Geroscience Press
Companion tool

Method Validation Planner

Generate a structured, CLSI EP-aligned validation plan for a biological aging assay. Adjust assay characteristics on the left; the right pane shows your study matrix, sample-size targets, and reporting checklist.

For educational use. Always confirm against current CLSI EP05-A3, EP06, EP07, EP09, EP17, EP25, EP28 documents and your local regulatory body.

Epigenetic Age (DNAmAge)

LDT · whole-blood

DNA methylation age (years) — reportable range 20 – 100 years. Decision point: Δage > 5 years.

Plan steps
8
Working days
155 d (~31 wk)
Approx. samples
2,323

Precision (repeatability & within-lab)

CLSI EP05-A3

Quantify random error at ≥2 concentrations spanning the reportable range (20 – 100 years).

Design
  • 2 levels (low, high) — add a 3rd at the clinical decision point
  • 2 runs/day × 2 replicates × ≥20 days
  • Same instrument, ≥2 operators, ≥2 reagent lots
Sample size

16 days × 2 runs × 2 replicates × 2 levels = 128 measurements per level pair

Acceptance

Total CV ≤ 4% at each level (95% CI upper bound).

~16 working days

Trueness via reference materials

CLSI EP15-A3

No comparator method: assess recovery using certified reference materials / proficiency samples.

Design
  • ≥3 certified reference materials or interlaboratory comparison panels
  • 5 replicates per material, single run
Sample size

3 materials × 5 replicates

Acceptance

Mean bias ≤ 5% at the decision point (Δage > 5 years).

~5 working days

Linearity / reportable range

CLSI EP06-Ed2

Verify linear response across the claimed reportable range (20 – 100 years).

Design
  • 5–11 equally spaced levels from low to high anchor
  • 4 replicates per level
  • Polynomial regression: prefer linear unless higher-order term is significant AND clinically meaningful
Sample size

9 levels × 4 replicates = 36 measurements

Acceptance

Deviation from linearity ≤ 5% across the range.

~2 working days

Limit of blank / detection / quantitation

CLSI EP17-A2

Define analytical sensitivity for low-concentration reporting.

Design
  • LoB: ≥60 blank measurements across 4 reagent lots / 3 days
  • LoD: ≥60 low-level samples spanning expected LoD ±50%
  • LoQ: lowest level meeting precision + bias goal
Sample size

≥120 measurements (60 blank + 60 low-level)

Acceptance

LoQ confirmed where total error ≤ allowable (4% CV + 5% bias).

~4 working days

Interference & specificity

CLSI EP07-Ed3

Screen common endogenous (hemolysis, lipemia, icterus) and matrix-specific interferents for whole-blood.

Design
  • Paired-difference design at 2 analyte levels
  • Spike each interferent at peak pathological concentration
  • n=5 replicates per condition
Sample size

8–12 interferents × 2 levels × 5 replicates

Acceptance

Interference effect ≤ 5% relative to control.

~3 working days

Reference interval establishment / verification

CLSI EP28-A3c

Aging biomarkers shift with age and sex — partitioning is usually required.

Design
  • De novo establishment: ≥120 reference subjects per partition (parametric or non-parametric)
  • Partition by age decade and sex; test Harris–Boyd criteria before pooling
  • Document inclusion/exclusion (fasting, BMI, comorbidities, medications)
Sample size

120 per partition × ≥4 partitions (e.g., age × sex) = 480 subjects

Acceptance

Non-parametric 2.5–97.5 percentiles with 90% CI per CLSI EP28.

~30 working days

Clinical / diagnostic accuracy

STARD 2015 + CLSI EP24-A2

Estimate sensitivity & specificity at the decision point (Δage > 5 years) with adequate precision (95% CI half-width ≤5%).

Design
  • Prospective consecutive sampling preferred over case-control
  • Blinded reference standard
  • Report ROC AUC, sensitivity, specificity, PPV/NPV at clinical prevalence
Sample size

1307 total subjects (196 cases at 15% prevalence)

Acceptance

Pre-specified performance targets met with 95% CI lower bound above minimum acceptable.

~60 working days

Sample & reagent stability

CLSI EP25-A

Define pre-analytical conditions that preserve the measurand.

Design
  • Specimen stability in whole-blood at RT, 2–8°C, −20°C, −80°C
  • Time points: 0, 4 h, 24 h, 7 d, 30 d (and longer for archive claims)
  • Freeze-thaw: 0, 1, 3, 5 cycles
  • On-board reagent + calibrator stability across one full lot
Sample size

3 patient pools × 4 conditions × 5 time points × 3 replicates

Acceptance

Change ≤ 5% relative to baseline.

~35 working days
Reporting checklist
  • State intended use, measurand traceability, and target population.
  • Report imprecision with 95% CI; flag any run with Westgard rule violations.
  • Disclose all interferents tested and the concentrations evaluated.
  • For aging clocks: report Pearson r, MAE, and median Δage with bootstrap CI.
  • Archive raw data, calibration lots, and operator IDs for ≥2 years.