HPLC method validation under ICH Q2(R2) and Q14: a practical guide for peptide purity and assay methods

A peptide HPLC result is only as good as the evidence that the method can produce it. Method validation is that evidence. Since 2024 it has been governed in the ICH regions by a revised guideline, ICH Q2(R2), and a new companion, ICH Q14, which moves much of the thinking into method development.

This guide applies both to the two methods that matter most for a peptide: the related-substances (purity) method and the assay. It covers what each validation characteristic means in practice, how system suitability under the chromatography chapter harmonised between the Japanese, European and US pharmacopoeias (JP 2.00, USP <621>) fits in, and what makes a method stability-indicating. For how peptide HPLC methods are built in the first place, see our guide to peptide purity testing.

What is ICH Q2(R2), and when did it take effect?

ICH Q2(R2) “Validation of Analytical Procedures” and ICH Q14 “Analytical Procedure Development” were both adopted by the regulatory members of the ICH Assembly at Step 4 on 1 November 2023 [1], [2]. Regional implementation followed:

Region	Q2(R2)	Q14	Ref
ICH	Step 4 adoption 1 November 2023	Step 4 adoption 1 November 2023	[1], [2]
European Union (EMA)	EMA/CHMP/ICH/82072/2006; legal effective date 14 June 2024	EMA/CHMP/ICH/195040/2022; effective from 14 June 2024	[3], [4]
United States (FDA)	Final guidance, March 2024	Final guidance, March 2024	[5], [6]

In the US, FDA’s 2015 guidance on analytical procedures and methods validation gives recommendations on how analytical procedures and validation data are submitted [7]. For synthetic peptides in the EU, EMA’s 2025 guideline sets “no additional requirements” for validation of analytical procedures beyond those for other chemical substances [8], so Q2(R2) is the rulebook. What the peptide guideline adds is a demanding view of specificity, covered below.

Q2(R2) applies to analytical procedures “used for release and stability testing of commercial drug substances and products”, its principles can be applied “in a phase-appropriate manner” during clinical development, and the guideline states that “approaches other than those set forth in this guideline may be applicable and acceptable with appropriate science-based justification” [1].

What changed in Q2(R2)?

Several concepts are new or renamed, all from the text of the guideline [1]:

• Performance characteristics. The guideline’s glossary notes that “previous ICH Q2 versions referred to this as VALIDATION CHARACTERISTIC”.
• Reportable range and working range. Range is now defined in terms of reportable results, with a working range describing what is presented to the instrument. The two “could be identical”.
• Response and calibration model. Linearity sits under a broader “response” characteristic that allows linear, non-linear and multivariate calibration models.
• Combined accuracy and precision. Accuracy and precision can be evaluated separately or “against a combined performance criterion”, using a prediction, tolerance or confidence interval.
• Technology-inherent justification. Specificity can sometimes be justified by technical parameters, such as “resolution of isotopes in mass spectrometry”.
• Platform procedures and lifecycle. Validation testing “can be abbreviated, if scientifically justified” for an established platform procedure used for a new purpose; revalidation, transfer and co-validation are framed as lifecycle activities.

Which characteristics does each type of peptide method need?

Q2(R2) Table 1 sets out what is normally tested for each kind of procedure [1]:

Characteristic	Identity (e.g. LC-MS, RRT)	Impurity, quantitative (related substances)	Impurity, limit test	Assay (content)
Specificity	Yes	Yes	Yes	Yes
Response (calibration model)	No	Yes	No	Yes
Lower range limit	No	Quantitation limit	Detection limit	No
Accuracy	No	Yes	No	Yes
Repeatability	No	Yes	No	Yes
Intermediate precision	No	Yes	No	Yes

Q2(R2) adds that a lack of specificity in one procedure “should be compensated by one or more other supporting analytical procedures, unless appropriately justified” [1]. For peptides, that footnote describes the normal case: EMA recommends at least two orthogonal identity methods [8].

