Abstract
Abstract
Objectives
To compare the premarket and postmarket evidence of safety and efficacy of direct oral anticoagulants approved for stroke prevention in atrial fibrillation patients across four major regulatory agencies.
Design
Cross-sectional.
Setting
European Medicines Association (EMA), US Food and Drug Administration (FDA), Health Canada and Australian Therapeutic Goods Administration (TGA).
Participants
Apixaban, dabigatran, edoxaban and rivaroxaban marketing authorisations.
Outcome measures
Concordance among regulatory agencies with respect to (1) premarket evidence used to establish efficacy and safety and (2) postmarket safety boxed warnings and postmarketing study requirements.
Results
Apixaban, dabigatran and rivaroxaban were approved by each of the four regulatory agencies; edoxaban was only not approved by TGA. For premarket efficacy evidence, there was concordance across all agencies in terms of phase 3 trials for three (75%) drugs, sample size for three (75%) drugs, primary endpoints for four (100%) drugs, numerical results for three (75%) drugs, agency interpretation of results for four (100%) drugs and number of phase 2 trials for three (75%) drugs. For the premarket safety evidence, there was concordance across all agencies in terms of phase 3 trials for three (75%) drugs, sample size for two (50%) drugs, primary endpoints for four (100%) drugs, numerical results for three (75%) drugs, agency interpretation of results for three (75%) drugs and number of phase 2 trials for zero (0%) drugs. For postmarket safety information, FDA was the only agency that issued boxed warnings (for three (75%) drugs). Additionally, EMA and TGA required postmarketing studies (for four (100%) and two (50%) drugs, respectively), while FDA and Health Canada did not have any postmarketing requirements.
Conclusions
There was a high degree of concordance in the phase 3 trial premarket evidence used to establish efficacy and safety of direct oral anticoagulant approvals across four major regulatory agencies, but discordance in the phase 2 trial premarket evidence used, as well as in postmarket safety boxed warnings and postmarketing study requirements. These discrepancies highlight opportunities for further harmonisation in the evaluation and regulation of medical products globally.
Keywords: safety, health policy, clinical trial, neurology
STRENGTHS AND LIMITATIONS OF THIS STUDY.
By examining premarket evidence and postmarket requirements across four major regulatory agencies, our study provides a comprehensive comparison of regulatory practices and standards.
While focusing on direct oral anticoagulants provides insights into the approval evidence for a commonly used drug class, regulatory agency alignment may differ for other drug classes.
Introduction
Atrial fibrillation, the most common cardiac arrhythmia, affects millions of individuals worldwide and remains a significant contributor to morbidity and mortality due to associated consequences, including stroke.1 Strokes associated with atrial fibrillation have poorer outcomes, including prolonged hospitalisation and increased morbidity and mortality.2 Traditionally, vitamin K antagonists like warfarin have been the cornerstone of oral anticoagulation therapy for stroke prevention in atrial fibrillation patients; however, the introduction of direct oral anticoagulants offers several advantages that have transformed anticoagulation therapy, including reduced monitoring needs, quicker onset and offset effects and fewer drug interactions.3 These drugs, including direct thrombin inhibitor (dabigatran) and factor Xa inhibitors (rivaroxaban, apixaban and edoxaban), have gained widespread acceptance in clinical practice, with new prescriptions surpassing those of warfarin for atrial fibrillation patients.4
As for all prescription medications, the market authorisation and postmarketing surveillance of direct oral anticoagulants are managed by regulatory agencies, which require evidence of safety and efficacy from well-designed clinical trials.5 6 While different regulatory agencies across the world have similar overall missions, there are differences in policies, processes, data requirements and timelines of regulatory decisions.7,11 With the globalisation of drug development, there has been a growing call for regulatory authorities to harmonise drug regulations and processes and collaborate more closely.12,14 These efforts could reduce clinical trial duplication and development costs, thereby improving efficiency, facilitating pharmaceutical innovation and ensuring timely access to safe and effective treatment.14
While one might expect regulatory agencies across the world to rely on the same clinical evidence when approving new drug applications, one study examining the market authorisation of extended-release methylphenidate found that approval decisions made by the regulatory agencies in Australia, Europe, Japan and the USA were based on different clinical trials.15 Other studies have shown overall high concordance between the decisions made by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) regarding drug approvals.16 17 The most common reason for discordance in initial marketing authorisation decisions was a difference in efficacy conclusions made by FDA and EMA despite evaluating similar clinical efficacy data.17 Moreover, there may be differences in the data available to inform regulatory decisions, as sponsors usually submit market authorisation applications earlier to the FDA compared with the EMA.7 18 19
Direct oral anticoagulants, approved in the early 2010s, by which time many regulatory agencies were making marketing authorisation documents publicly available, offer a unique opportunity to examine differences in premarket and postmarket evidentiary standards between regulatory agencies around the world. These drugs were expected to be rapidly adopted and have had large budget implications for countries given their costs, so the evidence establishing their safety and efficacy was widely scrutinised. Accordingly, we compared the safety and efficacy evidence supporting the approval of direct oral anticoagulants for stroke prevention in atrial fibrillation patients across four major regulatory agencies; EMA, FDA, Health Canada and Australian Therapeutic Goods Administration (TGA). We also examined the most serious safety warnings in addition to the required postmarketing studies issued by the regulatory agencies at the time of approval.
