A Quality by Design Process for Studies

Quality by Design (QbD) is a broad approach and way of thinking. During the process of evaluating Quality by Design, Critical-to-Quality (CtQ) factors can also be defined to drive the operational strategy and maximize quality.

Further below is a process that promotes critical thinking of the study design, which then allows the team to hone in on more specific attributes or factors that are critical-to-quality. At a high level the process includes:

1. QbD brainstorming is a way to critically acknowledge and analyze key challenges or unique aspects of the study. The output of this process should then be included as part of the protocol and overall study design, and could also support the definition of CtQ factors, KRIs or the RBQM strategy.

   
   

2. CtQ factors should also be identified. They are specific, are formally documented, and have actionable KRIs and QTLs linked to their success.

Again, the goal of this process is to help the team initiate discussion and review of Quality by Design for an upcoming trial, and also identify Critical-to-Quality factors that should drive the quality focus and quality-driven processes for the study.

A Quality by Design process:

1. Create a work instruction or formal process for the selection, documentation and implementation of QbD elements and CtQ factors for your studies. Some important considerations include:

   
   

   1. Cross-functional input: define the stakeholders involved e.g. Clinical Science, Medical Monitoring, Biostatistics, Clinical Operations, Data Management...

      
      

   2. Quality conventions: ideally a reference database, guidance document or wiki exists, categorized and searchable, where there are standard question prompts for teams to consider, design best practices, related CtQ factors and KRIs (perhaps in a separate database), and references to external guidances like CTTI, Transcelerate, FDA and ICH.

      
      

   3. Documentation, which could be a section in the Risk Management Plan (discussed here) and should include:

      1. CtQ factors as defined in the protocol

      2. Data points and expectations e.g. values or ranges, including Quality Tolerance Limits

      3. References to documents or plans that support risk management of the CtQ factor, e.g. Monitoring Plan and Data Review Plan

         Note: QbD elements may or may not include formal documentation, which could simply be meeting minutes and brainstorming documents.

         
         

   4. Finalization and signoff of CtQs and QTLs prior to protocol finalization.

      
      

   5. Processes for ongoing tracking, monitoring and review of the identified CtQ factors: discuss how CtQ factors will be defined in operational plans and how they will fall under the steps for heightened risk management within those plans.

      
      

   6. Analysis of the CtQ factors: how they will be referenced by the study Biostatistician and incorporated into the Statistical Analysis Plan (SAP). How they will be referenced by the scientific team writing the Clinical Study Report (CSR). Creating template sections within these documents is advisable.

      
      

2. Identify the QbD elements for your study.  This should be done during the early protocol writing stage once key requirements for efficacy, safety and analysis are known. To facilitate it can be helpful to have a cross-functional meeting:

   
   

   1. Create a QbD guidance document or slide deck that includes the following elements:

      1. Reason for QbD exercise to help everyone understand the big picture: review FDA and ICH recommendations, as well as driving forces behind quality and risk management.

      2. Objectives for the study as driven by the ideal results, a reflection of the target product profile and key attributes of the study treatment.

      3. Categories and tasks related to study milestones, operations and design: CTTI has some good references for this, but one way to organize the team review is by using a standard study timeline with all key tasks and milestones, or a CRO proposal with line items covering all major tasks, for instance:

         1. Feasibility: patient population, competing trials, countries, site selection (including technical requirements or experience)...

         2. Startup: site initiation, training, vendor and technology implementation...

         3. Conduct: recruitment, protocol implementation, safety monitoring, RBQM...

         4. Closeout: data cleaning/finalization, decommissioning, archival...

            Your list should be a lot more comprehensive and specific to your study.

      4. Use prompts and focus questions to brainstorm QbD considerations for each study category or major task: questions like "What are potential challenges with...?" and examples from prior studies that can help get the creative juices flowing.

         
         

3. Identify the CtQ factors for your study. As you are brainstorming your QbD elements, also ask "Is there a QtC factor here?" Here are some ways to evaluate:

   1. Is the proposed factor critical to understanding efficacy and safety for the treatment?

   2. Is the proposed factor required for eligibility and unique to the intended patient population? ...e.g. congenital blindness vs. acquired blindness.

   3. Is the proposed factor a major protocol deviation (read: participant removal from analysis) if missing, incorrect or out-of-range? ...e.g. required primary endpoint missing at baseline or pre-treatment.

   4. Are there critical study processes like blinding, randomization, treatment assignment or data falsification that could undermine the reliability of results?

   5. Are there moderate to severe safety concerns for participants based on the indication, treatment or study design? ...e.g. acute myocardial infarction, suicidal ideation.

   6. Are there logistics that are uniquely "risky" for the study? ...e.g. assessments done in an ER.

      
      

4. Define Quality Tolerance Limits (QTLs) for your CtQ factors. These would result in potential loss of data integrity for the study or cause excessive safety risk. Some concepts:

   1. % of missing key data points, e.g. primary endpoint assessment at baseline.

   2. % of out-of-range values, e.g. lab or vitals, that are deemed critical for safety.

   3. Key eligibility criteria are not met in greater than X% of enrolled participants.

   4. Y% of participants are not achieving the minimal treatment duration to determine appropriate response.

      
      

5. Document your CtQ factors, QTLs and related information.  See 1c above.

   
   

6. Reference your documented CtQ factors in your operational plans.  For example:

   1. Monitoring Plan: if appropriate, emphasize CtQ related data points as requiring 100% SDV and review. Include escalation plans where issues are discovered.

   2. Data Review Plan: emphasize additional notifications, reports and data review processes that will help discover issues with any CtQ factors early. Include escalation plans where issues are discovered.

      
      

7. Ensure study analysis within the SAP and CSR cover CtQs and QTLs.  The study Biostatistician and scientific writing team should be involved in selection and approval of CtQ factors, and will need to create tables, listings and narratives that cover them. Adding template sections to the SAP and CSR is recommended.

   
   

Quality is a culture as much as a process, however it is helpful to use frameworks, tools and guidelines to focus the team on critical analyses of quality for a given trial. The steps above can support this focused QbD evaluation as well as define CtQ factors for your study.

Back to all posts.