E. Ornago

The European Medicines Agency (EMA) is committed to continuously extending its approach to 

. In October 2014, the agency released Policy 0070/2014, with the purpose to make medicine development more efficient, to foster public scrutiny to clinical study information by the scientific community, and to develop knowledge in the interest of public health, while promoting a better-informed use of medicines.

 According to EMA, “A high degree of transparency will take regulatory decision-making one step closer to EU citizens, and promote better-informed use of medicines. […] access to clinical data will benefit public health in future.”

	The scope of the EMA policy on publication of clinical data for medicinal products for human use

 relates to proactively sharing study-level and patient–level clinical data, (i.e., clinical reports and Individual patient data (IPD), submitted under the centralized marketing authorization procedure after Jan. 1, 2015.

	The policy serves as a complementary tool ahead of the implementation of the EU Clinical Trial Regulation No. 536/2014.

		Clinical data held by EMA for marketing authorization application (MAA) submitted under the centralized procedure before Jan. 1, 2015, and for extension of indication applications and line extension applications submitted before July 1, 2015.

		Clinical data (either data provided to EMA before Jan. 1, 2015, or data not yet held by the agency) submitted to EMA for non-centrally authorized products.

	These clinical data continue to be made available to external requesters on a reactive basis in accordance with the EMA’s policy on access to documents related to medicinal products for human and veterinary use (POLICY/0043) effective from Dec. 1, 2010.

	The procedure goes through two sequential phases:

		The first phase of disclosure (i.e., the publication of clinical reports supporting centralized MAA submitted only after Jan. 1, 2015, and-starting from Jan. 1., 2015-extension of indications and line extension applications relating to existing centrally authorized medicinal products.

		The second phase, regarding the provision of IPD, was foreseen for 2016 but, while at the writing this paper (February 2017), no indication from EMA was released.

	The publication procedure for clinical reports is based on two pillars:

		Terms of use (ToU), which govern the access to and use of clinical reports.

		A user-friendly technical tool allowing access to such clinical reports.

	The policy establishes methods for balancing the protection of patient’s privacy, through the anonymization/ de-identification of the protected personal data (PPD), while sharing clinical trial data and topics considered potential commercially confidential information (CCI) for redaction.

Part 1 - Anonymization of clinical reports for publication

	On July 6, 2015, EMA issued a guidance to the pharmaceutical industry on anonymization of clinical reports,

 in the context of Phase I of the policy (i.e., the publication of clinical reports on the EMA website). The guidance, whose terms and indications were briefly anticipated during an EMA webinar of June 24  2015, aims at assisting companies by recommending methods, techniques, and processes that could be applied to clinical reports, for the purpose of achieving adequate anonymization while retaining a maximum of scientifically useful information on medicinal products for the benefit of the public. A new release of this EMA guidance, completed by a summary of changes,

	Marketing authorization holders (MAHs)/applicants have the responsibility for submitting clinical reports that were previously rendered anonymous/de-identified.

	The data in the clinical reports must be processed in such a way that they can no longer be used to identify a natural person by using “all the means likely reasonably to be used” by either the controller or a third party.

	The same data can be adequately anonymized in different ways, depending on the context of the data release. When selecting the most appropriate technique, the specificities of the clinical data should be taken into consideration.

	Anonymization, a field of active research and rapidly evolving, makes available to MAHs/applicants several techniques. Each of them has its strengths and weaknesses. According to the Article 29 Working Party Opinion,

the techniques that could be applicable to clinical reports are:

	Randomization and generalization techniques are recommended in order to optimize the clinical usefulness of the information published.

Options to establish data set anonymization

	Two options are available to establish if the dataset is anonymized:

Demonstrate that, after anonymization, the following actions are no longer possible:

		Singling out: possibility to isolate some records of an individual in the dataset.

		Linkability: ability to link at least two records concerning the same data subject or a group of data subjects (in the same database or in two different databases).

		Inference: the possibility to deduce, with significant probability, the value of an attribute from the values of a set of other attributes.

	It is up to a sponsor taking due account of the ultimate purpose and use of the clinical reports to decide which option to use (demonstrate that after anonymization all three criteria are fulfilled: singling out, linkability and inference, or perform a risk assessment).

