Repurposing · Clinical evidence · Patient scale
N > 1

Every testimonial is N=1. Every Reddit thread on BPC-157 is N=1. Every "it worked for me" is N=1. We are the institution that takes substances out of anecdote and into evidence — rigorous trials, real endpoints, ill patients, public results.

beyond the anecdote
Start a conversation Read the thesis
The thesis

A structural gap in the US drug system — and the technology to close it.

Substances with no patent moat — curcumin, BPC-157, low-dose naltrexone, allicin, NAC, urolithin A, peptides, food-derived bioactives — cannot attract the $300M+ required to prove they work through the traditional FDA pathway. So they stay stuck: sold as supplements with vague claims, or compounded in a regulatory gray zone. Meanwhile, the technology to compress drug discovery, patient matching, and trial operations by 60–80% exists today. The opportunity is to build the full vertical as a single venture-scale company — and use the resulting evidence and patient base to force regulatory reform.

N = 1
  • Anecdote & testimonial
  • Reddit threads, podcast guests
  • "It worked for me"
  • No endpoint, no rigor
>
N > 1
  • Rigorous, powered RCTs
  • Real clinical endpoints
  • Ill patients, published results
  • Regulator-grade evidence
Why now

An 18–30 month window is open. It will close.

Six forces are aligned simultaneously — a configuration that did not exist two years ago and will not persist.

Regulatory window open
FDA Drug Repurposing RFI published May 2026 (Docket FDA-2026-N-4492); comments due June 11.
Aligned principals
Bhattacharya at NIH naming repurposing a priority; MAHA strategy directs FDA + NIH to invest in repurposing.
Legislative momentum
FDA Modernization Act 3.0 (S.355) passed Senate Dec 2025; PCAC peptide vote July 2026.
AI compression proven
Insilico: 18 months and $150K to preclinical (was 4–6 years); Exscientia/Recursion $688M merger validates the platform thesis.
Trial infrastructure maturing
DCT market projected $12B+ by 2030; FDA accepts RWE-based synthetic controls.
Window closing
Leadership in flux; political support softening in competitive districts. The configuration is temporary.
What we build

A five-layer vertically integrated platform.

Each layer attacks one cost or time bottleneck in repurposing. Together they compound.

01
AI Repurposing Discovery Engine
Knowledge graph + LLM + structural biology, ranking candidates by predicted success at specific disease endpoints.
Compresses discovery from 4–6 yr → ~1 yr
02
Patient Matching + Recruitment
EHR + claims mining and opt-in disease registries for national reach.
Recruitment 10× faster than the CRO model
03
Decentralized Trial Operations
One product, one data model: eConsent, randomization, eCRF, ePRO, wearables, telehealth, AE reporting.
Replaces the CRO middle layer · 60–80% lower cost
04
Evidence + Synthetic Controls
RWE-driven control arms, already FDA-accepted in oncology and rare disease.
Reduces required N by 30–50%
05
Patient Access & Delivery
Compounding pharmacy + telehealth dispensing under valid prescription.
The political moat — patients become the constituency
The flywheel: every trial improves the AI · every match improves recruitment · every outcome strengthens the synthetic-control library · every delivered therapy expands the registry. Network effects in clinical research.
Three software products

Built for ourselves first. Licensable once the cost advantage is public.

01

Repurposing Discovery Engine

Knowledge-graph + LLM + structural-biology pipeline tuned specifically for repurposing — not de-novo design — against indications we can actually run trials in.

02

Patient Matching + Recruitment

Opt-in registries across 3–5 indications with EHR/claims integration. Target: 100K patients across registries by end of Year 2.

03

Decentralized Trial OS

One unified system replacing the Medable + Castor + Veeva + Florence integration sponsors pay CROs to operate. Used internally, then licensed.

Trial portfolio

Three classes of candidate, one selection framework.

