Also, if data are collected in units other than SI units, how can the original data and the converted data be stored within the SDTM dataset? The use of SI units is required for all variables and parameters of test results to be stored in the Findings class domain of the SDTM dataset, as long as SI units are applicable.
In the case of multicenter or multiregional studies, etc. Examples of how to store data in conventional units and SI units into SDTM [Example 1] When values in domestically conventional units and internationally conventional units both exist: Some values need to be displayed in Japanese on analysis results, and such values need to be stored in datasets in Japanese. In such cases, should only datasets with alphanumeric values be separately prepared for electronic study data submission?
Basically, it is only necessary to submit datasets with alphanumeric values for electronic study data submission. For variables that can be properly translated from those displayed in Japanese analysis results into English without loss of information, it is acceptable to store the translated variables in datasets, and to submit only the datasets with alphanumeric values.
In such cases, it is possible, in addition to the datasets with alphanumeric values, to submit together the original datasets with Japanese characters that were used for displaying analysis results instead of alphanumeric values, from the viewpoint of traceability between the original and translated datasets or to present reference materials. If certain information is likely to be lost when Japanese data is translated into English, follow Section 4.
In the case of datasets translated at the time of their creation, this should be described in the data guide. If datasets include languages other than English or Japanese, what procedures should be followed? Basically, the requirement is that submitted datasets be entirely translated into English.
In the event that datasets are translated, this should be described in the data guide. When translation is performed e. No written form is required, but the applicant is responsible for ensuring the accuracy of the translation. When Japanese datasets will be submitted in addition to alphanumeric datasets, is it necessary to create a dataset definition document for each dataset?
The definition document is needed only for the alphanumeric datasets. Please store the definition document in the same folder as the alphanumeric datasets. If analysis software is used that requires no explicit program creation, is it permissible to submit an operational log instead of a program?
Submission of an operational log will be sufficient instead of the program, if the log clarifies the analysis algorithm. If datasets in a format other than the CDISC standards are converted into a CDISC-conformant format for submission, and if the source datasets and the Annotated CRF are to be submitted together as an explanation of traceability, under which folder should they be stored? Basically, we expect that data from multiple studies subject to an integrated analysis will be combined into a single dataset, and that one analysis dataset will be submitted for each analysis.
When datasets from multiple studies are consolidated by a program for analysis, it is possible to submit the datasets of each study and the program containing the process of consolidation. In which folder should they be stored? This should be explained in the data guide. In order to explain the process by which the variables of ADaM datasets were created, reference data used to create ADaM datasets reference tables such as Lookup tables, Metadata, etc.
Data used to impute missing data e. The handling of missing data should be explained in the dataset definition document and in the data guide. Information on the character sets and the encoding system is needed to identify the characters intended by the dataset creator. Examples of character set information to be described in the data guide: In connection with Q above, how can information be obtained on the encoding system used?
In principle, encoding system-related information can be obtained from the property of the dataset. In the case of an unconventional encoding system used for relevant character sets, provision of detailed data may be additionally requested.
No further information is necessary. However, if the given character set-related information does not work, additional detailed information may be requested. If the application is to be made based on the results of an interim analysis, is it permissible to include data obtained after the cutoff of the interim analysis up to the time of application with data submitted for the application? If the application is to be made based on the results of an interim analysis, it is required that data up to the cutoff of the interim analysis be included in the electronic data submission; it is also acceptable, however, to include data after the cutoff.
For instance, if the application is made after a certain period of time from the interim analysis, it may be useful to include data after the interim analysis. If the application is made including data after the interim analysis, clearly distinguish data before and after the cutoff, and the handling of the relevant data must be explained in the data guide. Prefixes should be used to keep numbers in the range of 0. For a clinical study with interim analyses, when is it necessary to submit the interim analysis data in addition to the final analysis data?
Please indicate the format and method for submission, and the points to consider when the interim analysis data and the final analysis data are to be submitted together at the time of application. Use the clinical trial consultation to verify the necessity of electronic data submission of the interim analysis datasets and the scope of submission.
Under the circumstances described above to submit additional electronic data after the application , store the additional data Study No. For combined data from multiple studies, it will be necessary to specify in the dataset which study each data comes from. If different versions of the standard were used within the same clinical study, this situation and the reason for it must be explained in the data guide.
Applicants should also perform validation of study data based on the single CDISC standards version in the define. Are there any points to consider when the CDISC Controlled Terminology version used to create the dataset and that used at validation are different?
It is necessary to explain in the data guide if the CDISC Controlled Terminology version used to create datasets and the version used at validation are different, and to specify each.
Is it possible to submit data with blanks in the variables except for drug information converted by CRT Japan? A sponsor collects the adverse event terms in Japanese during the clinical study, creates the AE domain in both Japanese and English, and creates figures or tables based on the English dataset.
In this case, is it possible to submit only the English dataset? In the meantime, what should the sponsor do if the figures or tables are created based on the Japanese dataset? In both cases, it is possible to submit only the English dataset if all information on the Japanese dataset is included in the English dataset. It is also possible to submit the Japanese dataset along with the English dataset.
