light-black-logo
View Platform
Request Demo
light-black-logo
View Platform
Request Demo

Regulations and guidance · Bioanalysis

ICH M10 bioanalytical method validation, section by section.

ICH M10 sets the expectations for validating bioanalytical methods and analyzing the study samples that support regulatory decisions. This page sets out what it asks of the calibration curve and the analytical run, linked to the primary documents.

Calibration curve rulesRead the white papers

May 24, 2022ICH M10 adopted at Step 4
6Minimum calibration levels, LLOQ to ULOQ
75%Minimum share of standards within limits
±15%Chromatographic accuracy, ±20% at the LLOQ

01 · What ICH M10 is and where it applies

What ICH M10 is and where it applies.

ICH M10, Bioanalytical Method Validation and Study Sample Analysis, was adopted at Step 4 on May 24, 2022. Under Section 1.1, the objective of validation is to show that a method is suitable for its intended purpose1.

Document Date Note
ICH M10 guideline, Step 4 May 24, 2022 Final harmonized text1.
ICH M10 Questions and Answers November 16, 2022 Nine questions, one on investigating ISR failures2.
FDA M10 guidance for industry November 2022 Final; announced November 7, 2022; replaces the 2019 draft; not revised since3,4,5.
FDA M10 Questions and Answers June 2024 FDA’s issue of the ICH Q&As6.
EMA, ICH guideline M10 (EMA/CHMP/ICH/172948/2019) January 21, 2023 In effect; supersedes EMEA/CHMP/EWP/192217/2009 Rev. 1 Corr. 27,8.
FDA, Bioanalytical Method Validation for Biomarkers January 2025; current version dated April 2026 Final; makes the M10 approach the starting point for biomarker assays9.
  1. In scope. Quantitative ligand binding assays (LBAs) and chromatographic methods such as LC and GC, typically with mass spectrometry detection, for chemical and biological drugs and their metabolites in biological samples. The studies are nonclinical toxicokinetic studies under GLP principles, nonclinical pharmacokinetic studies that stand in for clinical studies, and all phases of clinical trials, including comparative bioavailability and bioequivalence (BA/BE) studies (1.3)1.
  2. Out of scope. Biomarker bioanalysis and immunogenicity methods. For exploratory studies not used for regulatory decisions, the applicant decides the level of qualification (1.3)1.

02 · How the guideline is organized

How the guideline is organized.

Section What it covers Why it matters for calibration
1 Objective, background and scope Which assays and studies the rules govern.
2 Method development; full, partial and cross validation A written method and validation procedure, set in advance, in the study matrix (2.2.1).
3.1 to 3.2 Chromatography: reference standards and validation 3.2.4 sets the curve, regression model and accuracy limits.
3.3 Chromatography: study sample analysis Run acceptance, standard rejection, range revision, reanalysis, reinjection, integration.
4.1 to 4.2 LBA: key reagents and validation 4.2.3 sets the curve, anchor points and wider limits.
4.3 LBA: study sample analysis Each plate’s QCs and curve meet the run criteria (4.3.1).
5 Incurred sample reanalysis (ISR) A failure triggers an investigation under an SOP.
6 Partial and cross validation Range changes, laboratory transfers and combined data.
7 Additional considerations Endogenous analytes: surrogate matrix, surrogate analyte, background subtraction or standard addition (7.1).
8 Documentation, including Table 1 Which calibration records stay at the site and which are reported.
9 Glossary Analytical run, batch, calibration standard, calibration range, reintegration.

Numbering follows the ICH Step 4 text1.

03 · What M10 asks of the calibration curve

What M10 asks of the calibration curve.

Sections 3.2.4 and 4.2.3 define the curve in validation, and the same accuracy limits apply in every study run (3.3.2, 4.3.2)1.