Methods
This study did not require institutional review board approval because it was based on publicly available information in accordance with Code of Federal Regulations under Title 45, Part 46. Informed consent was not required because no patient data were used. This cross-sectional study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology reporting guideline.20
Data sources
We included four main direct oral anticoagulants (apixaban, dabigatran, edoxaban and rivaroxaban) in this analysis. For each drug, we searched for new drug applications or supplemental indication approval documents on the following agencies’ websites: EMA’s European Public Assessment Report,21 Drugs@FDA,22 Health Canada’s Drug Product Database23 and TGA’s Australian Public Assessment Report.24 For each oral anticoagulant, we searched for the earliest version of the approval letter and/or medical review that described the evidence supporting the approved indication for stroke prevention among patients with atrial fibrillation. Because Health Canada’s website contained only the most recently revised product monograph and prescribing information, we corresponded with Health Canada to obtain historic documents dated to the time of the approval.
For each approval, we reviewed the medical review documents to identify the dates when the application was received by the agency and the date the approval was granted. Next, we determined the phase 2 and phase 3 clinical trials that established product efficacy and/or safety to support the agency’s approval, excluding trials that were conducted in non-atrial fibrillation populations, such as trials related to the prevention or treatment of venous thromboembolism, as well as trials that did not assess efficacy and/or safety, such as those that only examined pharmacokinetics or pharmacodynamics. Data extraction was performed and validated by two authors (TZ and MM), with disagreements being resolved by discussion among all three authors (TZ, MM and JSR).
Premarket evidence used to establish efficacy
For each phase 3 trial, we extracted the trial identifier, sample size, primary efficacy endpoint(s) and the corresponding numerical results and regulatory agencies’ interpretation of these endpoints. We used data extraction strategies consistent with prior studies.25,27 Sample size was defined as the number of individuals in the cohort used for the endpoint analysis informing the agencies’ approval. We identified endpoints’ numerical results and their measurement scale (eg, hazard ratio or other estimates) and categorised the agencies’ interpretations of clinical trials as positive or negative, based on whether the primary efficacy endpoint achieved statistical significance while taking into consideration the summary statements made by the agencies.
We also recorded the total number of phase 2 trials mentioned in the main text of the regulatory document for efficacy analyses. We excluded phase 2 trials that were not mentioned in the main text of regulatory documents but may have been identified in regulatory documents’ appendices.
Premarket evidence used to establish safety
Similar to the efficacy outcomes, for each phase 3 trial, we extracted the trial identifier, sample size, primary safety endpoint(s) and the corresponding numerical results and regulatory agencies’ interpretation of those endpoints. We also recorded the total number of phase 2 trials mentioned in the main text of the regulatory document for safety analyses.
Postmarket safety labelling
We reviewed the initial product labels to identify boxed warnings issued by the regulatory agencies. As a part of postmarket safety labelling, regulatory agencies use specific warnings to address serious safety concerns: FDA and TGA use ‘Boxed Warnings’, and Health Canada uses ‘Serious Warnings and Precautions Boxes’ to highlight critical risks associated with a drug. These mechanisms act as risk mitigation strategies to emphasise serious specific warning statements for healthcare providers and patients.28,30 EMA does not issue Boxed warnings or any equivalent safety labelling.
Postmarketing study requirements
We reviewed the approval documents to identify any clinical postmarketing studies that were required by the agency at the time of approval. We classified the postmarketing studies based on whether they involved conducting new studies or completion of ongoing studies. We excluded in-vitro studies, postmarketing requirements related to routinely mandatory paediatric studies and routine risk minimisation activities or pharmacovigilance activities such as educational materials, modified prescription event monitoring studies, prescriber or patient surveys and drug utilisation database studies.
Concordance analysis
We compared each of the following elements of premarket and postmarket information used by regulators to establish drug efficacy and safety and classified them as concordant, discordant or partially concordant: specific phase 3 trials, primary efficacy or safety endpoints, sample sizes to support primary efficacy or safety endpoints’ analyses, primary efficacy or safety endpoints’ numerical results, agencies’ interpretations of primary efficacy or safety endpoints and number of phase 2 trials. Sample size was considered concordant across agencies if they had numerical equivalence. Numerical results were considered concordant if effect estimates and/or 95% confidence intervals (CIs) or p values were numerically equivalent. Lastly, the number and content of boxed warnings stated in the safety labels in addition to the number and purpose of postmarketing studies required by the four agencies were compared.
Patient and public involvement
Patients and the public were not involved in the planning, design and implementation of the study, as this study used secondary data. No patients were asked to advise on the interpretation or writing of the manuscript.
Results
Apixaban
The application for apixaban for the indication of stroke prevention among patients with atrial fibrillation was received by EMA and FDA in September 2011 and by Health Canada and TGA in December 2011 (online supplemental table1). Following the agencies’ evaluation, apixaban was approved by TGA in June 2012, by EMA in September 2012, by Health Canada in November 2012 and by FDA in December 2012.