Validation characteristics applied to a peptide purity and assay method

Characteristic	What Q2(R2) asks for	How it is usually shown for a peptide RP-HPLC method	Ref
Specificity/selectivity	Absence of interference, or comparison with an orthogonal procedure; for separations, resolution of “the two components which elute closest to each other”	Resolution of the critical pair (often a deamidated or diastereomeric impurity next to the main peak); spiked and stressed samples; LC-MS or a second, orthogonal separation	[1], [8]
Response (linearity)	Plot of signal versus concentration; “a minimum of five concentrations appropriately distributed across the range”	Five or more levels across the reportable range for the main peak and for impurity reference standards where available	[1]
Reportable range	Derived from the specification (see the next table)	Assay 80–120%; impurities from the reporting threshold to 120% of the limit	[1]
Accuracy	Comparison with a reference material, spiking study or orthogonal procedure; for example 3 concentrations with 3 replicates each; mean recovery with a confidence interval	Spiking impurity standards into the peptide; for assay, comparison with a characterised reference standard	[1]
Repeatability	At least 9 determinations across the range, or at least 6 at 100%	Six preparations of one batch, or 3 × 3 across the range	[1]
Intermediate precision	Effects of days, analysts, equipment and environment; design of experiments encouraged	Second analyst, second instrument, different day and column lot	[1]
Reproducibility	Inter-laboratory trial; usually not required for submission	Needed for pharmacopoeial procedures or multi-site testing	[1]
Detection and quantitation limits	S/N 3:1 (DL) and at least 10:1 (QL), or DL = 3.3σ/S and QL = 10σ/S; QL at or below the reporting threshold	QL at or below 0.1% of the main peak for EU peptide impurity methods	[1], [8]
Robustness	Deliberate variations of parameters and stability of solutions; usually evaluated during development	Column temperature, gradient slope, mobile phase pH and ion-pair concentration, column lot, solution stability	[1], [2]

Specificity: the hard part for peptides

For small molecules, specificity is often a formality. For peptides it is the whole game, because the impurities are near-copies of the main molecule. Q2(R2) gives the general routes: “absence of interference or comparison of results to an orthogonal procedure”, and for a purity or impurity test, discrimination “by stressing or spiking product to achieve appropriate levels of impurities or related substances” [1]. Where impurities cannot be prepared or isolated, results can be compared with an orthogonal procedure [1].

EMA’s peptide guideline raises the bar in three ways [8]:

1. Methods for peptide-related impurities must at least meet the Ph. Eur. reporting threshold of 0.1% for synthetic peptides.
2. If one method cannot separate all peptide-related impurities, “additional independent method(s) may be needed”, and “when co-eluting impurities are observed as one peak the qualification threshold of 1.0% applies unless otherwise justified”.
3. “Stability indicating properties and mass balance of the method(s) need to be studied”, and “control of diastereomers may require the development of specific methods”.

Diastereomers and isoaspartate forms have the same mass as the main peptide, so mass spectrometry cannot rescue a method that does not separate them. Our guide to peptide degradation pathways lists which degradants a peptide method should be challenged with.

Reportable range for peptide methods

Q2(R2) Table 2 gives the recommended ranges [1]:

Use of the procedure	Low end of reportable range	High end of reportable range
Assay of a product	80% of declared content or of the lower specification criterion	120% of declared content or of the upper specification criterion
Impurity	Reporting threshold	120% of the specification acceptance criterion
Purity (as area %)	80% of the lower specification acceptance criterion	Upper specification criterion or 100%

Where assay and impurities are determined in a single run with one standard, “linearity should be demonstrated for both the reporting level of the impurities and up to 120% of the specification acceptance criterion for assay” [1]. Peptide methods are often built exactly that way, with the main peak and its related substances measured in one gradient.

Detection and quantitation limits

Q2(R2) offers three ways to set the lower range limits: signal-to-noise (3:1 for DL, at least 10:1 for QL), calculation from the standard deviation of the response and the slope (DL = 3.3σ/S, QL = 10σ/S), or direct validation of the QL by accuracy and precision at that level [1]. Estimated limits should then be confirmed with samples near the limit, though this can be omitted with justification when the QL is “approximately 10 times lower than” the reporting limit [1]. The binding requirement for impurity methods is that “the QL for the analytical procedure should be equal to or below the reporting threshold” [1].

