An eCTD viewer sounds like a solved problem — the submission is just folders, XML, and PDFs, so surely anything that renders a tree and opens files will do. That intuition is why so many teams start with a free viewer and quietly abandon it within a year. The gap between ‘opens a sequence’ and ‘supports how regulatory teams actually review submissions’ is wide, and most of it only becomes visible once real dossiers, real file sizes, and real FDA forms show up. This guide covers what to evaluate, roughly in the order the pain arrives.
Level 1: Rendering a Sequence
Every viewer handles the basics: parse the index.xml backbone, render the CTD tree, open leaf documents, follow the DTD structure. If you’re validating a single sequence you just published, a lightweight local viewer is genuinely fine. The eCTD structure was designed to be navigable, and it is.
Level 2: Lifecycle — Where Free Viewers Stop
Real review doesn’t happen against one sequence. It happens against an application with dozens or hundreds of sequences, where every document has a lifecycle: new, replace, append, delete. The questions reviewers actually ask are lifecycle questions:
- Cumulative view: show me everything ever submitted, organized by section, with superseded versions visible in their history.
- Current view: show me the dossier as the agency sees it today — replaced and deleted leaves resolved out.
- Version chains: for this document, show me every predecessor and what operation connected them.
Computing these views correctly across hundreds of sequences — including messy realities like imported legacy sequences and cross-referenced applications — is where viewer quality separates. Get it wrong and the ‘current view’ silently shows a superseded specification; the reviewer trusts the tool and reviews the wrong document. Accuracy here is a compliance property, not a convenience.
Level 3: The Problems Nobody Mentions in Demos
FDA forms that won’t render
US Module 1 is full of XFA forms — 1571, 356h, 2253, 3674 — built on Adobe’s LiveCycle technology. Open one in a stock browser PDF viewer or most embedded viewers and you get a single page reading ‘Please wait… If this message is not eventually replaced by the proper contents of the document, your PDF viewer may not be able to display this type of document.’ A serious review platform needs an XFA-capable rendering engine, including dynamic forms whose tables grow rows with the data. Test this specifically: bring a completed 356h to the demo.
Multi-hundred-megabyte files
Datasets, scanned legacy documents, and imaging-heavy files routinely run into hundreds of megabytes. A viewer that downloads whole files before rendering makes reviewers wait minutes; streaming with HTTP range support makes the same file open in seconds. Test with your largest real file, not the demo dossier.
Cross-dossier search
‘Where did we discuss this impurity across all our applications?’ is a real question with regulatory consequences. Answering it requires indexed full-text search across dossiers — increasingly with semantic search layered on — not per-PDF Ctrl+F.
Working documents, not just PDFs
Review increasingly touches source Word documents too. Viewing DOCX faithfully in the browser (via high-quality PDF rendition) saves the download-open-close loop hundreds of times per review.
Comparing Your Options
| Capability | Free/desktop viewers | Enterprise review platforms |
|---|---|---|
| Single-sequence rendering | Yes | Yes |
| Cumulative / current lifecycle views | Limited or manual | Core feature |
| XFA FDA form rendering | Usually broken | Varies — test it |
| Large-file streaming | No | Should be standard |
| Cross-dossier search | No | Yes, increasingly semantic |
| Team access, permissions, audit | Per-desktop installs | Browser-based, tenant-scoped, audited |
How DnXT Approaches Review
DnXT Reviewer is the review side of the DnXT platform: browser-based, read-only by design, and built for the whole-application questions above. It renders cumulative and current views resolved across full sequence histories (including imported legacy sequences), streams large files with range support, renders dynamic XFA FDA forms fully — including forms that defeat stock PDF.js viewers — and provides cross-dossier search with AI-assisted document suggestions, version diff, and a date-synced submission chronology. Because it shares a platform with DnXT Publisher, what reviewers see is computed from the same structures the publisher wrote — not a parallel interpretation. You can try it directly via our interactive Reviewer demo.
Frequently Asked Questions
What is an eCTD viewer?
Software that renders eCTD submissions — the XML backbone, CTD section tree, and leaf documents — for human review, ideally resolving lifecycle operations into cumulative and current views across an application’s full sequence history.
Are free eCTD viewers good enough?
For spot-checking a single published sequence, often yes. For ongoing team review across applications — lifecycle views, big files, FDA forms, search, access control — they run out of road quickly.
Why won’t FDA forms open in my PDF viewer?
They’re XFA (LiveCycle) forms: the PDF is a container for an XML form definition that most viewers, including browser-native ones, don’t execute. You need a viewer with an XFA-capable engine — or every Module 1 form becomes a download-and-open-in-Acrobat exception.
Viewer or full review platform — how do we decide?
Count the people who need to look at submissions: regulatory, quality, clinical, partners, consultants. Past a handful of users, browser access with permissions, audit trails, and shared views beats a per-desktop tool on both cost and compliance.
The fastest evaluation is looking at a real dossier: book a session and bring your hardest sequence — the one with the 400 MB file and the 356h.
