Mini Summit XIX: Update on Data Sharing/Data Transparency ... · an approved new drug application (NDA) or biologics license application (BLA); OR (2) an approved NDA or BLA would

The Fourteenth Pharmaceutical Compliance Congress and Best Practices Forum Mini Summit XIX: Update on Data Sharing/Data Transparency in Clinical Trials

38671373_3.pptx

Mark Barnes Caroline Stockwell Barbara Bierer, MDRopes & Gray LLP and Pfizer, Inc. Harvard Medical SchoolMulti-Regional Clinical Brigham & Women’s HospitalTrials Center at Harvard Multi-Regional Clinical Trials

Center at Harvard

2

Agenda

Background on ClinicalTrials.gov

FDAAA compliance problems

FDA request for comments on data sharing

European Medicines Agency (EMA) draft data sharing policy

Industry initiatives

Recurring questions and future recommendations

Insight from Multi-Regional Clinical Trials Center at Harvard (MRCT)

38671373_3.pptx

3

I. Background on ClinicalTrials.gov

Late 1990s:• Food & Drug Modernization Act of 1997 establishes

ClinicalTrials.gov, which went live in February 2000• Required registration of clinical trials for drugs treating “serious or

life-threatening conditions”• Registration was voluntary for other trials• Focus was on helping patients find a clinical trial dealing with their

disease

38671373_3.pptx

4

II. Background on ClinicalTrials.gov

2004: The World Health Organization (WHO) creates the International Clinical Trials Registry Platform (ICTRP)

• “The registration of all interventional trials is a scientific, ethical and moral responsibility.” WHO Statement on Clinical Trial Registries

2005: International Committee of Medical Journal Editors (ICJME) require clinical trial registration for publication

• Required whenever research involves assigning individuals to groups for “health-related interventions”

• Includes any intervention used to modify a biomedical or health related outcome (e.g., drugs, surgical procedures, devices, behavioral treatments)

38671373_3.pptx

5

III. Background on ClinicalTrials.gov

2007: Food and Drug Administration Amendments Act (FDAAA) requires registration on ClinicalTrials.gov of drug trials meeting the following requirements (Phase I trials excluded):

• Controlled (i.e., designed to permit a comparison of a test intervention with a control)

• Clinical investigation (i.e., an experiment in which the drug is administered to human subjects)

• One of the following is true: (1) drug being tested is the subject of an approved new drug application (NDA) or biologics license application (BLA); OR (2) an approved NDA or BLA would be required in order for that drug to be legally marketed

38671373_3.pptx

6

IV. Background on ClinicalTrials.gov

FDAAA also requires registration on ClinicalTrials.gov of device trials meeting the following requirements:

• Prospective clinical study of health outcomes• Comparing an intervention with a device against a control in human

subjects• Any of the following would be required before the device could be

legally marketed:–

A finding of substantial equivalence under section 510(k)–

An order under section 515 approving a premarket approval application

–

A humanitarian device exemption under section 520(m)• The trial is not a small clinical trial to determine feasibility of the

device

38671373_3.pptx

7

V. Background on ClinicalTrials.gov

Summary results submission on ClinicalTrials.gov is required for clinical trials that:

• Are required to be registered under FDAAA 801; and• Study drugs, biologics, and devices that are “approved, licensed,

or cleared by FDA”

38671373_3.pptx

8

VI. Background on ClinicalTrials.gov

Penalties for failure to register trials or report results, or for reporting false or misleading results can be severe

• Civil monetary penalty of up to $10,000 per violation• If violation not corrected within the 30-day period following

notification of failure to comply, an additional penalty of up to $10,000 per day could be assessed

• Withholding of grant funds for federally-funded studies

38671373_3.pptx

9

Agenda








38671373_3.pptx

10

I. Compliance problems with ClinicalTrials.gov

FDAAA compliance challenges• Getting investigators to report results to ClinicalTrials.gov has been

a challenge–

2012 BMJ article found that only 22% of completed clinical trials subject to mandatory reporting had delivered results to ClinicalTrials.gov within one year of study completion. See Andrew P. Prayle et al., Compliance With Mandatory Reporting of Clinical Trial Results on ClinicalTrials.gov: Cross Sectional Study, 344 BMJ (2012).

