if TOL-EoE's mechanism (cell-based tolerance induction / Tr1 reprogramming) qualifies it as a regenerative medicine therapy, RMAT confers the BTD benefits plus flexible post-approval evidence generation. Analysis item for regulatory counsel.

---

## 4. Disease-modification endpoint precedent (why the FDA will engage)

Teplizumab/Tzield is the pivotal precedent: the FDA approved a T-cell-directed immunomodulator on a **disease-trajectory** endpoint (delaying progression), establishing it as the **first disease-modifying medicine in autoimmune T1D** — not symptom palliation. That validates Tolera's core regulatory thesis: **the agency will approve an antigen-directed immune intervention whose value is changing disease course, provided the endpoint is objective and well-controlled.** For EoE the established objective endpoints are already in hand from the approved-drug era:
- **Histologic:** peak esophageal eosinophil count (≤6 eos/hpf remission threshold) — the endpoint on which dupilumab and budesonide were approved.
- **Endoscopic:** EoE Endoscopic Reference Score (EREFS) — note the field's caution that mechanistic biomarker changes must translate to EREFS (a recent pipeline agent was discontinued after failing to move endoscopic scores despite a mechanistic effect).
- **Symptomatic:** Dysphagia Symptom Questionnaire (DSQ).
- **Tolera-specific / disease-modification:** durable histologic remission **after cessation of dosing** and antigen-specific readouts (peTH2 clone contraction, tetramer⁺ frequency, IL-10 induction) — the differentiated endpoint that makes the "one-course, durable" claim.

---

## 5. Companion diagnostic (CDx) strategy

Stage 3 of PACT — the blood-based food-trigger + clonotype assay — is a **companion diagnostic**: it selects patients and defines each patient's cassette. Two regulatory consequences:
- **Co-development:** the CDx should be co-developed and co-submitted under the FDA's drug/device CDx framework (CDRH + CBER coordination), since dosing depends on its output.
- **Standalone value:** the assay can pursue its own clearance as a diagnostic that names the causal food from a blood draw — a revenue-generating, pipeline-independent product (see market model).
- **IP/FTO flag:** the functional T-cell-assay concept for identifying EoE-causal foods is **patented (Hill/Spergel)**. Tolera's freedom-to-operate on the diagnostic front end requires a **license or design-around**; this is a named diligence and BD item, not a hidden risk.

---

## 6. CMC by personalization tier (the hardest individualized-product problem)

The central individualized-product challenge is that **each product is a unique batch**. Tolera's shared-backbone design is the mitigation: comparability is anchored to a qualified backbone, and only the cassette/allele varies.

| Tier | CMC treatment | Release |
|---|---|---|
| **A — validated cassette** | Full comparability to qualified reference lot | Parametric / real-time release on pre-validated criteria |
| **B — computational-prior cassette** | Backbone comparability + cassette-specific identity/potency + **ex-vivo functional gate** | Release + functional confirmation before dosing |
| **C — rare HLA / novel epitope** | New backbone-allele characterization added to platform file | Full characterization workup; feeds back into the qualified set |

Control-strategy pillars mirrored from individualized-vaccine practice: **chain-of-identity / chain-of-custody** from biopsy to dose, **individualized release criteria**, potency assay tied to MHC-peptide presentation, and a validated selection algorithm the FDA can audit.

---

## 7. IND-enabling package (mapped from the existing preclinical roadmap)

The project's existing 5-phase preclinical roadmap already supplies the IND spine. Regulatory mapping:

| Roadmap phase | IND-enabling module | Regulatory purpose |
|---|---|---|
| 1. Human functional validation | Pharmacology / proof-of-mechanism | Confirms antigen-specific Tr1 induction; converts priors → validated cassettes |
| 2. Murine proof-of-concept | Primary pharmacology (efficacy models) | Disease-relevant activity |
| 3. Toxicology & biodistribution | GLP tox, biodistribution, immunogenicity | Safety for FIH; NP biodistribution is a specific review focus |
| 4. GMP manufacturing | CMC / Module 3 | The platform control strategy above |
| 5. Regulatory submission prep | Pre-IND + IND | INTERACT → pre-IND → IND |

**FDA interaction sequence:** **INTERACT** meeting (early, novel-modality advice on the platform-vs-product framing and CDx co-development) → **pre-IND** (endpoint + CMC control-strategy alignment) → **IND** → **FIH**.

---

## 8. Phase-by-phase regulatory timeline

| Milestone | Timing (from seed close) | Regulatory content | Value / de-risking |
|---|---|---|---|
| **Orphan Drug Designation filing** | Month 0–3 | ODD application | Exclusivity + fee/tax benefits; cheap early credibility |
| **INTERACT meeting** | Month 6–9 | Platform/product framing, CDx, algorithm audit | Confirms the regulatory architecture is viable |
| **Human functional validation done** | Month 9–15 | Proof-of-mechanism data | Validates Tier-A cassettes; supports BTD case |
| **Pre-IND meeting** | Month 15–20 | Endpoints + CMC control strategy | FDA alignment before spend on GLP/GMP |
| **GLP tox + GMP lots** | Month 18–30 | Module 3 + tox | The costliest gate; enables IND |
| **IND submission** | Month 24–33 | Full platform IND | Clears FIH; major value inflection |
| **First-in-human dosing (Phase 1)** | Month 30–39 | Safety + antigen-specific PD | Human safety + mechanism |
| **Fast Track / BTD filing** | at Phase 1 signal | Expedited designation | Accelerates and validates |
| **Phase 2 PoC (durable remission)** | Month 45–60+ | Histologic remission off-therapy | The "durable tolerance" claim; Series B/partnering inflection |

*Timelines are targets for an individualized biologic and will move with FDA interactions and manufacturing readiness. Ranges reflect the platform-IND (not per-patient-IND) strategy.*

---

## 9. Top regulatory risks & mitigations

| Risk | Mitigation |
|---|---|
| Individualized-batch CMC burden | Shared-backbone comparability; parametric release; tiered characterization |
| Endpoint translation (mechanism ≠ EREFS) | Power on histologic remission (validated approval endpoint); treat antigen-specific PD as supportive, not primary |
| FTO on the food-trigger assay (Hill/Spergel patent) | License or design-around; named BD diligence item |
| Autoimmunity / off-target presentation | Epitope selection excludes self-cross-reactive/leader sequences; tolerogenic (not immunogenic) mechanism; GLP immunogenicity |
| Novel-modality review uncertainty | Early INTERACT; lean on neoantigen-vaccine + teplizumab precedents |

*All non-dairy epitope binding values are computational priors pending wet-lab validation; the dairy anchor (β-casein aa59–78, DRB1*07:01) is tetramer-validated. TOL-EoE is a preclinical-stage design.*