Kathie Clark

Even though sponsors and CROs are working to create more and more documents as electronic originals, scanned paper still comprises a large portion of any given eTMF. If you take the number of documents scanned by your organization and multiply it by the cost to scan, index and QC each document, you may be surprised at the cost burden.  Recently, we have heard anecdotal evidence of an eTMF solution at a tier 1 pharma costing upwards of six dollars per document—obviously resulting in a significant cost to that organization. The cost impact is compounded by the negative effects of inefficiency on the overall business process (e.g. documents not available in a timely manner).

A scanning process integrated with eTMF can reduce cost and risk while increasing efficiency and quality.  This post will examine some key lessons learned and best practices related to scanning integrations.

Best Practice: Documents Indexed by Documentation Specialists, not Scan Operators

In the past, many systems were implemented that duplicated the business logic that existed in eTMF (or other EDMS) in scanning workstations.  In other words, the same indexing screens and rules that existed in eTMF were re-implemented as part of the scanning process, with the scanning operator expected to both scan and index the document.

There are several problems with this approach:

When business rules are duplicated, system implementation, test, and maintenance effort increases

Inconsistencies can occur between how rules are implemented

Scanning operators are not experts in eTMF documentation and are not usually trained to recognize document types, extract metadata, etc.

Instead, using the eTMF software itself—and the document specialists who are trained to handle eTMF documents—improves quality and decreases effort.

Best Practice: eTMF-Generated Scanning Cover Sheets to Identify Documents

The easiest mechanism for allowing eTMF document specialists to index scanned documents is through the use of scanning cover sheets, also known as transmittals.  These cover sheets display information about the document to be scanned along with a bar code that identifies the unique document in eTMF so that when the scanning system passes in the scanned document, minus the cover sheet, the eTMF can automatically identify the associated placeholder and its metadata.

Document specialists should be able to generate cover sheets for placeholder documents in eTMF.  Since most metadata is already known for placeholder documents, very little additional indexing is required.  Only metadata not known at planning time (such as the document date or expiration date) must be supplied.  The system should allow cover sheets to be generated for a batch of documents, allowing metadata to be entered once for the batch and applied to all documents in the batch.  The eTMF should also allow batch printing of cover.

Best Practice: Let Scanning Systems and Operators Do What They Excel At

The scanning system itself can contribute to the efficiency of the process.  Some key features that should be supported in the scanning system and its integration with eTMF include:

Optical Character Recognition (OCR) to support full-text searching

Automatic capture of scan-related metadata and release to eTMF along with associated documents (examples include scan center, scanning operator, scan station ID, scan time, OCR confidence, etc.).  This data can then be used to produce metrics and reports analyzing the scanning processes.

Automatic release into eTMF.  The process (which involves an eTMF web service) passes the content, scan metadata and identifier from the bar code back to eTMF so that processing of the document can continue in eTMF.  The process should handle failures related to individual documents gracefully so that the entire batch does not fail.

Scanning operators should be given clear direction on the aspects of quality for which they are responsible.  Typically, these include ensuring that all pages are present, checking for skewing, ensuring that pages are correctly oriented, and so forth.

Scanning operators should also be given guidance on whether to use features such as de-skewing and de-speckling.

Best Practice: Clarify and Optimize QC Practices

When documents are released into eTMF from the scanning solution, they may be released into one of two states:

Completely indexed and ready for QC.  In this case, the document should be sent to a QC process for a final check that ensures that the correct document was associated with the cover sheet.

Partially indexed, or error encountered.  In this case, the document should be sent to an individual or group to complete indexing or address errors.

Once the document enters the QC process, document specialists should be made aware of which quality checks they are responsible for.  For example, they will need to ensure that the document matches the placeholder it was associated with (including study, country and site), but they may also need to check for required wet ink signatures, and may need to do advanced checks for documents such as 1572s.   They should not have to repeat quality checks done during the scanning process.

The combination of eTMF metadata and scanning metadata should allow eTMF to produce reports that can analyze performance across many dimensions.  Reports should allow analysis of defect rates by scan center, operator, equipment, time frame, document type, and more.  The following is a simple example report comparing scanning accuracy across scan centers:

Scanning is integral to eTMF success.  Your eTMF should provide integration and automation capabilities to optimize efficiency and quality, and reduce risk of incorrect documents being filed in your system.

For a news release detailing Wingspan’s new eTMF product, click 

Articles

Even though sponsors and CROs are working to create more and more documents as electronic originals, scanned paper still comprises a large portion of any given eTMF.

