Skip to main content
NCBI home page
Clinical Pharmacology and Therapeutics logoLink to Clinical Pharmacology and Therapeutics
. 2025 Jan 3;117(4):910–919. doi: 10.1002/cpt.3540

PMDA Perspective on Use of Real‐World Data and Real‐World Evidence as an External Control: Recent Examples and Considerations

Junichi Asano 1, Hiromi Sugano 1, Hiroyuki Murakami 2, Atsushi Noguchi 3, Yuki Ando 1, Yoshiaki Uyama 4,✉
PMCID: PMC11924144  PMID: 39749966

Abstract

Recent discussions about the utilization of real‐world data (RWD) and real‐world evidence (RWE) have been more focused on drug development for regulatory approval rather than during the post‐marketing stage. In Japan, RWD/RWE have been practically utilized as an external control for drug approval. Most cases were related to orphan diseases where the feasibility of conducting randomized controlled clinical trials was generally low. The utilization of RWD/RWE as an external control provides additional information that can support regulatory review for drug approval. However, many points should be taken into consideration through all stages of a study that is based on RWD/RWE, including planning, analysis, and interpretation. In this article, we present our recent review experience focusing on efficacy evaluations with an external control based on RWD/RWE that were submitted as a part of new drug applications in Japan, and we describe our regulatory consideration of the utilization of RWD/RWE for drug evaluation and approval. Points described in this article promote appropriate drug development based on RWD/RWE and facilitate a proper discussion about RWD/RWE utilization with PMDA. Further accumulation of regulatory experience in PMDA with RWD/RWE utilization will enhance our knowledge and contribute to better regulatory decision making for drug approvals based on RWD/RWE.

Utilization of real‐world data (RWD) and real‐world evidence (RWE) has been actively discussed for the purpose of regulatory decision making on the benefit/risk assessment of a drug. 1 , 2 , 3 In recent years, discussions about RWD/RWE utilization have been more focused on drug development for regulatory approval rather than for the post‐marketing activities. 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 Especially in cases where the feasibility to conduct randomized controlled clinical trials is low such as trials targeted for orphan diseases or pediatric populations, RWD/RWE utilization as an external control (EC) is expected to increase the efficiency of drug development and accelerate drug approval for faster access to patients by providing better scientific evidence/information. 12 , 13 However, there are many challenges in such utilization of RWD/RWE. 14 , 15 , 16

This article presents our recent experiences focusing on the efficacy evaluation with an EC based on RWD/RWE submitted as a part of a new drug application in Japan, and describes our regulatory considerations in utilizing RWD/RWE to evaluate a drug for approval.

RECENT EXAMPLES OF RWD/RWE UTILIZATION AS AN EC IN JAPAN

Eight recent examples of RWD/RWE utilization as an EC for regulatory review of a new drug application are summarized in Table 1 . An outline and discussion/conclusion about RWD/RWE utilization in the PMDA review for each drug are presented below. Previous examples of RWD/RWE utilization have been reported by Nishioka et al., 3 including cases of RWD/RWE utilization as an EC (standard of care) for orphan drug approval such as alglucosidase‐alfa (genetical recombination) for glycogen storage disease type II (approved in April 2007), argatroban hydrate for heparin‐induced thrombocytopenia type 2 (approved in May 2011) and asfotase‐alfa (genetical recombination) for hypophosphatasia (approved in August 2015).

Table 1.

Recent examples of RWD/RWE utilization as an EC for drug approval in Japan

Case No. Approval Month/Year Non‐proprietary name Indication Source of RWD/RWE Types of EC Status
I Jun 2019 Defibrotide sodium Hepatic veno‐occlusive disease (VOD)

Data of foreign patients with VOD based on the historical medical records of trial sites (n = 32a).

Data of Japanese patients with VOD enrolled in the transplant registry unified management program in Japan (n = 107a)

 Standard of care Orphan drug
II Sep 2019 Cerliponase alfa (genetical recombination) Neuronal ceroid lipofusciosis type 2 disease (CLN2) Data of foreign patients with CLN2 enrolled in the patient registry for neuronal ceroid lipofuscin (n = 49a) Standard of care Orphan drug
III Mar 2022 Olipudase alfa (genetical recombination) Acidic sphingomyelinase deficiency (ASMD)

Data of foreign pediatric patients with ASMD in the prospective clinical study for collecting natural history data of ASMD patients (n = 14a).

Note: Additional data from placebo‐controlled trial in adults including Japanese patients were also available

 Standard of care Orphan drug
IV Mar 2022 Pertuzumab (genetical recombination)/ Trastuzumab (genetical recombination) Unresectable advanced or recurrent HER2‐positive colorectal cancer that has progressed after cancer chemotherapy Data of Japanese patients with unresectable advanced or recurrent HER2‐positive colorectal cancer enrolled in the patient registry for advanced solid tumors having specific genetic abnormalities (SCRUM‐Japan) (n = 5a) Standard of care –
V Jun 2022 Vosoritide (genetical recombination) Achondroplasia (ACH) prior to epiphyseal closure

Data of foreign pediatric patients with ACH in the clinical study for collecting natural history data of ACH patients (n = 1,329a).

Note: Additional data from placebo‐controlled trial in pediatrics including Japanese patients were also available

 Standard of care Orphan drug
VI Sep 2022 Selumetinib sulfate Plexiform neurofibromas (PN) in patients with neurofibromatosis type 1 (NF1) Data of foreign patients with NF1 who have PN enrolled in the clinical study for collecting natural history data of NF1 patients (n = 92a) Standard of care Orphan drug
VII Aug 2023 Trastuzumab deruxtecan (genetical recombination) Unresectable advanced or recurrent HER2 (ERBB2) mutation‐positive non‐small cell lung cancer (NSCLC) that has progressed after cancer chemotherapy Data of Japanese patients with unresectable advanced or recurrent HER2(ERBB2) mutation‐positive NSCLC enrolled in the patient registry for advanced lung cancers having specific genetic abnormalities (LC‐SCRUM‐Japan) (n = 140a) Standard of care Orphan drug
VIII Jan 2024 Lonafarnib Hutchinson‐Gilford Progeria Syndrome (HGPS) and the processing‐deficient progeroid laminopathies (PDPL) Data of foreign patients with HGPS enrolled in the international patient registry of the progeria research foundation or identified in the literature and the publicly available database (n = 173a) Standard of care Orphan drug
Open in a new tab

SCRUM‐Japan: cancer genome screening project for individualized medicine in Japan.

a

Shown as the number of patients included in analysis.