Analytical procedure development under ICH Q14

Q14 describes two ways to develop a method. The minimal (traditional) approach “remains an acceptable approach”; the enhanced approach adds systematic risk assessment, multivariate experiments and documented parameter ranges [2].

Its key tools [2]:

• Analytical target profile (ATP). “A description of the intended purpose of the analytical procedure, appropriate details on the product attributes to be measured and relevant performance characteristics with associated performance criteria”. Submitting an ATP is optional.
• Robustness. “A measure of its capacity to meet the expected performance criteria during normal use”, tested by deliberate variations, including the stability of sample preparations and reagents. If it was evaluated during development, “it does not need to be repeated during validation”.
• Proven acceptable ranges (PARs) and method operable design regions (MODRs). Once approved, “moving within an approved PAR or MODR does not require regulatory notification”.
• Analytical procedure control strategy. Defined before validation and confirmed after it, including the parameters that need control and the system suitability test.

For a peptide gradient method, the parameters that usually earn a place in a robustness study are the ones that shift the critical pair: column temperature, gradient slope, mobile-phase pH and ion-pair reagent concentration, and column batch. Q2(R2) explicitly lets data from such development studies, for example a design of experiments, count toward validation [1].

System suitability under the harmonised chromatography chapter (JP 2.00, USP <621>)

Validation shows the method can work; system suitability shows it worked on the day. Q14 defines the SST as part of the analytical procedure that verifies “the measurement system and the analytical operations” are fit for purpose “during the time period of analysis” [2]. The chromatography chapter, harmonised between the Japanese, European and US pharmacopoeias, states that “no sample analysis is acceptable unless the suitability of the system has been demonstrated” [9].

The harmonised text lists the parameters that may be used: “column plate number, retention factor (mass distribution ratio), system repeatability, signal-to-noise, symmetry factor and resolution/peak-to-valley ratio” [9]. Its default requirements [9]:

Parameter	Harmonised definition or default	Note for peptide methods
Resolution	Rs = 1.18 (tR2 − tR1) ÷ (wh1 + wh2), using peak widths at half height	Set on the critical pair; the Japanese text adds, as a local provision, that Rs ≥ 1.5 is baseline separation
Symmetry factor	AS = w0.05 ÷ 2d; unless otherwise stated, 0.8 to 1.8 for the peak used for quantitation	Applies to the peak used for quantitation
System repeatability (assay)	Maximum %RSD calculated for 3 to 6 replicate injections of the reference solution	See the table below
System sensitivity	QL (S/N 10) at or below the reporting threshold	For EU peptide impurity methods, at or below the 0.1% reporting threshold [8]
Signal-to-noise	S/N = 2H ÷ h, with noise measured over 20 times the peak width at half height (at least 5 times if 20 is not obtainable)	Matters for impurity peaks near the reporting threshold

Maximum permitted %RSD of replicate injections for an assay, where B is the upper content limit minus 100% [9]:

B (%)	n = 3	n = 4	n = 5	n = 6
2.0	0.41	0.59	0.73	0.85
2.5	0.52	0.74	0.92	1.06
3.0	0.62	0.89	1.10	1.27

USP has since proposed narrowing two of these: the peak symmetry requirement would apply only to organic impurities, related substances tests and assays, and be waived for impurity tests that prescribe a %RSD check; system sensitivity would apply only when the monograph states a reporting threshold [10]. The proposal is based on the version of <621> official as of 1 December 2024 [10].

Two harmonised provisions matter for complex peptide chromatograms. For “complex chromatographic profiles (e.g., for biotechnological/biological products)”, visual comparison of profiles can be used as a system suitability test when stated in the monograph [9]. And gradient methods are treated cautiously: adjustments are “more critical” than for isocratic methods because they “may shift some peaks to a different step of the gradient”, “potentially causing partial or complete coelution of adjacent peaks or peak inversion” [9]. For isocratic methods, column length and particle size may change provided the L/dp ratio stays within −25% to +50% of the prescribed value [9].

