Bioanalytical Method Validation Calculator (ICH M10)
Run acceptance, accuracy and precision, selectivity, carry-over and ISR, against ICH M10.
Try before you buy.
- Free preview: the method, the limits that apply and the verdict for your input; computed figures stay hidden until the pass.
- Locked until you unlock it: download and print.
- Unlock: Ultimate pass, ₹1,199 for 30 days or ₹449 for 1 week, a one-time payment unless you choose automatic renewal. Unlock: Ultimate pass, $24.99 for 30 days or $9.99 for 1 week, a one-time payment unless you choose automatic renewal.
Ways to unlock shows how to get the full result.
Printing this result is locked in the free preview.
Run-acceptance report
Locked in the free preview. Opens the ways to unlock this result.
Locked in the free preview. Batch runs unlock with a pass.
Locked in the free preview. Query results unlock with a pass.
About the Bioanalytical Method Validation Calculator (ICH M10)
Checks the calibration standards and QCs of bioanalytical runs against ICH M10 Bioanalytical Method Validation and Study Sample Analysis, for chromatographic methods (section 3) or ligand binding assays (section 4): each standard’s back-calculated concentration against ±15 % (±20 % at the LLOQ; for LBA ±20 %, ±25 % at the LLOQ and ULOQ), the 75 % and six-level rule, a rejected LLOQ or ULOQ that moves the run’s range, the QC rule of two-thirds overall and half at each level, within-run and between-run accuracy and precision, and incurred sample reanalysis with its two-thirds rule and minimum sample count.
Add what your validation holds besides the runs — selectivity of the blank sources, carry-over after the ULOQ, the matrix effect of QCs in each lot (and the matrix factor if your SOP asks for it), dilution integrity and stability — and the report judges each against its section. Paste the export of your data system or open its CSV or Excel file. The free preview shows every run’s decision, every verdict, the limits that apply and the method; the computed figures and the working stay hidden, and the PDF, Word, Excel and printed record open with an Ultimate pass. Worked out to the method you choose; your QA approves the result.
How to use it
- Choose a chromatographic method or a ligand binding assay, and whether the runs are accuracy and precision runs of a validation or runs of study sample analysis.
- Paste the runs, one sample per line — run, type, level, nominal and measured concentration, and the day — or open the CSV or Excel export of your data system; the header row is read by its names.
- Open the optional parts for selectivity and carry-over (with the responses at the LLOQ), the matrix effect, dilution, stability and incurred sample reanalysis, and paste their results.
- Read the board of runs, each run’s decision and reasons, the tables of standards, QCs and accuracy and precision, and the checks with the M10 section each comes from.
- Without a pass the report stays a free preview with its computed figures hidden; the PDF, Word, Excel and Print record unlocks with an Ultimate pass.
Examples
8 standards from 1 to 1,000 ng/mL and QCs at 1, 3, 400 and 800 ng/mL, 5 replicates each, on three days; run 2’s CS4 reads 23.6 for 20 ng/mL
All three runs accepted: run 2 keeps 7 of 8 standards within limits on 7 levels; every QC level meets ±15 % and a CV ≤ 15 % (±20 % at the LLOQ)
CS4’s relative error is (23.6 − 20) ÷ 20 × 100 = 18.0 %, outside ±15 %, but 7 of 8 is above 75 %.
CS1 at 1 ng/mL reads 1.26 (+26 %); the low QC is at 3 ng/mL
The run’s range starts at CS2 (2 ng/mL), which still covers the low, medium and high QCs, so the run can be accepted; reanalyse study samples below 2 ng/mL
Initial 41.0, repeat 52.6
% difference = (52.6 − 41.0) ÷ 46.8 × 100 = 24.8 %: outside ±20 % for a chromatographic method
Two-thirds of the repeats must be within ±20 % (±30 % for LBA); 8 of 9 within passes.
Common uses
- Judging each accuracy and precision run of a method validation, and the between-run figures over all of them.
- Accepting or rejecting the runs of a BA/BE or toxicokinetic study as they come off the instrument.
- Checking incurred sample reanalysis and its minimum count before the bioanalytical report is written.
- A second check of the run summary a LIMS prints, with the criteria cited by section.
The criteria ICH M10 sets
- Calibration (3.2.4, 3.3.2): a blank, a zero and at least 6 levels including the LLOQ and ULOQ; back-calculated concentrations within ±15 % of nominal, ±20 % at the LLOQ; at least 75 % of the standards, covering at least 6 levels, meet this. With replicates in validation runs, at least 50 % per level.
- Accuracy and precision (3.2.5): QCs at the LLOQ, within three times the LLOQ (low), around 30–50 % of the range (medium) and at least 75 % of the ULOQ (high); at least 5 replicates per level in each run and at least 3 runs over at least two days; mean within ±15 % and CV ≤ 15 % (±20 % and 20 % at the LLOQ), within each run and between runs; when a run misses the within-run criteria, an overall estimate of within-run accuracy and precision for each level.
- Run acceptance (3.3.2): at least 2/3 of all QCs and at least 50 % at each level within ±15 %; dilution QCs decide only the diluted samples.
- Selectivity (3.2.1): at least 6 individual sources; blank responses not more than 20 % of the analyte at the LLOQ and 5 % of the IS. Carry-over (3.2.6): the blank after the ULOQ within the same 20 % and 5 %.
- Matrix effect (3.2.3): low and high QCs, at least 3 replicates in each of at least 6 lots, accuracy within ±15 % and CV ≤ 15 % in each lot.