Case I: Defibrotide sodium approved for hepatic veno‐occlusive diseases 17

Outline

Defibrotide sodium designated as an orphan drug was approved in June 2019 for the treatment of hepatic veno‐occlusive diseases (VOD) in Japan. Two RWD were available for the PMDA review. One was the historical control data of foreign patients with VOD (n = 32) as an EC based on screening of historical medical records of trial sites for comparison with data of foreign patients with defibrotide sodium in a clinical trial (n = 102: 2005–01 [www.clinicaltrials.gov: NCT00358501] study). 18 The other was from Japanese patients with VOD (n = 107) enrolled in the transplant registry unified management program of the Japanese data center for hematopoietic cell transplantation. 19 This Japanese RWD was included to confirm the appropriateness of a pre‐determined threshold value for showing the efficacy of this drug in a Japanese single‐arm clinical trial and as the EC for comparing with data in that trial (n = 19: FMU‐DF‐002 study). The reason for using these RWD for drug approval was the low feasibility of a randomized controlled clinical trial. In the US, this drug has been used off‐label for more than 10 years for the treatment of VOD, which had a high fatality rate and no approved treatments. In Japan, hematopoietic stem cell transplantation (HSCT) has been typically carried out in small‐sized hospitals in a decentralized way, making it difficult to identify patients with VOD for whom the timing of onset was unexpected, and making it difficult to enroll many VOD patients in a hospital. For efficacy evaluation, comparison of the EC using RWD to defibrotide sodium using clinical trial data was based on survival rate at 100 days after HSCT in both foreign and Japan clinical trials.

Discussion and conclusion in the PMDA review

This drug was approved mainly based on consideration of the following points on the efficacy evaluation related to RWD.

  • VOD was a rare and a potentially life‐threatening complication of HSCT without any approved drugs in Japan.

  • No critical differences between Japanese and foreign populations were observed in the pharmacokinetic profiles and extrinsic factors such as diagnostic procedure and standard therapy for VOD.

  • For foreign data, the observed survival rate (an objective endpoint) in the clinical trial (2005–01 study) was significantly better than the rate in the EC, although the small number of patients in the EC (n = 32), which was lower than the target sample size (n = 80), would result in insufficient power for comparisons and would limit the adjustment of confounding factors. The results were also consistent with those from other clinical trial (FMU‐DF‐002, 99–118 [NCT00003966] and 2006–05 [NCT00628498]) and were better than results in a defibrotide sodium naive population in a meta‐analysis. 20 The results for a secondary endpoint (i.e., complete response rate) also supported the efficacy of this drug.

  • For Japanese data, the point estimate of results in a clinical trial was above the pre‐determined threshold value (30% based on the interim analysis for the Japanese RWD 21 ) and was similar to the foreign clinical trial results. In addition, the final analysis for the Japanese RWD provided supportive information for evaluating the threshold value prespecified in the clinical trial and for the efficacy assessment of this drug in comparison with clinical trial data, although differences between the EC and the clinical trial on the data period (1999–2010 for EC; 2014–2016 for clinical trial) and exclusion criteria (excluding high‐risk patients, such as complications with viral fulminant hepatitis, combination use with anti‐coagulant, having a history of organ transplantation, etc., in the clinical trial but not in the EC) might have limited comparisons due to different baseline patient characteristics of the populations.

Case II: Cerliponase alfa (genetical recombination) approved for neuronal ceroid lipofuscinosis type 2 disease 22

Outline

Cerliponase alfa (genetical recombination) designated as an orphan drug was approved in September 2019 for the treatment of neuronal ceroid lipofuscinosis type 2 disease (CLN2) in Japan. The historical control data of foreign patients with CLN2 (n = 49) enrolled in the patient registry for neuronal ceroid lipofuscin 23 were included as the EC for comparison with the data of cerliponase alfa from a foreign single‐arm clinical trial (n = 23: 190–201/202 [NCT01907087/NCT02485899]). The reason for using these RWD for drug approval was the low feasibility of a randomized controlled clinical trial for CLN2, which was an ultra‐rare disease and for which intraventricular injection of this drug to pediatric patients was ethically unacceptable for the control group. For efficacy evaluation, a comparison of the EC using RWD to cerliponase alfa using clinical trial data was based on motor and language domains of the CLN2 clinical rating scale (ML). 24

Discussion and conclusion in the PMDA review

This drug was approved mainly based on consideration of the following points on the efficacy evaluation related to RWD.

  • CLN2 was an ultra‐rare neurodegenerative disease leading to early death and for which there were no approved drugs in Japan.

  • No critical differences were observed in the pharmacokinetic profiles between Japanese and foreign populations.

  • The delayed time for disease progression attributed to the drug was observed based on the ML score, known as the international standard when comparing cerliponase alfa using clinical trial data with the EC using historical control data in matched populations for some factors such as ages and baseline ML score.

