Understanding the Characteristics and Establishing ... · Understanding the Characteristics and Establishing Acceptance Criteria ... •ICH Q6A: Specifications (Dec ... and Establishing

Understanding the Characteristics and Establishing Acceptance Criteria for

Analytical Methods Validation

Ying Verdi

IVT LAB WEEK EUROPE

June 2017

Partners in Health Since 1919

Regulatory View of Method Characteristics

Statistics Behind Method Validation

Statistical Tools

Establishing Acceptance Criteria Based on Method Characteristics

2

3

• Guidance for Industry: Analytical Procedures and Methods Validation for Drugs and Biologics (July 2015)

• ICH Q2(R1): VALIDATION OF ANALYTICAL PROCEDURES: TEXT AND METHODOLOGY

• ICH Q6A: Specifications (Dec 2000, Aug 1999)

• ICH Q3A, Q3B, ICH M7

• ICH Q8 (R2): Pharmaceutical Development (Nov 2009)

• ICH Q9: Quality Risk Management (Jun 2006)

• ICH Q10: Pharmaceutical Quality System (Apr 2009)

• 21CFR Part 211.165 (e)

• EMEA Guidance on Validation of Analytical Procedures: Text and Methodology

4

“The ATP is based on the understanding of the target measurement uncertainty, which is the maximum uncertainty that the data should have in order to maintain acceptable levels of confidence in data quality”

5

- USP Stimuli articles: Lifecycle Management of Analytical Procedures: Method Development, Procedure Performance Qualification, and Procedure Performance Verification

Original Results (Analyst A)

Level Prep 1 Prep 2 Prep 3 Mean(n=3)

%RSD (n=3)

80% 102.1 102.2 102.3 102.2 0.1

100% 100.6 101.6 101.1 101.1 0.5

120% 100.3 100.3 99.6 100.1 0.4

Mean (n=9)

%RSD (n=9)

101.1 1.0

Repeat Results (Analyst B)

Level Prep 1 Prep 2 Prep 3 Mean(n=3)

%RSD (n=3)

80% 102.6 103.2 102.3 102.7 0.4

100% 102.7 102.3 101.8 102.3 0.4

120% 101.9 102.1 101.0 101.7 0.6

Mean (n=9)

%RSD (n=9)

102.2 0.6

Acceptance Criteria: 98.0 – 102.0%Route Cause: Acceptance Criteria is set too tight. Really?

6

7

8

Validation

Parameter

RSD (n=5)

NMT 2.0%

RSD (overall)

NMT 2.0%

Tailing

NMT 2.0

Check

Standard

98.0 – 102.0%

Accuracy &

Specificity run0.1 0.1 1.3 99.9

9

Level Conc. Peak AreaNormalized Peak

Area%Bias from 100%

Level

40 0.02016 176778 8768737.351 2.2

80 0.04031 347232 8614039.618 0.4

100 0.05039 432536 8583766.759 0.0

120 0.06047 516484 8541164.445 -0.5

160 0.08062 676728 8394041.379 -2.2

10

Bivariate Fit Residual Normal Quantile Plot

R2 = 0.9997

11

• Sample Prep• Place 20 tablets into a

100-mL vol. flask

• Add 50 mL diluent, sonicate 10 min

• Cool to room temp

• QS with diluent, mix

• Dilute 10:50

12

• Unlikely causes of OOS results• Standard prep

• Spiking solution

• Instrument precision

• Analyst training

• Acceptance criteria

• Likely root cause of OOS results• Placebo displacement

• Placebo displacement study confirmed the root cause

• Sample preparation modification

13

Level Prep 1 Prep 2 Prep 3 Mean (n=3) %RSD (n=3)

80% 100.5 100.0 99.9 100.1 0.3

100% 99.6 99.3 98.9 99.3 0.3

120% 99.4 99.4 99.6 99.5 0.2

Mean (n=9) %RSD (n=9)