–

2013 Journal of Clinical Oncology article found that only 9% of cancer trials had reported results to ClinicalTrials.gov within one year of study completion. See Thi-Anh-Hoa Nguyen et al., Public Availability of Results of Trials Assessing Cancer Drugs in the United States, 31 J. Clinical Oncology 2998 (2013).

38671373_3.pptx

11

II. Compliance problems with ClinicalTrials.gov

FDAAA compliance varies by funding source of trial• Both articles cited in previous slide found that industry-funded trials

were the most compliant–

BMJ article found 40% compliance for solely industry-funded trials, 9% for mixed industry/non-industry funded trials, and 8% for NIH/government funded trials

–

Journal of Clinical Oncology article found 20% for solely industry-funded trials, 11% for mixed industry/non-industry funded trials, and 6% for solely academic-funded trials

38671373_3.pptx

12

Agenda








38671373_3.pptx

13

I. Proposed FDA initiative regarding “masked” data

FDA request for comments of June 4, 2013 (78 Fed. Reg. 33,421) • Proposal to make available for research participant-level data from

medical product applications • FDA recognizes a “potential to further advance regulatory science”

by allowing non-FDA experts to analyze data submitted to FDA• Commercially confidential information and trade secrets would be

excluded from any data release

38671373_3.pptx

14

II. Proposed FDA initiative regarding “masked” data

Data under consideration would be both masked and de-identified• “Masked data” = data stripped of information that could link them to

a specific product or application• “De-identified data” = data that do not identify an individual nor

provide a reasonable basis to believe that the individual could be identified

Variety of comments received in response to request• Concern about re-identification and “unmasking” of data• Fear that data release could lead to proliferation of low-quality

studies that would not advance scientific knowledge• Potential of competitive harm to sponsors submitting data• Fear that masking data could diminish the utility of the information

released, undermining the rationale for data sharing

38671373_3.pptx

15

Agenda








38671373_3.pptx

16

I. EMA draft data sharing policy

The European Medicines Agency (“EMA”) has proposed a more aggressive policy

• Interesting progression–

EMA was historically opposed to sharing of clinical trials data and clinical study reports

–

In 2007 a researcher requested access to clinical trials data held by EMA regarding anti-obesity drugs

–

EMA denied access on the grounds that granting access to clinical trials data would harm the commercial interests of drug manufacturers

–

The European Union Ombudsman reviewed the situation and ruled that EMA should release data from the studies

–

In late 2010, EMA began to release clinical study reports on request as part of its access-to-documents policy

38671373_3.pptx

17

II. EMA draft data sharing policy

November 2012: EMA announces that as of January 2014, it will require that participant-level clinical trials data used to support a marketing authorization of a medicine be made publicly available

EMA made clear its commitment to proactive publication of data• “We are committed to proactive publication of clinical trial data, once

a marketing authorisation decision has been taken. We will deliver this project in dialogue with our stakeholders.” Introductory Presentation - November 2012 EMA Workshop on Access to Clinical Trial Data

38671373_3.pptx

18

III. EMA Draft Data Sharing Policy

EMA Policy was Shaped by Five Working Groups Comprised of Experts from industry, academia, and patient advocacy organizations

• Legal Aspects–

Examined legal aspects beyond personal data protection, such as commercial confidentiality

• Protecting Patient Confidentiality–

Examined how EMA policy can ensure that patient and other personal information will be adequately protected

• Clinical-trial Data Formats–

Focused on developing a clear and understandable data format allowing for appropriate analysis

• Rules of Engagement–

Analyzed the rules and conditions that should govern access to data

• Good Analysis Practice–

Determined whether there are good-analysis-practice guidelines that data requester should be asked to consider38671373_3.pptx

19

IV. EMA draft data sharing policy

EMA draft data sharing policy was released in June 2013• Provides mechanisms for release of the following clinical trials (CT)

data:–

Clinical summary–

Clinical trial overview–

Clinical study reports with line-level data listings• Data submitted to the EMA on or after March 1, 2014 will be subject

to the policy

The policy was subject to a public comment period from the time of its release until September 30, 2013

