Implementing Quality by Design in Clinical Trials

Part of Good Clinical Practice (GCP) — ICH E6(R3): Risk Management and Quality by Design.

About 7 minutes. Free, in your language, no account needed.

Why this matters

Most quality problems at sites were designed in months earlier. A visit window too narrow for working participants. A questionnaire nobody can complete in the time allowed. A dose diary that assumes literacy the population does not have. Once the trial starts, all you can do is detect the damage. International Council for Harmonisation (ICH) guideline E6(R3) for Good Clinical Practice (GCP) asks the sponsor to prevent it instead.

Quality means fitness for purpose

Principle 6 states that "Quality should be built into the scientific and operational design and conduct of clinical trials."

Principle 6.1 then defines the word:

"Quality of a clinical trial is considered in this guideline as fitness for purpose."

This is a change of standard. Quality is not perfection, and not a complete absence of error. Quality is whether the trial is fit to answer its question and protect its participants. A trial with a few transcription errors in an exploratory variable can be fit for purpose. A trial with a flawless database and an unreliable primary endpoint is not.

Principle 6.2 gives the method: "Quality by design involves focusing on critical to quality factors of the trial in order to maximize the likelihood of the trial meeting its objectives."

Quality by design is prospective; quality control is detective

This is the practical distinction to carry away from the module.

• Quality by design — you change the design so the risk does not arise. It happens before the first participant is enrolled, and removes work rather than adding it.
• Quality control — you check work after it is done, to find and correct problems. It runs throughout the trial.

Both are required. Neither replaces the other. But they act at different times and cost differently. Designing out a redundant assessment costs one protocol decision. Detecting that same assessment going wrong at forty sites costs two years of queries.

E6(R3) is explicit about what quality control is in a trial. Section 3.11.3 says quality control "should be applied using a risk-based approach to each stage of the data handling to ensure that data are reliable and have been processed correctly." It then names the activities:

"Within clinical trials, monitoring and data management processes are the main quality control activities."

Section 3.11.4 adds that "Monitoring is one of the principal quality control activities."

Quality assurance is different again. Section 3.11.1 says it "includes implementing risk-based strategies to identify potential or actual causes of serious noncompliance" so that corrective and preventive actions can follow. Quality assurance asks whether the system works. Quality control checks its output.

Designing quality in

Section 3.1.2 puts the duty at design time: "Sponsors should incorporate quality into the design of the clinical trial by identifying factors that are critical to the quality of the trial and by managing risks to those factors."

Three further design duties support it.

• Section 3.1.3 — sponsors should consider inputs from a wide variety of interested parties, for example healthcare professionals and patients, when developing protocols and participant-facing materials.
• Section 3.1.4 — all aspects of the trial should be operationally feasible, avoiding unnecessary complexity, procedures and data collection.

Notice how much of quality by design is subtraction. The guideline repeatedly asks for less: less complexity, fewer procedures, less data, less burden.

Risk control, communication and review

Lesson 1 of this module covered risk identification and evaluation. Quality by design is where the remaining steps of section 3.10.1 are carried out.

• 3.10.1.3 Risk Control — control should be proportionate to the importance of the risk. The guideline states where mitigation lives: "in protocol design and implementation, monitoring plans, agreements between parties defining roles and responsibilities, and training." Protocol design is listed first.
• 3.10.1.4 Risk Communication — the sponsor should document and communicate identified risks and mitigating activities to those involved in taking action or affected by them. A risk nobody at the site has been told about is not controlled.
• 3.10.1.5 Risk Review — the sponsor should periodically review control measures to check whether they remain effective and relevant, "taking into account emerging knowledge and experience." Additional controls may be added as needed.

Principle 7.3 says the same thing in one line: risks to critical to quality factors "should be managed proactively and adjusted when new or unanticipated issues arise once the trial has begun."

Risk reporting into the clinical trial report

Quality management does not end when the last participant leaves. Two reporting duties close the loop.

Section 3.10 requires that "The sponsor should describe the quality management approach implemented in the trial in the clinical trial report," cross-referring to ICH E3.

Section 3.10.1.6 Risk Reporting is narrower and harder. The sponsor should summarize important quality issues, including instances where the pre-specified acceptable ranges of section 3.10.1.3 were exceeded, together with the remedial actions taken, and document them in the clinical trial report.

That is a transparency duty. Quality problems, and what was done about them, become part of the permanent record regulators read.

In practice

ABC Pharmaceuticals is designing a Phase III trial in chronic heart failure. Priya is the operations lead. She runs a quality by design workshop before the protocol is finalized. A cardiologist, a data manager, a statistician, a former trial participant and two site coordinators attend.

The draft protocol requires a six-minute walk test at every one of nine visits. The coordinators say the corridor is shared with a busy clinic, so the test often slips outside its window. The statistician confirms the walk test is a secondary endpoint and that four time points would answer the question.

The team applies section 3.10.1.2. Likelihood of deviation is high. Detectability is good, because deviations are logged. Impact on the primary endpoint is low, but impact on participant burden is real.

Two decisions follow. The walk test moves from nine visits to four, which is risk control by protocol design. The remaining visits get a wider window that the sites agree is feasible.

Compare the alternative. Keep nine walk tests, find the deviations through monitoring, raise queries, retrain sites, write it all up. That is quality control doing the work design should have done.

At the end of the trial, Priya's team reports the quality management approach under section 3.10 and the important quality issues and remedial actions under section 3.10.1.6.

Key takeaways

• Principle 6.1 defines the quality of a clinical trial as fitness for purpose, not as the absence of all error.
• Quality by design is prospective and changes the trial so risks do not arise; quality control is detective and finds problems after work is done.
• Section 3.11.3 identifies monitoring and data management processes as the main quality control activities in clinical trials.
• Section 3.10.1.3 places risk mitigation first in protocol design and implementation, then in monitoring plans, agreements and training.
• Section 3.10.1.4 requires risks and mitigations to be communicated to the people who must act on them or who are affected by them.
• Section 3.10 requires the sponsor to describe the quality management approach in the clinical trial report, and section 3.10.1.6 requires important quality issues and remedial actions to be reported there.
• Quality that is designed in costs one decision; quality that is inspected in costs the whole trial.