99.6 0.5

14

Cliffs of

Moher

15

• Assay: The procedure must be able to quantify [analyst] in [presence of X, Y, Z] over a range of A% to B% of the nominal concentration with an accuracy and uncertainty so that the reportable result falls within +/- C% of the true value with at least a 90% probability determined with 95% confidence

USP Stimuli articles: Lifecycle Management of Analytical Procedures: Method Development, Procedure Performance Qualification, and Procedure Performance Verification

16

17

• Assay: The procedure must be able to quantify [analyst] in [presence of X, Y, Z] over a range of A% to B% of the nominal concentration with an accuracy and uncertainty so that the reportable result falls within +/- C% of the true value with at least a 90% probability determined with 95% confidence

Specificity

Range

Accuracy

Precision (Repeatability, Intermediate Precision, Reproducibility)

Acceptable Risk

Acceptable Uncertainty

18

USP <1225> ICH Q2(R1) FDA Guidance

Accuracy

Precision• Repeatability• Intermediate Precision• Reproducibility

Specificity

DL

QL

Linearity

Range

Robustness * *

* Can be done during method development

19

Method Performance Characteristics Defined by ICH and USP

Accuracy, Specificity, Linearity System Variability

Precision, DL, QL Inherent Random Variability

Range, Robustness N/A

• There are two types of method performance characteristicso systematic variability (bias)o inherent random variability (noise)

20

21

Sample Preparation

Standard Preparation

Instrument Calibration

Analytical Measurement

Data Output(Acquisition and Processing)

Results Presentation

Manufacturing ProcessLaboratory

Sample

Test Portion

Drying & Weighing

Dispensing & Weighing

Representative

Homogeneity

IntegritySelectivity

22

• Error of Measurement• Difference between an individual result and the true value of

the measurand

• Types of Error• Random Error• Systematic Error• Gross Error

EURACHEM/CITAC Guide Quantifying Uncertainty in Analytical Measurement 3rd Edition, 2012

23

• Random Error (Noise)• In replicate measurements varies in an

unpredictable manner

• Systematic Error (Bias)• In replicate measurements remains

constant or varies in a predictable manner

• Gross errors• Only abandonment of the experiment

and a fresh start is an adequate cure

24

•A parameter associated with the result of a measurement, that characterizes the dispersion of the values that could reasonably be attributed to the measurand

EURACHEM/CITAC Guide Quantifying Uncertainty in Analytical Measurement 3rd Edition, 2012

25

Error Uncertainty

A single value A range or interval

The value of a known error can be applied as a correction to the result

The value of the uncertainty cannot be used to correct a measurement result

26

S.D. Phillips PQI / IIGD&T 8/11/2011 27

“ There are known knowns; there are things we know that we know. There are known unknowns; that is to say, there are things that we now know we don't know. But there are also unknown unknowns –there are things we do not know we don't know. ”

— United States Secretary of Defense, Donald Rumsfeld

28

Profit

Conforms to Specification

Passed Inspection

Loss

Does not Conform to Specification

Passes Inspection

Type II Error

Loss

Conforms to Specification

Fails Inspection

Type I Error

Loss

Does not Conform to Specification

Fails Inspection

Reality

Mea

sure

d R

esu

lt

29

• Type A Method of evaluation of uncertainty by the statistical analysis of series of observations• Normal distribution

• Type B Method of evaluation of uncertainty by means other than the statistical analysis of series of observations• Rectangular distribution

• Triangular distribution

30

DistributionRectangular (Uniform)

Triangular Normal

Shape

Most Conservative

31

Define

• Process Elements

Identify

• Error Sources

Estimate

• Individual Contributions

Combine

• Overall Uncertainty

32

• Accurately weigh approximately 100mg of reference standard into a 250-mL volumetric flask• Reference standard; Purity (99.46 ±0.25)

• Dissolve in water at a laboratory temperature of 20 ± 4°C

• Dilute to Volume and mix well

33

34

• A tolerance interval is a statistical interval within which, with some confidence level, a specified proportion of a sampled population falls.

• The endpoints of a tolerance interval are called tolerance limits.

• An application of tolerance intervals is to compare specification limits with tolerance limits.