Premarket efficacy evidence
The information used by the four agencies to establish apixaban’s efficacy, including the phase 3 trials, efficacy sample size, primary efficacy endpoints, numerical results and regulatory agencies’ interpretation of these endpoints, as well as the number of phase 2 trials, were concordant across all pairwise comparisons (table 1 and online supplemental table 2).
Table 1. Efficacy summary of concordance of clinical trials supporting regulatory approval for direct oral anticoagulants across four major regulatory agencies.
| Agency pair | Phase three trial | Sample size | Primary efficacy endpoint | Numerical result | Agency interpretation | Number of phase 2 trials |
| Apixaban | ||||||
| FDA-TGA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| FDA-HC | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| FDA-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| TGA-HC | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| TGA-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| HC-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| Dabigatran | ||||||
| FDA-TGA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| FDA-HC | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| FDA-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | DiscordantFDA: 0, EMA: 2 |
| TGA-HC | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| TGA-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | DiscordantTGA: 0, EMA: 2 |
| HC-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Discordant:HC: 0, EMA: 2 |
| Edoxaban | ||||||
| FDA-TGA* | NA | NA | NA | NA | NA | NA |
| FDA-HC | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| FDA-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| TGA*-HC | NA | NA | NA | NA | NA | NA |
| TGA*-EMA | NA | NA | NA | NA | NA | NA |
| HC-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| Rivaroxaban | ||||||
| FDA-TGA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| FDA-HC | DiscordantHC missing J-ROCKET trial | DiscordantFDA: ROCKET: 13,962,HC: ROCKET AF: 14 143 | Concordant | Partially concordant† | Concordant | Concordant |
| FDA-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| TGA-HC | DiscordantHC missing J-ROCKET trial | DiscordantTGA: ROCKET: 13,962,HC: ROCKET AF: 14 143 | Concordant | Partially concordant† | Concordant | Concordant |
| TGA-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| HC-EMA | DiscordantHC missing J-ROCKET trial | DiscordantHC: ROCKET AF: 14,143,EMA: ROCKET AF: 13 962 | Concordant | Partially concordant† | Concordant | Concordant |
Green: concordant
Yellow: partially concordant
Red: discrodant
No safety interpretation was made by Health Canada and EMA for one of the two trials. Edoxaban is not approved in Australia for stroke prevention in atrial fibrillation patients.
Concordant for the trial that was considered by both agencies. The other trial was not considered by Health Canada.
EMA, European Medicines Agency; FDA, Food and Drug Administration; HC, Health Canada; TGA, Therapeutic Goods Administration
Premarket safety evidence
The information used by the four agencies to establish apixaban’s safety, including the phase 3 trials, safety sample size, primary safety endpoints and numerical results of safety endpoints, was concordant across all pairwise comparisons (table 2 and online supplemental table 3). However, for one (AVERROES (NCT00496769)) of the two phase 3 trials, no safety interpretation was made by either EMA or Health Canada; the interpretation of the other trial (ARISTOTLE (NCT00412984)) was concordant across all agencies. Moreover, the number of phase 2 trials discussed by the agencies was discordant: while EMA and TGA used one phase 2 trial, FDA and Health Canada did not use any phase 2 trials for determining the apixaban’s safety.
Table 2. Safety summary of concordance of clinical trials supporting regulatory approval for direct oral anticoagulants across four major regulatory agencies.
| Agency pair | Phase 3 trial | Safety sample size | Primary safety endpoint | Safety numerical result | Agency safety interpretation | Number of phase 2 trials |
| Apixaban | ||||||
| FDA-TGA | Concordant | Concordant | Concordant | Concordant | Concordant | DiscordantFDA: 0, TGA: 1 |
| FDA-HC | Concordant | Concordant | Concordant | Concordant | Partially concordant * | Concordant |
| FDA-EMA | Concordant | Concordant | Concordant | Concordant | Partially concordant * | Discordant:FDA: 0, EMA: 1 |
| TGA-HC | Concordant | Concordant | Concordant | Concordant | Partially concordant * | Discordant:TGA: 1, HC: 0 |
| TGA-EMA | Concordant | Concordant | Concordant | Concordant | Partially concordant * | Concordant |
| HC-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Discordant:HC: 0, EMA: 1 |
| Dabigatran | ||||||
| FDA-TGA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| FDA-HC | Concordant | Concordant | Concordant | Concordant | Concordant | Discordant:FDA: 3, HC: 0 |
| FDA-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| TGA-HC | Concordant | Concordant | Concordant | Concordant | Concordant | Discordant:TGA: 3, HC: 0 |
| TGA-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| HC-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Discordant:EMA: 3, HC: 0 |
| Edoxaban | ||||||