Case III: Olipudase alfa (genetical recombination) approved for acid sphingomyelinase deficiency 25

Outline

Olipudase alfa (genetical recombination) designated as an orphan drug was approved in March 2022 for the treatment of acid sphingomyelinase deficiency (ASMD) in Japan. The historical control data of foreign pediatric patients with ASMD (n = 14) from a prospective clinical study that collected natural history data of pediatric ASMD patients 26 were included as the EC for comparison with the data of olipudase alfa from a foreign single‐arm clinical trial in a pediatric population (n = 15: DFI13803 [NCT02292654]). The reason for using RWD for drug approval was the low feasibility of a randomized controlled clinical trial for ASMD, which was a rare disease and for which it was ethically unacceptable to have a non‐treatment group in pediatric patients. It should be noted that data from a multi‐regional placebo‐controlled clinical trial in adult patients with ASMD including Japanese patients (DFI12712 [NCT02004691]) were also evaluated for the efficacy and safety of this drug. For the efficacy evaluation, comparison of the EC using RWD to olipudase alfa using clinical trial data was mainly based on splenic and liver volumes, platelet count, predicted % DLco, and height Z score.

Discussion and conclusion in the PMDA review

This drug was approved mainly based on consideration of the following points on the efficacy evaluation related to RWD.

  • ASMD was a rare and progressive genetic disorder without any approved drugs in Japan.

  • No critical differences were observed in pharmacokinetic as well as pharmacodynamic (decreased blood lyso‐sphingomyelin concentration) profiles between the Japanese and foreign populations.

  • The efficacy of this drug for ASMD based on objective endpoints such as splenic volume and predicted % DLco was clearly shown in a multi‐regional placebo‐controlled clinical trial in an adult population (DFI12712).

  • Observed efficacy in a pediatric population in a clinical trial (DFI13803) was similar to that in an adult population. In addition, comparisons of pediatric data between the EC and the clinical trial provided supportive information for the efficacy assessment of this drug, although only limited evaluation of the comparisons was possible due to the small number of patients.

Case IV: Pertuzumab (genetical recombination) and trastuzumab (genetical recombination) combination therapy approved for unresectable advanced or recurrent HER2‐positive colorectal cancer that has progressed after cancer chemotherapy 27

Outline

Pertuzumab (genetical recombination) was approved in March 2022 for combination therapy with trastuzumab (genetical recombination) for unresectable advanced or recurrent HER2‐positive colorectal cancer that has progressed after cancer chemotherapy. The historical control data of Japanese patients (n = 5) with similar baseline patient characteristics to clinical trial patients enrolled in the registry for patients with advanced solid tumors having specific genetic abnormalities (SCRUM‐Japan: Cancer genome screening project for individualized medicine in Japan) 28 were included as the EC for comparison with the data of the pertuzumab and trastuzumab combination therapy from the Japanese single‐arm clinical trial (n = 30: TRIUMPH [www.umin.ac.jp: UMIN000027887]). The reason for using RWD for drug approval was the low feasibility of a randomized controlled clinical trial for unresectable advanced or recurrent HER2‐positive colorectal cancer which was a rare target population among patients with colorectal cancer. For the efficacy evaluation, a comparison of the EC using RWD to pertuzumab and trastuzumab combination therapy using clinical trial data was based on response rate according to the response evaluation criteria in solid tumor (RECIST) ver1.1. 29

Discussion and conclusion in the PMDA review

This drug was approved mainly based on consideration of the following points on the efficacy evaluation related to RWD.

  • The cancer type targeted by this drug was a life‐threatening disease with limited treatment options in Japan.

  • The overall response rate in the single‐arm Japanese study (TRIUMPH study) was clinically meaningful because it exceeded the pre‐determined threshold value based on the results of the previous clinical studies. 30 , 31

  • Comparison of the EC using RWD to pertuzumab and trastuzumab combination therapy using clinical trial data also provided supportive information for efficacy assessment of this drug, although only limited evaluation of the comparisons was possible due to the small number of patients and a less established relationship between overall survival and overall response rate.

Case V: Vosoritide (genetical recombination) approved for achondroplasia (ACH) prior to epiphyseal closure 32

Outline

Vosoritide (genetical recombination) designated as an orphan drug was approved in June 2022 for the treatment of achondroplasia (ACH) prior to epiphyseal closure in Japan. The historical control data of foreign pediatric patients with ACH (n = 1,329) enrolled in the clinical study to collect natural history data of ACH patients based on medical records 33 were included as the EC for evaluating long‐term efficacy in comparison with the data of vosoritide from the foreign and multi‐regional single‐arm clinical trials in a pediatric population (n = 10 (≥5 years old): 202/205 [NCT02055157/NCT02724228]; n = 11 (<5 years old): 206/208 [NCT03583697/NCT03989947]). The reason for using RWD for drug approval was the low feasibility of a randomized controlled clinical trial for ACH over a very long‐term period in a rare disease for which non‐treatment of pediatric patients in the control group was ethically unacceptable. It should be noted that data from multi‐regional placebo‐controlled clinical trials in pediatric patients with ACH including Japanese patients (301 [NCT03197766]) were also evaluated for the efficacy and safety of this drug. For the efficacy evaluation, a comparison of the EC using RWD to vosoritide using clinical trial data was mainly based on annualized growth velocity and height Z‐score.

Discussion and conclusion in the PMDA review

This drug was approved mainly based on consideration of the following points on the efficacy evaluation related to RWD.

  • ACH was a rare and serious genetic disorder with limited symptomatic treatment (i.e., growth hormonal therapy) in Japan.

  • No critical differences were observed in pharmacokinetic and pharmacodynamic profiles (increased urine cGMP/creatinine and decreased serum collagen type X marker concentrations) between Japanese and foreign populations.

  • The efficacy of this drug at 52 weeks based on the objective endpoint (i.e., height growth velocity) was clearly demonstrated in a multi‐regional placebo‐controlled clinical trial in patients with ACH aged 5–18 years (301 study).

  • Comparisons of data between EC and clinical trials (≥5 years old: 202/205 studies, <5 years old: 206/208 studies) provided supportive information for efficacy assessment of this drug, although only limited evaluation of the comparisons was possible due to the small number of patients.

