To Top Page


Q&A on the gGuideline for Ethnic Factors in

the Acceptability of Foreign Clinical Datah





The Guideline uses general terms to be used in the countries (regions) participating in ICH. What should the terms gnew regionh and gregulatory authorityh in the Guideline particularly mean in Japan, when foreign clinical data are used according to the Guideline in a new drug application submitted to the Ministry of Health and Welfare of Japan?


The terms are interpreted in Japan as follows:

New region: Japan

Regulatory authority: In a narrow sense, the term means the Evaluation and Licensing Division, Pharmaceutical and Medical Safety Bureau, Ministry of Health and Welfare, and the Pharmaceuticals and Medical Devices Evaluation Center, National Institute of Health Sciences. In a broader sense, however, the Organization for Pharmaceutical Safety and Research (OPSR), which affords consultation with an applicant on bridging studies etc., should be included.




Likewise, what should the term gregulatory requirementsh be interpreted in Japan?


The term gRegulatory requirementsh means in Japan gMHW Ordinance on the Guideline for Good Clinical Practice (GCP)h pursuant to Article 14, Paragraph 3 of the Pharmaceutical Affairs Law (PAL), gStandards for Reliability of Application Documentationh (Article 18-4-3 of the Enforcement Regulations of PAL), the requirements on attached data described in gData to be Attached to Applications for Approval to Manufacture or Import Drugsh (PAB Director-General Notification No. 698 dated May 30, 1980), and the various guidelines issued by MHW concerning clinical studies. However, data obtained not strictly in compliance with the guidelines can be acceptable, if the clinical studies generating the data employed appropriate methods reflecting the progress of related sciences.


(Basic Principles of the Guideline)


The Guideline mainly supposes that most of the clinical data necessary for application were obtained outside Japan. How is the Guideline applied to a case where most of the clinical studies of a drug are conducted in Japan, necessitating, as a result, the foreign data to be only partially used, or where about the same amount each of domestic clinical data and foreign ones are used in the approval application?

A3: In both cases, it is necessary to prove that the foreign clinical data to be used in the application can be extrapolated to the Japanese population. If the data of the clinical studies conducted in Japan are usable as bridging data and the foreign datafs extrapolatability is assured, a new bridging study need not be conducted.


Q4: If a new regionfs regulatory authority assesses the extrapolatability of a foreign clinical data package after its completion, to suggest a bridging study, the approval in the new region can be delayed. How can a bridging study be best designed and conducted in new regions, while the data package is being completed in the original region, so that approval applications may be made simultaneously in the original and the new regions?

A4: As is pointed out in the question, the Guideline primarily assumes the existence of a clinical data package consisting mainly of foreign clinical data. The Guideline then suggests a bridging study to assess whether the data can be extrapolated. It is possible, however, to conduct clinical trials to construct a clinical data package to be bridged and a bridging study in parallel, adhering to the fundamental concept of the Guideline. Needless to say, however, the data supporting the appropriateness of the bridging study must be obtained before its initiation especially in terms of the safety of its trail subjects.


(On Policy on Accepting Foreign Clinical Data)


It is stated in the guideline that the judgement on the acceptance by the authority of a new region is influenced by the external factors such as the medical care practice of the new region as well as the regulatory authorityfs experience on the acceptance. What are the therapeutic/diseases categories for which the diagnostic criteria and practices in medical care are so similar among Japan, European nations, and USA that the foreign clinical data may be extrapolated without difficulty?

@A5: In the case of antibiotics whose effect-profile to clinically isolated strains is similar between the foreign countries and Japan, for example, the extrapolation should not be difficult, because trials for the class of drugs generally employ common endpoints and control groups among Japan and foreign countries. Also the data on antineoplastic agents are readily extrapolated, as the rate of tumor reduction is commonly employed in their evaluation.


Q6: As for the definition of the term gforeign countries,h is it allowable to use clinical data obtained in a country other than the ICH participating countries (regions) in approval application in Japan as far as the data meet Japanfs requirements?

A6: The Guideline in fact has been formulated at ICH, assuming the mutual acceptance of clinical data among the three regions, which possess the established regulations on clinical studies, such as GCP. However, if the data obtained in a region not participating in ICH meet Japanfs regulatory requirements (Refer to the answer A2 to Q2) and are deemed acceptable in Japan, the Guideline is applicable to the data.


