Transparency too little, too late? Why and how Health Canada should make clinical data and regulatory decision-making open to scrutiny in the face of COVID-19

Canada has become a global leader in publicly releasing clinical data behind therapeutic products since 2019. Disclosure of clinical data is, however, limited to the point of product approval. The COVID-19 pandemic has underscored the limitations of such a point-in-time approach to transparency. As interventions are rapidly authorized for clinical trials or clinical use through emergency mechanisms despite weak evidence of safety and effectiveness, we argue that the current level of transparency must be expanded in order to mitigate potential harms to trial participants and patients, improve the reliability of clinical trials, and ultimately preserve trust in regulatory decision-making. Using existing legal powers, we explain how the Canadian regulator can and should seek to expand data transparency by making pre-clinical studies, clinical trial protocols, informed consent forms, interim analyses, and other information transparent upstream in the research process and throughout the product’s lifecycle.

Hard-won gains in the transparency of therapeutic product data in recent years 1 have occurred alongside growing reliance by regulators upon expedited review processes. 2 The concurrence of these two trends raises fundamental questions for the future of pharmaceutical regulation about whether the institutionalization of transparency will foster improved oversight of drugs, biologics, vaccines, and other interventions, or else, provide cover for a relaxing of regulatory standards of safety, effectiveness, as knowledge about the product's safety and effectiveness continues to evolve. And, while the argument we develop applies in principle across jurisdictions, we zero in on Canada in particular where recently enacted transparency laws provide ample authority to implement our recommendations. Specifically, we detail how Canada's existing transparency laws can be deployed to ensure that data, which only the sponsoring company may hold during the research process, are made available 9 and facilitate independent scrutiny of information held by sponsors and the regulator alike in order to improve judgments about the safety and effectiveness of SARS-CoV-2 interventions. We begin by setting out the mechanisms by which COVID-19 therapeutic products are being authorized for clinical study and use, then develop arguments about why greater transparency is warranted before finally presenting how to do so precisely under current Canadian law. We close by considering how added transparency might better assure public trust in regulatory agencies, such as Health Canada.

I. CANADA'S EXPEDITED REGULATORY PATHWAYS FOR COVID-19 CLINICAL TRIALS AND CLINICAL USE
Regulators worldwide have mobilized existing and new temporary mechanisms to expedite clinical trial approval and facilitate access to therapeutic products with uncertain benefits and harms in order to combat COVID-19. 10 In Canada, this involves three mechanisms ( Table 1): two of which pre-date the COVID-19 pandemic, while the third has been developed as the global health crisis has unfolded. The pre-existing pathways by which unapproved interventions can enter clinical use in Canada are the Special Access Program (SAP) 11 and the Access to Drugs in Exceptional Circumstances (ADEC) pathway. 12 The SAP grants access for individual patients on a case-by-case basis upon physician request. 13 Prior to its conditional approval by Health Canada in late July, 14 the antiviral drug therapy remdesivir had 9 For example, Health Canada authorized a SARS-CoV-2 targeting drug (remdesivir) solely on the basis of 'study protocols and preliminary and/or topline results' . The CSRs that normally accompany a New Drug Submission to the regulator were apparently not provided. We discuss this and other examples of data that only the sponsor may have at the time in question and Health Canada's legal authority to compel the production of that information. See infra note 33 and accompanying text. 10 Health Canada, Health Canada's regulatory response to COVID-19: Access to health products, aem (2020), https://www.canada.ca/en/health-canada/services/drugs-health-products/covid19-industry/regula tory-response-health-product-access. been accessed by individual physicians under the SAP for the treatment of at least a dozen patients. 15 In contrast to the case-by-case nature of the SAP, the ADEC (which has not yet been invoked during COVID-19) allows Health Canada to authorize the distribution and use of drugs at a population level, provided that they have been previously approved by a regulator in the USA, European Union, or Switzerland. 16 The third mechanism known as Interim Orders, which authorize the federal Minister of Health to make temporary changes to the standard regulatory framework, has emerged as the option of choice for Health Canada, presumably because of the efficiency and flexibility that it provides. 17 Several Interim Orders have been enacted during the pandemic to date, including one that streamlines clinical trial authorization for both drugs and medical devices, 18 another that facilitates (or expedites) the clinical use of medical devices, 19 and, in September 2020, an Interim Order that creates several new ways for drugs (defined to include pharmaceuticals, biologics, and vaccines) to enter the Canadian market. 20 Pursuant to these Interim Orders, a number of clinical trials, testing hydroxychloroquine, remdesivir, and several vaccines, have been authorized in record time (Supplementary File), 21 and one medical device has been authorized for clinical use (although it was subsequently recalled). 22 No therapeutic products have been authorized under the most recent Interim Order to date, but a number of its features are worth noting.
