The Fast & The Flexible: How to Take Control of Your Clinical Data

Challenges abound with typical clinical data review. Here’s how to avoid them so that you can maintain control throughout the process.​​

The speed of evolution of pharmaceutical and medical device development is accelerating, and with each trial, more complex and disparate kinds of data are developed through varied methods. There is also increasing trial complexity, new data types, and a growing reliance on contract research organizations that handicap access and analyze the data sponsors need when they need it.

The chief goal is to ensure control of your data in any given situation and time point during a trial. A modular analytics system that is adaptable and flexible can help trial sponsors to achieve the goal of accelerated clinical development speed and increased agility while also preserving accuracy of incoming data.

The priority of clinical data review is to identify risks as early as possible, however, a widespread inability to perform self-service analytics in near real-time has long hampered clinical research. Further, a major drawback of many existing analytics tools used to review clinical data is that they can’t be molded to the unique aims of a study.

The use of conventional analytics tools creates common challenges, contributing to disruptions in the analysis pathway that can increase the risk of misinterpretation, and erroneous conclusions, ultimately nullifying the trial.

6 Common Data Analysis Disruptions​

1. Risk of missing an unexpected safety signal​

Ensuring the safety of the subjects enrolled in a study is critical, necessitating built-in processes enabling rapid identification of any unexpected safety signals or adverse events to minimize patient risk.

Reduced safety risk leads to a higher-quality trial, shorter timelines and, consequently, lower cost, highlighting the value of a near-real-time view as subjects undergo screening. Equally important is the ability to quickly analyze study-specific objectives and endpoints in accordance with changing needs since this can reduce time to insight for faster, safer study completion.

2. Rigid tools unable to be adapted for specific therapies or use cases​

A major drawback of many existing analytics tools used to review clinical data is that they can’t be tailored to the unique aims of a study. Study-specific endpoints, objectives, and data points must also be considered – but these can only be properly analyzed using a tool that can adapt to a particular use case and that is designed with clinical analysis in mind. The key lies in striking a balance between common metrics used across all studies while concurrently focusing on stated objectives, endpoints, and safety risks that are relevant to a specific study and/or therapeutic area. This is fundamental to supporting scalability throughout clinical development.

3. Difficulty in discerning relationships​

Another limitation of current technologies and processes used to review clinical trial data is that they prevent users from seeing the bigger picture. The practice of using independent line listings, basic spreadsheets, and static analyses to sift through data remains common.

Using such an outdated approach complicates the ability to view adverse events in parallel with the corresponding lab results and concomitant medications in a single view or on a patient profile. It also prevents users from analyzing all laboratory results at a population level before selecting abnormal data, drilling down to a patient level, analyzing the corresponding medical history and concomitant medications, then finalizing the review on a line listing.

Rapid identification of relationships and trends is vital to minimize risk and to avoid slowing down clinical development.

4. Existing technologies or processes slow down data review and decision-making​

Information is generated across numerous sites, geographies and by multi-site internal and external research groups specialized in many areas. As such, the ability to review data in near real-time without depending on outside organizations for analytics reporting -- be they another group within the same company or an external contract research organization -- is key to keeping clinical development on track.

5. Reliance on data reporting from outside groups​

Relying on a middleman rather than going directly to the data source tends to add considerable time between question and answer. As such, insourcing clinical data review is often a preferred approach to alleviate regulatory pressure while ensuring compliance with ICH guidelines stating that ultimate responsibility for the quality and integrity of trial data always resides with the sponsor. Moreover, the financial incentives to quickly determine safety and efficacy mean that performing clinical data review in-house provides a faster route to understanding success or failure and adapting accordingly.

Currently, many clinical data review teams are forced to request data in a standardized format from data management groups, biostatisticians or external vendors, a process which inevitably incurs delays and extra costs. In contrast, having direct access to source data that is easily manipulated into a workable format provides those performing clinical data review with faster time to actionable results.

6. Each team is reviewing their own reports​

When clinical data review teams rely on data management groups, biostatisticians, and external vendors to provide manual reports and analysis, they increase risk of inadvertent study biasing and also add complications. This interdependency invites significant delays, but it can also lead to the generation of individual analyses that are difficult to share, impossible to reproduce or compare across domains, and that typically must be recreated every time new data becomes available.

Since clinical development is a team sport that requires verifiable results, a centralized view and collaboration mechanism is essential to avoid any errors that can jeopardize the entire process.

The risk of these six potential disruptions may be mitigated with the appropriate clinical data management solutions that are fast and flexible.

5 strategies for successful clinical data review​

Five key strategies underpin a successful approach to resolving the challenges of clinical data review:

Speed​: Speed recognizes the need for a solution that can rapidly be deployed and that allows visualization of near real-time data as soon as a study begins enrolling. It also references the capacity to promptly identify safety, efficacy, or data quality issues, as well as the ability to adapt quickly to any given changes in a study.

Agility​: An agile solution highlights the requirement to handle protocol amendments, data changes, multiple data sources, and the specific metrics of a study or therapy. Agility allows for tailored analytics.

Adaptable analytics​: Striking the right balance between common and study-specific analysis solutions is critical. With the effectiveness of clinical data review hinging on being able to interrogate unique objectives, endpoints, and safety risks, as well as being fully amenable to scalability, it is vital that adaptable analytics can be used to augment standard safety metrics for faster time to actionable insights.

Workflow flexibility​: This means providing a user experience that helps highlight risk and relationships to streamline data discovery. Having the capacity to drill down from population to subject and then also across related safety domains naturally paves the way to more efficient clinical data review.

Collaboration​: With collaboration being pivotal to the success of any clinical data review process, analytics solutions must be designed with teamwork in mind. Users should be able to clearly track the data review process, share results, receive feedback from peers, and communicate findings to stakeholders. In combination, these strategies allow for rapid identification of adverse events and swift resolution of any issues with the potential to impact safety in a larger population.

Partnering with PerkinElmer Informatics​

PerkinElmer Informatics’ (PKI) solutions integrate data across the clinical trial development lifecycle – safety, operational, clinical and outcomes -- then present the data as clear visual informatics using the industry-leading clinical analytics platform, TIBCO Spotfire, providing a complete picture of your pipeline in real time.

Flexibility and speed are hallmarks of PerkinElmer Informatics Clinical solutions. PerkinElmer Informatics provides a hybrid solution that allows customers to decide what is best for them.  Leveraging TIBCO Spotfire, PerkinElmer builds various modules to answer common clinical trial use cases.  Each of these modules are connected to your data by our Services team (or your team) and are ready within a few weeks.  Additionally, PerkinElmer’s expert team can speak your language and rapidly extend the modules to your therapies, protocols, or other needs.

PerkinElmer Informatics empowers sponsors to make meaningful, data-driven decisions, dramatically reduce the costs associated with clinical development and guide new drugs and devices to market faster.