• For method validation, we can also compare accuracy study acceptance criteria with tolerance limits.

35

• Confidence limits are limits within which we expect a given population parameter, such as the mean, to lie.

• Statistical tolerance limits are limits within which we expect a stated proportion of the population to lie.

36

• Solid line represents population distribution

• Dotted line distributions result from the uncertainty of the sample mean

• The difference in location of the mean due to this uncertainty is defined by the confidence interval.

http://www.propharmagroup.com/blog/understanding-statistical-intervals-part-iii-tolerance-intervals 37

• ANOVA gauge R&R measures the amount of variability induced in measurements by the measurement system itself, and compares it to the total variability observed to determine the viability of the measurement system. (Wikipedia)

• The Gauge R&R method analyzes how much of the variability in your measurement system is due to operator variation (reproducibility) and measurement variation (repeatability).

• Gauge is the measurement device

38

• Collect a random sample of parts over the entire range of part

sizes from your process.

• Select several operators at random to measure each part

several times.

• Method validation: Precision

39

Ass

ay R

esu

lt

ProcessVariability

Method Variability

Accuracy(Bias)

Precision

Repeatability

Reproducibility

Gauge R&R

40

Analyst Spiking Level Prep 1 Prep 2 Prep 3 Prep 4

1 80% 100.2 109.0 98.2 95.5

1 100% 96.8 101.7 103.8 104.5

1 120% 101.3 105.6 99.9 97.3

2 80% 105.0 96.5 102.6 98.0

2 100% 104.1 105.6 98.0 96.8

2 120% 103.8 99.1 100.0 95.5

3 80% 108.7 94.3 105.6 104.5

3 100% 98.4 100.3 98.2 105.0

3 120% 94.9 98.0 96.4 105.7

41

42

43

44

Which one contributed more variability to the results, analyst or method itself?

45

46

•Based on combination of the following• Intended Use•Customer requirements•Historical data•Risk tolerance• Statistical analysis

•Specified in protocol

47

Test Result

Test Method Variability

Raw Material Attributes

Manufacturing Process

Variability

Sampling Variability

Storage Conditions &

Duration

Before method validation, ask yourself this question: How much variability can I have for my method?

48

Product Release Specification

Method Valuation

Method Precision

Instrument Precision Standard and Sample Preparation

49

USP <621> SYSTEM SUITABILITY

50

51

Analyst Spiking Level Prep 1 Prep 2 Prep 3 Prep 4

1 80% 100.2 109.0 98.2 95.5

1 100% 96.8 101.7 103.8 104.5

1 120% 101.3 105.6 99.9 97.3

2 80% 105.0 96.5 102.6 98.0

2 100% 104.1 105.6 98.0 96.8

2 120% 103.8 99.1 100.0 95.5

3 80% 108.7 94.3 105.6 104.5

3 100% 98.4 100.3 98.2 105.0

3 120% 94.9 98.0 96.4 105.7

52

Analyst 1 Analyst 3

Analyst 2 All Analysts

What will be the MV Acceptance Criteria?

53

Standard Concentration

Re

spo

nse

Standard Concentration

Re

spo

nse

Standard Concentration

Re

spo

nse

Standard Concentration

Re

spo

nse

54

• “The quantitation limit of an individual analytical procedure is the lowest amount of analyte in a sample which can be quantitatively determined with suitable precision and accuracy. The quantitation limit …… is used particularly for the determination of impurities and/or degradation products.”

- ICHQ2(R1)

• QL ≤ Reporting Threshold

55

56

DesignKnowledge

Control

Method Development Stage

ATP

Initial MD

Risk Assessment

DoE

Method Validation Stage

Pre-validation

Robustness Study

Method Validation

Lifecycle Stage

Tech. Transfer

Method verification

Method Improvement

57

• Starting with end goal in mind

• Identify critical method parameters, understand them and control them

• Set appropriate acceptance criteria for method validation to demonstrate quality performance characteristics

• Achieve cost-effective method and generate quality data

58

59