Q7: A clinical data package may contain evaluations based on criteria different from those employed in Japan, as criteria in evaluating the efficacy of drugs are generally different from country (region) to country (region). To what extent are such differences allowed?

A7: If the criteria employed to evaluate clinical data obtained in a foreign country are scientifically rational and deemed acceptable in Japan, the criteria need not be strictly identical with those used in Japan.


Q8: When a drug shows similar efficacy and safety in the Japanese and the foreigners at different doses resulting from different pharmacokinetic behaviors, can the foreign data be extrapolated? And if so, naturally the dosage and administration in the approval application to MHW should be different from those approved abroad. Are such differences permissible?

A8: Similarity in the pharmacokinetic profile of a drug between populations is not a necessary condition for the foreign clinical data to be accepted. Even with a difference in the profile, foreign data may be extrapolated to the Japanese if similar efficacy and safety are shown in the Japanese trial subjects by modifying the dose (See Section 3.2.2 of the Guideline). In such a case the dosage and administration approved in the foreign country can be different from those in Japan.

It should be noted that even if there is no difference in a drugfs pharmacokinetic profile between the foreigners and the Japanese, the fact alone is not enough to ascertain the extrapolatability.


Q9: When (an) active control study(ies) of a drug conducted abroad employed, as a comparator or the dose thereof, those not approved in Japan, are the data of the study(ies) usable in the approval application to MHW?

A9: The data may be used for approval application in Japan as far as it can be confirmed that the comparator has been approved in the foreign country based on appropriate clinical data. The cogency of the data, however, depends on the comparatorfs characteristics such as how established it is as the therapy, and how its mechanism of action resembles that of the investigational drug.


Q10: What are the concrete standards to determine whether the dose-response, safety, efficacy, etc. of a drug are gsimilarh or gnot greatly differenth between ethnically different populations when the result of the bridging study is assessed to judge whether the foreign clinical data can be extrapolated to the Japanese?

A10: It is impossible to suggest concrete standards to judge whether the cited features of a drug are gsimilarh or gnot greatly differenth across populations. It is of note, however, that to so judge, neither statistical gequivalenceh nor strict identicalness is requisite.

For drugs with a steep pharmacodynamic (effect-concentration) curve and/or a narrow therapeutic range, it is necessary to evaluate carefully the differences in dose-responses, safety and efficacy, for total assessment of the extrapolatability.



What should be the recommended measures to take when a bridging study reveals that it is impossible to extrapolate the foreign clinical data to the Japanese?

A11: For example, a considerable difference in a drugfs dose-response between the foreign and the Japanese subjects, or that in incidences of unexpected ADRs can result in a different recommended clinical dose for the Japanese. In such cases, the data of comparative studies employing the dose preferred in the foreign country are generally difficult to extrapolate. Such are the circumstances where a clinical data package consisting of foreign data is proven to be not extrapolatable to the Japanese.

When a package cannot be extrapolated as a whole, it is recommended to review the data of each trial constituting the package to judge their usability in the approval application to be submitted to MHW. Since the salvaged (bridgeable) data are naturally insufficient to constitute a data package to be submitted, the applicant should conduct necessary trials to supplement and structure the clinical package for submission.


Q12: How does MHW handle a bridging study for an orphan drug?

A12: Small patient populations receiving orphan drugs often make it difficult to conduct bridging studies as suggested in the guideline, which demonstrate the extrapolatability clearly. In such cases, the judgment by the authority should be made based on the limited bridging data with a condition that the applicant should conduct post-marketing clinical studies and/or special surveillance studies to reconfirm the validity of the approval.


Q13: The Paragraph 2.1 of the guideline cites ga clinical trial using different comparators at the new regionfs approved dosage and dose regimenh as an example of an additional clinical study to fulfill the regulatory requirements in the new region. Is an active-control study always required in Japan even when the applicant already possesses the data of placebo-control study conducted abroad?

A13: In a therapeutic category where a standard drug/therapy has been established, the investigational drug is clearly located in terms of its medical relevance when it is compared with the standard drug to afford the information concerning relative efficacy and safety. Such information is usually useful to promote appropriate use of the drug after its marketing. This idea is reflected in Japanfs guidelines for clinical evaluation in specific therapeutics categories.