Integrated Scanning: A Key to eTMF Efficiency

Even if your eTMF includes high-quality tools for planning the documents needed for a specific clinical trial, it’s a given that events will happen during the course of the trial that will require not just new documents, but new sets of documents.  Some common events include: 

In this article, we’ll take a closer look using an example of a protocol amendment event.  According to the article 

  “Nearly 60% of the 3,410 protocols required one or more amendments. Completed protocols across all phases had an average of 2.3 amendments, though later-stage Phase II and III protocols averaged 2.7 and 3.5 amendments respectively. ” Given the frequency of this event, and the complexity of the document sets involved, protocol amendments make a good use case for the purposes of discussing TMF events.  In addition, missing documents related to amendments are a common inspection finding.

A protocol amendment always includes the amendment itself (TMF Reference Model ID 02.01.04, Protocol Amendment) and a signature page from each active site documenting investigator agreement to the protocol amendment (TMF Reference Model ID 05.02.03, Protocol Amendment Signature Page).  Beyond that, the documents needed to support the amendment vary greatly depending on the nature of the protocol changes and the countries and sites in which the studies are being conducted.   Some examples of other documents that may be needed to support an amendment include:

Country Level Amendment and/or Translation of Amendment.

 A modified amendment may be needed for a country, for example, to allow for specific assessment techniques permissible in that country.  The amendment may need to be translated to the local language, and the local version back translated, both of which may require certificates of translation.

IRB/IEC Submission (04.01.01) and Approval (04.01.01).

 FDA requires that “A sponsor shall submit a protocol amendment describing any change in a Phase 1 protocol that significantly affects the safety of subjects or any change in a Phase 2 or 3 protocol that significantly affects the safety of subjects, the scope of the investigation, or the scientific quality of the study.”  They also require that “The change has been approved by the IRB with responsibility for review and approval of the study.”  Therefore, most amendments would have to be submitted to an IRB, but a sponsor could not submit an amendment outside of the FDA’s requirements as, strictly speaking, it would not require IRB approval.  In addition, the IRBs/IECs affected may be a combination of central and local committees, not necessarily one per site.

Regulatory Submission (03.01.01) and Approval (03.01.01).

 For similar reasons, these would often but not always be required.  In addition, country specific regulations may result in submissions and/or approvals being needed in some countries but not others.  Local language versions would generally require back translation to English.

Protocol supporting documents such as Informed Consent Form (02.02.03), Informed Consent Form (02.02.03), etc.

 These documents are closely related to the protocol, and changes to them may result in an amendment and vice versa.  They may also require country or even site-level versions and translations.

To complicate handling the event in eTMF, some of these documents might not be needed for all countries, or for all sites (for example, sites that are terminated or completed).

It’s clear that significant expertise is needed to determine exactly which documents are needed to fully support a protocol amendment.  For large studies, an amendment could result in the need to collect hundreds or even thousands of documents.  

eTMFs range from handling study events elegantly to not handling them at all.  At the top end, users will be able to initiate the event in eTMF, select the exact documents, countries and sites involved, and create placeholders to track the expected documents.  eTMF will also collect an event date that allows a due date to be assigned for collecting and approving the documents in eTMF (generally based on an offset such as 60 days after the event).  This allows for a repeatable process and prevents documents from being overlooked or not collected in a timely manner.

In more basic systems, no special tools are provided and a manual procedure is needed instead.  In this case, it’s highly advisable to create a work instruction with an associated checklist.  An event-specific checklist can then be created (probably in Excel) with a list of documents that must be collected for each country and site.  This process is certainly workable, but is more labor-intensive and error-prone.

Finally, it’s important to recognize that, as the study progresses, additional documents may be needed for countries or sites that are added later.  If a site is initiated after the protocol has been amended several times, signature pages (and potentially IRB/IEC submissions and approvals) must still be collected.  If the eTMF doesn’t support this in an automated manner, it will be necessary to revisit past events to see what documents need to be collected for new countries and sites.

In conclusion: process remains all-important.  A sophisticated eTMF can remove much of the manual labor for you, but it can’t tell you what exact documents are needed for each event.  A well-designed and documented process works with technology to ensure quality and compliance.

It's a given that events will happen during the course of the trial that will require not just new documents, but new sets of documents.

Study Events: Keeping your Trial Master File Up-To-Date as Your Trial Evolves

	The life sciences industry has been evolving towards the use of electronic Trial Master Files (eTMFs) to manage the documentation associated with clinical trials.  The promise of enhancing process efficiency, improving access to studies for increasingly global and virtual teams, and providing timely access to the TMF for internal and agency audits has proven attractive, especially to Tier 1 Pharma and large CROs.   Smaller organizations have begun to express interest as well.  However, progress was accompanied by lingering concerns about the expectations of regulatory agencies around the management and auditing of eTMFs and the records they contain.
	
