Sponsors, CROs, and industry providers learn the pros and cons of switching to the electronic common technical document.
By Ann Neuer
May 19, 2009 | Consider the enticing prospect of eliminating the blizzard of paper associated with regulatory submissions through a process in which applications are built once from the beginning, and then managed throughout their lifecycle. Imagine enhancing the quality and speed of those submissions, while establishing a history of the submission in an easily searchable form.
“Anything is better than paper,” says Nancy Smerkanich, VP Global Regulatory Affairs for Octagon Research, who sums up a widely held opinion in the industry. “Just being able to pull up a 3-year-old investigational new drug (IND) application and look at it in a current view as well as a historical view with a couple of clicks is an enormous advance over spending hours in a file room looking through boxes, pulling out volume after volume—in some cases, hundreds of volumes.”
The key to this significant change is the electronic common technical document (eCTD), which represents a major advance in the march toward electronic regulatory submissions across the globe. Put forth by the International Conference on Harmonisation (ICH) back in 2003, the eCTD specifies how electronic submissions are to be created, reviewed, and archived.
Following the 2008 mandate by the Center for Drug Evaluation and Research (CDER), a division of FDA, Bio•IT World set out to gauge the reaction of the major stakeholders—pharmaceutical companies, government, and industry providers—to the rising tide of eCTD adoption, and assess the benefits and challenges in making eCTD adoption a successful effort. Users are finding that creating the infrastructure to participate in the eCTD world is complicated, requiring a commitment to process change and long-range strategic planning. Regulatory bodies are seeing frequent errors in submissions that can hinder and possibly delay review, although many of these are preventable.
Looking ahead to future trends, industry insiders anticipate an impact of the Regulated Product Submission standard (RPS) on the eCTD, just as it is gaining traction. RPS is a Health Level 7 standard designed to facilitate processing and review of all regulated products, and within several years, eCTD is expected to be folded into that standard.
On the March
Long burdened with volumes, even truckloads, of paper submissions that must be created, shipped, reviewed and archived, the drug industry is actively embracing the eCTD, signaling a new paper-free era for the regulatory submission process. North America, Europe, and Japan are experiencing continual growth, and other parts of the world, notably Australia, which issued a draft guidance in January, are hoping to follow suit.
On January 1, 2008, CDER mandated that electronic submissions had to use the eCTD format. That decision sparked a sharp jump in the cumulative number of eCTD submissions, which tripled in a single year, from nearly 15,000 (October 2007) to more than 43,000 by October 2008. The number of total applications soared tenfold over the two years from October 2006 (390) to October 2008 (4,051). An application is defined as an investigational new drug application (IND), a new drug application (NDA), an amended NDA (ANDA), a biologic license application (BLA), or a drug master file (DMF).
Europe is also seeing strong growth in eCTD submissions in advance of the European Medicine’s Agency’s) (EMEA) recommendation for eCTD-format electronic-only submissions starting July 1. According to a survey conducted by the Telematics Implementation Group for Electronic Submission, a group of representatives from European Union National Competent Authorities and EMEA, the number of eCTD submissions spiked almost 50% in a six-month period spanning 2007 and 2008. The volume of eCTD submissions grew from 1,435 in the second half of 2007 to 2,122 by the end of the first half of 2008.
By January 2010, the agency will mandate eCTD format for all electronic submissions using the Centralised Procedure, which is a method for medicinal products seeking approval for use in all EU countries, and applies to all applications and all submission types.
Pharma sponsors and outsourced providers are driving the expanding eCTD marketplace. Terri Booth-Genthe, senior director of global regulatory submission management for Wyeth Research, says the company was an early eCTD adopter, dating back to a 2003 submission to FDA. Since then, Wyeth has submitted nearly 5,000 sequences, which are multiple submissions in support of a drug in clinical trials, mostly to FDA, but also to EMEA and Health Canada. Last year, Wyeth submitted a total of 2079 eCTDs, a 6% increase over the previous year.
Currently, Wyeth submits everything in eCTD that FDA is able to accept. Booth-Genthe explains that in 2007, Wyeth set about converting all of its active NDAs, INDs, DMFs and BLAs into the eCTD format. “By the end of 2007, all of our active U.S. files were in eCTD.” The following year, Wyeth converted all of its European marketing authorisation applications (MAA) to eCTD format. That effort is paying dividends. Currently, there is just a 1-4 week lag between Wyeth's U.S. and EU submissions; prior to the eCTD initiative, Booth-Genthe says the gap between submissions was often 6-12 weeks.
GlaxoSmithKline has also embraced the eCTD, and prepares them using three different publishing tools—one for the United States, one for Europe and another for Canada and Japan—but is in the process of transitioning to a single global tool. Andrew Marr, Director Global eRegulatory Development, comments that in the U.S., over the past 18 months, Glaxo has transitioned all of its NDAs to the eCTD, bringing the total to approximately 140. In Europe, the company has submitted 28 MAAs in eCTD format to the European Medicines Agency through the Centralised Procedure. The company’s current focus is on expanding the use of the eCTD to include the IND process. “We’ll be piloting some INDs into CDER, trying to get the process right. We have already submitted a handful of eINDs into the Center for Biologics Evaluation and Research (CBER),” says Marr.
