DATA INTEGRITY (ALCOA ++)

Stratified Sampling In-Process 

(Powder Blends)

INTRODUCTION

 This guidance is intended to assist manufacturers of human drug products in meeting the requirements of 21 CFR 211.110 for demonstrating the adequacy of mixing to ensure uniformity of in-process powder blends and finished dosage units. This guidance describes the procedures for assessing powder mix adequacy, correlating in-process dosage unit test results with powder mix test results, and establishing the initial criteria for control procedures used in routine manufacturing.

FDA's guidance documents, including this guidance, do not establish legally enforceable responsibilities.  Instead, guidances describe the Agency's current thinking on a topic and should be viewed only as recommendations, unless specific regulatory or statutory requirements are cited.  The use of the word should in Agency guidances means that something is suggested or recommended, but not required.

Scope

Stratified sampling is the process of sampling dosage units at predefined intervals and collecting representative samples from specifically targeted locations in the compression/filling operation that have the greatest potential to yield extreme highs and lows in test results.  These test results are used to monitor the manufacturing process output that is most responsible for causing finished product variability.  The test results can be used to develop a single control procedure to ensure adequate powder mix and uniform content in finished products.

The methods described in this guidance are not intended to be the only methods for meeting US-FDA requirements to demonstrate the adequacy of powder mix.  Traditional powder blend sampling and testing, in conjunction with testing for uniformity of content in the finished product, can be used to comply with current good manufacturing practice requirements

Assessment of Powder Mix Uniformity

We recommend the assessment of powder mix uniformity using the following procedures:

·         Conduct blend analysis on batches by extensively sampling the mix in the blender and/or intermediate bulk containers (IBCs). 

·         Identify appropriate blending time and speed ranges, dead spots in blenders, and locations of segregation in IBCs.  Determine sampling errors. 

·         Define the effects of sample size (e.g., 1-10X dosage unit range) while developing a technique capable of measuring the true uniformity of the blend.  Sample quantities larger than 3X can be used with adequate scientific justification.  Appropriate blend sampling techniques and procedures should be developed for each product with consideration to various designs of blend powder sampling and the physical and chemical properties of the blend components.

·         Design blend-sampling plans and evaluate them using appropriate statistical analyses.

Quantitatively measure any variability that is present among the samples.  Attribute the sample variability to either lack of uniformity of the blend or sampling error.  Significant within-location variance in the blend data can be an indication of one factor or a combination of factors such as inadequacy of blend mix, sampling error or

Exhibit/Validatation batch Powder mix Homogeneity

This section describes sampling and testing the powder mix of exhibit and process validation batches used to support implementing the stratified sampling method described in this guidance. 

We recommend that during the manufacture of exhibit and process validation batches, you assess the uniformity of the powder blend, the in-process dosage units, and the finished product independently.  We recommend you use the following steps to identify sampling locations and acceptance criteria prior to the manufacture of the exhibit and/or validation batches. 

1.      Carefully identify at least 10 sampling locations in the blender to represent potential areas of poor blending.  For example, in tumbling blenders (such as V-blenders, double cones, or drum mixers), samples should be selected from at least two depths along the axis of the blender.  For convective blenders (such as a ribbon blender), a special effort should be made to implement uniform volumetric sampling to include the corners and discharge area (at least 20 locations are recommended to adequately validate convective blenders). 

2.      Collect at least 3 replicate samples from each location.  Samples should meet the following criteria:  

·         Assay one sample per location (number of samples (n) ³  10)

(n = 20 for ribbon blender).

·         RSD (relative standard deviation) of all individual results £ 5.0 percent.

·         All individual results are within 10.0 percent (absolute) of the mean of the results. 

If samples do not meet these criteria, we recommend that you investigate the failure according to the flow chart in Attachment 1.   We also recommend that you not proceed any further with implementation of the methods described in this guidance until the criteria are met.

Sampling errors may occur in some powder blends, sampling devices, and techniques that make it impractical to evaluate adequacy of mix using only the blend data.  In such cases, we recommend that you use in-process dosage unit data in conjunction with blend sample data to evaluate blend uniformity