Case VI: Selumetinib sulfate approved for plexiform neurofibromas (PN) in patients with neurofibromatosis type 1 (NF1) 34

Outline

Selumetinib sulfate designated as an orphan drug was approved in September 2022 for the treatment of plexiform neurofibromas (PN) in patients with neurofibromatosis type 1 (NF1). The historical control data of foreign patients with NF1 who had PN (n = 92) that were enrolled in the clinical study (NCI‐08‐C‐0079 [NCT00924196]) to collect natural history data of NF1 patients were included as the EC for comparison with the data of selumetinib sulfate from the foreign single‐arm clinical trial (n = 50: D1532C00057 [NCT01362803] 35 ). The reason for using RWD for drug approval was the low feasibility of a randomized controlled clinical trial for PN in patients with NF1, which was a rare disease for which non‐treatment of patients in the control group was ethically unacceptable. For the efficacy evaluation, comparison of the EC using RWD to selumetinib sulfate using clinical trial data was conducted mainly based on response rate according to the response evaluation criteria in neurofibromatosis and schwannomatosis (REiNS). 36

Discussion and conclusion in the PMDA review

This drug was approved mainly based on consideration of the following points on the efficacy evaluation related to RWD.

  • NF1 with PN was a rare and progressive genetic disorder that may have serious complications without any approved drugs in Japan.

  • No critical differences were observed in the pharmacokinetic profiles of Japanese and foreign populations.

  • The response rate (the objective endpoint) in the clinical trial was much higher than the prespecified threshold that was based on natural history data and was significantly better than that in the EC in age‐matched populations.

Case VII: Trastuzumab deruxtecan (genetical recombination) combination therapy approved for unresectable advanced or recurrent HER2 (ERBB2) mutation‐positive NSCLC that has progressed after cancer chemotherapy 37

Outline

Trastuzumab deruxtecan (genetical recombination) was approved in August 2023 for unresectable advanced or recurrent HER2 (ERBB2) mutation‐positive NSCLC that has progressed after cancer chemotherapy. The historical control data of Japanese patients (n = 140) with baseline patient characteristics similar to the clinical trial patients enrolled in the registry for patients with lung cancers having specific genetic abnormalities (LC‐SCRUM‐Japan: Lung cancer genomic screening project for individualized medicine in Japan) 28 , 38 were included as the EC for comparison with the data of trastuzumab deruxtecan combination therapy from the multi‐regional clinical trial (n = 152: U206 [NCT04644237]). The reason for using RWD for drug approval was the low feasibility of a randomized controlled clinical trial for unresectable advanced or recurrent HER2(ERBB2)‐positive NSCLC which was a rare target population among patients with NSCLC. For the efficacy evaluation, comparison of the EC using RWD to trastuzumab deruxtecan combination therapy using clinical trial data was based on response rate according to the response evaluation criteria in solid tumor (RECIST) ver1.1. 29

Discussion and conclusion in the PMDA review

This drug was approved mainly based on consideration of the following points on the efficacy evaluation related to RWD.

  • The cancer type targeted by this drug was a life‐threatening disease with limited treatment options in Japan.

  • The value of the overall response rate in the multi‐regional clinical trial (U206 study) was clinically meaningful based on the pharmacological and biological mechanisms of the target tumor and was higher than that of the previous clinical study. 39

  • Comparison of the EC using RWD to trastuzumab deruxtecan combination therapy using clinical trial data also provided supportive information for the efficacy assessment of this drug, although only limited evaluation of the comparisons was possible due to a less established relationship between overall survival and overall response rates.

Case VIII: Lonafarnib approved for Hutchinson‐Gilford progeria syndrome (HGPS) and the processing‐deficient progeroid laminopathies (PDPL) 40

Outline

Lonafarnib designated as an orphan drug was approved in January 2024 for the treatment of Hutchinson‐Gilford Progeria Syndrome (HGPS) and processing‐deficient progeroid laminopathies (PDPL). The historical control data of foreign patients with HGPS but no prior treatment experience with lonafarnib (n = 173), mainly enrolled in the international patient registry of the progeria research foundation, 41 were included as the EC for comparison with the data of lonafarnib from HGPS patients in the foreign single‐arm clinical trials (n = 62: 07‐01‐0007 [NCT00425607] and 09‐06‐0298). The reason for using RWD for drug approval was the low feasibility of randomized controlled clinical trials in HGPL and PDPL, which was an ultra‐rare genetic disorder for which non‐treatment of patients in the control group was ethically unacceptable. For the efficacy evaluation, comparison of the EC using RWD to lonafarnib using clinical trial data was conducted mainly based on survival rate.

Discussion and conclusion in the PMDA review

This drug was approved mainly based on considerations of the following points on the efficacy evaluation related to RWD.

  • HGPS and PDPL were an ultra‐rare genetic disease leading to early death without any approved drugs in Japan.

  • No critical differences were observed in pharmacokinetic profiles between Japanese and foreign populations.

  • The survival rate (the objective endpoint) in the clinical trial was significantly better than that in the EC in populations matched with some factors such as age, sex, and region (if possible) at baseline among living patients who were born in 1991 (the same year as the oldest patient enrolled in the clinical trial) or later, and whose condition was not severe enough to prevent traveling. It should be noted that the index date for calculating the survival time for both the EC using RWD and lonafarnib using clinical trial data was set at the start date of the treatment for HGPS and PDP in a patient.

  • The results in the supplementary analyses, taking into consideration the impacts of possible biases, basically supported the efficacy of this drug, although potential biases in patient selection and adjustment of confounding factors may exist.

