ICH Q3A/Q3B: Why 0.1% Isn't a Universal Impurity Threshold

Ask a CMC team what the ICH impurity threshold is and most will answer with a single number: 0.1%. That figure does not appear anywhere in ICH Q3A(R2)'s default threshold table. Q3A(R2), which governs organic impurities in new drug substances, and Q3B(R2), its counterpart for degradation products in new drug products, both set three separate thresholds — reporting, identification, and qualification — and every one of them shifts with the drug's maximum daily dose. Treating any single figure as universal is the error that shows up in method validation packages and CMC sections alike.

Three thresholds, not one number

For a new drug substance with a maximum daily dose of 2 grams or less — the band that covers most oral small-molecule products — Q3A(R2) sets the reporting threshold at 0.05%, the identification threshold at 0.10% or 1.0 mg total daily intake (whichever is lower), and the qualification threshold at 0.15% or 1.0 mg total daily intake (whichever is lower). Cross the 2 g/day line and every figure tightens: reporting drops to 0.03%, identification and qualification both fall to 0.05%. Q3B(R2) applies the same three-tier logic to degradation products in the finished drug product, using its own dose-banded table rather than borrowing Q3A(R2)'s numbers directly.

The dose tier most teams forget to check

The 2 g/day breakpoint is not a footnote. A high-dose product — an antibiotic or a metformin-class combination, for instance — can cross it, and every threshold that applies drops when it does. A team that carries the ≤2 g/day figures into a high-dose product's specification is not being conservative; it is applying the wrong table. The same discipline applies on the Q3B(R2) side: degradation products in the finished product follow their own dose-tiered thresholds, distinct from the drug-substance figures in Q3A(R2), because a degradation product forming in the tablet is not the same regulatory question as an impurity carried in from synthesis.

• Reporting threshold. The level above which an impurity must be listed in the specification and analytical results — the lowest bar, and the one most labs already test to.
• Identification threshold. The level above which the impurity's structure must be characterized — a laboratory and analytical-method question.
• Qualification threshold. The level above which the impurity needs safety justification — a nonclinical study, documented clinical exposure, or a literature-based argument — before it can be accepted.
• The dose breakpoint. Every one of the three thresholds above changes value once maximum daily dose crosses 2 g/day; confirm the dose band before quoting a percentage.

Crossing the qualification threshold is not a rejection. It is a data requirement — and the teams that treat it as the end of the conversation are the ones who file late. Why qualification is a workstream, not a verdict

Three impurity frameworks, three different logics

Q3A/Q3B, ICH Q3D, and ICH M7 are frequently treated as one impurity control program with three names. They are not. Q3A/Q3B's thresholds are dose-relative percentages scaled to maximum daily dose. Q3D sets elemental impurity limits as fixed permitted daily exposures (PDEs) tied to the route of administration, independent of the drug substance's own dose. M7 sets mutagenic impurity limits from a toxicological threshold of concern, and a mutagenic or unusually toxic impurity requires qualification regardless of where it sits on the Q3A/Q3B percentage scale — the ordinary reporting and identification thresholds do not exempt it. A CMC impurity control strategy that runs all three through a single dose-relative percentage logic will misapply at least one of them.

What to do now

None of this is exotic once the tables are in front of the team; the failure mode is almost always a remembered percentage substituting for the actual table. Building a CMC regulatory affairs impurity control strategy around the correct dose band, applying quality risk management principles to route each impurity type to its governing framework, and confirming the strategy holds under a continuous manufacturing control strategy where impurity profiles can shift with process changes are what keep the threshold conversation from becoming a filing-week surprise.