Requirement Chromatography (3.2.4) LBA (4.2.3)
Matrix Same as the study samples. Same; a surrogate matrix needs scientific justification.
Levels Blank, zero sample (blank with internal standard) and at least 6 levels, LLOQ to ULOQ. Blank and at least 6 levels, LLOQ to ULOQ; anchor points outside the range carry no criteria.
Left out of the fit Blank and zero samples. The blank.
Regression model A simple model that adequately describes the concentration-response relationship, chosen under written procedures; model, weighting and transformation set in validation. Most often a 4- or 5-parameter logistic model; other models need justification.
Back-calculated accuracy Within ±15% of nominal; ±20% at the LLOQ. Within ±20% of nominal; ±25% at the LLOQ and ULOQ.
Proportion within limits At least 75%, across at least 6 levels; with replicates, at least 50% per level. At least 75%, anchor points excluded, across at least 6 levels including the LLOQ and ULOQ.
Failing standard Rejected and the curve re-evaluated. In accuracy and precision runs, losing all LLOQ or ULOQ replicates rejects the run. No rule in 4.2.3; in study runs, 4.3.2 rejects it and re-evaluates the curve when more than 6 standards are used.
Validation runs All acceptable curves from at least 3 independent runs over several days are reported. At least 6 independent runs over several days.
Fresh or frozen Freshly spiked in at least one assessment; frozen within stability. Freshly spiked preferred; frozen within stability.
Reported Curve parameters (such as slope and intercept) and back-calculated concentrations with mean accuracy and precision. Curve parameters for accepted runs (Table 1).

ICH M10 Sections 3.2.4, 4.2.3 and Table 11.

M10 names no weighting factor; the case for the chosen model comes from the validation data1. See ICH M10 calibration curve acceptance criteria, explained and Regression and weighting under ICH M10: building the justification.

04 · How an analytical run is accepted

How an analytical run is accepted.

Run acceptance criteria are defined in the protocol, the study plan or an SOP (3.3.2)1. A chromatographic run is accepted when:

  1. It is complete. A blank, a zero sample, standards at 6 or more levels, and low, medium and high QCs in duplicate or at 5% of study samples, whichever is higher, bracketing the study samples (3.3.1)1.
  2. Its standards are within limits. At least 75% back-calculate within ±15% of nominal (±20% at the LLOQ), across at least six levels (3.3.2)1.
  3. Failures follow the rules. With more than six levels, a failing standard is rejected and the curve refitted. A rejected LLOQ or ULOQ moves the run’s limit to the next acceptable standard (a new LLOQ keeps ±15%; with replicates, one failing LLOQ or ULOQ replicate leaves the range unchanged), the revised range still covers the low, medium and high QCs, and study samples outside it are reanalyzed (3.3.2)1.
  4. Its QCs are within limits. At least two thirds of all QCs, and 50% at each level, fall within ±15% of nominal; otherwise the run is rejected and its study samples go into a new batch (3.3.2)1.
  5. Batches and analytes stand alone. Standards in a failed batch cannot support other batches, and a run can be accepted for one analyte and rejected for another (3.3.2)1.

LBA runs use the same proportions with wider limits, ±20% for standards (±25% at the LLOQ and ULOQ) and 20% for QCs, and a plate with its own standards and QCs is assessed on its own (4.3.1, 4.3.2)1.

Section 8 asks that the basis for changing or reprocessing data be documented and the original record kept1. See Excluding calibration standards: the 75% rule and the record behind it.

05 · Reanalysis, reinjection and reintegration

Reanalysis, reinjection and reintegration.

Reasons for reanalysis, the number of replicates and the rule for choosing the reported value are predefined before study samples are analyzed (3.3.4)1. In comparative BA/BE studies, reanalysis for a pharmacokinetic reason is not acceptable1.

Processed samples may be reinjected after equipment failure if reinjection reproducibility has been demonstrated, but not because standards or QCs failed without an identified analytical cause (3.3.5)1.