To begin, the Interim Order specifies three new pathways to authorization by Health Canada. One is an expedited authorization procedure that allows for a 'rolling application' in which the sponsor submits information to the regulator based upon an agreed-upon schedule. 23 This mirrors rolling application processes elsewhere and, by reducing the amount of information required initially upon submission, is intended to trigger faster decision-making about whether the benefits of the therapeutic product outweigh its risks. 24 Another pathway allows for authorization where the product has been previously approved by one or more trusted foreign regulators, opening up the list of such regulators from the three eligible regulators under ADEC to at least seven foreign regulatory authorities. 25 Finally, the Interim Order also describes how Health Canada can, without waiting for an application from the sponsor, seek to expand the indication of a previously approved drug to encompass treatment for COVID- 19. 26 Secondly, the drug-focused Interim Order aims to prioritize the review of applications by a sponsor for a new or modified 'establishment license' in order to lawfully produce a COVID-19 drug. 27 Granting the regulator the discretion to alter the requirements typically applied to establishment license applications, the Interim Order's stated that aim is to equip Health Canada with the 'agility to facilitate rapid access to COVID-19 drugs while mitigating risks' . 28 The third notable change introduced by the Interim Order is the creation of a 'prepositioning' option to allow a drug to be imported into Canada by an establishment license holder and be prepared for distribution prior to market authorization. 29 To be pre-positioned, the Government of Canada must have a procurement contract in place with the sponsor in respect of the Covid-19 drug in question, 30 and the Chief Public Health Official of the Public Health Agency of Canada must provide notice to the federal Minister of Health.
While Health Canada has, with the enactment of several Interim Orders, demonstrated a responsiveness to the urgency of the pandemic, it is notable that these efforts to expedite access to experimental COVID-19 drugs, vaccines, and other interventions have not been accompanied by parallel increases in transparency. 31  conducted abroad, and crucial details about the design of proposed clinical trials, such as clinical trial protocols, which specify inclusion/exclusion criteria for participants, primary and secondary outcomes that will be used to assess the product's safety and efficacy, randomization and blinding procedures to ensure the integrity of the findings, and a variety of other information. A published summary of Health Canada's decision to approve remdesivir, moreover, raises more questions than it answers given that the sponsor (Gilead Sciences) has yet to provide CSRs-the key document that regulators rely upon to assess safety and efficacy-for any of the COVID-19 remdesivir clinical trials. 33 And, while several publicly available lists (identifying, for example, any applications that the regulator has received for marketing authorization against Covid-19 34 ) were launched in conjunction with the September 2020 Interim Order, no meaningful changes were made to the regulator's approach to transparency. On the contrary, Health Canada will only release data pertaining to a therapeutic product's safety and effectiveness at the point of authorization. 35 Given that the regulator has, in one case, reduced the level of information required for authorization (eg granting Gilead approval for remdesivir even in the absence of CSRs) and relaxed when information is due pursuant to the Interim Order (by creating a rolling application process for all sponsors), there is reason to worry that Health Canada may not possess much data to share. Before outlining how the regulator can take a more proactive and dynamic approach to transparency under Canadian law, we explain in depth why the absence of transparency during the research and development process and beyond approval may precipitate a range of harms in the context of COVID-19.