38671373_3.pptx

20

V. EMA draft data sharing policy

Data release• Some data are “open access” and will be downloadable from the

EMA website, whereas others are subject to a “controlled access” policy

• Method of access depends on type of data – data types are organized based on privacy concerns

• EMA policy places data into three categories:–

Category 1: Data containing commercially confidential information (CCI)

–

Category 2: Data without protection of personal data (PPD) concerns

–

Category 3: Data with PPD concerns; essentially “raw CT data”

38671373_3.pptx

21

VI. EMA draft data sharing policy

Category 1 – data containing CCI• Documents containing CCI will not be made available (but may be

available under the Policy on Access to Documents)• CCI is defined as information:

–

That is not in the public domain; and –

Disclosure of which may undermine the legitimate economic interest of the information’s owner

• EMA takes the position that only a “small number of CT data/documents” contain CCI

38671373_3.pptx

22

VII. EMA draft data sharing policy

Category 2 – data without Protection of Personal Data (PPD) concerns• Classified as “open access” data• Data will be available as downloads from the EMA website• Subcategories:

–

Documents that lack personal data (e.g., summary tables presenting only aggregate data)

–

Documents in which personal data have been “adequately de- identified”

–

Instances where public health needs override PPD considerations

38671373_3.pptx

23

VIII. EMA draft data sharing policy

Category 3 – data with PPD concerns • Available only upon request • Requesters must identify themselves; EMA will verify their identity• Requesters must be “established” in the EU• Data must be de-identified• Requesters must enter a legally binding data-sharing agreement

requiring the requester to, among other things:–

Access data for the sole purpose of addressing a question or conducting analyses in the interest of public health

–

Refrain from any attempt to retroactively identify participants–

Refrain from using data for any purposes outside the boundaries of the patients’ informed consent

38671373_3.pptx

24

IX. EMA draft data sharing policy

EMA draft policy leaves several questions unanswered• Level of de-identification required for Category 2 vs. Category 3

data• Method and nature of evaluation of informed consent from studies

for which data are shared• Extent to which secondary analyses will be placed in the public

domain• Which types of documents will be deemed to contain CCI and thus

unavailable for release

38671373_3.pptx

25

X. EMA draft data sharing policy

38671373_3.pptx

Summary of EMA data sharing categories

26

XI. EMA draft data sharing policy

Conflict between EMA draft policy and current FDAAA requirements• As a result of regulations promulgated under FDAAA, the following

language must be inserted in all informed consent forms for trials subject to ClinicalTrials.gov registration requirements:

“A description of this clinical trial will be available on http://www.ClinicalTrials.gov, as required by U.S. Law. This Web site will not include information that can identify you. At most, the Web site will include a summary of the results. You can search this Web site at any time.” 21 C.F.R. §

50.25(c).

• EMA’s draft policy allowing for data sharing of participant-level data is directly at odds with this language

38671373_3.pptx

http://www.clinicaltrials.gov/

27

XII. EMA draft data sharing policy

Additional transparency efforts in the EU• Clinical Trial summary results will be made available via the

EudraCT database, which is similar to clinicaltrials.gov

38671373_3.pptx

28

XIII. EMA draft data sharing policy

Litigation Threats• EMA’s future policy is likely to spawn litigation• The current access-to-documents policy has already led to

lawsuits by AbbVie and Intermune, seeking to prevent the release of participant-level data by EMA

• AbbVie seeks to “protect [its] confidential and commercially- sensitive information” by opposing disclosure that “does not meaningfully contribute to the scientific review or evaluation of our products”

• The EU General Court issued a preliminary injunction earlier this year barring the EMA from releasing the participant-level data

• Interestingly, the clinical study reports in the AbbVie case were not requested by an academic researcher, but rather were requested by a rival pharmaceutical manufacturer (UCB)

38671373_3.pptx

29

Agenda








38671373_3.pptx

30

I. Industry initiatives

PhRMA/European Federation of Pharmaceutical Industries and Associations – Principles for Responsible Clinical Trial Data Sharing (2013)