Questions on electronic study data on clinical pharmacology Q Are all clinical pharmacology studies including phase I studies in the data package attached to a new drug application subject to electronic study data submission? For clinical pharmacology studies including phase I studies that provide major evidence for dosage and administration or dose adjustment, electronic data must be submitted. To ensure the need for electronic submission of study data, it is recommended that applicants use the clinical trial consultation.
In the case of a population analysis being subject to submission, which data used to create the population analysis datasets - e.
For a population analysis, only electronic data related to the population analysis such as analysis datasets for the population analysis are subject to submission. Yes, it is possible. The above statement in the Notification means that it is necessary to submit datasets in a manner that clarifies why data was excluded from analyses; thus, submission can be made in other ways as long as it fulfills this aim.
When a sponsor conducts clinical trials to evaluate the effect of intrinsic or extrinsic factors such as age, sex, weight, genetic factors, severity of disease, disease complications, or dietary, alcohol or smoking habits on the pharmacokinetics of an investigational drug, are these study data subject to submission even if the study result indicates negative effect of these factors on the pharmacokinetics of the drug? It is not necessary to submit study data in that case.
In some cases, study data might not need to be submitted even though the ratio is not within the range of 0. To ensure the necessity of submission of study data evaluating the effect of intrinsic or extrinsic factors, we recommend using a clinical trial consultation if necessary. This refers to population analysis that includes data from a phase 3 trial, or which is the basis of dosage in a phase 3 trial. Such analyses are generally considered to be subject to submission.
Examples include the results of an analysis used as the basis of the content in the section on drug interactions contraindication or caution for combination in a package insert, or used as the basis of the judgment not to conduct a clinical drug interaction study.
When a sponsor submits results from all phase II and III studies including long-term studies , which are generally regarded as the major evidence for evaluation of efficacy, safety, and dosage and administration [ 2. Basic Principles on Electronic Submission of Study Data for New Drug Applications], is it necessary to submit data for clinical pharmacology analyses such as drug concentrations or the pharmacokinetic parameters included in these studies?
If it is necessary, in what format should it be submitted? It is necessary to submit the data for clinical pharmacology analyses included in the clinical studies shown above. Please submit data following the data standards shown in 3. In the case where a sponsor creates an analysis dataset in a format other than ADaM, and converts the dataset to ADaM for submission for a new drug application, is it necessary to submit the original dataset used for the analysis?
No, it is not necessary to submit the original dataset used for the analysis. In cases where the original dataset is considered useful for explaining traceability between datasets or to show analysis specifications, it may be submitted and used for the explanation. Either a single or separate datasets is acceptable for submission if the dataset can be used for analysis without additional handling so called Analysis Ready.
Is it recommended that Analysis Results Metadata be submitted when a sponsor creates a dataset on clinical pharmacology studies in the CDISC standards format and submit the dataset?
If yes, please show the scope of the Analysis Results Metadata subjected to submission. Even for clinical pharmacology studies, submission of the Analysis Results Metadata for the analyses performed to obtain the main results of the clinical study is recommended if the dataset was created in CDISC standards format in those studies. As the pharmacokinetic parameters are derived data, and not the accrual data collected in clinical study, is it necessary to include the PP domain in the SDTM dataset?
The pharmacokinetic parameters themselves are considered as data to capture the characteristics of the drug and should be included in database.
Creating a RELREC dataset would be preferable, but explaining the procedures by which the variables in the PP domain were created from the variables in the PC domain in a document such as a data guide would be also acceptable, if the logic needed to create a RELREC dataset were too complex.
In the case where a PC domain is created by converting a dataset from a format other than CDISC standards, is it necessary to secure traceability by recalculating PK parameters from the data stored in the PC domain, and then storing the recalculated PK parameters into the PP domain? In the case where only summary statistics of PK parameters are derived, is it necessary to submit the analysis dataset to output the summary statistics? No, it is not necessary to submit the analysis dataset in that case.
When a sponsor submits a standard pharmacokinetic analysis dataset in the ADaM format, what variables other than those that are required in ADaM IG should be included in the dataset? Please include the variables necessary for the analysis, and which can be used for analysis without any additional data handling, as well as the variables that are required in the ADaM IG.
For example, derived values needed for the analysis such as AUC used as a PK index, the change in a PD marker from baseline, and flags to extract records for analysis should be included in the dataset. As the basic model and final model represent the basic scope of submissions, it is not necessary to submit program or output files for covariate models or model evaluation. However, if the process of covariate models or model evaluation is complicated, and the sponsor considers that such files would aid understanding of the analysis method, it is possible to submit these files.
Is it necessary to submit the files even if the major results are described in the analysis report? The files in question are output files with calculated values by the base model or the final model. If major results are written in the analysis report, it is not necessary to submit the output files, but this should be mentioned and explained in the data guide. Simulations used for decision-making are recommend to be submitted, for example, determination of the patient population to be administrated, or the dosage.
Regarding Q, which kind of data should be submitted for each simulation?