| FDA-TGA | Not applicable† | Not applicable† | Not applicable† | Not applicable† | Not applicable† | Not applicable† |
| FDA-HC | Concordant | Discordant:FDA: 21,026,HC: 14 024‡ | Concordant | Partially concordant§ | Concordant | Discordant:FDA: 5, HC: 0 |
| FDA-EMA | Concordant | Concordant | Concordant | Concordant | Concordant | Discordant:FDA: 5, EMA: 6 |
| TGA -HC | Not applicable† | Not applicable† | Not applicable† | Not applicable† | Not applicable† | Not applicable† |
| TGA -EMA | Not applicable† | Not applicable† | Not applicable† | Not applicable† | Not applicable† | Not applicable† |
| HC-EMA | Concordant | Discordant:EMA: 21,026,HC: 14 024‡ | Concordant | Partially concordant§ | Concordant | Discordant:HC: 0, EMA: 6 |
| Rivaroxaban | ||||||
| FDA-TGA | Concordant | Concordant | Concordant | Concordant | Concordant | DiscordantFDA: 0, TGA: 2 |
| FDA-HC | Concordant | Concordant | Concordant | Concordant | Concordant | Concordant |
| FDA-EMA | Partially concordant¶ | Discordant¶FDA: 14,236,EMA: 15 514 | Concordant | Concordant** | Concordant | Concordant |
| TGA-HC | Concordant | Concordant | Concordant | Concordant | Concordant | DiscordantTGA: 2, HC: 0 |
| TGA-EMA | Partially concordant¶ | Discordant¶TGA: 14,236,EMA: 15 514 | Concordant | Concordant** | Concordant | Discordant:TGA: 2, EMA: 0 |
| HC-EMA | Partially concordant¶ | Discordant¶HC: 14,236,EMA: 15 514 | Concordant | Concordant** | Concordant | Concordant |
Green: concordant
Yellow: partially concordant
Red: discrodant
No safety interpretation was made by Health Canada and EMA for one of the two trials. The interpretation for the other trial was concordant across all agencies.
Edoxaban is not approved in Australia for stroke prevention in atrial fibrillation patients.
Safety analyses done by FDA and EMA included three trial arms (Wwarfarin, Eedoxaban 30 mg mg, and Eedoxaban 60 mg mg); however, Health Canada’s safety analysis included only two arms (Wwarfarin and Eedoxaban 60 mg mg).
Concordant for the analysis corresponding to the trial’s arms that were considered for both agencies. One of the trial arms was not considered by Health Canada.
EMA considered the pooled analysis from two trials (ROCKET AF and J-ROCKET) while the other agencies only considered the analysis from ROCKET AF trial.
There was a minimal difference between the values (HR ([95% CI CI)]=1.03 (0.96, 1.11)] versus HR ([(95% CI CI)]=1.04 (0.97, 1.11)]).
EMA, European Medicines Agency; FDA, Food and Drug Administration; HC, Health Canada; TGA, Therapeutic Goods Administration
Postmarket safety labelling and requirements
FDA issued one boxed warning for apixaban, for the increased risk of stroke on discontinuation without adequate continuous anticoagulation, whereas neither TGA nor Health Canada issued a boxed warning, for any safety risk, for apixaban (table 3). As a postmarketing requirement, EMA required the completion of one clinical study for apixaban; however, the other three agencies did not require any studies to confirm clinical safety or efficacy after market authorisation (online supplemental table 4).
Table 3. Serious safety or boxed warnings issued by regulatory agencies for direct oral anticoagulants*.
| Drug | FDA’s boxed warnings | Health Canada’s serious warnings and precautions boxes | TGA’s boxed warnings |
| Apixaban |
|
None | None |
| Dabigatran | None | None | None |
| Edoxaban |
|
None | None |
| Rivaroxaban |
|
None | None |
No equivalent to Bboxed warnings issued by EMA was found.
FDAFood and Drug AdministrationTGATherapeutic Goods Administration
Dabigatran
The application for dabigatran for the indication of stroke prevention among patients with atrial fibrillation was received by FDA and Health Canada in December 2009 and by EMA and TGA in January 2010. Following the agencies’ evaluation, dabigatran was approved by FDA and Health Canada in October 2010, TGA in April 2011 and EMA in June 2011.
Premarket efficacy evidence
The information used by the four agencies to establish dabigatran’s efficacy, including the phase 3 trials, efficacy sample size, primary efficacy endpoints and numerical results and regulatory agencies’ interpretation of these endpoints, was concordant across all pairwise comparisons. Additionally, the number of phase 2 trials was concordant among the FDA, Health Canada and TGA but discordant with EMA: while EMA used two phase 2 trials, other agencies did not use information from any phase 2 trials.
Premarket safety evidence
The information used by the four agencies to establish dabigatran’s safety, including the phase 3 trials, safety sample size, primary safety endpoints and numerical results and agencies’ interpretation of these endpoints, was concordant across all pairwise comparisons. Additionally, the number of phase 2 trials was concordant across the EMA, FDA and TGA but discordant with Health Canada: while EMA, FDA and TGA used three phase 2 trials, Health Canada did not use any phase 2 trials to establish the dabigatran’s safety.
Postmarket safety labelling and requirements
FDA, Health Canada and TGA did not include any boxed warnings for dabigatran. As postmarketing requirements, EMA and TGA required the conduct of two new clinical studies for dabigatran; however, the other two agencies did not require any studies to confirm clinical safety or efficacy after market authorisation.
