Bioequivalence and Bioavailability Forum

Dear All,

I have a question about ICH M13A & FDA Bioequivalence Guideline ' missing data,as below:

ICH M13A :
M13A stipulates that data should be removed from the statistical analysis because of high pre-dose concentrations (see Section 2.2.3.3) and may be removed because of low exposure in exceptional cases (see Section 2.2.1.1).
In addition to the reasons specifically stated above, study protocol deviations may necessitate removal of data from the statistical analysis. The following are a few examples that may support such removal:
1. A subject does not complete the pre-dose meal in a fed study.
2. A subject completes a study period but is deemed to have insufficient number of samples to allow for an accurate estimation of the primary PK parameters.
3. A subject experiences emesis within 2 times the expected median tmax.
4. In rare cases, a subject experiences an adverse event that may change GI motility during the study period that may affect drug absorption, e.g., diarrhea within 2 times the expected median tmax.
5. A subject who does not complete the study due to AEs, non-compliance or withdrawal of consent due to personal reasons.
The specific reasons for protocol violation that may lead to subject removal from statistical analysis should be pre-specified in the protocol. Exclusion of data from the statistical analysis for any reason other than those specifically stated in Section 2.2 of M13A, should be documented prior to bioanalysis.
In a 2-way crossover design, if data from one period are excluded, the subject should not be included in the statistical analysis. In more complex study designs, removal of subject data from only one period may not result in the complete removal of the subject from the statistical analysis.

FDA Bioequivalenc Guideline:
The protocol or the protocol and the SAP should include plans to minimize missing data. The trial protocol or SAP should prospectively define anticipated causes of missing data, the corresponding statistical assumptions about reasons for the missing data, and how missing data will be treated in the statistical analysis. The treatment of missing data in the statistical analysis should be justified such that valid statistical inferences can be made under the assumptions about the missing data mechanism.
Statistical methods for handling missing data include complete case analysis, available case analysis, weighting methods, imputation, and model-based approaches. For example, in a twoway crossover study, a complete case analysis could be a general linear model as implemented in SAS PROC GLM, which removes all subjects with any missing observations for any variables included in the GLM model (i.e., removes subjects missing one or both periods). An available case analysis could be done using SAS PROC MIXED, which uses all observed data (e.g., in a two-way crossover study, uses all subjects with one or two complete periods of data).
Approaches for handling missing data and the statistical methods for the primary BE analysis (e.g., GLM vs. MIXED) should be prespecified in the study protocol or SAP. Depending on the nature of the assumed or likely missing data mechanism, statistical methods from any of these categories may be appropriate. The validity of a statistical approach to handle missing data depends on a variety of factors, including, but not limited to, the mechanism for missingness, the fraction of incomplete cases, the values that are missing, specifics of the analysis, and definition of the estimand. Sensitivity analyses should be prespecified in the trial protocol to evaluate the robustness of conclusions to deviations from the assumptions about the missing data mechanism.
The applicant should provide detailed information about reasons for missing data and any observed intercurrent events.

For a 2×2×2 crossover trial, if a subject completes only one period, do ICH M13A and the FDA have consistent considerations regarding whether this subject should be excluded from the analysis set? If not, how should this be handled when submitting an FDA ANDA?