	The sponsor is also in charge of deciding which anonymization techniques to use in order to achieve adequate anonymization, while retaining a maximum of scientifically useful information. The legislation is not prescriptive about the techniques to be used by data controllers.

In the context of phase 1 of policy 0070, dataset is the set of clinical reports published by the EMA.

2. Perform an analysis of re-identification risk. 

	It is important to note that de-identification does not reduce the risk of re-identification of a data set to zero. Rather, the process produces data sets for which the risk of re-identification is very small.

	There are in fact three plausible re-identification attacks on the data by an adversary that need to be protected against, as summarized in Table 2.

	Measuring the risk of re-identification involves selecting an appropriate metric, a suitable threshold and the actual measurement of the risk in the clinical data information to be disclosed. The choice of a metric depends on the context of data release.

	Setting an acceptable threshold encompasses:

		The evaluation of the existing mitigation controls (none in the context of public disclosure).

		The extent to which a particular disclosure would be an invasion of privacy to the trial participant.

		The motives and the capacity of the attacker to re-identify the data.

	Once a threshold has been determined, the actual probability of re-identification can be measured. MAHs/applicants are encouraged to use quantitative methods to measure the risk of re-identification as soon as they are in a position to do so.

EMA recommendation to best achieve anonymization of PPD of trial participants

	There are several sections with data results in clinical reports that may contain personal data of trial participants. These include:

		Disposition (recruiting, pre-assignment, period/arms of the trial, etc.) of trial participants

		Safety (adverse events, laboratory findings, and vital signs).

	In general, clinical overviews (CTD mod. 2.5) and clinical summaries (CTD mod. 2.7) do not contain personal data related to trial participants, with the exception of the Narratives of the Clinical Summary. In addition, some of the tables included in the clinical overviews and clinical summaries may also contain personal data.

Anonymization of direct identifiers and quasi identifiers

	A classification of variables into direct and quasi-identifiers for clinical trials has been completed by a PhUSE (Pharmaceutical Users Software Exchange) working group.

	Any direct identifiers (e.g., name, email, phone number, social security number, signature, full address, clinical trial participant numbers, and medical device serial numbers) should be removed, or in the case of unique identifiers, like Patient ID numbers, at least pseudonymized

. There are established standards for such pseudonymization.

	The Guidance Regarding Methods for De-identification of Protected Health Information in Accordance with the Health Insurance Portability and Accountability Act (HIPAA) identifies 18 direct identifiers within the frame of the Safe Harbor method.

Pseudonymization consists of replacing one attribute (typically a unique attribute) in a record by another. The natural person is still likely to be identified indirectly. Pseudonymisation reduces the linkability of a dataset with the original identity of a data subject.

	Quasi-identifiers, which consist mostly of dates, location information, demographics, socioeconomic information, rare diagnoses, concomitant illnesses and medications, and serious adverse events such as death, hospitalization, and birth defects, cannot just be removed as these variables are very useful for the analysis.

More sophisticated techniques (e.g., generalization) need to be applied to retain the value of these variables but also reduce the probability that these variables can re-identify participants.

 The need to redact quasi identifiers will depend on the following aspects:

		Number of quasi identifiers per trial participant

		Frequency of trial participants with same category/value on a set of the quasi identifiers (group size)

Articles

Outlining the techniques for anonymization of clinical study reports and the identification and redaction of commercially confidential information to comply with EMA's Policy 0070 on trial data disclosure and transparency.     


Anonymization and Redaction of Clinical Trials According to the EU Regulation

In 2014, the European Parliament released the Regulation 536/2014 on clinical trials on medicinal products for human use.

 reinforced the concept of clinical trial transparency. Both the Regulation and the Policy took regulatory decision-making one step closer to European Union (EU) citizens, and promoted better-informed use of medicines.

	As a first step to promote clinical trial data disclosure, the European Medicines Agency (EMA) implemented a public European database in which to collect and retrieve all relevant information on clinical trials submitted to the agency (as per Article 81), establishing a right balance between disclosure of information and protection of anonymity and commercially confidential information. Therefore, the Regulation enables the public, patients, and healthcare professionals to identify clinical trials in an easier way and to evaluate their relevance toward an individual patient’s condition.