We don't choose trials by anecdote. We choose across substance class and evidence maturity, so a single null result strengthens the institution instead of cratering it. The first portfolio is built with our founding CSO.

Supplement / nutraceutical / the quick win
low friction
Lowest regulatory exposure — sold today as supplements, so the trial itself carries little legal risk. Agents with real mechanistic rationale and an existing human signal, run as pragmatic-endpoint RCTs.
Why it mattersThe fastest path to a publishable, guideline-relevant result — or a definitive null that still moves the field.
Repurposed generic drug / the medium bet
IIT at AMC
Investigator-initiated trials at academic medical centers: off-patent, established safety, a hard clinical endpoint, an organized patient population.
Why it mattersWhere rigorous evidence directly changes prescribing — on a trial model that's already proven.
Peptide / the high-upside bet
PCAC-gated
The frontier: enormous patient demand, almost no regulator-grade evidence, and a regulatory path that opens with the PCAC ruling.
Why it mattersThe first credible RCT in the category defines it — the highest-leverage evidence we can produce.
Each trial in the $2–10M range against a traditional $300M+ path — the compression is the thesis.
Structure & discipline

Four entities, one mission. Win first, ask permission later — only where it's legal.

A C-Corp PBC holds the platform IP and capital; a wholly-owned pharmacy is the regulator-facing surface; a platform LLC isolates licensable, HIPAA-bound IP; and a paired 501(c)(3)/(c)(4) Foundation runs the credibility flank and lobbies. The structure absorbs regulatory friction without bankrupting the operating company. We hold strict lane discipline:

Push hard

  • IIT-style trials at AMCs on generic drugs
  • Build the DCT platform; use it ourselves
  • Opt-in patient registries with matching
  • Compounded therapeutics under valid Rx (on-bulks-list)
  • Sharp public communication, preprints, op-eds

Gray zone — play smart

  • RWE as synthetic control arms
  • Async telehealth for non-controlled
  • Right-to-Try expansion paths
  • With standing regulatory counsel, always

Never

  • Disease claims on unapproved drugs
  • Sterile compounding without licensure
  • Trials without IRB / IND when required
  • Compounding patent-protected molecules at scale
Capital plan

Seed to a $1B+ company in 5–7 years.

Comparables: Insilico ($1B+), Recursion ($2B+ post-Exscientia), Zoe ($250M), Cost Plus Drugs ($1B+ revenue run rate).

Year 1
$15–25M
Seed + Series A
40–50 FTE
2 trials, 1 registry
Year 2–3
$50–100M
Series B
120–200 FTE
8–12 parallel trials
Year 4–5
$200–500M
Series C+
500+ FTE
Platform licensing
Year 6–7
$1B+
Cash-flow positive or
IPO / strategic exit
Risks, named

Every failure mode in the category has a precedent. So does every mitigation.

Single-donor capture (the NUSI failure mode)
MitigationOpen Science Guidelines from day 1; pre-commit to publishing negatives; diversify funders by Y3.
MAHA capture / crank-adjacency
MitigationExplicit pro-vaccine, pro-RCT masthead; correct factual errors publicly; refuse co-platforming with discredited figures.
Founder-vs-brand civil war (the Cochrane problem)
MitigationSeparate methodology from any individual from day 1; codified dissent rights; co-equal voices.
Commercial COI eroding c3 credibility
MitigationFour-entity wall; pre-published sponsor lists; royalty caps with surplus to mission.
FDA criminal referral on gray-zone tactics
MitigationStrict green/yellow/red lane discipline; standing regulatory counsel; never cross into red.
Get in touch

This is the artifact that begins the work.

If the inequality lit something up for you, reach out. We're having the first conversations now.

> Potential co-founders — CSO, regulatory, clinical ops, comms, policy
> Board members & advisors — nonprofit governance, platform economics, PBC structure
> Investors — pre-seed and Series A
Message @pwidden on X
DMs open · @pwidden on X