As described above, RWD/RWE have been utilized in regulatory review for drug approval in Japan, although such cases were still limited. Most of the examples that utilized RWD/RWE as an EC for drug approval in Japan were related to orphan diseases. From our perspective, orphan diseases could be one of the major areas for RWD/RWE utilization, because the conduct of randomized controlled clinical trials, which is the gold standard for the benefit/risk assessment of a drug, is generally not feasible due to the limited number of patients eligible for clinical trials and the ethical concerns for non‐treatment of patients without alternative therapies. In the past, the efficacy of a drug targeted for orphan diseases has sometimes been evaluated only in non‐randomized single‐arm clinical trials in Japan. 42 , 43 From this standpoint, the submitted RWD/RWE even with a small sample size were useful for PMDA review for drug approval because they relatively increased the accuracy and robustness of scientific evidence. However, challenges remain in utilizing RWD/RWE as an EC. 4 , 44 , 45 One common limitation seen in the above cases was that RWD was only available from a small number of patients. This situation made the comparison of evidence between the EC using RWD and a drug using clinical trial data more difficult because adjustment of confounding factors was usually incomplete. Since such situation is usually observed in RWD of orphan diseases, further efforts will be needed including the establishment of a registry that intensively accumulates RWD in patients with orphan disease 46 and/or active use of the innovative clinical trial design using Bayesian approaches. 13 , 47 In addition, the impact of unmeasured confounding factors on data evaluation was also a point that should be generally taken into consideration.

PMDA'S CONSIDERATIONS FOR UTILIZING RWD/RWE AS EC

Based on our review experiences, major points for considerations in utilizing RWD/RWE as an EC are summarized in Table 2 .

Table 2.

Points to consider in utilizing RWD/RWE as an EC based on PMDA experiences

Points for considerations Backgrounds General recommendations
Frequent communications with PMDA

  • Regulatory experiences on RWD/RWE utilization as an EC are currently limited

  • Common understanding between PMDA and a sponsor about a concept and strategy of RWD/RWE utilization facilitates proper discussions

  • Drug development strategy based on RWD/RWE including limitations, compliance status to relevant regulation/standards and a plan of clinical data package for new drug application should be shared and discussed with PMDA from an early stage
Feasibility study

  • Limited RWD/RWE from small number of patients make comparisons difficult due to incomplete adjustment of confounding factors between populations

  • A feasibility study is usually conducted at the planning stage of a study but careful considerations are necessary

  • Conduct of a feasibility study may be useful for understanding the possible number of patients for comparisons and considering the limitations more appropriately, but examination of outcome itself and related indexes should be avoided to maintain scientific integrity of a study
Minimize bias for evaluations

  • Biases cannot be completely eliminated in the case of EC based on RWD, and inflation of the type I error rate due to the biases is a critical issue in the efficacy evaluation of a drug

  • Careful consideration is necessary about effects of potential biases and limitations including a bias that is difficult to adjust in the planning stage of a study

  • Analytical methods including adjustment of confounding factors should be prespecified in a protocol for appropriate interpretation of study results

  • Any amendments of the protocol during a study should be clearly recorded and described
Consideration of specific points
Outcome/Endpoint
  • Different degree of interventions between EC and clinical trial may cause a bias
For example,
  • Frequent communications with medical professionals under controlled situations in a clinical trial may add more positive effects on drug response
 Use an objective or clearly defined outcome/endpoint for data comparisons between EC and the target clinical trial
Index date

  • Index date is usually defined as an initiation date of drug administration in a clinical trial, but such date may be difficult to define in case of EC based on RWD

  • Inappropriate definition of index date may cause a bias, especially for a time‐to‐event outcome (e.g., immortal time bias)

 Set a clear and relevant index date for EC based on careful consideration about potential biases
Data period

  • Temporal time difference between RWD and data in a target clinical trial may cause a bias due to changes with the times of clinical practice such as differences in use of standard/concomitant medications and other treatment options

 Set the same data period for EC with that of a target clinical trial
Baseline patient characteristics
  • Imbalance of baseline patient characteristics between EC and clinical trial may cause a bias
For example,
  • Differences in diagnostic procedures, regions of patients, concomitant medications, and comorbidities between populations may cause a bias (e.g., confounding bias)
 Match the population of EC with a target clinical trial population based on inclusion/exclusion criteria of clinical trial as much as possible
Patient enrollment

  • Same patient enrolling to both EC and clinical trial may cause a bias (e.g., allocation bias)

 Avoid duplicated enrollment of a patient into both EC and clinical trial
Sensitivity/supplementary analysis
  • Because of the difficulty in eliminating effects of all potential biases in EC based on RWD, different approaches in terms of study conditions are important to confirm a robustness of evidence
For example,
  • Differences on a frequency of missing values and follow‐up periods between populations may cause a bias
 Sensitivity/supplementary analysis should be planned and conducted with different conditions, analytical methods, and assumptions
Personal information

  • Regulatory review for drug approval may be suspended if critical issues relating to the privacy protection of RWD are identified after the submission of new drug application

 Check with a data holder as early as possible about whether RWD can be used for regulatory submission of new drug application and are appropriately managed in terms of compliance with relevant regulations on privacy protection of personal health information
Data reliability

  • Regulatory review for drug approval may be suspended if critical issues relating to the reliability of RWD are identified after the submission of new drug application

 Check with a data holder as early as possible about whether the management practice of RWD by the data holder is fit for purpose and meet regulatory standards on data reliability
Open in a new tab

EC, external control; RWD, real‐world data; RWE, real‐world evidence.

The first point for consideration is communications with PMDA during drug development. Common understanding between PMDA and a sponsor (i.e., pharmaceutical industry) about a concept and strategy of RWD/RWE utilization is an essential step to facilitate proper discussions for promoting drug development with RWD/RWE under the limited regulatory experience. We encourage a sponsor to communicate with PMDA continuously from the early stage when discussing a drug development strategy based on RWD/RWE. These discussions should include limitations, compliance status to relevant regulations/standards in terms of privacy protection and data reliability, as well as a plan for the clinical data package for a new drug application in Japan.

The second point is about a feasibility study that is usually conducted at the planning stage of a study for EC. Such feasibility study may be useful for understanding the possible number of patients needed for comparisons, and for appropriately considering the limitations. However, examination of the outcome itself and related indexes may lead to inappropriate study planning, which would result in data dredging. 48 Therefore, only a minimum of points that focus on the number of patients rather than on the outcome should be examined in a feasibility study to maintain its scientific integrity.