- Dilution integrity (3.2.7): at least 5 replicates per factor, mean within ±15 %, CV ≤ 15 %. Stability (3.2.8): the mean of each QC level within ±15 %.
- Incurred sample reanalysis (section 5): % difference = (repeat − initial) ÷ mean × 100, within ±20 % for at least 2/3 of the repeats; at least 10 % of the first 1,000 study samples and 5 % of those above.
Ligand binding assays (section 4)
Standards within ±20 % (±25 % at the LLOQ and ULOQ), anchor points left out of the 75 % rule; QCs at five levels including the ULOQ, at least 3 replicates in at least 6 runs; accuracy within ±20 % and CV ≤ 20 % (25 % at the limits), and a total error — |relative error| + CV — of at most 30 % (40 % at the LLOQ and ULOQ); runs accept with 2/3 of the QCs and half per level within ±20 %; ISR within ±30 %; selectivity in at least 10 sources with 80 % of them meeting each criterion; dilution linearity within ±20 %.
How the figures are worked out
- Relative error = (measured − nominal) ÷ nominal × 100, for every standard and QC and for each level’s mean.
- CV = standard deviation (n − 1) ÷ mean × 100. Between-run figures combine every result of the accepted runs; a one-way analysis of variance over the runs gives beside them the pooled within-run CV (the overall estimate of within-run precision) and the intermediate precision.
- Each percentage is rounded to one decimal, as it is reported, before it is compared with its limit: 15.04 % reads 15.0 % and is within ±15 %.
- A level whose standards are all rejected moves the run’s range to the next acceptable standard in study runs, and rejects an accuracy and precision run when it is the LLOQ or ULOQ.
What your device needs
Everything runs in this browser. Checking a list of 20,000 rows — about 110 runs of standards, QCs and study samples — takes under a tenth of a second on a laptop; the report then draws a table row for every standard and QC, so a study of that size reads best on a computer. Lists of up to 20,000 rows and files of up to 25 MB are taken.
Sources
- ICH M10 Bioanalytical Method Validation and Study Sample Analysis — sections 3.2 to 3.3 (chromatographic methods), 4.2 to 4.3 (ligand binding assays) and 5 (incurred sample reanalysis)
Limitations
- It judges the back-calculated concentrations your data system reports; it fits no calibration curve, so after a standard is rejected refit the curve there and enter the new values.
- Specificity, recovery, the haemolysed and lipaemic evaluations and the documentation of section 8 are judged from your own records; partial and cross validation (section 6) and endogenous analytes (section 7) are not covered.
- Criteria that M10 leaves to the SOP or protocol, such as the matrix factor’s limit, are yours to enter.
- For a ligand binding assay, carry-over compares each blank’s signal with the signal of the LLOQ standard, which fits formats whose signal rises with the concentration; judge a competitive format by the blanks’ back-calculated results instead.
- MySmartCoPilot is not affiliated with or endorsed by ICH. The record applies the criteria of the method you choose; your QA approves the result.
Privacy
On this device; its files download once
Your runs and results stay in this browser and are never sent anywhere. The PDF, Word and Excel files are made on your device.
Frequently asked questions
What do I get without a pass?
Without a pass, Bioanalytical Method Validation Calculator (ICH M10) shows the method, the limits that apply and the verdict for your input; computed figures stay hidden until the pass. Until you unlock it, the result can’t be downloaded or printed. An Ultimate pass, a one-time payment unless you choose automatic renewal, unlocks the full result. The pricing page lists the passes and their prices.
What does the Ultimate pass add here?
This tool in full, and every other Pro, Premium and Ultimate tool, for 7 days, 30 days or a year (a pass bought with PayPal leaves out email-address extractors). The pass is a one-time payment unless you choose automatic renewal.
What is the 75 % rule for calibration standards?
At least 75 % of the calibration standards must be within ±15 % of nominal (±20 % at the LLOQ), and those standards must cover at least six concentration levels. With 8 standards, 6 must pass; with 7 of 8 passing on 7 levels the curve stands.
When is a study run rejected for its QCs?
When fewer than two-thirds of all its QCs, or fewer than half of the QCs at any one level, are within ±15 % of nominal (±20 % for ligand binding assays). With 6 QCs, 4 must pass, and no level may lose both of its QCs.
How is the ISR percent difference calculated?
As (repeat − initial) ÷ the mean of the two × 100. For 41.0 and 52.6 it is 11.6 ÷ 46.8 × 100 = 24.8 %. At least two-thirds of the repeats must be within ±20 % for chromatographic methods and ±30 % for ligand binding assays.
How many incurred samples must be reanalysed?
At least 10 % of the study samples up to 1,000, then 100 plus 5 % of the samples above 1,000: 44 for 437 samples, 125 for 1,500. Enter the study’s sample count to check it.
What happens when the LLOQ standard fails?
In a study run, the next lowest acceptable standard becomes the run’s lower limit and keeps its ±15 % criterion; the revised range must still cover the low, medium and high QCs, and study samples below it are reanalysed. In an accuracy and precision run of a validation, a run whose LLOQ or ULOQ standards are all rejected is rejected.
Can I open the export of my LC-MS/MS software?
Yes: paste it or open its CSV, TXT or Excel file. The header row is read by its names — Run or Batch, Sample Type, Sample Name, Actual or Nominal Concentration, Calculated Concentration, Date or Day — so the columns may come in any order and others are left out. Solvent, wash and system suitability injections are left out like blanks, results such as BLQ, No Peak or < 1.00 count as no result, and a date with a time counts as its day. Numbers are read in the decimal mark the list itself uses (a comma-separated export writes decimal points), whatever your own number format; where a list could mean either, your own format decides.