II. HARMS STEMMING FROM A LACK OF TRANSPARENCY SURROUNDING EXPERIMENTAL COVID-19 INTERVENTIONS
Poor data transparency, in any circumstance, poses risks to trial participants and patients, jeopardizes clinical trial quality, 36 and has the potential to undermine trust in health professionals, pharmaceutical manufacturers, and regulatory decisionmaking. 37 There is also some evidence that a perceived lack of transparency is one of the drivers of 'vaccine hesitancy' . 38 Many have therefore long agitated for greater transparency across the entire spectrum of pharmaceutical research and regulation, with some notable successes, including the creation and uptake of clinical trial registries such as ClinicalTrials.gov; 39 the development of searchable databases documenting financial relationships between physicians and sponsors; 40 and the passage of a variety of laws and policies meant to condition compliance with nascent transparency norms. 41 Transparency remains a work in progress 42 but the extent to which the treasure trove of previously undisclosed data from clinical trials, which the EMA and Health Canada now release upon product approval, 43 will be regularly subject to independent scrutiny which is uncertain. 44 It has the potential for transformative engagement by actors beyond the nexus of regulator and sponsor in the work of pharmaceutical governance, in theory, involving diverse expertise and relevant publics in the product's appraisal.
36 By 'quality' we mean the trial's ability to answer the most important public health or scientific questions. We use this term in contrast to clinical trial 'reliability' , ie the trial's ability to accurately answer the questions it has selected however important to public health they may be. Even so, the strides toward transparency that have been made in the pharmaceutical sphere in the recent years have diminishing value in the context of COVID-19, where experimental interventions are being rapidly trialed and/or entering clinical use through various expedited regulatory pathways. Delaying disclosure of the data underlying these regulatory decisions until the point of approval may be too late to guide rational decision-making given the misinformation available online about remedies that are not the subject of a clinical trial and, at the same time, intense political pressure to demonstrate progress towards a cure even if the intervention's scientific merits are uncertain. 45 In this particular context, transparency needs to happen upstream, prior to the therapeutic product authorization or approval, and across the product lifecycle, in order to mitigate or avoid several significant harms.
First, consider the potential harms to trial participants. Several trials authorized under the Interim Order pertaining to clinical trials aim to truncate or combine the different phases of a trial. This poses increased risks to trial participants when the safety of the intervention-normally first tested in a small group of healthy volunteers as part of a Phase 1 trial 46 -remains unknown as the larger Phase 2 trial, meant to assess efficacy in a larger sample of participants, begins. Results from a small Phase 1 trial in China of the COVID-19 vaccine candidate, known as Ad5-nCoV, sponsored by CanSino Biologics Inc., 47 likely served as the basis for authorizing a larger Phase 2 trial in China, 48 as well as a combined Phase 1/2 trial in Canada. 49 Given the lack of public access to data from the Phase 1 trial, or to the template consent forms used for the combined Phase 1/2 trial in Canada, it is not possible to assess whether patients will be informed of these risks, or what risk mitigation measures, if any, were built into the design of the larger trial. 50 Rather than reviewing the trial protocol for consistency with ethical norms, Health Canada  the trial's design as quickly as possible. Embedded within the very institutions in which the research is being conducted, Canadian research ethics boards have routinely failed to apply and enforce ethical requirements in ordinary times, 52 a reality potentially exacerbated by COVID-19. Public access to the trial consent forms could help train attention upon the risks endured by trial participants while also ensuring that consent forms are updated as risks associated with a particular COVID-19 candidate are newly identified. 53 For instance, a participant in a trial of a vaccine under development by researchers at Oxford and AstraZeneca experienced severe neurological symptoms. 54 But it is unclear whether that potential risk was incorporated into consent processes when trials of the vaccine resumed. Further, the prospect of running 'challenge trials' in which a healthy volunteer is intentionally exposed to SARS-CoV-2, in an effort to more efficiently assess the effectiveness of the experimental candidate, underscores the importance of informed consent. 55 There has been growing interest in challenge trials, marked by a grass roots campaign to enroll in such a trial were one to be initiated despite the absence of existing treatments for SARS-CoV-2, over the course of the pandemic. 56 To ensure that the public is confident that participants are adequately informed of the risks involved, template consent forms used in the course of the trial should be publicly available, especially in light of lax regulatory oversight of clinical trials of late. 57 Second, a lack of transparency stands to undermine the quality of clinical studies if the trials are not designed to answer the most pressing public health questions. For example, vaccine trials are expected to determine specific outcomes, eg the potential to reduce the incidence of COVID-19 infections and related complications, hospitalizations, or deaths. 58 But without access to the trial protocol, which registries like Cli nicalTrials.gov do not provide, it is impossible to assess whether the trial design can achieve robust results. For instance, one of the largest international COVID-19 vaccine trials, involving a novel mRNA vaccine, seeks to enroll 30,000 participants. 59  if that target is met, the trial may not be large enough to assess whether the vaccine actually reduces the rates of hospitalization-one of the study's secondary outcome measures-due to the infrequent rate of COVID-19 hospitalization in the general population. 60 Several trials have already terminated early as new cases of COVID-19 waned in the region; new trials must have a viable plan for enrolling sufficient numbers. 61 But without consistent transparency of study documents such as trial protocols or correspondence between sponsors and regulators, the opportunity is missed for timely evaluation when inadequacies can in theory still be corrected. 62 Trial design modification may be unlikely even though regulators have the authority to compel changes to study design, but-at a minimum-the public can be more informed about the potential weaknesses of a given study.
