Your Trusted 24 Hours Service Provider!
Mivado GlobalPerformance
  • Number #1 Provider
    of Industrial Solution
  • Professional Certification
    Lean Six Sigma Black Belt | Project Management (PMI)
  • Quebec , Canada
    Certified Trainer
  • QMS Certification
    ISO 9001 : 2015
By - Kossi Molley (he/him)

Innovation in Pharmaceutical Risk Assessment

Innovation in Pharmaceutical Risk Assessment: Using FMEA to Identify Critical Material Attributes (CMAs).

A Step-by-Step Walkthrough: Learn how to apply FMEA in pharmaceutical development to identify CMAs, with a worked example of wet granulation before and after DoE.

How to Use FMEA to Identify Critical Material Attributes (CMAs)?  Why Is FMEA Still the Most Practical Innovation in Pharmaceutical Risk Assessment?

In Quality by Design (QbD), the real innovation isn’t a new tool; it is using an established tool at the right moments with the right data. Failure Mode and Effects Analysis (FMEA) is exactly that. Originally standardised for engineering use (IEC 60812:2018), FMEA is cited in ICH Q9 as a recognised quality risk management tool, and it fits neatly into the product development logic of ICH Q8(R2) and ICH Q11 [1–4].

The most effective teams do not run FMEA once and file it. They run it twice:

1. Before experimentation, to screen many material attributes and decide which deserves a Design of Experiments (DoE).
2. After experimentation, to confirm that the control strategy has genuinely reduced risk to an acceptable level.

This article walks through both passes using a worked wet granulation example, so you can apply the method straight away to your own CMA identification exercise.

Key Definitions (Quick Refresher)

• CQA (Critical Quality Attribute): a physical, chemical, biological or microbiological property that should be within an appropriate limit to ensure the desired product quality [1].
• CMA (Critical Material Attribute): a property of an input material (API or excipient) whose variability has a meaningful impact on a CQA. CMAs are commonly discussed together with Critical Process Parameters (CPPs) in the QbD literature [1, 5].
• FMEA: a structured method that scores each failure mode on Severity (S), Occurrence (O) and Detectability (D). The product of the three gives the Risk Priority Number (RPN) [3, 6].

The Worked Example: A High-Shear Wet Granulation Process

Scenario (illustrative, not a real product): an immediate-release tablet containing a poorly soluble, low-dose API, manufactured by high-shear wet granulation, fluid-bed drying, milling, lubrication and compression.

Target CQAs: dissolution, content uniformity, assay and degradation products, tablet hardness and friability.

All scores and results below are hypothetical and are provided to teach the method. Always base your own ratings on prior knowledge, literature, supplier data, and experimentation [2, 3].

Step 1: Define the Scope and the Scoring Scales

Before scoring anything, agree on the following as a cross-functional team (formulation, analytical, manufacturing, QA, regulatory):

• Scope: material attributes of the API and excipients only (process parameters are assessed in a separate or parallel FMEA).
• Scales: 1–10 for each of S, O and D, with written anchor definitions. For example, Severity 9–10 means a direct impact on patient safety or efficacy; Detectability 10 means the failure cannot be detected before release.
• Action thresholds: for example, RPN above 100, or any Severity of 8 or higher regardless of RPN, triggers further study.

The second threshold matters. A high-severity item with a low occurrence can produce a deceptively small RPN, so a severity override is good practice. ICH Q9(R1) also stresses managing subjectivity in risk scoring, which is why documented scale definitions and a facilitated team session are essential [2].

Step 2: List Material Attributes and Failure Modes

Using a process map and the Quality Target Product Profile (QTPP), list each input material and its potentially variable attributes. For each, describe the failure mode (what could go wrong) and its effect on the CQAs.

Step 3: First FMEA Pass – Screening Potential CMAs

At this stage, the team relies on prior knowledge, supplier information, and literature. Detectability reflects current controls (for example, supplier certificate of analysis only).

Result of pass one: from seven candidates, five are carried forward as potential CMAs. Notice that no attribute is called “critical” yet. Criticality is confirmed with data, not with opinion [1, 5].

Step 4: Use the FMEA Output to Design the DoE

The FMEA tells you what to put in the experimental design. A practical approach:

1. Choose a screening design (for example, a fractional factorial) that includes the five flagged attributes plus the two or three highest-risk process parameters, such as binder liquid amount and wet massing time.

2. Define responses linked to the CQAs: dissolution at a discriminating time point, content uniformity, granule size distribution, hardness.

3. Analyse main effects and interactions, and judge significance against the analytical and process variability.

This links the risk assessment directly to the design space and knowledge-generation approach described in ICH Q8(R2) [1].

Hypothetical DoE conclusions:

• API D90 and binder viscosity grade significantly affect dissolution ? confirmed CMAs.

• Mg stearate surface area interacts with lubrication time on hardness and dissolution ? confirmed CMA, controlled together with the lubrication CPP.

• MCC moisture significantly affects granule size ? confirmed CMA, controlled via a loss-on-drying limit.

• API polymorph showed no change within the tested range, but severity remains high ? retained as a controlled attribute by specification.

Step 5: Define the Control Strategy

Translate the findings into actions, in line with ICH Q10 and Q11 expectations for a lifecycle control strategy [4, 7]:
• API D90: specification NMT 40 µm (illustrative) with supplier agreement and incoming testing.
• Binder: defined viscosity range in the material specification.
• MCC: incoming loss-on-drying limit and an in-process moisture check.
• Mg stearate: specific surface area range and a fixed lubrication time.
• API polymorph: confirmed on each API batch or on a justified periodic basis.

Step 6: Second FMEA Pass – Verifying Risk Reduction

Now repeat the FMEA using the experimental knowledge and the new controls. Here is a key principle: Severity usually does not change, because the consequence of a failure is inherent to the product. Risk is reduced by lowering Occurrence (better control, tighter specifications) and Detectability (better testing).

All residual RPNs fall below the 100 threshold, and the high-severity items have a low combined O × D. The team documents the rationale, records the confirmed CMAs, and feeds them into the process validation strategy and ongoing monitoring [8].

Common Pitfalls to Avoid

• Treating RPN as absolute truth. RPN is a ranking aid, not a measurement. Use severity overrides and scientific justification [2, 6].
• Scoring without data. Every score needs a documented rationale (literature, experience, experiment).
• Running FMEA only once. Without the second pass, there is no evidence that the control strategy worked.
• Mixing material and process risks in one table. Keep CMAs and CPPs distinct, then link them through interactions.
• Forgetting the lifecycle. Revisit the FMEA when suppliers, scale, or site change, as expected under ICH Q9(R1) and Q10 [2, 7].

Key Takeaways

1. FMEA is a practical, regulator-recognised way to screen and confirm CMAs.
2. Use it twice: first to focus your DoE, then to demonstrate residual risk reduction.
3. Reduce risk through Occurrence and Detectability, not by rescoring Severity.
4. Document scales, rationale, and decisions to keep the assessment defensible and reproducible.

Take the Next Step

Ready to make your FMEA and CMA identification more consistent, traceable, and audit-ready?

Visit  Mivado GlobalPerformance to discover how Mivado can support your pharmaceutical risk assessment and QbD activities, and get in touch to discuss your project.

Leave a Reply