Edoxaban
Edoxaban has not been approved by TGA. The application for edoxaban for the indication of stroke prevention among patients with atrial fibrillation was received by EMA and FDA in January 2014 and by Health Canada in September 2015. Following the agencies’ evaluation, edoxaban was approved by FDA in January 2015, by EMA in April 2015 and by Health Canada in November 2016.
Premarket efficacy evidence
The information used by the three agencies to establish edoxaban’s efficacy, including the phase 3 trials, efficacy sample size, primary efficacy endpoints, numerical results and regulatory agencies’ interpretation of these endpoints, as well as the number of phase 2 trials, was concordant across all pairwise comparisons.
Premarket safety evidence
The information used by the three agencies to establish edoxaban’s safety, including the phase 3 trials, primary safety endpoints and agencies’ interpretation of these endpoints, was concordant across all pairwise comparisons. However, the sample size supporting primary safety endpoint analysis by Health Canada was discordant with those supporting primary safety endpoint analyses by EMA and FDA: while the safety analyses done by EMA and FDA included three trial arms (warfarin, edoxaban 30 mg and edoxaban 60 mg), Health Canada’s safety analysis included only two arms (warfarin and edoxaban 60 mg). Therefore, the safety numerical results were considered partially concordant in the pairwise comparisons between FDA-Health Canada and EMA-Health Canada: the numerical results were concordant for the analyses corresponding to the trial’s arms that were considered by both agencies but could not be evaluated for the other trial arm that was not considered by Health Canada. Additionally, the number of phase 2 trials was overall discordant across the three agencies: FDA and EMA used five and six phase 2 trials, respectively, while Health Canada did not use any phase 2 trials to establish the safety of edoxaban.
Postmarket safety labelling and requirements
FDA issued one boxed warning for edoxaban, specifying three warnings (for risk of reduced efficacy in patients with certain creatinine clearance, risk of ischaemic events on premature discontinuation and risk of spinal or epidural haematoma), whereas Health Canada did not issue a boxed warning, for any safety risk, for edoxaban. As a postmarketing requirement, EMA required the conduct of one new clinical study for edoxaban; however, FDA and Health Canada did not require any studies to confirm clinical safety or efficacy after market authorisation.
Rivaroxaban
The application for rivaroxaban for the indication of stroke prevention among patients with atrial fibrillation was received by EMA and FDA in January 2011, by Health Canada in February 2011 and by TGA in March 2011. Following the agencies’ evaluation, rivaroxaban was approved by EMA in September 2011, by FDA in November 2011, by Health Canada in January 2012 and by TGA in April 2012.
Premarket efficacy evidence
The information used to establish rivaroxaban’s efficacy regarding the phase 3 trials was concordant across EMA, FDA and TGA but not with Health Canada: Health Canada only considered one (ROCKET AF trial (NCT00403767)) of the two trials (ROCKET AF (NCT00403767) and J-ROCKET (CT00494871)) that was considered by the other three agencies. Moreover, there was discordance in terms of the sample size of the ROCKET AF trial between Health Canada and the other agencies. Information corresponding to primary efficacy endpoints of the phase 3 trials and agencies’ interpretation of these endpoints were concordant across all pairwise comparisons. The numerical results of the efficacy endpoints were concordant among EMA, FDA and TGA for both trials and were also concordant with that of the trial considered by Health Canada. Lastly, the number of phase 2 trials was concordant across all pairwise comparisons.
Premarket safety evidence
The information used to establish rivaroxaban’s safety regarding the phase 3 trials was concordant across FDA, Health Canada and TGA but not with EMA: while EMA considered the pooled analysis from two trials (ROCKET AF (NCT00403767) and J-ROCKET (NCT00494871)), the other agencies considered only the analysis from the ROCKET AF trial. This also caused discordance in terms of sample size between EMA and the other three agencies. Information corresponding to primary efficacy endpoints, numerical results and agencies’ interpretation of these endpoints was concordant across all pairwise comparisons. Lastly, the number of phase 2 trials was concordant across the EMA, FDA and Health Canada but not with TGA: while TGA used two phase 2 trials, EMA, FDA and Health Canada did not include information from any phase 2 trials to establish rivaroxaban’s safety.
Postmarket safety labelling and requirements
FDA issued two boxed warnings for rivaroxaban (for risk of ischaemic events on discontinuation and risk of spinal or epidural haematoma), whereas neither TGA nor Health Canada issued a boxed warning, for any safety risk, for rivaroxaban. As the postmarketing requirement, EMA required the conduct of one new clinical study and TGA required the conduct of two new clinical studies for rivaroxaban; however, the FDA and Health Canada did not require any studies to confirm clinical safety or efficacy after market authorisation.