	In 2012, the UK’s Medicines and Healthcare products Regulatory Agency (MHRA) included a discussion of TMF/eTMF in their published Good Clinical Practices (GCP) Guide.  The Guide discussed issues such as security, records management and audit access.  It provided significant clarification and was viewed as a bellwether for the practices of other agencies.  Recently, the European Medicines Agency (EMA) issued a draft “

Reflection paper on GCP compliance in relation to trial master files (paper and/or electronic) for management, audit and inspection of clinical trials

”.   This paper is the first published framework for eTMF for an entire region.   EMA states that the paper is provided in response to the numerous questions they have been receiving regarding TMF / eTMF, as well as in response to inspection findings.
	
	EMA states that the aim of the paper is to “set out the requirements for the TMF as covered in directives and guidance and to give recommendations to assist organisations in maintaining a TMF that facilitates trial management, GCP compliance and inspection. The paper also addresses archiving of the TMF, clarifying retention times and gives some recommendations regarding destruction of paper documentation.”  Although it answers many of the questions posed by industry, it also raises some additional questions.  Industry will have a chance to provide comment until the end of April 2013.
	
	This article will examine the most significant portions of the paper and their impact on people, processes and technology. 

	The paper emphasizes the differences between the sponsor TMF and investigator TMF (investigator site file).  Although ICH Good Clinical Practice CPMP/ICH/135/95 provides details on which documents are required for each TMF, confusion is not uncommon and presentations have been made suggesting that a common archive could manage both files.  This is not clearly acceptable:

		In organizing the TMFs, it is essential to segregate some documents that are generated or held by the sponsor from those of the investigator and vice versa as some documentation held by the investigator should not be provided to the sponsor, for example those documents that would result in breach of subject confidentiality.

		Exceptions are possible only when the sponsor and the investigator are the same organization, for example when the sponsor is a hospital/health institution.

	The role of the CRO also impacts the organization and control of the TMF.  The sponsor remains responsible for the trial and needs to maintain – and document - oversight. Sponsors should ensure that the role of the CRO is formally defined and documented, including specific details around TMF/eTMF structure, filing processes, access, applicable procedures and documents, and audit.  For an eTMF managed by a CRO, the CRO will either have to provide access to the sponsor or determine how TMF documents will be provided without direct access.  Details in the refection paper can serve as the basis for contract review.

	The sponsor should identify the location of all of the potential documentation that is part of the TMF.  In large organizations, the TMF could include documents from across a variety of different departments and systems other than clinical operations, for example, Data Management, Statistics, Pharmacovigilance, Clinical Trial Supplies, Pharmacy, Legal, Regulatory Affairs, etc.    These documents need not be physically present in the TMF/eTMF, but if they are not, access to them by auditors during inspection is necessary no matter where they reside.
	EMA also stresses that sponsors should establish a common TMF structure used across trials regardless of location, with unnecessary sections deleted from the master structure for each trial.  Investigators may use a sponsor’s structure if they wish.

	EMA emphasizes that any documentation which has been created during the trial and that helps reconstruct and evaluate trial conduct must be filed in the TMF, irrespective of whether it is explicitly listed in published guidelines.   Superseded documents must be retained in the TMF so that trial activity can be reconstructed for any point in the trial.   Any quality record produced from following a quality system procedure must be retained in the TMF to demonstrate compliance.
	Specific guidance on trial correspondence includes:

		Electronic correspondence (emails) may be retained electronically, provided the requirements for eTMF and electronic archiving are considered. Emails should be saved to ensure that the associated metadata is retained, for example as .pst files rather than pdf documents or being printed and signed.

		Correspondence (paper and emails) should be effectively organized and filed in chronological order in an appropriate section in the file.

		Duplication of any documentation in the TMF is recommended to be avoided, but this often occurs with email correspondence and with attachments.

		Sections including correspondence must be complete, that is, both sent and received correspondence must be included.

	The TMF should to be up to date, with documents placed in the TMF in a timely manner with the aim to maintain the TMF “inspection ready”.
	The overall implication is that sponsors and CROs will need to review their TMF master lists to see if they are really complete or if they have been including only documents listed in guidance.  Specific procedures for handling correspondence and quality records should be reviewed to ensure that documents are collected in a timely manner and contain all of the information expected during an inspection.

	As expected, EMA emphasizes the need for the sponsor and investigator TMFs to be readily available for inspection at any time during a trial.  Direct access is required for all systems comprising the TMF.  A new consideration for sponsors is whether they would be willing to provide remote eTMF access to agency auditors – EMA has expressed interest in this capability but sponsors would need to determine how comfortable they would be in offering remote access as it is not currently required.

