Publishers know the pattern: the sequence comes together in the last week, validation runs the night before the deadline, and the final day goes on repair. Good publishing software changes when the work happens. Here is what that looks like, step by step.
Key takeaways
- Envelope values should come from data, not typing.
- Lifecycle should be chosen from the current view, not from a file path.
- Validation should run while you build, at the agency’s severity.
- Software can suggest; a person confirms.
The problem with the last week
Most technical rejections are mechanical: an envelope value typed by hand, a replace that points at the wrong document, a PDF with an unembedded font, a link that opens the right file on the wrong page. None is hard to fix. They are expensive because they are found late, when the sequence is assembled and validated for the first time. Good tools move each check to the moment the problem is created.
Here is how a sequence comes together in DnXT Publisher, and why each step is designed the way it is.
1. Open the dossier, not a blank form
Everything starts from the dossier for a product and an authority. Its application details — applicant, application number, products and substances, contacts — are held once and used everywhere. When you create a sequence, its envelope is built from that record and from the region’s own controlled vocabulary. You choose a submission type from a list the agency accepts; you do not type it. Envelope errors are the most common high-severity finding, and this removes their most common cause.
Filing the same product in a new region? Clone the dossier to carry products, substances and applicant across. Module 1 is specific to each region, so it is reviewed before anything is published — a clone is a starting point, not a submission.
2. Bring documents in from where they live
Documents come from DnXT eDMS, SharePoint, Box or Egnyte through the Content Manager, or by upload. Document QC checks each one as it arrives — PDF properties, fonts, bookmarks, Fast Web View — and Auto Link Verification checks every hyperlink against where it actually lands. Problems are found while the author can still fix the source.
3. Place documents and choose lifecycle
Auto-Assign makes a first pass at placing documents in their CTD sections, and a publisher confirms each one. For documents that change something already filed, the operation — new, replace, append or delete — is chosen against the application’s current view, so you pick the document being replaced from what is actually in force. Cumulative and current views sit side by side.
Suggestions are made by software. Decisions are made by people.
4. Validate as you go, at the agency’s severity
Validation runs the target region’s own criteria and grades each finding the way that agency grades it. A high-severity FDA finding blocks publishing until it is fixed; a best-practice EU finding is shown but does not. Every run is kept, and each one can be compared with the last, so you see what is new, what is resolved and what has not moved.
5. Let the sponsor confirm the plan
For publishing partners and CROs, the sponsor can review the plan for the sequence through a secure link: each document, its section and whether it is new or replaces something already filed. They confirm or flag each row and check the application details on file. Their notes become the change, and the published sequence is delivered through the same portal.
6. Publish the package and its evidence
The output is a standard eCTD sequence with its validation report. Afterwards you can show what was in force at any point in the application’s life, with status and validation history.
Covered regions
The same workflow covers 16 eCTD regions — including the US, EU, Japan, Canada, Australia, Switzerland, the GCC and China — in eCTD 3.2.2 and 4.0, with more than 2,500 region-specific validation criteria.
Try it on your own documents
Book a walkthrough and build a sequence from your own content.