Discussion
In this cross-sectional study of the evidence used to establish premarket efficacy and safety, as well as postmarket safety labelling and study requirements, of direct oral anticoagulants by four regulatory agencies, we found high rates of concordance, particularly with respect to information used to establish premarket efficacy. For safety, regulators were less likely to report relying on the same premarket evidence and less likely to include the same postmarket safety labelling information. Additionally, there were inconsistencies in postmarketing study requirements across the agencies. These discrepancies, which might stem from variations in assessment criteria and regulatory standards across regulatory agencies, highlight the opportunities for further harmonisation in the evaluation and regulation of direct oral anticoagulants globally.
With the exception of one drug (edoxaban), which has not been approved in Australia, direct oral anticoagulants were approved by the four regulators within a relatively short timeframe in Australia, Canada, Europe and the USA. These findings expand on and confirm the results of previous research showing high concordance between the decisions made by EMA and FDA regarding many drug approvals.16 17 19 On examining the premarket evidentiary basis of the approvals, we found that phase 3 trials and their primary endpoints were broadly concordant across the four agencies. There was only one instance where a phase 3 trial was not considered and another where an arm of the phase 3 study (corresponding to the lower dose of the drug) was not considered by Health Canada.
On the other hand, the phase 2 trials used as premarket evidence were frequently discordant across the agencies, with EMA considering phase 2 trials more frequently compared with the other three agencies. Although the International Conference on Harmonisation’s guidelines and its member countries’ regulations recommend or require drug sponsors to submit ‘pertinent’ and relevant data,31 none of the agencies examined in our study explicitly require the submission of all clinical trials conducted.32,35 Thus, our findings could be explained by differences in drug sponsors’ submissions of phase 2 trials, as well as differences in what trials regulatory agencies considered pertinent and relevant. For instance, some phase 2 trials may not have been complete at the time of submission to one regulator but were available for later submission to another regulator. In addition, certain phase 2 trials may have been initiated in response to certain regulatory concerns or jurisdictional needs, such as studies that enrolled patients of specific race, ethnicity or citizenship.
Regulatory agencies were more frequently discordant with respect to the premarket safety evidence and postmarket safety labelling and study requirements. While the specific phase 3 trials and their primary safety endpoints were often concordant across regulatory agencies, there were some discordances in terms of sample size and agencies’ interpretation of safety results. Moreover, FDA was the only agency that issued boxed warnings on the drugs’ labels to emphasise the most serious associated risks. These findings are consistent with previous studies showing a low level of concordance among regulators in issuance of safety warnings for approved drugs.36,38 The absence of boxed warnings on the labels in countries other than the USA may suggest different risk tolerance thresholds or regulatory standards for safety labelling across different agencies. These could also be justified by the variation in regional characteristics, including the potential differences in susceptibility to certain side effects, that might affect safety considerations.39 Nonetheless, while regulatory agencies may employ alternative mechanisms for communicating safety information, the lack of standardised warnings could potentially lead to inconsistencies in risk perception among clinicians and patients in different regions. By promoting standardised approaches to risk communication and labelling, regulatory agencies could improve safety messaging and facilitate informed decision-making in healthcare.
Our analysis also uncovered inconsistencies in postmarketing study requirements across the four regulatory agencies. Specifically, EMA and TGA required postmarketing studies more frequently compared with FDA and Health Canada. Given the potential for postmarketing requirements to further evaluate the drug safety and efficacy and address uncertainties regarding the drugs’ risk-benefit profiles, regulatory agencies could consider aligning postmarketing requirements and mutual recognition of postapproval studies conducted in different regions. Such efforts would enhance efficiencies for sponsors and potentially maximise the likelihood that these clinical studies are completed in a timely manner, which has not happened consistently,40 so that the evidence is generated and available to inform clinical care decisions.
Our study should be considered in the context of its limitations. First, our study was limited to four regulatory agencies and our findings may differ if we had included agencies serving other prominent jurisdictions, including the UK, Japan and China. Second, we relied on publicly available data in the agencies’ approval documents, though other information may have been considered by the agencies when making approval decisions, including clinical trials not listed in the marketing authorisation documents. Third, for postmarket safety labelling, we exclusively examined boxed warnings and did not consider other forms of safety communication, such as contraindications, precautions or adverse reactions listed on the drug labels. While focusing on boxed warnings provides insight into the most critical safety issues, comparing less severe but clinically important safety information was not feasible due to the significant variation in how each agency categorises and reports this information. Lastly, this study was limited to approval of direct oral anticoagulants as an illustrative case example. Regulatory agency alignment may differ for other drug classes or other therapeutic uses. Future research should examine differences among regulatory agencies for other therapeutic classes and products, such as antineoplastic agents, medications used for diabetes management, or drugs indicated to treat rare diseases.
Conclusion
This cross-sectional examination of direct oral anticoagulant approvals across four major regulatory agencies found overall high concordance in the phase 3 trials used for premarket evidence of efficacy, but discordance in the phase 2 trial information, postmarket safety labelling and postmarketing requirements. Although the similarities in the information used by EMA, FDA, TGA and Health Canada suggest that continuous collaborative efforts in regulatory science may have led to harmonisation across the regulatory agencies, opportunities exist to ensure more consistent standards in the evaluation and regulation of medical products globally.
supplementary material
Footnotes
Funding: This work was supported by the Yale University-Mayo Clinic Center of Excellence in Regulatory Science and Innovation (CERSI) Scholars Program. However, the funder had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review or approval of the manuscript and decision to submit the manuscript for publication.
Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2024-090376).
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Patient and public involvement: Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.
Contributor Information
Maryam Mooghali, Email: maryam.mooghali@yale.edu.
Tianna Zhou, Email: tianna.zhou@yale.edu.
Joseph S Ross, Email: joseph.ross@yale.edu.
Data availability statement
Data are available upon reasonable request.
References
- 1.Iwasaki Y, Nishida K, Kato T, et al. Atrial Fibrillation Pathophysiology. Circulation. 2011;124:2264–74. doi: 10.1161/CIRCULATIONAHA.111.019893. [DOI] [PubMed] [Google Scholar]
- 2.Lip GYH, Lane DA. Stroke prevention in atrial fibrillation: a systematic review. JAMA . 2015;313:1950–62. doi: 10.1001/jama.2015.4369. [DOI] [PubMed] [Google Scholar]
- 3.Chen A, Stecker E, A Warden B. Direct Oral Anticoagulant Use: A Practical Guide to Common Clinical Challenges. J Am Heart Assoc. 2020;9:e017559. doi: 10.1161/JAHA.120.017559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zhu J, Alexander GC, Nazarian S, et al. Trends and Variation in Oral Anticoagulant Choice in Patients with Atrial Fibrillation, 2010-2017. Pharmacotherapy . 2018;38:907–20. doi: 10.1002/phar.2158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.U.S. Food and Drug Administration Development & Approval Process | Drugs. https://www.fda.gov/drugs/development-approval-process-drugs#:~:text=A%20team%20of%20CDER%20physicians,drug%20is%20approved%20for%20sale Available.
- 6.European Medicines Agency Authorisation of Medicines. https://www.ema.europa.eu/en/about-us/what-we-do/authorisation-medicines Available.
- 7.Downing NS, Aminawung JA, Shah ND, et al. Regulatory review of novel therapeutics--comparison of three regulatory agencies. N Engl J Med. 2012;366:2284–93. doi: 10.1056/NEJMsa1200223. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lexchin J. New drug submissions in Canada and a comparison with the Food and Drug Administration and the European Medicines Agency: Cross-sectional analysis. PLoS One. 2023;18:e0286802. doi: 10.1371/journal.pone.0286802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Van Norman GA. Drugs and Devices: Comparison of European and U.S. Approval Processes. JACC Basic Transl Sci. 2016;1:399–412. doi: 10.1016/j.jacbts.2016.06.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hakariya H, Moriarty F, Ozaki A, et al. Continued cancer drug approvals in Japan and Europe after market withdrawal in the United States: A comparative study of accelerated approvals. Clin Transl Sci. 2024;17:e13879. doi: 10.1111/cts.13879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hwang TJ, Kesselheim AS, Tibau A, et al. Clinical Benefit and Expedited Approval of Cancer Drugs in the United States. JCO Oncol Pract Sep. 2022;18:e1522–32. doi: 10.1200/OP.21.00909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Food and Drug Administration International regulatory harmonization. https://www.fda.gov/drugs/cder-international-program/international-regulatory-harmonization n.d. Available.
- 13.Murakami M, Narukawa M. Matched analysis on orphan drug designations and approvals: cross regional analysis in the United States, the European Union, and Japan. Drug Discov Today. 2016;21:544–9. doi: 10.1016/j.drudis.2016.02.016. [DOI] [PubMed] [Google Scholar]
- 14.Zeukeng MJ, Seoane-Vazquez E, Bonnabry P. A comparison of new drugs approved by the FDA, the EMA, and Swissmedic: an assessment of the international harmonization of drugs. Eur J Clin Pharmacol . 2018;74:811–8. doi: 10.1007/s00228-018-2431-7. [DOI] [PubMed] [Google Scholar]
- 15.Boesen K, Jørgensen KJ, Gøtzsche PC. Clinical trials were missing from regulatory documents of extended-release methylphenidate for ADHD in adults: a case study of public documents. J Clin Epidemiol. 2022;143:242–53. doi: 10.1016/j.jclinepi.2021.10.027. [DOI] [PubMed] [Google Scholar]
- 16.Dalla Torre Di Sanguinetto S, Heinonen E, Antonov J, et al. A Comparative Review of Marketing Authorization Decisions in Switzerland, the EU, and the USA. Ther Innov Regul Sci . 2019;53:86–94. doi: 10.1177/2168479018764660. [DOI] [PubMed] [Google Scholar]
- 17.Kashoki M, Hanaizi Z, Yordanova S, et al. A Comparison of EMA and FDA Decisions for New Drug Marketing Applications 2014-2016: Concordance, Discordance, and Why. Clin Pharmacol Ther . 2020;107:195–202. doi: 10.1002/cpt.1565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Lythgoe MP, Desai A, Gyawali B, et al. Cancer Therapy Approval Timings, Review Speed, and Publication of Pivotal Registration Trials in the US and Europe, 2010-2019. JAMA Netw Open . 2022;5:e2216183. doi: 10.1001/jamanetworkopen.2022.16183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Larochelle M, Downing NS, Ross JS, et al. Assessing the potential clinical impact of reciprocal drug approval legislation on access to novel therapeutics in the USA: a cohort study. BMJ Open. 2017;7:e014582. doi: 10.1136/bmjopen-2016-014582. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for reporting observational studies. Int J Surg. 2014;12:1495–9. doi: 10.1016/j.ijsu.2014.07.013. [DOI] [PubMed] [Google Scholar]
- 21.European Medicines Agency Medicines. https://www.ema.europa.eu/en/medicines n.d. Available.