Small to mid-size players that lack the volume or infrastructure to justify developing internal capability often outsource this function to providers with established expertise in eCTD preparation and submission. Nancy Smerkanich of Octagon Research says the company’s eCTD volume has tripled since 2003-2004. “We have 80 active eCTDs that we have filed and maintained, and we recently celebrated our 1000th sequence going through the electronic submissions gateway at FDA. The majority of those 80 applications are for small pharma or those in the biotech area.”
Similarly, Apyx is seeing a significant uptick in its eCTD business. According to president and CEO Ken VanLuvanee, eCTD represented less than 20% of revenues a couple of years ago, but that has now risen to 75% to 80%. “In approximately three years, when Regulated Product Submission (RPS) becomes a consideration, Apyx anticipates that 65% to 70% of company business will be electronic publishing, and of that, 90% will be in eCTD format,” VanLuvanee explains. Approximately 70% of Apyx’s eCTD submissions go to FDA, 25% to EMEA, and 5% to Health Canada.
Benefits Come With Challenges
Wyeth’s Booth-Genthe sees the eCTD as providing major benefits beyond its role as an electronic submission tool. The information it contains can also bring substantial efficiencies to the internal organization. “More of our authors and regulatory liaisons are depending on eCTD files to give them background and they are starting to rely on these files as a reference,” she says. “They can easily go back and review individual components and their lifecycle. This adds significant value as many people in the quality organizations can now easily see changes in specifications we’ve made over time.”
But reaching the point that an eCTD becomes integral to the operation—both as a submissions method and as a reference tool—requires an understanding of how a compliant eCTD is structured and the process changes needed for successful implementation. Presently, many organizations are submitting improper eCTDs, either because they failed to learn the intricacies of what constitutes a validated eCTD, or have not adopted the necessary process changes. FDA can reject improper eCTD submissions for technical reasons, meaning that the clock for review does not begin, thereby delaying approval. When this happens, the submitter is notified to make corrections.
Antoinette Azevedo, president and CEO of e-SubmissionsSolutions.com, says a key barrier to eCTD acceptance is companies’ lack of budget, followed closely by non-compliant content preparation. The three most common errors she sees are: invalid XML content; inability to navigate the PDF files that constitute the narrative part of the submission; and datasets submitted in invalid format. These problems are slowing the adoption of eCTD and hampering regulatory review. But Azevedo attributes these errors to fixable rookie mistakes and to the fact that the system is still flooded with paper. “The challenge is that because of the paper, there is a lot of legacy content, and there are some people who think that the best way to get to eCTD format is through scanning,” she says.
Much of what Azevedo says is confirmed by FDA representatives, who peg scanned documents as particularly frustrating to reviewers, as they are not searchable and are often illegible. Other common errors are a missing or inaccurate table of contents, inoperable hyperlinks, and missing files (see "Common Errors in eCTDs").
According to Specifications for eCTD Validation Criteria, an FDA publication, incoming eCTDs receive a programmatic validation step that checks the backbone against more than 100 rules. Errors are rated as high, medium, low, or ignore, which determines the level of reviewability. A “high” error would be missing files, the lack of the study report, protocol, and methods validation, without which the review cannot proceed. Similarly, if the submission does not have a US-regional.xml file, which identifies application type, it is impossible to process it. In July 2008, EMEA established its own validation criteria using 44 rules, rating violations as serious, medium, or low.
While submitters have the opportunity to make corrections, a better route is to avoid these problems altogether. Improving performance requires process changes. Booth-Genthe says that Wyeth identified its first challenge as getting documents from the authoring department in an acceptable format. This required ensuring the documents used the correct font, had appropriate hyperlinks and margins, and had the granularity that eCTD requires.
Granularity refers to the level of the smallest file unit or building block within the eCTD. As a highly defined structure, the eCTD includes a hierarchy of files and folders as laid out in the table of contents. The idea of granularity is to avoid combining several documents that violate the eCTD structure, confuse reviewers, and hinder life cycle management down the road.
“There was some resistance in the chemistry, manufacturing, and control (CMC) organization to break down the documents to that granular level,” says Booth-Genthe. “There was a very long-standing way of writing those sections, and we had to change that. We had to change the templates, the guidances, and the SOPs. It was a gradual process.”
As companies begin implementing the necessary changes, there is optimism that the quality of submissions is starting to improve. Errors still abound, but there is growing awareness among trial sponsors that electronic submissions cannot be reviewed if they are of poor quality.