Notably, some trial protocols have-at the sponsor's discretion-been publicly disclosed during the course of the pandemic, including both Moderna and Pfizer's mRNA vaccine trials, which include approximately 30,000 and 44,000 participants, respectively, as well as AstraZeneca's US trial of its adenovirus-based vaccine. 63 The details of these protocols raise some critically important questions, in particular, about whether the integration of 'interim analyses' in the design of the trials (the Pfizer, Moderna, and AstraZeneca trials include 4, 2, and 1, interim analyses, respectively) may lead to the trials being halted or modified prematurely (eg administering the experimental vaccine to the control group) if the efficacy of the vaccine in question is trending in a positive direction. 64 Precisely how those decisions will be made is not delineated in the protocols that have been made publicly available; rather, those decision will fall to 'data monitory committees' (also known as 'data safety monitoring boards' or 'DSMBs'), which typically operate under conditions of strict secrecy in order to both prevent false hope based on preliminary data, 65 and to shield their members from outside influence. 66 But the basic point remains that, unless trial protocols are-as a rule-open to scrutiny (as opposed to only when sponsors see fit to disclose them) as well as other key materials such as the 'charters' used by DSMBs to guide their decisions about stopping or modifying an ongoing trial (either because a safety issue has arisen or the intervention is showing significant therapeutic promise), it is not possible to fully interrogate the quality of COVID-19 trial designs and the reliability of trial outcomes. And, in the absence of trial protocol and DSMB charter transparency, before full data from the trial is released, there is a significant risk that downstream decisions by physicians about whether to administer the intervention in question to patients will be made without the benefit of strong evidence.
Third, the rush to promote partial, preliminary findings, has the potential to propagate misinformation. That is the main reason why data that are considered by DSMBs during the course of a trial are normally kept confidential. 67 Over the course of the pandemic, however, sharing preliminary data has become the norm, not the exception. And while preprints have become an important vehicle for rapidly sharing scientific findings, they also have the potential to be misinterpreted or accepted as accurate before they have been subjected to rigorous peer review. 68 Without the benefit of a peerreviewed publication, let alone access to the underlying data which provide a more comprehensive record of a trial's results, 69 there is a significant chance that the riskbenefit profile of a given product will be misunderstood by both clinicians and the public.
