Producing a DSUR and a PBRER to deadline
A development safety update report and a periodic safety update report or periodic benefit-risk evaluation report are complex safety documents authored at defined intervals to strict submission deadlines, from many types of data held by many stakeholders. Producing them on time involves the way the two reports' periods and deadlines interact, the complexities of their safety data, and the project set-up, communication and regulatory knowledge around them.
Scope, reporting periods and submission requirements
The DSUR is required during clinical development and summarises safety information across all the ongoing and completed trials for a product; it is submitted annually to the relevant authorities. Its reporting period runs from the development international birth date — the date of the first approval or authorisation to conduct an interventional clinical trial in any country — to the data lock point, the last day of the period. The PSUR or PBRER is required after marketing and gives an interval-based evaluation in the context of the marketed product's cumulative benefit-risk profile; its period runs from the international birth date, the date of the first marketing authorisation for the active substance granted to any company worldwide, to the data lock point. These complex safety documents are authored at defined intervals and must meet strict submission deadlines.
Once a drug is approved the DSUR can, if desired, be prepared on the international birth date so that its lock point matches the periodic report's. In practice both documents must then be authored at the same time and still meet their own submission deadlines. Both collect data from safety, clinical, non-clinical, regulatory, literature and post-marketing sources where applicable, and their modular structures are harmonised so that corresponding sections can be similar and sometimes identical in content.
Although the two reports' lock points can be aligned, their periodicity can vary. A DSUR's reporting period should be no longer than one year, and the report is due 60 days after its lock point. The periodicity of a PSUR/PBRER can depend on several factors, including how long the product has been marketed, and can range from six months to three years; it must be submitted within 70 days of the lock point if the period is twelve months or less and within 90 days if it is longer. So the writer should consider whether the interval data in a DSUR suit the harmonised sections of a periodic report on a six-month, two-year or three-year cycle, and the reverse.
The regions requiring each report may also vary. To comply with national or regional requirements, additional region-specific appendices are required for DSURs submitted to the UK, Canada, China, the US and the EU, and for PSURs/PBRERs submitted to the US and the EU. The writer should be mindful of the regions requiring submission, be familiar with their regional guidelines, consider factoring in time for translation, and consider the need to submit several regional versions of each report with their regional appendices.
Failure to submit on time can lead to warning letters, fines, penalties and reputational damage. To author these reports successfully, the writer must demonstrate strong project leadership, regulatory knowledge, and the ability to interpret and summarise complex safety data.
Safety data sources
Compiling the reports can be challenging and involves analysing various types of data source from numerous stakeholders. Of the several sources the reports summarise, the primary source for adverse event data is the global safety database of the sponsor or marketing authorisation holder; the primary sources for exposure, disposition and efficacy data are the sponsor's clinical trial databases and the sales team.
Significant safety information from sources outside the sponsored trials and post-marketing reports, such as investigator-initiated trials and named-patient or compassionate-use programmes, should also be included. Such information is not necessarily provided through the standard programmed outputs from the global safety database, so the writer should establish early which stakeholders can provide it.
Medicinal products can often be developed, and then marketed, by several companies — co-development sponsors, or distinct holders responsible for marketing authorisation in different regions or countries. A safety data exchange agreement should then be in place, setting out each party's responsibilities for collecting, managing and exchanging safety information. The writer should be aware of which sponsors or holders are involved, whose standard operating procedures are to be followed, who will provide which source documents, and who is ultimately responsible for authoring and submitting the reports under that agreement.
Complexities of the safety data
Standardised, validated outputs should be generated from the global safety database: cumulative tabulations of serious adverse events and serious adverse reactions, a line listing of serious adverse reactions, and a cumulative tabulation of adverse drug reactions from post-marketing sources. Summarising a topic of interest — important identified and potential risks, targeted medical events, or a response to a regulatory request that uses a defined standardised MedDRA query — requires additional outputs or listings to be generated from the global safety database. These non-standardised outputs can include an extensive list of cases, often in Excel, which requires detailed analysis to extract the relevant information, especially for products with a high case volume or with multiple or complex safety concerns. The writer should be proficient in Excel, and seek additional training where needed, so as to generate accurate pivot tables for a high-level summary of cases or events. For more complex safety concerns the writer may need to review each individual case to determine what merits inclusion in the report. Determining what qualifies as clinically relevant information can be challenging for a writer who is not a medically qualified professional, and support should be sought from the sponsor's or holder's safety physician when needed.
The type of product strongly influences the complexity of the safety data reported. In general, products with more complex risk profiles have larger volumes of safety data to summarise per reporting interval. Risk profiles also vary between diagnostic agents, used to help diagnose disease, such as contrast agents used in X-rays and radioactive tracers used in nuclear medicine scans, and therapeutic agents, used to treat or prevent disease. Pharmacovigilance for both involves detecting and reporting adverse events and assessing the products' benefit-risk profile, but the specific types of adverse events differ. For a diagnostic agent, adverse event reporting often occurs over a narrower window and may include events related to both the product and the diagnostic procedure, such as allergic reactions to contrast agents, injection site reactions or complications of the procedure. For a therapeutic agent, where administration and adverse event reporting occur over a longer period, the focus is identifying adverse drug reactions. Typically, owing to their one-time use and lower dose, diagnostic agents have a less complex safety profile than therapeutic agents.