Stability-indicating methods and forced degradation

Q2(R2) defines a stability-indicating procedure as a “validated quantitative analytical procedure that can detect changes in relevant quality attributes of a product during storage” [1]. To demonstrate it, “samples containing relevant degradation products should be included” in the specificity study: spiked samples, samples “exposed to various physical and chemical stress conditions”, and aged or stressed product [1].

The stress conditions come from ICH Q1A(R2): heat in 10 °C increments above the accelerated condition, humidity of 75% RH or more, oxidation, photolysis and hydrolysis across a wide pH range, with stress testing intended in part to “validate the stability indicating power of the analytical procedures used” [11]. ICH Q1B adds photostability, where forced degradation testing serves “method development purposes and/or degradation pathway elucidation” [12]; for peptides and proteins more broadly, standard forced degradation procedures “are largely unavailable, except for photostability” [13]. Degradation of 5–20% “has been accepted as reasonable for validation of chromatographic assays” [14]. Bakshi and Singh list separation of all degradation products and establishment of mass balance among the critical issues in developing stability-indicating methods [15], which matches EMA’s requirement that stability-indicating properties and mass balance be studied [8].

Published peptide examples

Two recent papers show what validated peptide methods report, and why the method’s purpose matters.

Liraglutide related substances. Zhang and colleagues developed a gradient RP-HPLC method on a C8 column with phosphate buffer and acetonitrile that separated five specified impurities from the drug substance. They reported resolution between the main peak and each impurity greater than 1.0 and a main-peak tailing factor below 1.0, and validated specificity, linearity, LOD, LOQ, accuracy, precision and robustness per ICH [16].

Semaglutide and tirzepatide assay. Khalil and colleagues used an isocratic method on a C18 column with 0.1% formic acid and acetonitrile (30:70) at 220 nm, with the two peptides eluting at 1.42 and 1.68 minutes, linearity of 1–500 µg/mL (r > 0.9999) and LODs of 16 ng/mL for semaglutide and 10 ng/mL for tirzepatide; acid, base, oxidative and photolytic stress were applied to show quantification without interference from degradants [17].

Both were reported as validated under ICH guidelines, but they answer different questions. The first is a related-substances method built to separate named impurities. The second is an assay of the main peak. Q2(R2) validates a method for its intended purpose [1], and EMA expects peptide impurity methods to separate peptide-related impurities down to the 0.1% reporting threshold [8]. A method validated for assay is not automatically fit to report purity.

Validation over the method lifecycle

Validation is not a one-off event. Q2(R2) says changes to a validated procedure may need “partial or full revalidation”, with the extent based on the performance characteristics affected [1]. Transfer to another laboratory calls for partial or full revalidation and/or comparative analysis of representative samples, and co-validation across sites “could also satisfy the requirements of analytical procedure transfer” [1]. Q14 recommends ongoing monitoring of method outputs to spot trends [2].

A validation checklist for a peptide HPLC method

1. Define the purpose (assay, related substances, or both in one run) and, ideally, an ATP [2].
2. Identify the critical pair and the degradants from forced degradation; show they are resolved or detected by an orthogonal method [1], [8].
3. Set the reportable range from the specification, including the reporting threshold if impurities are measured [1].
4. Show response with at least five levels [1].
5. Show accuracy (for example 3 × 3) and precision (repeatability, intermediate precision) with confidence intervals [1].
6. Confirm the QL at or below the reporting threshold [1].
7. Document robustness from development [2].
8. Fix the SST: critical-pair resolution, symmetry, replicate-injection %RSD and sensitivity [9].
9. Plan the lifecycle: change control, transfer and revalidation triggers [1].

When you assess a peptide supplier, ask which method generated the purity figure on the COA and whether it was validated for related substances; see how to qualify a peptide API supplier.

How Winhycare Pharma supplies peptides

Winhycare Pharma supplies research peptides from Hyderabad, India, as research-use-only material, at purity of ≥99% as stated in our product brochure. Semaglutide, tirzepatide, retatrutide, cagrilintide and BPC-157 are also supplied as API to licensed manufacturers; for semaglutide, we support US-DMF and CTD Module 3 documentation. See semaglutide API, tirzepatide API and peptide APIs, or enquire about semaglutide API. For research material, request research peptides.

For laboratory research use only. Not for human or veterinary use.