Consider, for example, remdesivir's path to market. No trial designs were publicly available before participant enrolment began. Weeks before 'preliminary findings' were published in high-profile journals, 70 details of the drug's purported benefits began to appear in the media, which Gilead promoted by press release. 71 US government officials, specifically Dr Anthony Fauci, suggested remdesivir would become the 'standard of care' treatment for COVID-19. 72 Trial details emerged, only after these press releases and announcements, raising significant concerns including the 'serious methodologic error' of prematurely censoring data of deceased patients, as well as design weaknesses such as lack of blinding and control, and limited sample size which may have produced unreliable results. 73 76 Fourth, limited transparency-coupled with politicized and expedited regulatory decision-making-risks significant patient harm. Early adoption of remdesivir may, for instance, carry significant opportunity costs, halting or slowing the investigation of other treatments that could eventually prove more effective against SARS-CoV-2 because of patients' reluctance to risk the chance of receiving a placebo in a new trial when there is an apparently effective drug already available. The United States' Food and Drug Administration (USFDA) premature decision surrounding the antimalarial drugs chloroquine and hydroxychloroquine is also illustrative. In March 2020, the USFDA granted 'Emergency Use Authorizations' (EUAs) for the two drugs to treat hospitalized COVID-19 patients. 77 Numerous news sources, including Presidential tweets, referred to hydroxychloroquine's emergency authorization as an 'approval' which sowed confusion and inflated perceptions of safety and efficacy. 78 Subsequent randomized trials of hydroxychloroquine in hospitalized COVID-19 patients reported no evidence of benefit, and the EUA was subsequently revoked by the USFDA in mid-June, citing a lack of efficacy and concerns over side effects. 79 Lack of transparency around the EUA prevented independent scrutiny of the USFDA's initial determination that 'the totality of scientific evidence' made it 'reasonable to believe that [the drugs] may be effective in treating COVID-19' . 80 Under constant pressure from the Trump administration, the USFDA is holding firm to its stated standard of not authorizing a vaccine-even on an emergency basisuntil at least two months have elapsed since trial participants have received the vaccine in question, and its observed efficacy meets or surpasses a pre-defined threshold. 81 If and when an EUA is granted, it will be critical to ensure that the knowledge which continues to accumulate during the remainder of the trial is proactively and publicly shared post-authorization. The worry that underpins this call for continuous data transparency is two-fold: if a vaccine appears sufficiently effective, the DSMB may elect to allow it to be administered to participants in the control arm of the trial, nullifying the trial's randomized design; or, even if no changes to the trial design are made, many participants will withdraw from the study once the decision has been made to grant an EUA. 82 In either eventuality, the integrity of the trial's results will be jeopardized. Therefore, ensuring that any changes made to trial designs and the trial results are open to scrutiny, both as the trial proceeds and after an EUA is granted, is essential. On a literal reading of the Interim Order issued in September, it appears that Health Canada can rubberstamp an EUA granted by the USFDA through the expanded foreign authorization mechanism that was introduced. 83 Whether the Canadian regulator does so, or utilizes the expedited mechanism by which remdesivir was authorized in July, 79

III. HEALTH CANADA'S CURRENT APPROACH TO DATA DISCLOSURE AND HOW TO IMPROVE IT DURING COVID-19
Canada's regulator has a long history of secrecy, electing to keep data confidential as a matter of practice. 85 The introduction of 'Vanessa's Law' in 2014, 86 however, promised fundamental change. Adding a variety of patient safety measures to Canada's Food and Drugs Act, including the power to unilaterally recall drugs from the market and compel acute care hospitals to share adverse event data with the federal regulator, 87 one of the express purposes of Vanessa's Law was to 'promote greater confidence in the oversight of therapeutic products by increasing transparency' . 88 To achieve that goal, several new legal authorities were added to the legislation, which, taken together, positioned Canada to become a global leader in transparency. 89 Implementation of these transparency provisions has been gradual, 90 but with the launch of Health Canada's 'Clinical Information Portal' in March 2019 significant progress has been made. 91 To date, the Canadian regulator has published important safety and efficacy data related to 75 drugs (including biologics and vaccines) and 15 medical devices. Some data are notably exempt from disclosure, including Case Report Forms (CRFs) that document outcomes for individual trial participants, and postmarketing experience in other jurisdictions. 92 Still, a wide range of other data, including consent forms, outcome measures, blinding and randomization procedures, statistical analysis plans, and trial results are readily accessible via the Portal, which extends not only to interventions that have been approved (or rejected) since the Portal's launch online, but also previously approved/rejected interventions. 