Calibrators and Controls Traceability for IVD Kits

Why Calibrator Traceability Determines Whether Your IVD Results Are Trustworthy

When two hospitals measure the same patient's troponin I using different IVD kits from different manufacturers, and get results that differ by 40%, the consequences are not academic — they are clinical. A patient may be diagnosed with myocardial infarction at one hospital and sent home from another. This inconsistency is not rare. It happens because troponin I is a complex protein measurand without a single universally accepted reference measurement procedure, and different manufacturers calibrate their assays against different reference materials, or in some cases against proprietary in-house standards with no documented traceability chain.

Metrological traceability is the framework that connects the value assigned to a calibrator in an IVD kit to a higher-order reference — ideally a primary reference material or a reference measurement procedure recognized by an international body. ISO 17511:2020, "In vitro diagnostic medical devices — Requirements for establishing metrological traceability of values assigned to calibrators, trueness control materials and human samples," is the foundational standard that defines how this chain must be constructed and documented. FDA recognizes ISO 17511:2020 under Recognition Number 7-305, and compliance with this standard is expected for IVD manufacturers marketing in both the US and EU.

This guide covers the full scope of calibrator and control traceability for IVD kits: the calibration hierarchy models defined by ISO 17511, the reference materials and reference measurement procedures available through the JCTLM database, the concept of commutability and why it matters, measurement uncertainty budgeting, value assignment protocols, and the practical steps an IVD manufacturer must take to build and maintain a defensible traceability chain.

The Calibration Hierarchy: Six Models in ISO 17511:2020

ISO 17511:2020 defines six calibration hierarchy models (CH1 through CH6), each representing a different level of available reference infrastructure. The model applicable to a given measurand depends on what higher-order references exist.

CH1 — Traceability to SI Unit via Primary Reference Material and Reference Measurement Procedure

This is the gold standard. A primary reference material (typically a highly purified, well-defined chemical substance) is available, and a primary reference measurement procedure exists to value-assign a primary calibrator. The traceability chain runs:

Primary RM (m.1) → Primary calibrator (m.2), value-assigned by primary RMP (p.2) → Secondary RM / calibrator (m.3), value-assigned by secondary RMP (p.3) at a calibration laboratory → Manufacturer's working calibrator (m.4), value-assigned by manufacturer's standing measurement procedure (p.4) → Manufacturer's product calibrator (m.5), value-assigned by manufacturer's standing MP (p.5) → End-user sample result

This model applies to well-defined, low-molecular-weight analytes such as glucose, creatinine, electrolytes (Na⁺, K⁺, Cl⁻), and some hormones (thyroxine, cortisol). Metrological traceability extends to the SI unit (usually mol/L or g/L) with the lowest achievable measurement uncertainty.

CH2 — SI Traceability via Reference Measurement Procedure Only (No Primary RM)

For some measurands, no primary reference material exists in purified form, but an internationally agreed reference measurement procedure defines the measurand. This model is common for enzymes (where catalytic activity, not mass, is the measured quantity) and some proteins. The measurand is defined operationally by what the reference method measures. The calibration hierarchy still achieves traceability to SI units, but through the reference method rather than a pure substance standard.

CH3 — Traceability to an International Conventional Calibrator

When neither a primary RM nor a primary RMP exists for a given measurand, but an international conventional calibrator is available — typically a WHO International Standard or an IFCC-endorsed material — traceability is established to the assigned value of that conventional calibrator. The assigned value is arbitrary (not expressed in SI-derived units per mole of substance) but is agreed upon internationally.

This model applies to many complex protein measurands, including:

• Coagulation factors (WHO International Standards for Factors II, VII, VIII, IX, X)
• Some tumor markers where no pure substance reference exists
• Various cytokines and growth factors

The critical requirement under CH3 is that the international conventional calibrator must be commutable with human samples for the measurement procedures it is intended to calibrate. Commutability must be validated by the calibrator producer before first release.