The third point is about how to minimize bias on the evaluations. 49 As previously reported, 44 , 45 many biases may occur in comparison of EC using RWD to a drug using clinical trial data. It should be recognized that bias cannot be completely eliminated in the case of an EC based on RWD, and inflation of the type I error rate due to bias is a critical issue in the efficacy evaluation of a drug. Therefore, careful consideration is necessary about the effects of potential biases and limitations including a bias that is difficult to adjust in the planning stage of a study. Analytical methods including adjustment of confounding factors should be prespecified in a protocol for appropriate interpretation of study results. If any amendments of the protocol have been made during a study, all changes should be clearly recorded and described. Those efforts will reduce the chance of erroneous conclusion and avoid data dredging. 48

Some more specific points to consider on bias minimization are presented in Table 2 , including general recommendations for (i) use of an objective or clearly defined outcome/endpoint for data comparisons between an EC using RWD and a drug using target clinical trial data (e.g., Cases of I, III, V, VI and VIII in Table 1 ), (ii) setting a clear and relevant index date for an EC based on careful considerations about potential biases (e.g., Case VIII in Table 1 ), (iii) setting the same data period for an EC as that for a target clinical trial (e.g., Cases of I and VIII in Table 1 ), (iv) avoiding duplicate enrollment of the same patient into both an EC and a clinical trial (e.g., Case VIII in Table 1 ), and (v) matching the population of an EC with a target clinical trial population based on inclusion/exclusion criteria of the clinical trial, as much as possible (e.g., Cases of I, II, IV, VI, VII, and VIII in Table 1 ). Sensitivity and supplementary analyses should also be planned and conducted with different conditions, analytical methods, and assumptions because of the awareness about difficulties in eliminating the effects of all potential biases in an EC based on RWD (e.g., Case VIII in Table 1 ). Generally, controlling the frequency of missing values and follow‐up periods in an EC is impossible which may lead to bias. Results from sensitivity and supplementary analyses will be useful to confirm the robustness of evidence, which in turn promotes the appropriate interpretation of study results. When any differences exist even after these careful considerations, the impact of such differences on the data evaluation should be considered and be discussed. It may be of interest to note that insufficient adjustment of confounding factors in propensity score matching have been discussed in the recent unapproved case of autologous myoblast sheet in Japan. 50 , 51 Potential biases caused by inappropriate patient selection, different study periods, and inadequate definition of index data have been also discussed in the review of 131I‐omburtamab in the US. 52

In addition, as early as possible the sponsor should discuss with the holder of RWD about the management of medical and personal information in compliance with relevant regulatory standards on the privacy protection and data reliability. This would ensure that RWD can be properly used for the review and approval of a drug without an inappropriate situation in which the regulatory process is suspended due to any critical issues on these points that are identified after the submission of the new drug application.

Finally, if major points that are difficult to manage or consider appropriately in utilizing RWD/RWE remains even after substantial considerations, alternative approaches such as conduct of a clinical study may be needed and should be discussed with PMDA.

CONCLUSION

In Japan, RWD/RWE have been practically utilized as an EC for drug approval. Most cases have been related to orphan diseases, where the feasibility for conducting randomized controlled clinical trials was low. RWD/RWE utilization as an EC provides additional information that can support regulatory review for drug approval. To appropriately utilize RWD/RWE as an EC, many points should be taken into consideration through all stages of a study, including planning, analysis, and interpretation. Points described in this article promote appropriate drug development based on RWD/RWE and facilitate a proper discussion with PMDA about RWD/RWE utilization. Further accumulation of regulatory experience in PMDA about RWD/RWE utilization will enhance our knowledge and contribute to better regulatory decision making for drug approvals based on RWD/RWE.

FUNDING

All members are employees of PMDA and have received no external funding.

CONFLICT OF INTEREST

The authors declared no competing interests for this work.

ACKNOWLEDGMENTS

The views expressed herein are the result of independent work and do not necessarily represent the views and findings of the Pharmaceuticals and Medical Devices Agency. We also thank all members of the RWD working group in PMDA for their continuous efforts in promoting RWD/RWE utilization in regulatory decision making in Japan.