Although a sponsor or holder should take care that the data it holds are accurate, live databases will inevitably have inconsistencies, especially while trials are ongoing. Cases can be duplicated in the global safety database, as when a spontaneous case from a healthcare professional is later reported in a literature article, leading to potential over-reporting, and the case narrative can disagree with the listings because of possible data entry errors or delays in updating a case after follow-up information arrives. The writer should be aware of these potential inconsistencies when analysing the data and discuss them with the sponsor or holder to determine the necessary course of action.
Complexities can arise in analysing fatal cases from clinical trials. Fatal cases should be included in the global safety database, but where a trial's endpoint is overall survival — in an indication where disease progression is expected, for example — a participant's death may not always be captured as an adverse event there, in accordance with the trial's protocol. It is therefore prudent to request listings of patient deaths from both the global safety database and the clinical trial databases to assess for any discrepancies. Deaths occurring before the lock point but added to the clinical or safety databases after it may not be captured in the interval listing when data are extracted by the date of death, so the writer should consider requesting outputs of the period's fatal cases by the date they were entered into the databases rather than the date they occurred.
An aggregate report takes a snapshot of the product's safety data at a given point in time, and while a sponsor is conducting an ongoing blinded trial, care should be taken not to unblind certain team members. The DSUR's required line listing of serious adverse reactions may include unblinded data, and a periodic report may need to discuss events bearing on important identified and potential risks from blinded trial data. The writer should be cognisant of the data they are reviewing and should document at the kick-off meeting which team members must remain blinded. Where unblinded data are included in a report, care should be taken to distribute and store them separately from blinded team members so that they do not review an unblinded report.
Project set-up and the kick-off meeting
An effective kick-off meeting is one of the most vital steps in setting a project up, and the writer should come prepared, having had prior discussions with the sponsor's or holder's point of contact, such as a safety scientist or project manager, to:
- review the relevant standard operating procedures,
- acquire background knowledge of the product,
- collate information on the reference information, ongoing and completed trials, signals and regulatory requests (the Pharmacovigilance Risk Assessment Committee's assessment report), and
- confirm the stakeholders' details.
The meeting should be held about 30 days before the lock point to give stakeholders enough time to prepare, and a joint meeting is advisable when a DSUR and a PSUR/PBRER share a lock point.
The kick-off meeting should be logically structured and include:
- Writer introductions. The meeting is an opportunity for the writer to meet the wider team and set expectations for the writer's role.
- Background information. This should include the report's scope, the reporting period, the regions for submission, which determine the appendix content, the standard operating procedures and template used, and where sources will be provided and the report authored.
- List of stakeholders. It clearly defines which stakeholders will provide the required sources and who will review and approve the report. For a very large team, defining a point of contact for each group is essential. If blinded trials are ongoing, a list of blinded and unblinded stakeholders should be documented.
- Timelines. Both reports have short, defined timelines, so stakeholders need to know when to provide source information, when draft reviews will be conducted, when approval is required and when any translation will be performed, in order to meet the submission deadline. Staggered timelines should be considered when a DSUR and a PSUR/PBRER with the same lock point are authored, to allow adequate time for the team to review both, and the writer should ensure that changes made in reviewing one are addressed in the corresponding section of the other, as applicable.
- Discussion points. The meeting is an opportunity to discuss the Pharmacovigilance Risk Assessment Committee's requests, and those of other regions, to determine whether additional safety database outputs are required to satisfy them, and to resolve any disagreements between sponsors or holders.
- Early source requests. To frontload as much as possible, sources independent of the lock point should be requested early — relevant literature from weekly searches, the reference safety information, completed clinical and non-clinical study reports, signal evaluations, regulatory requests, and the corresponding draft safety outputs, to ensure that the outputs are satisfactory.
After the meeting it is helpful to circulate an authoring contribution document — a source document checklist or a blank template — tagging stakeholders for the sources and contributions agreed at the meeting. The writer can also provide regulatory guidance text to help stakeholders supply appropriate content. A tracker of what has been provided can be useful when several documents are prepared in parallel and can also help with quality control of the final report.
Team meetings and lessons learned
Weekly team meetings with key stakeholders, such as the safety physician or a pharmacovigilance scientist, held throughout the authoring are a valuable way to build team relationships, discuss issues with the data, obtain the safety physician's feedback on more complex safety concerns, and keep abreast of new information the sponsor or holder may have from internal discussions — a decision to alter the benefit-risk following a signal evaluation, for instance. Meetings should be documented for reference and to ensure that any actions are addressed. The writer should manage the creation of placeholders for the weekly meetings, the comment-resolution meetings and the draft and review periods so that the wider team is aware of the deadlines, and as these milestones approach, reminders should be sent to ensure compliance. After submission it is advisable to hold a lessons-learned meeting at which stakeholders give feedback on what in the project's delivery can be improved, which also strengthens team relationships and sets the foundations for future work.
Regulatory guidance
Regulatory requirements for aggregate safety reporting are continually evolving, and sponsors and holders often look to the writer for guidance and leadership. As the experts in authoring these reports, writers should stay up to date with new regulations and regulatory updates to ensure compliance. The writer must be familiar with the ICH, Good Pharmacovigilance Practices and regional guidelines for these reports, including:
- ICH E2F guidance on authoring DSURs,
- ICH E2C (Revision 2) guidance on authoring PSURs/PBRERs, and its questions-and-answers guidance,
- the guideline on GVP Module VII, on PSURs (Revision 1),
- the questions-and-answers document on Regulation (EU) 536/2014,
- the region-specific DSUR requirements for Canada and the UK,
- the EMA's post-authorisation procedural advice for users of the centralised procedure, and
- the region-specific DSUR requirements for China.