However, an active-control study is not an absolute requisite; when the drugfs efficacy has been demonstrated in placebo-control studies abroad and its usefulness in Japanfs medical settings confirmed, an active-control study is not established in Japan, should be additionally conducted.



How does MHW handle extrapolation of foreign clinical data in a special population such as the elderly and the renally or hepatically impaired, to the corresponding Japanese subpopulation?

A14: In general, when the foreign clinical data on a drugfs general patient population are demonstrated to be extrapolatable, so deemed are those on special populations without the necessity of additional bridging studies.

Furthermore, while the gGuideline on Clinical Trials in Geriatricsh requires the total of about 100 elderly subjects included in trials of a drug, an applicant is allowed to combine the numbers of the elderly in the foreign studies and that in Japanese ones, if the extrapolatability is confirmed as above. This principle applies to other guideline's requirements as well.


(Evaluation of the Influence of Ethnic Factors)

Q15: How are the effects of the medical practice and the diet of a region, the main extrinsic ethnic factors, assessed? Is accumulation of the experiences necessary to clarify those?

A15: There is not general method to evaluate the effects of the medical practice and the diet, because their influence on a drugfs pharmacokinetic and pharmacodynamic behavior varies from drug to drug. Thus the evaluation should be performed on a case by case basis, though it is sometimes advisable to seek reference with drugs of a similar structure or a mechanism of action.



Is it possible to conduct a clinical study of a drug with the participating institutions in plural countries (regions), or an international multi-center cooperative study), and perform a race-stratified analysis on the result to assess the ethnic factors?

A16: If the study is appropriately conducted with single protocol and a sufficient number of trial subjects to allow the analysis of ethnic differences are included, the analysis in the question should be feasible. By making a drugfs principal trials designed with this concept, the necessary information to assess the extrapolatability of the drugfs foreign clinical data can be obtained.



Clinical data contained in new drug applications submitted to MHW have not been analyzed taking the trial subjectsf ethnicity into consideration. Will it become necessary to take into consideration ethnic factors?

A17: It is beyond saying that an analysis on ethic factors should be performed in a clinical study conducted on plural ethnic populations.


Q18: When the inclusion criteria of a foreign trial are different from those accepted for the class of drugs in Japan, is it necessary to perform an analysis based on a sub-dataset comprising cases meeting the Japanese criteria?

A18: As stated in the gGuideline for Structure and Content of Clinical Study Reportsh (PAB/PCD Notification No. 335 dated May 1, 1996), and gStatistical Principles for Clinical Trials (PMSB/ELD Notification No. 1047 dated November 30, 1998), the validity of such analysis as mentioned in the question is conditioned on the similarity of the result of the analysis on full data set (containing all the cases) and that on the partial data set (consisting of the cases fulfilling certain conditions, extracted from the full set).

If, however, the inclusion criteria and diagnostic criteria employed in a foreign clinical trial are scientifically valid and acceptable from the viewpoint of Japanfs medical practice, the data of the study can serve as those attached to a new drug application, even if the criteria etc. are not strictly identical to those used in Japan.


(Bridging Studies on Efficacy)


When the result of Phase II (dose response) studies of a drug conducted in Japan indicates that the foreign clinical data on the drug can be extrapolated to the Japanese population, a confirmative study comparing the drug with an existing drug or a placebo is not required in Japan?

A19: In general, when dose-response studies conducted in Japan have proven extrapolatability of a drugfs foreign clinical data, confirmative comparative studies need not be conducted in Japan.



Is it possible to give a concrete description of the appropriate bridging study on efficacy, including its size, design, etc., on which the Guidelinefs Section 3.2.3 touches, to meet a specific situation expected?

A20: Section 3.2.3 cites three typical situations and the design etc. of necessary bridging studies for each situation. In the present situation, MHW proposes a dose-response study to be considered as the first candidate of the study (Paragraph gControlled Clinical Trialsh in Section 3.2.3. specifies gIt will usually be necessary to carry out controlled clinical trial, often a randomized, fixed dose, dose-response study, in the new regionh). Information on dose-response in the Japanese often plays an important role in assessing the extrapolatability. It is of note, however, that a comparative study is sometimes unnecessary as a bridging study especially for drugs insensitive to ethnic factors. It is impossible to categorically specify the sample size of a study (i.e. the required number of subjects), because the trialfs object and endpoint primarily determine its size.