	The second step was the requirement of a layperson summary that should complement the summary of clinical trials results. The layperson summary is a novel document specifically designed for non-medically expert patients. Both clinical trials results and layperson summary should be submitted to the European Union (EU) database within one year from the end of a clinical trials in all member states (MS) (Article 37).

 The regulation will apply six months after the European Commission will publish the notice in the Official Journal of the European Union that the EU portal and EU database are fully functional.

 This communication was originally expected on May 28, 2016 (Article 99), but, while writing this paper (January 2017), no indication from EMA was released.

	In its Annex V, the Regulation 536/2014 provides a list of items that should be included in the layperson summary. However, the Annex V does not provide any hierarchy of information (the order that each item should have in the document) and cannot be considered as guidance for writing a layperson summary.

 Since 2015, the debate on the items listed in Annex V and how they should be disclosed to average-literacy people has highlighted their potentially critical aspects.

 In 2016, a consultation document on layperson summary, released by the Medical Research Council Technology Return of Results Toolkit, specified the structure of each item and sub-item (see Table 1 below). The Regulation (Article 39) states that the trial sponsor should guarantee the clarity of layperson summaries and their proper submission to EU database; penalties for breach of transparency provisions will be proposed.

Layperson text: Motivation and characteristics of readers

	The target audience for a layperson summary is expected to include not only the general public and the participants in the clinical trials and their relatives, but also healthcare professionals, pharmaceutical researchers, and academics who may be interested in the study but are not genuine experts on the elements involved. Therefore, a balance should be reached to accomplish the average literacy level of the general population and, at the same time, provide useful information for more health-literate people.

	To correctly define the target audience, two important notions should be taken into account: the expected reader’s characteristics (age, education, culture, scientific, and trial knowledge)

 and the reason why he/she is willing to read that specific layperson summary.

		Allow discriminating what should be explained and what could be tacit (level of implicit) in the layperson summary.

		Profoundly change from one reader to another.

	We can reasonably suppose that the most frequent motivation for a layperson to read a clinical trials summary may be related to his/her personal health status or the health of one of their loved ones.

	The readers’ age, education, and culture contribute to defining his/her ability to understand a written document, but in the particular case of a layperson summary concerning a clinical trial, it is quite likely that the health motivation driving the reader is much more relevant than his/her biographic and cultural traits. All together, these aspects represent the cognitive profile of the reader. 

	Therefore, while projecting the hierarchy of a layperson summary, it may be useful to keep into account the motivation and the cognitive profile of the expected reader. 

	First, it should be defined which information is at the top of the reader’s interest: most likely, “Is this drug effective in treating the (my) condition? Is it safe?” The answers to these questions are the discriminant information

 for the reader; they define the rationale for the specific content of the text, justifying the drafting of the summary and motivation for reading.

	Therefore, in our opinion, the investigational medicinal products used in clinical trials and the overall results should have the maximum evidence and should be discussed as the first topic. Then, general information on study procedures, patient populations, and adverse effects should be described in detail to locate the results in their proper context. More technical information required to guarantee the transparency may be summarized at the end in a few sentences (see Figure 1 below). The proposed hierarchy agrees on the regulatory requirements, as described in Annex V, with readers’ interest and expectations.

	Once the hierarchy of information is finalized to address to the readers’ primary interests, the body of the summary may be designed according to some general principles of any health-related text.

	In our proposed hierarchy (right column of Figure 1), we have deleted point 8 (“Comments on the outcome of the clinical trial”). The reason for this deletion is that, in our opinion, it is useless to perform comments related to a drug’s development based on a single clinical trial. We agree with some authors’ opinions 

 that adding any comment to the overall result of the study is, ultimately, useless or even disturbing for the layperson reader. Annex V does not clarify what the sponsor is supposed to comment; besides, any possible qualitative summary statement seriously runs the risk of being perceived as promotional.

	On the contrary, we believe that some comments on the particular result of the single trial, written with clarification purposes, could be included in the overall results, so as to keep the document as readable as possible.

		Adopt a low level of implicit in the summaries.