References

  • 1. Sherman, R.E. et al. Real‐world evidence—what is it and what can it tell us? N. Engl. J. Med. 375, 2293–2297 (2016). [DOI] [PubMed] [Google Scholar]
  • 2. Cave, A. , Kurz, X. & Arlett, P. Real‐world data for regulatory decision making: challenges and possible solutions for Europe. Clin. Pharmacol. Ther. 106, 36–39 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Nishioka, K. , Makimura, T. , Ishiguro, A. , Nonaka, T. , Yamaguchi, M. & Uyama, Y. Evolving acceptance and use of RWE for regulatory decision making on the benefit/risk assessment of a drug in Japan. Clin. Pharmacol. Ther. 111, 35–43 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Franklin, J.M. , Glynn, R.J. , Martin, D. & Schneeweiss, S. Evaluating the use of nonrandomized real‐world data analyses for regulatory decision making. Clin. Pharmacol. Ther. 105, 867–877 (2019). [DOI] [PubMed] [Google Scholar]
  • 5. Sola‐Morales, O. et al. Effectively leveraging RWD for external controls: a systematic literature review of regulatory & HTA decisions. Clin. Pharmacol. Ther. 114, 325–355 (2023). [DOI] [PubMed] [Google Scholar]
  • 6. Campbell, U.B. , Honig, N. & Gatto, N.M. SURF: a screening tool (for sponsors) to evaluate whether using real‐world data to support an effectiveness claim in an FDA application has regulatory feasibility. Clin. Pharmacol. Ther. 114, 981–993 (2023). [DOI] [PubMed] [Google Scholar]
  • 7. Alipour‐Haris, G. , Liu, X. , Acha, V. , Winterstein, A.G. & Burcu, M. Real‐world evidence to support regulatory submissions: a landscape review and assessment of use cases. Clin. Transl. Sci. 17, e13903 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Carrigan, G. et al. External comparator groups derived from real‐world data used in support of regulatory decision making: use cases and challenges. Curr. Epidemiol. Rep. 9, 326–337 (2022). [Google Scholar]
  • 9. Mishra‐Kalyani, P.S. et al. External control arms in oncology: current use and future directions. Ann. Oncol. 33, 376–383 (2022). [DOI] [PubMed] [Google Scholar]
  • 10. Izem, R. , Buenconsejo, J. , Davi, R. , Luan, J.J. , Tracy, L. & Gamalo, M. Real‐world data as external controls: practical experience from notable marketing applications of new therapies. Ther. Innov. Regul. Sci. 56, 704–716 (2022). [DOI] [PubMed] [Google Scholar]
  • 11. Hernán, M.A. & Robins, J.M. Using big data to emulate a target trial when a randomized trial is not available. Am. J. Epidemiol. 183, 758–764 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Ministry of Health, Labour and Welfare . Basic principles on utilization of registry for applications, PSEHB/PED Notification No. 0323–1 /PSEHB/MDED Notification No. 0323‐1 <https://www.pmda.go.jp/files/000239821.pdf> (2021). Accessed December 6, 2024.
  • 13. Weberpals, J. & Wang, S.V. External controls to study treatment effects in rare diseases: challenges and future directions. Clin. Pharmacol. Therap. 116, 1521–1524 (2024). [DOI] [PubMed] [Google Scholar]
  • 14. O'Connell, P. , Ridolfi, A. & Fretault, N. Case study using RWD in the context of a pivotal trial for regulatory approval in a rare disease. J. Biopharm. Stat. 33, 812–819 (2023). [DOI] [PubMed] [Google Scholar]
  • 15. Lasky, T. & Chakravarty, A. Real world data (RWD) in pediatrics. J. Biopharm. Stat. 33, 875–880 (2023). [DOI] [PubMed] [Google Scholar]
  • 16. Yue, L.Q. , Lu, N. & Xu, Y. Designing premarket observational comparative studies using existing data as controls: challenges and opportunities. J. Biopharm. Stat. 24, 994–1010 (2014). [DOI] [PubMed] [Google Scholar]
  • 17. Pharmaceuticals and Medical Devices Agency . Review report: Defibrotide sodium approved for hepatic veno‐occlusive diseases [in Japanese] <https://www.pmda.go.jp/drugs/2019/P20190703002/530263000_30100AMX00006_A100_1.pdf> (2018). Accessed December 6, 2024.
  • 18. Richardson, P.G. et al. Phase 3 trial of defibrotide for the treatment of severe veno‐occlusive disease and multi‐organ failure. Blood 127, 1656–1665 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. The Japanese data center for hematopoetic cell transplantation . Transplant registry unified management program <https://www.jdchct.or.jp/trump/>. Accessed December 6, 2024.
  • 20. Coppell, J.A. et al. Hepatic veno‐occlusive disease following stem cell transplantation: incidence, clinical course, and outcome. Biol. Blood Marrow Transplant. 16, 157–168 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Yakushijin, K. et al. Sinusoidal obstruction syndrome after allogeneic hematopoietic stem cell transplantation: incidence, risk factors and outcomes. Bone Marrow Transplant. 51, 403–409 (2016). [DOI] [PubMed] [Google Scholar]
  • 22. Pharmaceuticals and Medical Devices Agency . Review report: Cerliponase alfa (genetical recombination) approved for ceroid lipofuscin type 2 disease [in Japanese] <https://www.pmda.go.jp/drugs/2019/P20191010003/641173000_30100AMX00236_A100_2.pdf> (2018). Accessed December 6, 2024.
  • 23. Nickel, M. & Schulz, A. Natural history studies in NCL and their expanding role in drug development: experiences from CLN2 disease and relevance for clinical trials. Front. Neurol. 13, 785841 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Wyrwich, K.W. et al. An adapted clinical measurement tool for the key symptoms of CLN2 disease. J. Inborn Errors Metab. Screen. 6, 2326409818788382 (2018). [Google Scholar]
  • 25. Pharmaceuticals and Medical Devices Agency . Review report: Olipudase alfa (genetical recombination) approved for acidic sphingomyelinase deficiency [in Japanese] <https://www.pmda.go.jp/drugs/2022/P20220322002/780069000_30400AMX00191_A100_1.pdf> (2022). Accessed December 6, 2024.
  • 26. Fournier, M. , Msihid, J. , Willemze, A. , Laredo, F. & Pulikottil‐Jacob, R. S295 clinical relevance of spleen volume and platelet count with bleeding events in patients with acid sphingomyelinase deficiency (ASMD). HemaSphere. 6, 398–399 (2022). [Google Scholar]
  • 27. Pharmaceuticals and Medical Devices Agency . Review report: Pertuzumab (genetical recombination) approved for the combination therapy with trastuzumab (genetical recombination) for advanced or recurrent unresectable HER2‐positive colorectal cancer [in Japanese] <https://www.pmda.go.jp/drugs/2022/P20220323001/450045000_22500AMX01001_A100_1.pdf> (2022). Accessed December 6, 2024.