: Section 3.2.3 of the Guideline cites, as one of the cases where a comparative study is required as a bridging study, g(when) there is little or no experience with acceptance of controlled clinical trials carried out in the foreign region.h Is this to be understood, under the present Japanese situation, that a controlled study is required as a bridging study?


As is pointed out, MHW has limited experience of basing its new drug approvals mainly on the data of comparative clinical studies conducted abroad. Conduct of a comparative study in Japan, therefore, should often be necessary for the time being.



What is the definition of the term gpharmacologically related compoundsh and gdrug classh cited in Section 3.2.3 gControlled Clinical Trials?h


These terms refer to the compounds which are judged to be similar to the drug in the light of chemical structure, pharmacological action, and other characteristics combined.


(Bridging Studies for Safety)


In what cases is it necessary to conduct a separate bridging study for safety?


As for the extrapolation of a drugfs data on its safety, the incidences of side effects (adverse events) need not be identical between populations, nor is precise comparison required. In fact, a small difference in the incidence of a mild side effect, for example, between populations neither cause serious problem nor affect the extrapolatability. A separate bridging study for safety need not be conducted unless the bridging study for efficacy or the findings on analogous drugs arouses a grave safety concern on the drugfs safety in the Japanese.

Detection of a low-frequency side effect (advance event) and accurate determination of its incidence should be done in a drugfs post-marketing surveillance.



Section 3.2.4 of the Guideline gives specific figures in its discussion on the accurate determination of the rates of relatively common adverse events and the detection of a serious adverse events (adverse events occurring with a frequency around 1% generally need about 300 patients to detect). What is the ground of the figures?


The figures (1% and 300 patients) are calculated and indicated based on the idea that adverse events with an incidence of 1% or more should be detected in the clinical studies conducted before approval. The idea is employed in another MHW Guideline gThe Extent of Population Exposure to Assess Clinical Safety for Drugs Intended for Long-term Treatment of Non-life Threatening Conditionsh (PAB/PCD Notification No. 592 dated May 24, 1995) based on ICH E1 Guideline.

However, the E1 Guideline does not require the long-term safety data be composed solely of those obtained in a single ethnic group. The exposition in Section 3.2.4 should be interpreted that the total number of the long-term cases regardless of ethnicity should exceed the given figure.

Thus it is generally needless to conduct a bridging study on 300 subjects only to collect data on safety (to detect, for example, a side effect occurring at a relatively high incidence, etc.), unless in exceptional situations where the safety profile of a drug such as the adverse events caused and their incidences is totally different between the Japanese and the foreigners. Utilization of the foreign clinical data on a drug naturally reduces the number of the Japanese cases included in the drugfs safety evaluation, which the applicant should supplement in the post-marketing phase.



When appropriate bridging studies prove the extrapolatability of a drugfs foreign clinical data to the Japanese population in terms of its safety, and the long-term safety data in accordance with ICH E1 Guideline have been obtained abroad, is it right to assume that no long-term trial in Japan is required ?


When foreign clinical data on safety, including those of long-term administration trials, are judged to be extrapolatable to the Japanese, it is not necessary to carry out in Japan a long-term exposure trial.


(Clinical Pharmacology Studies)


: Are foreign pharmacokinetic data generated in the following populations usable as those in the Japanese population?
  1. Those who had been born of Japanese parents and brought up in Japan, and were living in the foreign countries when the trials were conducted
  2. Second-or-third generation Japanese (those who had been born of Japanese parents living in foreign countries, or those with their Japanese grandparents having emigrated to the foreign country and both of their parents being Japanese living in the country).


The data described in the question are usable, except where difference in the diet and the environment is expected to alter the drugfs pharmacokinetic behavior.



What are the points to consider when the pharmacokinetic profiles are compared between the Japanese and the foreigners.