CH4 — Traceability to an International Consensus Protocol or Method

Some measurands have neither a reference material nor a reference measurement procedure, but an international consensus protocol defines how the measurement should be performed. The calibration hierarchy traces to the protocol specification rather than to a physical reference material.

CH5 — Traceability to a Manufacturer's In-House Calibrator with No Higher-Order Reference

For measurands where no higher-order reference system exists at all — no primary RM, no RMP, no international conventional calibrator — the manufacturer establishes traceability to its own in-house calibrator. The manufacturer must document why no higher-order reference is available, describe the in-house calibrator's characterization, and estimate measurement uncertainty based on available data.

This is the reality for many emerging biomarkers, proprietary assays, and some multiplexed panel tests. While CH5 provides the weakest traceability chain, the standard requires that manufacturers nevertheless document the hierarchy and estimate uncertainty.

CH6 — Traceability Through the Manufacturer's Own Measurement Procedure

In this model, the manufacturer defines the measurement procedure and the calibrator simultaneously. The measurand is operationally defined by the manufacturer's method. This model is typical for qualitative tests and some rapid tests where the output is a binary result (positive/negative) rather than a quantitative value.

The JCTLM Database and Reference Materials Infrastructure

The Joint Committee for Traceability in Laboratory Medicine (JCTLM) was established in 2002 as a collaboration between the Bureau International des Poids et Mesures (BIPM), the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC), and the International Laboratory Accreditation Cooperation (ILAC). Its mission is to promote global standardization of clinical laboratory test results through a shared reference measurement system.

The JCTLM maintains a publicly searchable database (jctlmdb.org) that lists:

• Category I: Reference Materials. Certified reference materials (CRMs) and international conventional calibrators that have undergone independent review and been found compliant with ISO 15194. This includes materials from NIST (USA), IRMM/EC/JRC (EU), NIBSC (UK, WHO International Standards), and other national metrology institutes.
• Category II: Reference Measurement Procedures. Methods that have been validated and reviewed for compliance with ISO 15193. These are the highest-order measurement procedures available for specific measurands.
• Category III: Reference Measurement Services. Calibration laboratories accredited to ISO 15195 and ISO 17025 that offer reference measurement services.

For an IVD manufacturer, the JCTLM database is the starting point for establishing a calibration hierarchy. The workflow is:

1. Search the database for the target measurand
2. Identify the highest available reference system component (CRM, RMP, or reference laboratory service)
3. Design the calibration hierarchy from the highest available component down to the product calibrator
4. Document the hierarchy in the technical file

If no JCTLM-listed reference exists for the measurand, the manufacturer must use CH5 or CH6 and document the rationale for the absence of higher-order references.

Commutability: The Bridge Between Reference Materials and Patient Samples

Commutability is the property of a reference material that means it behaves the same way as a native human clinical sample when measured by different measurement procedures. A commutable calibrator produces equivalent calibration regardless of which measurement procedure is used. A non-commutable calibrator may introduce bias that differs between methods, making results from different IVD kits non-equivalent even though both claim traceability to the same higher-order reference.

Commutability is assessed experimentally by measuring both the reference material and a panel of native clinical samples using two or more measurement procedures, then comparing the relationship between the reference material results and the clinical sample results. If the reference material falls within the prediction interval of the clinical sample regression, it is commutable for that pair of procedures.

Key facts about commutability:

• Commutability is measurement procedure-specific. A CRM may be commutable for 12 out of 15 measurement procedures and non-commutable for the remaining 3. The JCTLM listing includes commutability information for listed CRMs.
• Many older CRMs listed in the JCTLM database before 2010 have not been assessed for commutability. ISO 17511:2020 requires that manufacturers assess commutability before using a CRM as a trueness-transferring calibrator, even if the CRM is JCTLM-listed.
• When a CRM is found to be non-commutable for the manufacturer's measurement procedure, the manufacturer may apply a correction factor to the CRM's assigned value, but the correction must be validated and documented, and the additional uncertainty from the correction must be included in the uncertainty budget.
• Commutability assessment follows CLSI EP14-A3 and IFCC Working Group recommendations published in a series of papers in Clinical Chemistry (2018).