93 According to one analysis, Health Canada's Portal offers the broadest scope of, and most timely access to, clinical data relative to the EMA and the USFDA. 94 COVID-19, however, has revealed significant limitations to Health Canada's newfound transparency. The first limitation concerns timing. Currently, disclosure of trial design and safety and efficacy data generated during human trials is sequestered by a specific point-in-time, ie if and when Health Canada formally approves or rejects a product. 95 As a result, most data pertaining to an experimental intervention are treated as 'confidential business information' (CBI) ( Table 2) before a regulatory decision is reached. To its credit, the recent Interim Order extends this point of disclosure to include authorizations granted under the Order. 96 But in the context of COVID-19, trials are being rapidly authorized and misinformation about merits of various experimental interventions is prevalent. Delaying disclosure of clinical trial designs, correspondence between regulators and sponsors about those designs, and the basis for crucial decisions to be made by DSMBs about whether to halt or modify a trial at the point of interim analysis until after the decision to authorize or approve the intervention pre-empts the correction of potential flaws in trial designs and limits the opportunity to build public understanding of the knowledge and uncertainties behind a given COVID-19 intervention. 97 Secondly, even when disclosure is permitted, Health Canada may not possess the data to make it available. In ordinary circumstances, Health Canada does not, for example, require sponsors to submit CRFs unless a serious adverse event has occurred in the course of a study. 98 Unless the regulator compels companies to submit and/or disclose such patient-level data, it is not possible for independent researchers to carry out a re-analysis of a trial-a labor-intensive act of independent scrutiny that continues to be undervalued but has, at times, revealed deep inconsistencies between published 94 Egilman et al., supra note 9. Importantly, though, the transparency of Health Canada's decision-making relative to the USFDA and EMA is limited. The USFDA has, since 1997, published all of the scientific reviews connected to a given drug approval. The US agency has recently shifted its review process, threatening to undermine the transparency of its , https://www.canada.ca/en/health-canada/services/drugs-health-products/covid19-industry/dru gs-vaccines-treatments/interim-order-import-sale-advertising-drugs/guidance.html#a12. 97 Health Canada's Guidance with respect to the Portal explicitly notes that disclosure of interim analyses might bias trial results, therefore, decisions about whether to disclose interim analyses will be made on a case-by-case basis. Given how much misinformation is available about various COVID-19 interventions, however, the exclusion of interim analysis data from public disclosure marks a missed opportunity to inform wider audiences about the nuanced evidence under consideration. See   Any CSRs, including methodological details, specifications, and validation information.

Reports of biopharmaceutic studies
Studies that evaluate the rate and extent of release of active substance from the medicinal product (PK and BA data) Reports of studies pertinent to pharmacokinetics (PK) using human biomaterials In vitro studies to assess PK using biological systems

Reports of human PK studies
In vivo PK studies studies and what the trial actually found. 99 In the context of the current pandemic, however, Health Canada has shown a willingness to accept data on a piecemeal basis and even approved one drug (remdesivir) without the benefit of CSRs. Until those CSRs are submitted to Health Canada, little   Finally, the Portal does not incorporate product safety and efficacy data that accumulates post-approval. 101 Health Canada, like the USFDA, tends to provide little information about what post-market studies must be completed, sponsors' progress in fulfilling them, or the regulator's evolving understanding of the product's safety and efficacy in light of those post-market studies. 102 It is anticipated that many of the trials investigating COVID-19 drugs and vaccines will continue following authorization, especially if authorization occurs after an interim analysis; however, unless that postauthorization evidence is incorporated into the Portal, the evidence that will be open to independent scrutiny may represent only a fraction of what is known.
Fortunately, Canada's Food and Drugs Act provides several options to address the foregoing gaps, in turn, improving the overall transparency of the data and decisionmaking process surrounding SARS-CoV-2 interventions. Depending on who possesses the information at a particular interval, different legal mechanisms can be invoked (or modified by way of an Interim Order) to ensure that data disclosure occurs through one of the several existing information-sharing platforms. Furthermore, transparency need not always mean disclosure to the public writ large; instead, the approach to disclosure should be specific to the type of information involved, purpose driven, and complemented by the necessary resources (Table 3).

