Clinical development · Responsible AI
A methodological second opinion on your trial protocol. Every finding comes with the protocol passage, the page and the standard it is measured against.
The problem
Your protocol is written by many hands, under time pressure, and your team rarely has the capacity for an independent methodological read. Vague estimands, unaddressed intercurrent events or optimistic recruitment assumptions then surface at the site or with the regulator, when the fix is an amendment. Nearly 60% of protocols need at least one substantial amendment, each adding months, and sponsors themselves judge almost half of them avoidable.1
Your team may already use an AI assistant, and it helps. But an assistant answers the questions you ask. Amendments come from the questions nobody asked. A protocol review needs a defined method that covers every relevant point for your indication and phase, every time, shows the passage and the standard behind each finding, and has a person who stands behind the result.
DrFC brings responsible AI into clinical development, with people making the decisions. Our review applies a versioned checklist for your indication and phase, benchmarks your design against comparable public trials, and links every finding to the protocol passage, the page and the standard. Every run is recorded. In our expert packages, named specialists confirm or withdraw each finding.
1 Tufts Center for the Study of Drug Development; Getz KA et al. The impact of protocol amendments on clinical trial performance and cost. Ther Innov Regul Sci 2016;50(4):436–441. ↩
Protocol review
Our Protocol Review Assistant (PRA) finds the issues. In the expert packages, experts in biostatistics and medicine decide which ones stand.
Verbatim protocol passage, page and the standard it is measured against. Where a page cannot be verified, the finding says so.
Critical, major or minor by the effect on safety, endpoints, conduct and acceptability. Editorial points stay information.
In the expert packages, named experts turn each PRA hypothesis into a confirmed verdict or withdraw it with a reason.
Eligibility, drop-out and recruitment against comparable trials on ClinicalTrials.gov, counted the same way, with the method disclosed.
The reviewed file is identified by its SHA-256 checksum, result records are locked after each run, and checklists change only through a documented review and sign-off.
Dedicated check packs per indication, and every protocol is measured against phase-appropriate standards.
Three axes, applied to the whole protocol.
Available: gastroenterology (ulcerative colitis, Crohn’s disease); allergy and immunotherapy (respiratory, food and venom allergy).
In preparation: oncology.
Other indications: core review plus indication-specific guidelines by agreement.
Packages
Every package runs the PRA on your full protocol; you choose the expert review on top. One fixed price per protocol and a fixed scope. Prices and delivery times on request.
A fast, evidenced overview of where your protocol is methodologically exposed. Hypotheses for your team, not yet confirmed verdicts.
Verdicts instead of hypotheses: an expert-reviewed findings set you can base decisions on and share onwards.
For protocols where an independent expert judgement makes the difference.
The two expert packages are currently offered for allergy and immunotherapy protocols. For other indications: PRA Core, or talk to us about the reviewer match.
How it works
Under a confidentiality agreement, with the documents the review needs. We confirm scope and start date.
Against the check pack for your indication and phase, section by section.
In the expert packages, reviewers confirm or withdraw each finding.
Report and Excel workbook with a column for your answers. On your revision, a delta report.
Responsible AI
The AI proposes. In the expert packages, named experts confirm each finding or withdraw it with a reason. In PRA Core, findings stay labelled as unconfirmed.
No finding without the protocol passage and the standard behind it. What could not be verified is marked as such.
Versioned checklists, a checksum for the reviewed file, results locked after each run.
Unconfirmed findings are labelled as such, and we say plainly what the review is not.
About
Dr. Christoph Franz, MBA
Founder & Managing Director
I founded DrFC to give sponsors a second methodological opinion before a protocol goes out, not after the first amendment.
I am an auditory neuroscientist by training, with a PhD (Dr. rer. nat.) in Biology and an MBA, and have worked in clinical research since 2013: first as In-House CRA (Project Management Associate) on a Phase III trial, then in business development for international CROs.
DrFC combines that experience with a review assistant built for traceability: every finding evidenced, and in the expert packages every verdict made by a person.
Tell us the indication, phase and timeline. We reply with scope and a start date.
Heddesheim, Germany