  • 28. Ohtsu, A. , Goto, K. & Yoshino, T. Improvement of patient care using cancer genomic profiling: SCRUM‐/CIRCULATE‐Japan experience. Proc. Jpn. Acad. Ser. B Phys. Biol. Sci. 99, 241–253 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Eisenhauer, E.A. et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur. J. Cancer 45, 228–247 (2009). [DOI] [PubMed] [Google Scholar]
  • 30. Mayer, R.J. et al. Randomized trial of TAS‐102 for refractory metastatic colorectal cancer. N. Engl. J. Med. 372, 1909–1919 (2015). [DOI] [PubMed] [Google Scholar]
  • 31. Grothey, A. et al. Regorafenib monotherapy for previously treated metastatic colorectal cancer (CORRECT): an international, multicentre, randomised, placebo‐controlled, phase 3 trial. Lancet 381, 303–312 (2013). [DOI] [PubMed] [Google Scholar]
  • 32. Pharmaceuticals and Medical Devices Agency . Review Report: Vosoritide (genetical recombination) Approved for Achondroplasia (ACH) Prior to Epiphyseal Closure [in Japanese] <https://www.pmda.go.jp/drugs/2022/P20220708001/641173000_30400AMX00212_A100_1.pdf> (2022). Accessed December 6, 2024.
  • 33. Hoover‐Fong, J.E. et al. Growth in achondroplasia including stature, weight, weight‐for‐height and head circumference from CLARITY: achondroplasia natural history study‐a multi‐center retrospective cohort study of achondroplasia in the US. Orphanet J. Rare Dis. 16, 522 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Pharmaceuticals and Medical Devices Agency . Review Report: Selumetinib approved for neurofibromatosis type 1 (NF1) with plexiform neurofibromas (PN) [in Japanese] <https://www.pmda.go.jp/drugs/2022/P20220926004/870056000_30400AMX00430000_A100_1.pdf> (2022). Accessed December 6, 2024.
  • 35. Kim, A. et al. Characteristics of children enrolled in treatment trials for NF1‐related plexiform neurofibromas. Neurology 73, 1273–1279 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Wolters, P.L. et al. Patient‐reported outcomes in neurofibromatosis and schwannomatosis clinical trials. Neurology 81, S6–S14 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Pharmaceuticals and Medical Devices Agency . Review Report: Trastuzumab deruxtecan (genetical recombination) for unresectable advanced or recurrent HER2‐positive NSCLC that has progressed after cancer chemotherapy [in Japanese] <https://www.pmda.go.jp/drugs/2023/P20230911001/430574000_30200AMX00425_A100_1.pdf> (2022). Accessed December 6, 2024.
  • 38. Miyamoto, S. et al. Clinical development of molecular‐targeted therapies for non‐small cell lung cancer through nationwide genome screening in Japan (LC‐SCRUM‐Japan). Ann. Oncol. 30, v604 (2019). [Google Scholar]
  • 39. Garon, E.B. et al. Ramucirumab plus docetaxel versus placebo plus docetaxel for second‐line treatment of stage IV non‐small‐cell lung cancer after disease progression on platinum‐based therapy (REVEL): a multicentre, double‐blind, randomised phase 3 trial. Lancet 384, 665–673 (2014). [DOI] [PubMed] [Google Scholar]
  • 40. Pharmaceuticals and Medical Devices Agency . Review Report: Lonafarnib approved for Hutchinson‐Gilford Progeria Syndrome (HGPS) and the Processing‐Deficient Progeroid Laminopathies (PDPL) [in Japanese] <https://www.pmda.go.jp/drugs/2024/P20240116001/111298000_30600AMX00019_A100_1.pdf> (2023). Accessed December 6, 2024.
  • 41. Progeria Research Foundation . International progeria patient registry <https://www.progeriaresearch.org/international‐registry‐2/>. Accessed December 6, 2024.
  • 42. Nakayama, H. & Tsukamoto, K. Unique characteristics of regulatory approval and pivotal studies of orphan anticancer drugs in Japan. Invest. New Drugs 36, 702–708 (2018). [DOI] [PubMed] [Google Scholar]
  • 43. Sakushima, K. , Takeda, H. & Aoi, Y. Orphan drug designation and development in Japan: 25 years of experience and assessment. Nat. Rev. Drug Discov. 20, 893–894 (2021). [DOI] [PubMed] [Google Scholar]
  • 44. Burger, H.U. et al. The use of external controls: to what extent can it currently be recommended? Pharm. Stat. 20, 1002–1016 (2021). [DOI] [PubMed] [Google Scholar]
  • 45. US Food and Drug Administration . Guidance for industry (Draft): Considerations for the design and conduct of externally controlled trials for drug and biological products <https://www.fda.gov/media/164960/download> (2023). Accessed December 6, 2024.
  • 46. National Center of Neurology and Psychiatry Translational Medical Center . Remudy (Registry of Muscular Dystrophy) <https://remudy.ncnp.go.jp/>. Accessed December 6, 2024.
  • 47. Hampson, L.V. , Whitehead, J. , Eleftheriou, D. & Brogan, P. Bayesian methods for the design and interpretation of clinical trials in very rare diseases. Stat. Med. 33, 4186–4201 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Berger, M.L. et al. Good practices for real‐world data studies of treatment and/or comparative effectiveness: recommendations from the joint ISPOR‐ISPE special task force on real‐world evidence in health care decision making. Value Health 20, 1003–1008 (2017). [DOI] [PubMed] [Google Scholar]
  • 49. Gray, C. et al. Use of quantitative bias analysis to evaluate single‐arm trials with real‐world data external controls. Pharmacoepidemiol. Drug Saf. 33, e5796 (2024). [DOI] [PubMed] [Google Scholar]
  • 50. Pharmaceuticals and Medical Devices Agency . Review report for heartsheet (autologous myoblast sheet) [in Japanese] <https://www.pmda.go.jp/regenerative_medicines/2024/R20240808001/470034000_XXXXXXXXXXXXX_A100_1.pdf> (2024). Accessed December 6, 2024.
  • 51. Ministry of Health, Labour and Welfare . Discussion summary of the advisory committee on regenerative medicine and biotechnological products on HEARTSHEET [in Japanese] <https://www.mhlw.go.jp/content/11121000/001277653.pdf> (2024). Accessed December 6, 2024.
  • 52. US Food and Drug Administration . The Oncologic Drugs Advisory Committee Meeting (October 28, 2022), 131 I‐Omburtamab for neuroblastoma with central nervous system or leptomeningeal metastases, FDA Introductory Comments <https://www.fda.gov/media/162701/download> (2022). Accessed December 6, 2024.

Articles from Clinical Pharmacology and Therapeutics are provided here courtesy of Wiley and American Society for Clinical Pharmacology and Therapeutics

RESOURCES