	Acceptability of an eTMF depends on the practices around its implementation and use.  Storage of TMF documents within folders in a file system is unlikely to be considered a substitute for a true eTMF due to lack of controls around security and accessibility.
	Good practices expected for an eTMF include:

		Version control applied to electronic documents

		Control over timely creation and deletion of user accounts, and password-controlled access

		Protection of documents against unauthorized modification

	EMA has not yet seen the use of eTMF at investigator sites, but it would be permissible.  However, a situation where all the site records are sent to the external sponsor for uploading onto an eTMF system, which the investigator then accesses via a portal, would breach the requirement that the records stay under the control of the investigator.
	When agency auditors examine an eTMF, they expect to be able to work with it after no more than an hour’s training.  To facilitate review, the organization should provide adequately sized monitors and ideally provide tools to compare and mark documents.

	EMA provides a set of detailed requirements for quality control (QC) of scanned images. Sponsors and CROs should review these requirements to ensure they are complying.   A certified copy of a paper document can replace the original paper record.  Although 100% QC is not required, justification is needed for lower levels of QC, including following a validated process.

	EMA summarizes a variety of records retention requirements and their impact on TMF.  Factors that influence retention include the type of record (medical record, study report, IMP document, etc.), the status of the product (approved, marketed, or retired) and national retention requirements that may override EC directives.  After reviewing these complicated requirements, it’s clear that a record retention schedule should be established by a records management expert.  A named individual must be responsible for an archived TMF, and regulations and risks studied closely before destroying paper documents transferred to an eTMF.

	The reflection paper provides a specific set of requirements that every sponsor or CRO responsible for TMFs should review in detail, as they will serve as the basis for any agency inspection.  For those organizations using an eTMF, a self-audit is in order; for those contemplating eTMF, a clearer picture of the process requirements should assist in determining project scope and organizational readiness.

EMA's Reflection on eTMF

Electronic Trial Master Files (eTMFs) are growing in popularity.  A recent 

 summarized by eClinica reported that in 2012, 32% of survey respondents maintain inspectable TMF’s in paper only, down from 50% in the corresponding 2010 survey. While there is only a modest increase in the number of respondents maintaining inspectable electronic TMFs (8%), the increase in the number of respondents maintaining a combination of paper and electronic was significant – 48% in 2012, compared to just 27% in 2010, indicating a continuing transition to eTMF in the industry.
  
  As with most innovations, the benefits gained with eTMF also require new ways of working.  One of the most important considerations for companies maintaining eTMFs or hybrid paper/eTMFs is how to handle agency audits.  When paper TMFs are audited, the sponsor, CRO or investigator must ensure that all of the paper records are available, and assist the auditor in accessing them and locating information.  Usually, all of the parties involved have many years’ experience in assessing the quality and completeness of the paper TMF.
  
  With an eTMF, an inspector has new concerns:  

 Was a validated QC process used when paper documents were scanned and uploaded into the TMF to ensure that the final records were accurate, complete and legible?  

Does the system maintain records to prove that this process was followed?

Have all documents in a completed study completed the defined QC process? 

Has justification been provided for expected documents that were not received?

Are records in the eTMF secured against unauthorized access or alteration?

Are all significant actions recorded in an audit trail that can be understood by users and auditors?

Are associated paper records preserved and retrievable unless a certified process has been put in place for their destruction?  (see the recent position paper 

Are records retention policies documented and followed?

Can the full TMF be provided including both electronic documents and those retained only in paper?  

Can the inspectors access all electronic documents relevant to the inspected site or sites directly, using a user-friendly interface with reasonable performance?  Are the records indexed in a manner that makes it clear that all essential documents have been collected?

Can the validation records for the system be provided?
    
  If these requirements were considered when the eTMF was designed and implemented, ensuring the audit-readiness of a specific TMF should be straightforward. Compliant processes for scanning upload, and QC should be standard, not optional. Queries and reports should provide a catalog of the documents comprising the TMF, revealing any missing documents or documents that have not completed the QC process.
  
    Although eTMFs are new, authorities are beginning to signal their willingness to – and even preference for – conducting TMF audits in an eTMF environment.  (For comments on eTMF from a MHRA inspector, see the DIA presentation 

.)  They expect sponsors to take the steps needed to provide them with the tools and information needed to conduct quality audits in a timely manner.    

Kathie Clark

Articles

Integrated Scanning: A Key to eTMF Efficiency

Study Events: Keeping your Trial Master File Up-To-Date as Your Trial Evolves

EMA's Reflection on eTMF

Is your eTMF Ready for an Audit?