• Enhance data sharing with researchers upon request–

Scientific review board composed of scientists who are not employees of the company will review requests

–

Any patient-level data shared will be anonymized• Share results with patients who participate in clinical trials

38671373_3.pptx

31

II. Industry initiatives

(continued from previous slide)• Enhancing public access to clinical study information

–

Companies will make publicly available, at a minimum, synopses of CSRs following approval of new medicines or a new indication of an existing medicine

• Certification of compliance on a public website• Reaffirm commitment to publish clinical trial results

–

All company-sponsored clinical trials should be considered for publication regardless of results

• Some individual pharmaceutical companies have begun to develop their own data sharing policies

38671373_3.pptx

32

III. Industry initiatives

GlaxoSmithKline (GSK)• Will make available to researchers:

–

Raw data set of patient-level information–

Full protocols–

Annotated case report forms–

Data set specifications–

Clinical-study reports• Data will be released through a controlled access process requiring:

–

Submission of a brief research proposal–

Presence of a statistician on the research team

38671373_3.pptx

33

IV. Industry initiatives

GSK continued• Evaluation of research proposal

–

Panel of GSK-appointed experts will evaluate the analysis plan–

Unlike the EMA proposal, the panel will evaluate the qualifications of the investigators and the relevance of the research proposal to patient care

–

Reasons for acceptance/rejection of research proposal will be shared with investigator

38671373_3.pptx

34

V. Industry initiatives

GSK continued• Data-sharing agreement requirements

–

Data can be used only for research described in the accepted research proposal

–

GSK obtains a non-exclusive license to use any new intellectual property derived from the research

–

Data recipients must not attempt re-identification• Privacy Protections

–

Removal of personally identifiable information–

Revision of all dates through use of random anchor–

Destruction of key code linking data set provided and original data set

38671373_3.pptx

35

VI. Industry initiatives

GSK continued• Second level of privacy protection

–

Data are available exclusively through password protected site; controls in place to prevent downloading or sharing of data

–

Data cannot be downloaded or transferred and can only be accessed for 12 months

• Research uses must be in line with informed consent used in study–

This means that in many cases, the research must be related to the medicine/condition that was the subject of the study in which data were collected

For more on the GSK policy, see:• Perry Nisen & Frank Rockhold, Access to Patient-Level Data from

GlaxoSmithKline Clinical Trials, 369 N. Eng. J. Med. 475 (2013)

38671373_3.pptx

36

VII. Industry initiatives

Roche has also announced a data sharing policy• Applies to both clinical study reports and patient-level data• Two-tiered process: one process for CSRs and a second

process for Participant-Level Data• Clinical Study Reports

• If the CSR has been released to a regulatory agency, Roche will direct requests for the CSR to the relevant regulatory agency

• If the CSR has not been released to a regulatory agency, Roche will make it available to the requester within 3 months

• CSRs will not be made available until the CHMP review/European Commission decision are complete, or until the FDA review/decision is complete

• CSRs for withdrawn applications are not released

38671373_3.pptx

37

VI. Industry initiatives

Roche continued• Access to participant-level data will require submission of a

research proposal containing–

Scientific rationale–

Objectives–

Methods, including identifiers of clinical studies–

Statistical analysis plan–

Publication plan–

CVs of all researchers–

Names of three independent experts in the field who could discuss scientific merit of proposal

• An independent review panel composed of global experts outside of Roche will evaluate the research proposal

38671373_3.pptx

38

VIII. Industry initiatives

Roche continued• The IRP review process can have four outcomes

–

Approval–

Approval, subject to some conditions–

Rejection, with advice on issues to address in a possible revised request for access

–

Rejection• Data requesters must also enter a data-sharing agreement

agreeing to, among other things:–

Only use data for agreed patients–

Never attempt to re-identify participants–

Complete analyses within 24 months of access–

Affirm that the analyses will not ever be used by the data requester for commercial purposes

38671373_3.pptx

39

IX. Industry initiatives

Roche continued• Any dataset provided must be “appropriately anonymized,”

which requires that following:–

Removal of details that could help identify the patient (e.g., patient initials, patient date of birth, admission and discharge dates, hospital name, investigator name, text in comments field, genomic data, MRI and other imaging)