Measurement Uncertainty in the Calibration Hierarchy

Every value assignment in the calibration hierarchy carries measurement uncertainty. The uncertainty of the end-user calibrator (the calibrator that ships with the IVD kit) includes contributions from every step above it in the chain. ISO 17511:2020 requires that manufacturers estimate the combined standard uncertainty of values assigned to product calibrators.

The uncertainty budget for a product calibrator typically includes:

These contributions are combined using the root-sum-of-squares method (for independent contributions) or a more sophisticated model (for correlated contributions). The combined standard uncertainty (uc) is typically reported as an expanded uncertainty (U = k × uc, where k ≈ 2 for approximately 95% confidence).

For well-established measurands under CH1 (e.g., glucose, electrolytes), the expanded uncertainty of a product calibrator value might be 1–3% relative. For complex protein measurands under CH3 or CH5, the expanded uncertainty can be 10–20% or higher.

Value Assignment Protocols

Calibrator Value Assignment

The product calibrator's assigned value is determined by the manufacturer's standing measurement procedure, which traces upward through the calibration hierarchy. The value assignment protocol must specify:

• The number of replicates and runs
• The measurement procedure (instrument, reagent lot, operator)
• The reference calibrator used for calibration of the measurement procedure
• The acceptance criteria for replicate agreement
• The statistical method for determining the assigned value (mean, median, or weighted mean)

Typically, a product calibrator lot is value-assigned using at least 10 replicates across at least 3 independent runs. The between-run variability is captured in the uncertainty budget.

Control Value Assignment

Trueness control materials are value-assigned using the same traceability chain as calibrators, but their assigned values are used for monitoring (not for calibrating) the measurement system. Controls verify that the calibrated system produces results within expected limits.

Internal quality control (IQC) materials — the controls shipped with IVD kits — are typically assigned target values and acceptable ranges based on replicate measurements across multiple runs, days, operators, and reagent lots. The acceptable range accounts for both the measurement system's imprecision and the control material's own variability.

External quality assessment (EQA) / proficiency testing materials are produced by independent organizations and are used to compare results across laboratories and methods. ISO 15189 requires that laboratories participate in EQA programs where available.

Third-Party Controls

Microbiologics and other third-party control manufacturers emphasize that controls independent of the assay manufacturer provide more objective quality assessment. Assay manufacturer-supplied controls may share the same raw materials, processing, and value assignment methods as the kit itself, which means they can amplify systematic bias rather than detect it. ISO 15189 states that a "laboratory shall use quality control materials that react to the examining system in a manner as close as possible to patient samples," and third-party controls are specifically designed to fulfill this requirement.

Calibrator and Control Manufacturing Considerations

Matrix Selection

The matrix of a calibrator or control must be chosen to approximate the behavior of clinical samples in the measurement system. Common options include:

• Human serum or plasma pools. The most commutable matrix for serum/plasma assays, but limited in supply, subject to donor screening requirements, and potentially variable between pool lots.
• Animal serum or plasma. Less expensive and more available than human pools, but may not be commutable for all measurands (especially those with species-specific protein interactions).
• Buffer-based matrix. Defined chemical composition, reproducible, but generally not commutable with clinical samples for immunoassays.
• Spiked synthetic matrix. Analytes added to a defined base matrix to achieve target concentrations. Reproducible, but commutability must be verified for each analyte-matrix-measurement procedure combination.