		Develop the layout and content in line with the needs of the public.

		Follow health literacy and numeracy principles.

	In the development and review of the summary, specific attention should be dedicated to patients, patient representatives, and advocates. Indeed, both patients’ expectation and health literacy are crucial in checking the efficacy of the summary.

	After designing the reader-centered summary, it is essential to adopt an appropriate writing style and language that aligns with the cognitive profile of the readers. Therefore, even if readers’ characteristics regarding age, education, and culture may not affect the interest on the topic and the willingness to read and understand the summary, they may interfere with the complete understanding of the text. This aspect should be kept top of mind; according to the Regulation 536/2014, the layperson summary must be understandable, and the sponsor must guarantee its clarity.

	Many studies have described both general and health-specific literacy of readers. The health literacy includes both the ability to access, understand, and elaborate on information to promote and maintain good health and the cognitive and social skills that motivate an individual to approach health content.

 A recent European Health Literacy Survey

 shows that almost half (29% to 62%) of the population involved in the analysis had inadequate or problematic health literacy, with great differences among countries (see Figure 2).

	Based on Figure 2, it may be important to remove or limit at the minimum any medical jargon and explain any medical objective or outcome in easy, daily words.

	Concerning the general literacy, two adult literacy surveys (the Adult Literacy and Lifeskills Survey [ALL]

, and the International Adult Literacy Survey [IALS])

 are currently available, and a third (the OECD’s Program for the International Assessment of Adult Competencies [PIAAC]) is in progress.

 The Summary of Clinical Trial Results refers to the IALS that identifies five levels of proficiency ranging from level 1 (lowest) to 5 (highest) and points out that communications written for the general public should use simple everyday language to ensure ease of reading and understanding.

 Text should be suitable for people with a low to average level of literacy. Across Europe, the average proficiency level is 2-3, whereas level 3 corresponds roughly to high school completion.

 Sponsors should consider testing the readability of an initial version of the study results summary with a small number of people, representative of the target audience (e.g., patients with a particular disease, patient association representatives, key influencers of public opinion).

	The readability of texts can be formally determined using different metric tools currently available in many European languages (see Table 2 below). For a layperson summary written in English, the Flesch Reading Ease Score should be used: the higher the test score, the easier the text is to read; anything scoring 70 and above is considered easy to read.

 can be useful in multi-country studies where summaries are first drafted in English and then translated into other languages.

 Sponsors are encouraged to consider including an English version if the trial did not include the UK.

 However, the Regulation 536 states that the layperson summary should be provided in the local language of each of the EU countries where the trial had been conducted. Therefore, it is reasonable to expect that soon the regulatory authorities will request the layperson summaries in all the 28 EU languages.

	Neither the Regulation 536/2014 nor the Policy 0070/2014 provides any instruction on the format and the overall length of the lay summary.

 While all the authors recommend using short sentences, only some of them agree with the suggestion of keeping the text short.

 Table 3 (below) mentions the writing style indications to maximize the readability of the layperson summaries.

	The following proposes our experience-based contribution to the debate concerning the contents of the layperson summaries.

	According to our experience, reporting all available identifiers, such as the compound code when it comes to early trials or the international non-proprietary name and the trade names for later stages studies, is not a priority for the reader. Given this observation, we suggest using only generic names of the study drugs, avoiding incomprehensible acronyms or trade names that can vary in different countries.

Description of adverse reactions and their frequency

	Adverse drug reactions (ADRs) represent only a small part of the complete safety information contained in the technical documents (CRS, protocol, patients’ listings, etc.) of a study, as they refer only to drug-related adverse events. Reporting only the ADRs in the lay summary can provide the reader with an incomplete and misleading overview of the safety profiles and the tolerability of the study drugs. In our opinion, it is important to describe the safety profile of the drugs, clearly highlighting that this information is related to a single trial, in a specific population. While we consider appropriate reporting of the adverse events related and not related to the drug candidate to maintain consistency with the study sources, together with some authors,

 we do not consider that the use of MedDRA-preferred terms and EudraCT-system organ class is useful to categorize the study information. Moreover, we believe that adding the frequency (in figures and/or percentage) for all the adverse events can be confusing. As far as our lay summaries writing activity is concerned, we are reporting the most frequent and relevant adverse events, providing them in lay terms and specifying (also numerically whereas needed) if they:

		Resolved spontaneously or required any medical intervention/ hospitalization.