- 22.U.S. Food and DrugAdministration Drugs@FDA: FDA-Approved Drugs. https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm Available.
- 23.Government of Canada Drug Product Database: Access the database. https://www.canada.ca/en/health-canada/services/drugs-health-products/drug-products/drug-product-database.html Available.
- 24.Therapeutic Goods Administration Australian Public Assessment Reports (AusPAR) https://www.tga.gov.au/resources/auspar Available.
- 25.Janda G, Khetpal V, Shi X, et al. Comparison of Clinical Study Results Reported in medRxiv Preprints vs Peer-reviewed Journal Articles. JAMA Netw Open . 2022;5:e2245847. doi: 10.1001/jamanetworkopen.2022.45847. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Turner EH, Matthews AM, Linardatos E, et al. Selective publication of antidepressant trials and its influence on apparent efficacy. N Engl J Med . 2008;358:252–60. doi: 10.1056/NEJMsa065779. [DOI] [PubMed] [Google Scholar]
- 27.Zou CX, Becker JE, Phillips AT, et al. Registration, results reporting, and publication bias of clinical trials supporting FDA approval of neuropsychiatric drugs before and after FDAAA: a retrospective cohort study. Trials . 2018;19:581. doi: 10.1186/s13063-018-2957-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Government of Canada Guidance document: Product Monograph. https://www.canada.ca/en/health-canada/services/drugs-health-products/drug-products/applications-submissions/guidance-documents/product-monograph/guidance-document-product-monograph-2020.html Available.
- 29.U.S. Food and DrugAdministration Guidance for Industry: Warnings and Precautions, Contraindications, and Boxed Warning Sections of Labeling for Human Prescription Drug and Biological Products — Content and Format. https://www.fda.gov/media/71866/download Available.
- 30.Department of Health and Aged Care Therapeutic Goods Administration (TGA) https://www.tga.gov.au/resources/resource/guidance/boxed-warning-guidance Available.
- 31.ICH Harmonised Guideline Revision of M4E guideline on enhancing the format and structure of benefit-risk information in ICH efficacy M4E(R2) https://database.ich.org/sites/default/files/M4E_R2__Guideline.pdf Available.
- 32.Electronic Code of Federal Regulation Title 21, chapter 1, subchapter d, part 314, subpart B, §314.50. 2017. https://www.ecfr.gov/current/title-21/chapter-I/subchapter-D/part-314/subpart-B/section-314.50 Available.
- 33.Health Canada Submission Certificate for a New Drug Submission (NDS), Supplement to a New Drug Submission (SNDS), Supplement to an Abbreviated New Drug Submission (SANDS), Abbreviated New Drug Submission (ANDS), or Notifiable Change (NC) n.d.
- 34.European Medicines Agency ICH M4E Common technical document for the registration of pharmaceuticals for human use - efficacy - Scientific guideline. https://www.ema.europa.eu/en/ich-m4e-common-technical-document-registration-pharmaceuticals-human-use-efficacy-scientific Available.
- 35.Therapeutic Goods Administration Appendix A - Specific mandatory requirements n.d.
- 36.Hooimeyer A, Bhasale A, Perry L, et al. Regulatory post-market drug safety advisories on cardiac harm: A comparison of four national regulatory agencies. Pharmacol Res Perspect . 2020;8:e00680. doi: 10.1002/prp2.680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Perry LT, Bhasale A, Fabbri A, et al. Comparative Analysis of Medicines Safety Advisories Released by Australia, Canada, the United States, and the United Kingdom. JAMA Intern Med . 2019;179:982–4. doi: 10.1001/jamainternmed.2019.0294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Zeitoun JD, Lefèvre JH, Downing N, et al. Inconsistencies among European Union pharmaceutical regulator safety communications: a cross-country comparison. PLoS One. 2014;9:e109100. doi: 10.1371/journal.pone.0109100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Tanaka M, Idei M, Sakaguchi H, et al. Rationales of delay and difference in regulatory review by Japan, the USA and Europe among new drugs first approved in Japan. Br J Clin Pharmacol . 2021;87:3279–91. doi: 10.1111/bcp.14749. [DOI] [PubMed] [Google Scholar]
- 40.Wallach JD, Egilman AC, Dhruva SS, et al. Postmarket studies required by the US Food and Drug Administration for new drugs and biologics approved between 2009 and 2012: cross sectional analysis. BMJ . 2018;361:k2031. doi: 10.1136/bmj.k2031. [DOI] [PMC free article] [PubMed] [Google Scholar]