Homogeneity

ISO 15194 and ISO Guide 35 require that reference materials demonstrate homogeneity — the property that any subsample taken from the bulk lot has the same analyte value within a defined uncertainty. For liquid calibrators and controls, homogeneity is assessed by testing multiple vials from throughout the fill sequence (beginning, middle, end) and analyzing the between-vial variability using ANOVA or equivalent methods. For lyophilized calibrators, both pre-lyophilization homogeneity and post-reconstitution homogeneity must be assessed.

Stability

Calibrator and control stability must be established under defined storage conditions. The stability claim must be supported by real-time stability data, with accelerated stability data used as supportive evidence (following ICH Q1A(R2) principles, adapted for IVD reagents).

Stability is assessed by comparing the measured value at defined time points to the value at time zero. Acceptance criteria typically require that the measured value remains within ± the expanded uncertainty of the assigned value throughout the claimed shelf life.

For lyophilized calibrators, stability is typically much longer than for liquid calibrators. A lyophilized calibrator stored at −20°C may have a shelf life of 3–5 years, while a liquid calibrator stored at 2–8°C may have a shelf life of 12–18 months.

Regulatory Documentation Requirements

EU IVDR (Regulation 2017/746)

Under the EU IVDR, the manufacturer's technical documentation (Annex II) must include:

• The calibration hierarchy, identifying all reference materials and reference measurement procedures used
• The measurement uncertainty of calibrator assigned values
• Evidence of commutability assessment for any CRM used as a trueness-transferring calibrator
• The metrological traceability chain, documented from the product calibrator to the highest available reference

The IVDR also requires that the Instructions for Use (IFU) include information about the metrological traceability of calibrators (Annex I, Chapter II, Section 12.1).

FDA Requirements

FDA recognizes ISO 17511:2020 (Recognition Number 7-305, entered December 21, 2020). For 510(k) submissions, the calibration traceability approach should be described in the device description. For PMA submissions, the calibration hierarchy and measurement uncertainty are part of the manufacturing section. FDA may request traceability documentation during pre-submission meetings or as part of an information request during review.

ISO 13485

ISO 13485:2016 Clause 7.5.6 requires validation of processes for production and service provision where the resulting output cannot be verified by subsequent monitoring or measurement. Value assignment of calibrators is typically considered a special process that requires validation. Clause 7.4.3 requires verification of purchased product — which includes the CRMs and reference materials used in calibration hierarchies.

Practical Pitfalls and How to Avoid Them

Using a CRM without commutability assessment. Many CRMs, especially those listed before 2010, have not been assessed for commutability. Using such a CRM as a trueness-transferring calibrator without your own commutability assessment is a regulatory gap. This is flagged explicitly in the JCTLM database entries, which indicate the edition of ISO 15194 under which the CRM was reviewed.

Ignoring the uncertainty contribution from value transfer. The uncertainty of the product calibrator is not just the uncertainty stated on the CRM certificate. It must include the uncertainty of every value transfer step in the hierarchy — the between-run precision of the manufacturer's standing MP, the fill variability of the calibrator vials, and the stability uncertainty. An incomplete uncertainty budget will not withstand regulatory scrutiny.

Changing calibrator lots without bridging. Each new lot of product calibrator must be value-assigned independently, and lot-to-lot equivalence must be demonstrated before the new lot is released to the field. The bridging protocol should include measurement of the new lot against the reference calibrator using the validated standing MP, with predefined acceptance criteria.

Failing to update the traceability chain when reference materials change. WHO International Standards are periodically replaced. When a WHO IS for a given measurand is replaced by a new preparation, the assigned unitage may change (sometimes significantly), and all manufacturers using that IS must update their calibration hierarchies, re-assign calibrator values, and assess the impact on clinical results. Failure to track these updates can result in silent shifts in reported patient values.

Confusing precision controls with trueness controls. Precision controls (sometimes called internal quality controls) monitor the repeatability and intermediate precision of the measurement system. Trueness controls have value assignments traceable to higher-order references and monitor accuracy (bias). Both are needed, but they serve different purposes, and regulatory expectations for their traceability documentation differ.