III.A. Targeted Data Disclosure: DSMB Interim Analyses and Individual Patient Data from Completed Trials
Existing Canadian law grants the federal Minister of Health the discretion to disclose information that is deemed CBI to eligible persons, that is, persons who are engaged in the protection or promotion of human health and/or public safety, 103 provided they intend to use the CBI for a health-or public safety-related purpose rather than a commercial one. 104 This legal authority could be used, in the context of COVID-19, to share a number of different data with independent researchers before a decision is made to authorize a COVID-19 intervention for broader clinical use. Upon request from an eligible person (or government body), 105 for example, the Minister could disclose completed COVID-19 pre-clinical studies and trials conducted outside Canada that are submitted to Health Canada as part of an application to conduct a new clinical trial.
With several candidate vaccines now under study in large, Phase 3 clinical trials, this same discretionary authority could also be invoked to share the Interim Analyses conducted by DSMBs and shared with the regulator. As described above, it is plausible that such Interim Analyses will be used to justify an expedited authorization of a vaccine. While the trials will likely continue after such an authorization, any decision by a DSMB to alter a trial should be open to scrutiny by independent trialists and other researchers.
Finally, at a later point in time, when the trials are completed, Health Canada can make the individual patient data (IPD) available to eligible researchers in order to validate the safety and efficacy findings that are reported by the sponsor. In the eventuality that Health Canada does not possess the necessary IPD (eg CRFs) from completed trials, an additional provision in the Food and Drugs Act can be invoked to compel sponsors to disclose that information-as prescribed by regulations-in respect of products that have been imported into Canada for the purpose of a clinical trial and/or received an authorization or approval from Health Canada. 106 The regulator would need to define how that would occur using a new Interim Order and, in collaboration with other government agencies (eg Canadian Institutes of Health Research), marshal the resources to fund the labor-intensive work of re-analyzing and validating previous trial results using the IPD. 107 Provided that the recipients of the IPD protect the privacy of trial participants', 108 allocating even 1-2 per cent of the >$850 million already earmarked for the development of SARS-CoV-2 interventions for independent assessment would appear to be a worthwhile investment. Other types of data merit disclosure to the wider public. In order for the public to know that participants are being appropriately informed about the risks of participating in rapidly designed and authorized clinical trials, template consent forms that serve to enroll participants should be publicly shared before trials begin and in the event that the forms are updated in response to an adverse event. Challenge trials, if sanctioned, must be accompanied by disclosure not only of the consent forms but also the regulator's underlying ethical justification for authorizing a challenge trial design, given the absence of an effective treatment against SARS-CoV-2 in order to assuage public concerns about this type of trial. 109 Where the goal is to improve clinical trial reliability and promote scrutiny of trial results as well as regulatory decision-making, Health Canada should leverage its existing Clinical Information Portal and Drug and Health Product Register (DHPR) 110 to make these and other data publicly available. The Portal can serve to publicly share the pre-clinical and clinical evidence behind all clinical trial authorizations and all expedited authorizations granted pursuant to ADEC or the Interim Order. The DHPR can provide a repository for consent forms used in trials while also serving as a space for the regulator to document its rationale for each decision, including authorizing challenge trials, key correspondences with sponsors about trial designs, trial results, studies required to be carried out post-approval, and any other considerations that factored into the decision to authorize a product.