Virginia Ventura, regulatory information specialist in CDER’s Office of Business Process Support, has noticed improving submission quality because more industry representatives are sending email questions to the ESub support staff (email: firstname.lastname@example.org), submitting sample eCTDs, keeping abreast of updated specifications, and reviewing recent FDA eCTD presentations. “Documents are being placed in the correct or more appropriate locations within the eCTD, eCTD metadata is accurate most of the time, and leaf titles are more descriptive in the eCTD tree as compared to previous submissions,” Ventura remarks. Sponsors who obtain clarification from the Esub Support Team or ensure eCTD format is included in the pre-NDA or pre-IND meetings have a higher incidence of making successful submissions.
Get Ready for RPS
As sponsors cope with the best way to expand eCTD submissions, they will soon be confronting what many see as the next big trend—implementation of the Regulated Product Submission (RPS) standard. The objective of RPS is to define a single message structure that can be used globally for all regulated products, including biopharmaceuticals, medical devices, veterinary medicines, and food.
RPS is an HL7 standard that offers many of the same advantages of the eCTD plus additional strengths. It offers two-way communication between the submitter and the regulatory agency, whereas eCTD is a one-way communication tool. In addition, with RPS, previously submitted documents will not have to be resubmitted in marketing applications, as it will be possible to apply previously submitted documents to the marketing application. This will spare submitters from having to retest hyperlinks and bookmarks because the agency will already have those documents. Lifecycle management will also improve with RPS as it will be possible to change granularity and move documents, steps that are not possible with eCTD.
Jason Rock, CTO of GlobalSubmit explains how RPS will impact eCTD. “The next major version of eCTD is going to transition to the RPS standard. The same content submitted in eCTD, such as the PDF documents and the SAS datasets, will be submitted in RPS, but the inner workings of XML will be completely changed.”
According to Rock, by September 2012, FDA must have RPS implemented because it has Prescription Drug User Fee Act (PDUFA) commitments for two-way communication, and is planning on using RPS to meet those needs. The next milestone is January 2010, the date by which standards are to be in Draft Standard for Trial Use (DSTU) form, or ready for implementation on a trial basis. FDA is slated to develop an RPS Implementation Guide by the end of 2010, and expects to be able to accept RPS Release 2 submissions in spring 2011.
Although RPS is looming, eCTD adoption should continue to rise, especially as more regulatory agencies build an infrastructure to accept them and begin encouraging or mandating their use. Anticipated growth is driven by the fact that the majority of submissions are still made in paper. According to the EU eCTD Implementation Survey, by June 2008, a mere 1% of submissions used eCTD format. In the U.S., as of the end of 2007, only 11% of the 168,000 submissions to FDA used some form of electronic method.
The ability of companies to make process changes that will enable increased eCTD submissions, either through building in-house expertise or through outsourcing, is being hampered by economic concerns and the need to contain costs. Toban Zolman, director of consulting services at Image Solutions, says, “Companies are trying to solve those challenges without laying out a lot of capital investment, and by taking some novel approaches to the technology.” Larger companies are more likely to implement their own solutions in house, and may outsource portions based on region or submission type. Smaller players are taking a blended approach, looking at Software as a Service (SaaS) and hosted options. Helping drive down costs are the movement toward harmonization and fewer regional nuances, and an increase in the number of vendors providing eCTD services. These factors will eventually create more of a commoditization of the eCTD.
As eCTD becomes more integral to the global submission process, companies stand to gain the largest benefit as they start building in eCTD requirements from the initiation of the Phase I studies. This will result in higher quality electronic submissions from the beginning, cutting out paper, the costs associated with it, and allowing for faster regulatory review.
Here are a few basic facts about the eCTD structure to shed light on why and how it holds promise to foster significant change. The eCTD specification, developed by the M2 Expert Working Group (EWG) within the International Conference on Harmonisation (ICH), defines the criteria for the overall architecture of how the eCTD is to be structured so electronic submissions are technically valid.
In its most basic form, the eCTD is composed of three elements: a directory structure; an Extensible Markup Language (XML) backbone that creates links to leaf files; and content files. Essentially, it is the electronic version of its predecessor, the common technical document. The XML backbone enables management of metadata for the entire submission and for each document within the submission. It also provides a table of contents (TOC) and navigation aids to the individual files. Reports and forms use a PDF format, and datasets are transported using SAS XPORT files.
A submission is organized using five modules (with appropriate navigational links):
Module 1 - Administrative information and prescribing information unique to each region
Module 2 - Summaries
Module 3 - Quality
Module 4 - Nonclinical study reports
Module 5 - Clinical study reports
Each document should contain bookmarks and hypertext links from the TOC to all tables, figures, publications, and appendices. To enable navigation, documents should be generated from electronic source documents and not from scanned material.
Common Errors in eCTDs Reported by FDA
- Documents scanned instead of using PDF, and are often illegible
- No table of contents or an inaccurate one lacking bookmarks
- Inoperable hyperlinks
- Missing files or empty folders
- Incorrect or repeated use of application number
- No US-regional.xml file
- Level of granularity is not in keeping with the table of contents
- Use of spaces and characters not permitted by the Electronic Submissions Gateway
- Sending submissions to CDER instead of CBER, and vice versa