Integration and reintegration follow a study plan, protocol or SOP. The bioanalytical report lists reintegrated chromatograms, manual integrations included, with reasons; original and reintegrated chromatograms are kept, and submitted for comparative BA/BE studies (3.3.6)1. See FDA data integrity actions against bioanalytical CROs: what the notifications share.

06 · Incurred sample reanalysis

Incurred sample reanalysis.

ISR verifies the reliability of reported concentrations. It is expected at least once per species in nonclinical studies, in all pivotal comparative BA/BE studies, and in specified clinical trials such as the first trial in subjects (Section 5)1.

The sample is 10% of the first 1,000 study samples plus 5% of the rest, analyzed in separate runs on a different day. The difference between repeat and original values, as a percentage of their mean, is within ±20% for chromatography or ±30% for LBAs in at least two thirds of repeats1.

A failed ISR calls for an investigation and remediation, an SOP directs how investigations are triggered and conducted, and ISR data do not replace the original results1. The ICH Q&A adds that the investigation covers the entire process, including sample handling, processing and analysis2. See When incurred sample reanalysis fails, check the calibration.

07 · Partial and cross validation

Partial and cross validation.

Partial validation evaluates changes to a fully validated method, such as a transfer between laboratories, a change of detection system or platform, a changed calibration range or a change of anticoagulant. The parameters tested should meet full validation criteria (6.1)1.

Cross validation is required when data within a study come from different methods or laboratories, or when data from different methods are combined or compared across studies to support special dosing regimens or regulatory decisions (6.2)1. Both sides measure the same low, medium and high QCs, at least in triplicate, and study samples spanning the range (at least 30, if available). Bias can be assessed by Bland-Altman plots or Deming regression, and Section 6.2 sets no pass or fail limit1.

See Method transfer and cross validation under ICH M10: comparing calibration across sites.

08 · What the records and reports must show

What the records and reports must show.

Section 8 asks for everything needed to reconstruct a study: contemporaneous records, changes that do not obscure the original data, and a documented basis for any reprocessing1.

Record What M10 expects Where
Run summary tables 100% of tables for accepted and failed runs, including calibration curve, regression and weighting function Analytical site; reported for comparative BA/BE studies
Calibration standard results Concentrations and response function results (curve parameters) for accepted runs, with accuracy and precision Validation and bioanalytical reports
Run list All runs, failed runs included, with status, reason for failure and dates; reinjected runs with reasons Bioanalytical report
Reintegration Electronic audit trail, all original and reintegrated chromatograms, and the reason and mode of reintegration Analytical site; list and reasons in the bioanalytical report (3.3.6)
Reanalysis All repeat data with contemporaneous reasons; initial, repeat and final values with justification Analytical site; bioanalytical report (3.3.4)
Deviations Contemporaneous record, investigation and impact assessment Analytical site; described in both reports

From ICH M10 Table 1 and Sections 3.3.4 and 3.3.61.

Bioanalysis white papers

Go deeper on calibration under M10.

Six papers, each tied to the M10 sections above, with worked examples and primary sources. Every paper is free to read online or download as a PDF.

FDA, ICH and EMA updates

What changed, and what it means for calibration.

  1. October 6, 2026

    FDA revises its GLP compliance program for nonclinical laboratories

    FDA’s Bioresearch Monitoring compliance program 7348.808, Good Laboratory Practice (Nonclinical Laboratories), now carries an implementation date of October 6, 2026. The change history describes the update as a partial…

    Read the update

  2. June 3, 2026

    ICH adopts E6(R3) Annex 2 and publishes a consolidated GCP guideline

    The Regulatory Members of the ICH Assembly endorsed Annex 2 of E6(R3), the guideline for good clinical practice, under Step 4 on June 3, 2026, and ICH consolidated the principles, Annex 1 and Annex 2 into a single E6(R3)…

    Read the update

  3. April 10, 2026

    FDA reissues its final guidance on bioanalytical method validation for biomarkers

    FDA’s Center for Drug Evaluation and Research, in cooperation with the Center for Biologics Evaluation and Research, has issued Bioanalytical Method Validation for Biomarkers as a final Level 2 guidance dated April 2026…

