Starting a Risk Assessment? Start Early

Sign up to attend Medidata’s GLOBAL NEXT in New York City from Nov 8th - 10th and join us for a fantastic customer panel with BMS, Celgene and KaryoPharm to discuss the role of technology in ensuring inspection readiness and ICH E6 compliance.

When it comes to risk assessment, it’s critical to look beyond what a clinical research associate (CRA) does. Last week, we saw how cross-functional perspectives like supply chain management or statisticians enhance the discussion by accounting for and addressing all potential trial risks. Additionally, identifying and training internal candidates whose core competency is risk assessment will be key in conducting a robust risk assessment plan. Today, we will explore when you should start your risk assessment and what it really means to document your risk assessment.

Start Early

Start your assessment early, before you finalize your protocol, at the compound level. When you start researching a certain compound, you will understand certain things about it – its associated bench research, animal research, etc. – and you will therefore have certain hypotheses about risks and side effects. Start building the risk associated with the compound early in the process, with the idea that all risks can be floated down to a study level risk assessment so that you can take what you know, and apply it at a study level.

It’s not unusual to see a protocol with 10-12 primary endpoints, which can lead to long and tedious trial enrollment. Studies are getting more complex, but using a Quality by Design (QbD is a systematic approach to development that begins with predefined objectives and emphasizes product and process understanding and process control, based on sound science and quality risk management) approach allows you to find answers to your questions in the most efficient way possible.

• Although de-Risking is key, we’re not always thinking about how to de-risk the protocol. This includes anything you can do to make your data more accessible remotely, from a central monitoring protocol, such as:
  • Use of central vs local labs - so you have access to a lab result without needing to visit a site
  • Endpoint frequency - your endpoint may require you to look at an image to assess tumor size, for example. If that endpoint is occurring frequently enough, you must review it often enough to draw conclusions early.
• Better Late than No Risk Assessment. In the event that you have waited to assess risk in your study, there is still a solid case for doing so. Not doing it at all can lead to costly and time-consuming protocol amendments, which you may not have the budget to manage mid-study. 

We have worked with customers who assessed risk later on in the cycle. Although not a best practice, it’s never too late to track and document risks, as the process always helps with compliance. Be cautious to make sure you are not creating additional risks.

Conducting risk assessment mid-stream may increase risk, because it may change monitoring plans and instantly make different individuals responsible for monitoring indicators of which they were unaware. While it might be difficult to get a central monitoring strategy underway, it is still worth conducting a risk assessment to stay in compliance.

Document, Document, Document

According to the FDA, if you didn’t document it, it didn’t happen. Regulators do not expect perfect data. They want you to focus on what’s important, which, as mentioned, relates to trial endpoints, patient safety, and always ensuring adequate informed consent.

While assessing protocol risk is not a new process, documentation of these cross-functional risks in one consolidated place is new to many organizations. We now better understand what really helps from a monitoring perspective in a clinical trial; which is why we believe so strongly that it’s important to do this now.

SDV is not the most cost effective or efficient way to monitor data, and we are learning more and more everyday about what machine learning brings to anomaly detection. Machines can analyze data much more effectively than the human eye and can categorize it much more efficiently.

• What should you document
  •  ​First and foremost, what is the risk, how it is being measured, and who is responsible for tracking and mitigating the risk. Of the tools available to document risk assessment, regulators are in general agreement that RACT works best, when used correctly.
• • When Transcelerate first developed its RACT Tool, the feedback was that many people were overwhelmed about where to start. In the multiple iterations released since then, they tool has been simplified. The tool is readily available on their website, along with other directional information into what seems most successful for implementing risk assessment.
• • At Medidata, we have built our Risk Assessment Planning System (RAPS) based on guidance provided by Transcelerate and ICH E6 addendum. When combined with RACT Professional Services, it truly simplifies the process, keeping in mind the endpoints of the study. This, in turn, provides a blueprint for how to adequately monitor safety and to ensure that subjects are well consented.

Constantly Evolving, Constant Re-evaluation

The result of a proper risk assessment is a fluid document that must constantly evolve, be version controlled, and be re-evaluated. This may include:

• Setting up a cadence for document review
• Turning off certain KRIs when they don’t apply anymore - for example: Closing risks associated with enrollment periods, and not trying to assess for them upon the end of the enrollment period

• Re-evaluate your risk plan when the protocol is complete, to ensure everything still applies, is de-risked appropriately, and nothing was missed before beginning enrollment

Continuing on with our blog series on RBM, next week we’ll discuss centralized monitoring. Download a copy of our "Ultimate Starter Kit for RBM" eBook here to learn more.