–

All sites with fewer than 10 patients are combined into one artificial “umbrella” site and all references to the original sites are removed from the dataset

• Access to data sets will be given via password-protected internet site hosted by a third party

• Datasets will not be downloadable from the site; analytical software will be provided for use in the site

38671373_3.pptx

40

Agenda








38671373_3.pptx

41

I. Recurring questions and future recommendations

Privacy concerns• Sharing of participant-level data introduces possibility of participant

identification• Technology is making re-identification of participants easier

Informed consent• Past consent forms should be evaluated prior to sharing of data• Going forward consent forms should be revised to advise

participants about data sharing

38671373_3.pptx

42

II. Recurring questions and future recommendations

Protection of pharmaceutical companies’ intellectual property rights• Protection of commercially confidential information/trade secrets

should be a priority• Competitors may use released data for commercial advantage

Potential legal challenges• EMA currently faces lawsuits over its existing access-to-documents

policy• Responses to FDA request for comments suggest potential legal

challenges

38671373_3.pptx

43

Agenda








38671373_3.pptx

44

I. Insight from MRCT

MRCT Seeks to• Provide a multi-stakeholder perspective on potential solutions and

criteria for access to participant-level data for public health and scientific research purposes

• Identify potential areas of collaboration on these issues among stakeholder groups

38671373_3.pptx

45

II. Insight from MRCT

Core Principles of MRCT• Protect research participants• Advance innovation and public health• Balance risks with benefits of data sharing • Treat all data generators equally• Make data disclosure practicable by avoiding undue burdens on

data generators and requesters• Provide timely access to data • Ensure adequate transparency • Ensure accountability

38671373_3.pptx

46

III. Insight from MRCT

Core principles continued• Requests and decisions posted on the web• Requesters pre-commit to an analytical plan

• Requester’s identity and scientific plan are publicly disclosed

• Requester signs a data use agreement• Decisions about data releases include both the Data

Generator and other parties.

38671373_3.pptx

47

IV. Insight from MRCT

38671373_3.pptx

48

V. Insight from MRCT

Data Sharing ModelsOpen Access• Data Generator routinely posts data from trials when results are publicly

reported or submitted to regulator, along with documentation to facilitate use of data

• Responsible-use attestation when researcher downloads materialData Generator

• Data Generator reviews request, decides, and publicly documents rationale for decision.

• Denials are appealable to independent Appellate Board, whose decision is final.

38671373_3.pptx

49

VI. Insight from MRCT

Data Sharing Models (continued)Learned Intermediary Model – ***MRCT Preferred Model***

• Review Board independent of Data Generator• Board reviews request, collects input from Data Generator, decides, and

publicly documents rationale for decisionDatabase Query Model

• Requester submits a research query to the Data Holder• Data Holder runs the query and returns results—not data

38671373_3.pptx

50

VII. Insight from MRCT

38671373_3.pptx

51

VIII. Insight from MRCT

MRCT Point of Consensus (Continued)6.Many of the rationales/benefits require participant-level datasets

– Facilitate secondary analysis to verify results, regulatory decisions, public policy

– Improve safety surveillance– Speed new discoveries

7.Important mechanisms for a data sharing system:– Ensure adequate scientific expertise among the analytical team– Provide technical support sufficient to permit users to understand the

data8.Some benefits are difficult to achieve in a sponsor-controlled model9.Timing of availability for both summary and participant-level data should be 1 year after primary study completion.

– Assuming an adjudicated process to obtain participant-level datasets, evaluation of the purpose for the participant-level datasets could be different (‘tighter’) prior to product approval.