    Read the update

  4. July 7, 2025

    European Commission consults on revised Annex 11, Chapter 4 and a new AI Annex 22

    On July 7, 2025, the European Commission’s Directorate-General for Health and Food Safety opened a stakeholders’ consultation on three drafts for EudraLex Volume 4, the EU GMP guide: a revised Chapter 4 on documentation, a…

    Read the update

  5. March 28, 2025

    FDA notifies pharmaceutical companies that certain Raptim Research studies are unacceptable

    On March 28, 2025, FDA notified pharmaceutical companies that certain studies conducted by Raptim Research Pvt. Ltd., a contract research organization in Navi Mumbai, India, are unacceptable because of data integrity and…

    Read the update

See all accreditation updates

Questions

Frequently asked questions

Does ICH M10 apply to biomarker assays?

No. Section 1.3 excludes biomarker bioanalysis and immunogenicity methods. FDA’s biomarker guidance makes the M10 approach the starting point, scales validation to fitness for purpose, and expects full validation as M10 describes when biomarker data support regulatory decisions.

What percentage of calibration standards must pass under ICH M10?

At least 75%, across a minimum of six levels (Sections 3.2.4, 3.3.2, 4.2.3 and 4.3.2). The limits are ±15% of nominal for chromatography (±20% at the LLOQ) and ±20% for LBAs (±25% at the LLOQ and ULOQ), with anchor points excluded.

Can a failing calibration standard be excluded from the curve?

Yes. Under Section 3.3.2, with more than six levels, a failing standard is rejected and the curve refitted, provided at least 75% still pass. A rejected LLOQ or ULOQ narrows the range, which should still cover the low, medium and high QCs. Section 8 expects the basis for reprocessing to be documented and the original record kept.

Does ICH M10 specify a regression model or weighting factor?

No. Section 3.2.4 asks for a simple model that adequately describes the concentration-response relationship, chosen under written procedures, with the model, weighting scheme and transformation determined during validation. Section 4.2.3 notes that LBAs most often use 4- or 5-parameter logistic models and that other models should be justified.

Does ICH M10 set acceptance criteria for cross validation?

Section 6.2 sets no pass or fail limit. Both methods or laboratories measure the same low, medium and high QCs at least in triplicate, plus at least 30 study samples if available, and bias can be assessed by Bland-Altman plots or Deming regression.

Where WLTR fits

How WLTR helps.

WLTR is a verification layer for the calibration data your chromatography data system (CDS) already produces, built for LC-MS/MS, HPLC and UPLC methods that report against a calibration curve. It reads the raw instrument export, records a SHA-256 hash of the original file, and evaluates every candidate calibration model against the same 15 acceptance criteria per analyte, each set by your lab to its SOPs. The method is covered by U.S. Patent 12,038,448 B217.

WLTR checks back-calculated accuracy at every level, so bias at the LLOQ and ULOQ surfaces before it reaches a study report, and recommends the simplest model that meets every criterion; the analyst decides. Each exclusion is recorded with the user and the reason and travels into QA approval. WLTR does not replace your CDS, LIMS or method validation. WLTR verifies. Your lab validates.

Before your next sponsor audit

See how your calibration curves hold up under M10.

Bring a recent run. We will show you how each analyte’s curve holds up against 15 acceptance criteria, set to your SOPs, with the record an auditor will ask for.

Get Compliance Gap AssessmentJoin the pilot program

Sources

Primary documents.