The pharmacokinetic profiles of the two populations are compared in terms of the appropriate parameters (AUC, Cmax, clearance, etc.), which are obtainable through standard pharmacokinetic studies or using the principles of population pharmacokinetics. This presupposes the comparability of the parameters, which is ensured by the trialsf sufficient sample size, the calculation based on the blood level obtained at plural (at least three) doses, the same trial conditions (blood sampling time points, quantification of the drug, etc.), the same method of calculation, etc. The factors above should be taken into consideration from the planning stage of pharmacokinetic studies.

It should be heeded in interpreting the results that absence of statistically significant difference in pharmacokinetic parameters at a particular dose does not indicate gabsence of differenceh in the pharmacokinetic profiles between populations. In order to detect difference in the profiles, at least confidence intervals of the gdifferenceh or gquotienth between parameters obtained from the two populations should be calculated and analyzed, after (non)linearity of the drugfs pharmacokinetic behavior is examined.


What are the points to consider in utilizing foreign clinical data of Phase I studies of antineoplastic agents?


A full-scale study to obtain the maximal tolerable dose need not be re-performed in Japan for an agent with clear cytocidal effect, if appropriate foreign data are available on the drug. Instead, the applicant is required only to confirm the Japanese patientsf safety treated at the dose determined on the foreign data. Also a pharmacokinetic study affording pharmacodynamic information should be conducted to assess the necessity to adjust the dose for the Japanese.



Are results of foreign bioequivalence studies conducted in non-Japanese acceptable as clinical data attached to new drug applications?


Unless there is a major difference in the pharmacokinetic profiles between the Japanese and the foreigners, foreign clinical data are usable to prove the bioequivalence between dosage forms of a drug with different strengths or those before and after the change in its formulation, necessitated in the course of the drugfs development. This principle applies also to changes and additions in strengths, dosage forms, and formulations made after the drugsf approval.

As for bioequivalence studies on generic drugs, refer to the gGuideline for Judgment of Bioequivalence of Generic Drugsh (PMSB/ELD Notification No.487 dated December 22, 1997) and the Questions &Answers (issued on October 30, 1998) on the Notification.




Is it mandatory to attach all the foreign clinical data concerning a drug to its new drug application submitted to MHW or are the applicants allowed to submit only the data they judge necessary?


In addition to the foreign clinical data which the applicant uses to support the drugfs usefulness, a list of all the foreign clinical studies on the drugs regardless of whether the data thereof are used in the application, as well as the summaries of the major studies listed should be submitted. (This applies also to the foreign trials completed after the application concerning the drug is submitted to MHW.) Moreover, taking into account the increase in the number of drugs developed simultaneously in plural regions of the world, MHW requests the lists etc. cited above to be submitted, even when an applicant, having foreign clinical data on a drug, can constitute the clinical data package on the drug to be submitted without using foreign clinical data.

It should be noted here that when an applicant possesses data that cause doubt on the drugfs safety and efficacy, the applicant must submit them with the results of scrutiny and assessment thereof, as stipulated in Article 18-4-3 of the Enforcement Regulations of PAL.



How should the side effect information be presented in the precautions for use or package insert, when foreign clinical data are utilized in new drug application ?


Information on safety in overseas clinical data submitted as the application data is dealt in the same manner as domestic data, and both foreign and domestic data should be incorporated in the precautions for use, etc. If there is a rational reason, for example when it has been clarified that an ethnic factor has a great influence on evaluation of safety, then it is possible to describe domestic data and overseas data separately.


(Discussion with the Regulatory Authority)


This Guideline recommends a discussion between the applicant and the regulatory authority on the necessity, the type, etc. of a bridging study, and the regulatory authority in Japan will be OPSR. We hope that the structure of the regulatory authority participating in the discussion is strengthened and the results of discussion will be made public to the extent possible.


: General principles about the conduct of bridging studies are described in this Guideline, though the design of individual bridging studies should be determined based on the characteristics of each drug, medical care environment around the target disease, accumulation of clinical experience with analogous drugs, etc. Therefore it is advisable to utilize the clinical trial consultation system of OPSR to discuss about the type of a bridging study necessary for individual drugs. The details of clinical trial consultations cannot be disclosed from the standpoint of maintaining company confidentiality, but we would like to make rules for disclosure of the results in generalized cases.