38671373_3.pptx

52

IX. Insight from MRCT

MRCT Identification of Risks – Risk to Privacy of Participants and Clinical Study Personnel

• Research Participants• Fear that employers or insurance companies might access data and

re-identify participants• Persons involved in studies for sensitive conditions may fear being

identified due to possible stigma/discrimination• Genetic information is of special concern, as many studies have an

“add on” component in which genetic information is derived and stored

• Patients/participants might identify themselves in shared data sets• Clinical Study Personnel

• May fear retribution or stigma if their participation becomes known to others

• Risks to discrete and insular communities

38671373_3.pptx

53

X. Insight from MRCT

Does de-identification of data solve the problem of risks to participant privacy and confidentiality? ¨

• De-identification is not consistently defined; EMA definition is more vague, less detailed and thus possibly quite different than the HIPAA definition

• Removing HIPAA identifiers does not necessarily anonymise data• De-identification is a moving target due to improving technology,

e.g., genetic information is becoming increasingly identifiable, which may make the HIPAA de-identification standards obsolete

• Degree of de-identification is inversely related to data usefulness: the more identifiers removed, the less useful the data become to subsequent researchers

38671373_3.pptx

54

XI. Insight from MRCT

MRCT Identification of Risks – Possible Adverse Effects on Property Rights, Proprietary Interests and Competitive Concerns

Three primary concerns:

1.Patentability issues

2. Regulatory approval and data exclusivity issues

3. Economic “free rider” issues

38671373_3.pptx

55

XII. Insight from MRCT

Mandated disclosure of certain regulatory files could have secondary consequences that will affect public health

• May allow competitors to obtain regulatory approval in other countries without having invested in the research (“free-rider”), thus affecting incentives for investment in biomedical research

• Many data exclusivity provisions protect only data that have not been disclosed to the general public

• Additional negative effect on incentives to invest in biomedical research

38671373_3.pptx

56

XIII. Insight from MRCT

Possible solutions to concerns related to investment incentives.• Restricting access of data to “qualified” individuals / institutions (not

competitors)• Restricting the use of data through a contractual arrangement

– Specify acceptable uses of data for research purposes– Specifying requirements for keeping data set confidential / not

transferable– Prohibiting the filing of patent applications on inventions made

from data set– Potential penalties for misuse

• Delay disclosure of data• “Learned intermediary” model

38671373_3.pptx

57

XIV. Insight from MRCT

MRCT Identification of Risks – Changes to Regulatory Process1.Will there be obligations imparted onto the FDA, or other regulatory bodies as a result of any secondary analyses?

2.For example, should the information be sent to an FDA Advisory Committee?

3.What are the implications for drug or device labeling?

5.What are the regulatory processes that need to be followed by sponsors?

38671373_3.pptx

58

XV. Insight from MRCT

Changes to Regulatory Process (continued)6.Will there be a regulatory mechanism for researchers conducting secondary analyses to provide their respective findings to regulators?

7.Since drug companies and medical device manufacturers have certain reporting obligations (i.e., adverse events or patient safety issues) to regulatory agencies, what will be the minimum reporting requirements to sponsoring companies for unaffiliated researchers conducting secondary analyses?

8.Journals may become inundated with publications from those outside the company performing sub-studies or post hoc analyses and this may lead to second guessing of labeling, etc.38671373_3.pptx

59

XVI. Insight from MRCT

MRCT Identification of Risks – Informed Consent1.Informed consent form (“ICF”) and contract with the subject, should be honored

2.If unclear whether, how or what data sharing is allowed, ethics committee of the data generators should decide

3.If regulations require data sharing that is inconsistent with the ICF, data generators should not be liable for breach of contract or failure to comply

4.Prospective consent should explain process, benefits and risks of sharing

5.Compound consent should be avoided; “choice” will impact representation and/or statistical validity of study

6.Public education essential

38671373_3.pptx

60

XVII. Insight from MRCT

MRCT – Conclusions Regarding Data Sharing1.Data sharing, under the right conditions and with appropriate protections, is a laudable and achievable goal2.Patient privacy and confidentiality, and the terms of the ICF, should be paramount3.Company confidential information should be withheld from public view4.An independent, multi-stakeholder body should be charged with the process (“learned intermediary”)5.Data generators should be held harmless for compliance6.Data requestors are accountable for quality, scientific integrity, and expertise; honoring specific requests, confidentiality of participants, and held to same standards as data generators7.Responsibilities of regulators for results and analysis of secondary data should be determined prior to implementation38671373_3.pptx

6138671373_3.pptx

Questions???