  1. ICH, M10 Bioanalytical Method Validation and Study Sample Analysis, Step 4, May 24, 2022, Secs. 1 to 9 and Table 1. database.ich.org/sites/default/files/M10_Guideline_Step4_2022_0524.pdf
  2. ICH, M10 Bioanalytical Method Validation and Study Sample Analysis, Questions and Answers, dated November 11, 2022, adopted by the ICH Assembly November 16, 2022, Sec. 5 question. database.ich.org/sites/default/files/ICH_M10_QAs_2022_1111.pdf
  3. FDA, M10 Bioanalytical Method Validation and Study Sample Analysis, Guidance for Industry, November 2022. www.fda.gov/media/162903/download
  4. FDA, M10 Bioanalytical Method Validation and Study Sample Analysis; International Council for Harmonisation; Guidance for Industry; Availability, Federal Register 87(214), November 7, 2022, 67037–67039, Docket FDA-2019-D-1469. www.govinfo.gov/content/pkg/FR-2022-11-07/html/2022-24113.htm
  5. FDA, M10 Bioanalytical Method Validation and Study Sample Analysis, guidance document page, content current as of June 12, 2024, accessed October 26, 2026. www.fda.gov/regulatory-information/search-fda-guidance-documents/m10-bioanalytical-method-validation-and-study-sample-analysis
  6. FDA, M10 Bioanalytical Method Validation and Study Sample Analysis Questions and Answers, Guidance for Industry, June 2024. www.fda.gov/media/179296/download
  7. EMA, ICH guideline M10 on bioanalytical method validation and study sample analysis, EMA/CHMP/ICH/172948/2019, Step 5, final adoption by CHMP July 21, 2022, date of coming into effect January 21, 2023. www.ema.europa.eu/en/documents/scientific-guideline/ich-guideline-m10-bioanalytical-method-validation-step-5_en.pdf
  8. EMA, Implementation strategy of ICH Guideline M10 on bioanalytical method validation, EMA/449486/2023, April 4, 2024, adopted by CHMP April 25, 2024. www.ema.europa.eu/en/documents/scientific-guideline/implementation-strategy-ich-guideline-m10-bioanalytical-method-validation_en.pdf
  9. FDA, Bioanalytical Method Validation for Biomarkers, Guidance for Industry, January 2025; current version dated April 2026, Secs. I to III. www.fda.gov/media/185276/download
  10. 21 CFR Part 58, Good Laboratory Practice for Nonclinical Laboratory Studies, Secs. 58.130(e) and 58.185(a), eCFR, up to date as of October 22, 2026. www.ecfr.gov/current/title-21/chapter-I/subchapter-A/part-58
  11. 21 CFR Part 11, Electronic Records; Electronic Signatures, Secs. 11.1(a), 11.10(e) and 11.50, eCFR, up to date as of October 22, 2026. www.ecfr.gov/current/title-21/chapter-I/subchapter-A/part-11
  12. FDA, Electronic Systems, Electronic Records, and Electronic Signatures in Clinical Investigations: Questions and Answers, Guidance for Industry, Revision 1, October 2024, Q12. www.fda.gov/media/166215/download
  13. FDA, Data Integrity and Compliance With Drug CGMP: Questions and Answers, Guidance for Industry, December 2018, Q14. www.fda.gov/media/119267/download
  14. FDA, Data Integrity for In Vivo Bioavailability and Bioequivalence Studies, Draft Guidance for Industry, April 2024, Secs. III.C.2.c and III.C.5. www.fda.gov/media/177404/download
  15. 21 CFR 320.29, Analytical methods for an in vivo bioavailability or bioequivalence study, eCFR; text checked against the April 1, 2025 CFR edition. www.ecfr.gov/current/title-21/chapter-I/subchapter-D/part-320
  16. EMA, Guideline on computerised systems and electronic data in clinical trials, EMA/INS/GCP/112288/2023, March 9, 2023, Secs. 2, 6.2.1 and 6.2.2. www.ema.europa.eu/en/documents/regulatory-procedural-guideline/guideline-computerised-systems-electronic-data-clinical-trials_en.pdf
  17. U.S. Patent 12,038,448 B2, inventor Walter “Chip” Hinchman Sr. patents.google.com/patent/US12038448B2

Last reviewed October 26, 2026. Plain-language summary, not regulatory advice.