4 Ways to Optimise Clinical Development with Pharmacovigilance
Throughout my years working in clinical development, I have lost count of the number of discussions with colleagues where pharmacovigilance (PV) has been characterised as nothing more than the expedited reporting of Suspected Unexpected Serious Adverse Reaction (SUSARs) to regulatory authorities. More so than this is the common impression that PV is able to function in isolation and can just be slotted into an otherwise fully formed clinical system. Not only does this approach make it more difficult for PV to remain compliant (the most recent Medicines & Healthcare products Regulatory Agency (MHRA) Good Clinical Practice (GCP) Inspection Metrics Report[1] showing PV as the area with the most critical findings for commercial sponsors), it also stymies the additional benefits a fully integrated PV system can have on the development programme.
Here, I look at four areas where increased cross-functional collaboration and understanding can yield benefits, both immediately and also downstream. These problems are particularly exacerbated when a sponsor is using multiple vendors for different activities, who have no incentive or marry up their processes, unless prompted. Therefore, these efficiencies may have to be driven by the sponsor against some resistance.
1) Case Report Form Design
As previously mentioned, PV isn’t just about reporting cases to regulatory authorities, from the safety database. Signal detection, data safety monitoring boards (DSMBs), development safety update reports (DSURs) etc. all fall under the remit of PV and require more data than may be routinely captured on the safety database. As such, it’s a surprise just how often the clinical database case report forms (CRF) can be designed and finalised before any conversations with PV have taken place.
At a minimum, there should be agreement on the structure of the data being collected directly by PV in Serious Adverse Event (SAE) Forms. I have seen all too many examples of the same field offering different options for results (e.g. causality) between an SAE form and the CRF. Not only is a nightmare at reconciliation (more on that later) but is an early indicator to GCP inspectors of a lack of joined up thinking. Sponsors don’t want to be giving inspectors an easy ‘in’ on where to start digging, especially when PV is such a source of critical findings. Further changes in CRFs should involve all stakeholders, including PV, to assess for any impact on their function or even just call on their expertise.
Following on from the design of the CRF…
2) Listings and Tables
In my opinion, the single biggest improvement that could be made in the relationship between the clinical database and PV is in the provision on line-listings and tables. Signal detection and aggregate safety reports cannot solely rely on the data held in the safety database. All safety data must be considered for these activities, not just serious events. Therefore, it’s imperative that relevant data can be provided in a user-friendly format, to allow manipulation and interrogation. Many are the hours that I, and my teams, have spent manually reviewing hundreds of hard-copy pages of bulk adverse event (AE) exports to create usable tables and listings. Not only is this an inefficient use of time, it also introduces myriad opportunities for errors and inconsistencies across reports, which are then all but impossible to identify the source of further down the line. Detailed discussions with programming and data management (DM) at the design stage of the clinical database will ensure that the data included in and format of reports required by PV are available from the start. Not only does it make it easy for programming to schedule time for the activity, this approach also forces PV to really think about the data needed for a study upfront, rather than being reactive. Ad hoc reports and changes can then only be requested in exceptional circumstances, rather than being the norm.
3) Serious Adverse Event Reconciliation
Serious Adverse Event reconciliation is a vital process which can appear simple but can end up involving a whole raft of functions including PV, DM, medical monitors (MM), clinical research associates (CRA), site staff and investigators. The aim of reconciliation is to make sure the data held on SAEs, on the clinical and safety databases, are accurate and up to date to, ultimately, ensure patient safety. However, frequently, I have seen functions lose sight of this and adopt an adversarial relationship, blaming the other functions rather than focusing on understanding and rectifying the discrepancies. There will inevitably be differences, within reason, between the data in the databases due to different routes and timelines into them so there is no slight being passed on any function when discrepancies are found. Because it is an ongoing process it is one that can prove most tricky when multiple vendors are being used and requires a specific, robust, reconciliation plan, agreed between all stakeholders.
As well as a need for clear ownership of the process, it is important that all parties are trained on exactly what their roles may be. Each party just needs to be aware of how they fit in and when and WHY they may be called upon. I have seen many occasions where a query has been escalated to a CRA or MM and all that initially happens is a check of the PV and clinical databases and a response that there is indeed a mismatch. This is an unnecessary duplication of work and has got us no further to a solution. It is easily addressed by the CRA/ MM knowing before a study starts the kind of queries they will receive (i.e. genuine discrepancies that standard querying hasn’t been able to clarify) so they can get straight on with solving it and also DM/ PV being clear in their descriptions of the issues they need assistance with.
As previously mentioned, ensuring the design of the CRFs and SAE report forms match is an easy win in making the whole process run more smoothly and should be planned in advance of the study commencing. Both DM and PV should run dummy reconciliation reports to check that the formats and data included are fit for purpose. This shouldn’t be left until the first reconciliation is taking place particularly as, by then, it may be too late to develop and program the necessary listings leading to a problem for the duration of the study.
In a large, global, Phase III study, these small efficiencies in the reconciliation procedure can add up to huge time savings. Most importantly, they can avoid last minute scrambles to tidy up issues at the critical end of study stage.
4) Monitoring Visits
Clinical Research Associates have a tough job. Even in the nascent world of risk-based monitoring (RBM), the number of documents to be reviewed, cross referenced is massive, not to mention the frequent hand-holding sites may need. However, the PV lead for a project would be well advised finding a gap in a CRAs busy schedule to sit down and discuss what is required and expected from both sides throughout a study.
Whether it be something as simple as what communication PV need in the event a CRA uncovers an unreported SAE or as complex as using CRA insight into individual sites to tailor how follow-up requests are made, improved cross-functional communication and trust will lead to improved and smoother running processes.
It will also lead to better understanding of PV from sites. While it is best practice for PV to attend site initiation visits, to provide training, cost considerations mean that frequently the PV ‘training’ is limited to a couple of slides delivered by CRAs. The closer CRAs work with PV, the greater their understanding of the area beyond just ‘reporting SUSARs’ and the more effective the initiation training can be, even when not delivered by PV. Equally, the more PV understand site dynamics, the more it can amend processes and procedures to make things easier for sites, while increasing engagement and data quality.
This blog merely scratches the surface of the kind of cross-functional efficiencies that can be made during clinical trials, with each of these topics alone worthy of a white paper. However, hopefully it has given you something to thing about and a fresh eye when next reviewing your standard operating procedures (SOP)!
Drive Phase PV is a pharmacovigilance consultancy and provider of pharmacovigilance services, supporting patient safety throughout the life-cycle of our clients' products. Contact us to discuss optimising your PV clinical development processes. Whether via SOP development or provision of training to cross-functional teams or vendors, we deliver a tailored packaged to seamlessly fit to your company structure and strategy.
[1] https://www.gov.uk/government/publications/good-clinical-practice-inspection-metrics-2007-to-present