Currently, both the Portal and DHPR encompass information pertaining only to approved drugs and devices. However, Health Canada has broad authority to alter what information is considered CBI, and when to deem it no longer to be CBI in order to release the information. Amendments to the regulations, or an Interim Order, can be used to expand the information to be publicly disclosed to points upstream in the research process as well as after authorization as the evidence base continues to evolve. 111 The challenge concerning the clinical evidence to be posted on the Portal is timing. Unlike pre-clinical studies and other data that the regulator possesses as part of an application to conduct a clinical trial, the main documentation summarizing the safety and efficacy findings for completed trials (ie CSRs) can take weeks or months to prepare. The example of remdesivir suggests that Health Canada is prepared to accept and approve a drug submission without CSRs and more complete data in hand, and the recent introduction of a rolling application process appears to codify that same approach for the remainder of the pandemic. Therefore, to improve scrutiny of clinical trial designs and findings, Health Canada should disclose all the information it has regarding new trials, especially trial protocols, within 15 days of trial authorization (the same time frame in which the regulator is reviewing trial applications during the pandemic). The regulator should also invoke its authority-similar to patient-level data-to expedite the preparation and transfer of CSRs of all completed trials to Health Canada. 112 The regulator can then disclose these data to the public via the Portal within a reasonable time frame following trial completion (eg 60 days) as opposed to the current 120 days. 113 Finally, the DHPR, which presently contains only dates of meetings and limited summaries of Health Canada's interpretation of the evidence after approval, 114 similarly requires expansion to provide a complete record of each regulatory decision. No change in law is needed for this, as the reasons for Health Canada's decisions have long been theirs to share. 115 The DHPR should include all substantive correspondence between the regulator and sponsors, particularly with respect to trial designs, as well as a detailed explanation about why a given intervention is being trialed, made available through ADEC, authorized under the Interim Order's provisions, or approved through another pathway. The DHPR should also include any information pertinent to conditions attached to an approval, especially those pertaining to post-market studies, with frequent and timely updating as studies are designed and completed. 116

IV. REAL-TIME TRANSPARENCY, COMPETING CONSIDERATIONS AND THE QUESTION OF TRUST
By expanding transparency in the above ways, the Health Canada can enable deeper, science-based deliberations regarding what is known about a COVID-19 intervention, as the evidence evolves, rather than at a single point-in-time. With one intervention already approved in Canada (remdesivir) and one or more vaccines approaching authorization probably within the next 3-6 months, the federal government may need to overcome additional barriers to knowledge sharing in order to scale up production of the most promising products as no one manufacturer likely has the capacity to produce enough doses to meet the world's needs. 117 This may entail issuing a compulsory license in order to allow other manufacturers to produce a patented vaccine 118 as well as encouraging firms to share manufacturing know-how, proprietary adjuvants, and assays, which may be protected as trade secrets. 119 The prior task, however, is to ensure that the evolving knowledge about COVID-19 product candidates is meaningfully and continuously opens to scrutiny.
In principle, rendering clinical and decision-making data open to scrutiny carries competing considerations but none appear persuasive. The first concerns compliance with international law. Canada, like other signatories to international treaties governing intellectual property rights, is required to protect data against unfair commercial use. That commitment is an effort to balance the interests of first-mover and generic firms within the industry as they compete for market share. 120 In the context of COVID-19, however, this balancing can and has already been addressed in other ways. Specifically, the Interim Order passed in September stipulates that any product (and, by extension, its underlying data) that is authorized under the Order cannot be cited as a reference product by a generic firm in an effort to obtain regulatory approval. 121 As such, sponsors that reach the market first will not face generic competition unless they are unable to meet market demand for their product; in any event, international law allows data to be protected against unfair commercial use in a variety of ways and, insofar as the data are being made openly available to verify its reliability, as opposed to supporting a competing product, such use should not be characterized as commercially unfair. Enhanced disclosure of the data behind a COVID-19 intervention would not, as a result, appear to violate Canada's international commitments.
Secondly, enhancing the level of transparency will require resources. Health Canada has, however, already devoted substantial resources to the implementation of Vanessa's Law; moreover, the Clinical Information Portal and DHPR are already in place and, taken together, could incorporate all of the information to be made publicly available. Prior to the pandemic, Health Canada developed a set of review procedures to mitigate the risk of proprietary or personal information from being inadvertently disclosed via the Portal. In the usual course, this process is expected to consume 120 days. 122 Given the urgency of COVID-19 and the potential harms that flow from a lack of openness, an Interim Order designed to expand transparency might temporarily waive this review process altogether, allowing officials to efficiently post template consent forms, protocols, correspondence, CSRs, and decision letters to sponsors that accompany trial or product authorization. In contrast, the targeted disclosure of DSMB interim analysis data and anonymized IPD will require some resources, in part, to review requests with a view to ensuring that the intended recipients have the requisite capacity to provide independent scrutiny of the data in question. Allocating funding to support independent scrutiny will also be important if such targeted disclosures are to be completed in time to help inform patient understanding of a drug or vaccine's safety and efficacy and, assuming its risk-benefit profile proves favorable, improve uptake in the