		Led to patient withdrawal or early closure of the trial.

		Were related or not to the study drug. 

	In most cases, trials-especially when it comes to pivotal studies-contain more than one primary endpoint and a long series of secondary endpoints (e.g., efficacy, safety, and pharmacokinetic in the case of Phase II studies), whose meaning is difficult to explain to laypersons and whose value for them is doubtful. Our experience suggests limiting the full and in-depth description of the results to the primary endpoints. We also advise to place secondary and tertiary endpoints in the general chapters on the outcomes of the study and to add additional information (not numerical, unless necessary) only if they represent an added value for the reader compared to the general framework.

	The information provided offers one perspective on the debate around writing and constructing layperson summaries. We propose a “reader-centered” approach to guarantee both transparency and clarity, and fulfil the requirements of the current regulation. We believe that the points described in this paper could contribute to the discussion on layperson summary structure and could be considered a practical tool for sponsors on how to develop comprehensible layperson summaries. Further analysis is needed to describe the impact that layperson summaries will have on clinical trial disclosures. We believe that the process of sharing information has started and regulatory authorities will go ahead in stimulating transparency across clinical development.

			1. Regulation (EU) n. 536/2014 of the European Parliament and of the Council of April 16, 2014, on clinical trials on medicinal products for human use, and repealing Directive 2001/20/EC. Off. J of the European Union L 158, 27.5.2014, p.1. Available online: 

http://ec.europa.eu/health/sites/health/files/files/eudralex/vol1/reg_2014_536			/reg_2014_536_en.pdf

			2. EMA policy on publication of clinical data for medicinal products for human use (Policy 0070). Available online: 

http://www.ema.europa.eu/docs/en_GB/document_library/Other/2014/10			/WC500174796.pdf

			3. Sroka-Saidi K, Boggetti B, Schindler TM. Transferring regulation into practice: The challenges of the new layperson summary of clinical trial results. EMWA. 2015;24(1):24–27. Available online: 

http://www.tandfonline.com/doi/full/10.1179/2047480614Z.000000000274

			4. Gillow, C. Layperson summaries of clinical trial results: Useful resources in the vacuum of regulatory guidance, 

 2015; 24 (4): 205-9. Available online: 

http://www.tandfonline.com/doi/full/10.1179/2047480615Z.000000000324

			5. Brauburger K, New European clinical Trial Regulation: The Requirement for lay summaries and its impact on medical communicators, 

http://www.ctrials.org.il/files/regulation/ 2015v30n2_online_RegulatoryInsights.pdf

			6. MarÒ«al, A, Clinical Trial Transparency, The European Perspective in “Proceedings Harvard Multi-Regional Clinical Trials Center (MRCT) 3

 Annual Meeting, 2014; pp. 6-7. Available online: 

http://mrctcenter.org/wp-content/uploads/2016/02/2014-12-03-MRCT-Annual-Meeting-Proceedings.pdf

			7. Gross A, Lay Summary. Writing for Lay Audiences - Questions & Answers, pp.6-8. Available online:  

http://webcache.googleusercontent.com/search?q=cache:nVbf5uaUj2sJ:www.clinesco.co.uk/

			8. Summary of Clinical Trial Results for Laypersons. Available online: 

https://ec.europa.eu/health/files/clinicaltrials/2016_06_pc_guidelines			/gl_3_consult.pdf

. This document does not necessarily reflect the views of the European Commission and should not be interpreted as a commitment by the Commission to any official initiative in this area.

			9. Bierer B, Introduction to Harvard MRCT and the Return of Results Guidance Document, in Proceedings Harvard Multi-Regional Clinical Trials Center (MRCT), 2014, pp. 5-6. Available online: 

			10. World Health Organization, WHO/PAHO 2009. Available online: 

http://www.who.int/healthpromotion/conferences/7gchp/track2/en/

			11. Health literacy - The solid facts, World Health Organization, 2013. Available online: 

http://www.euro.who.int/__data/assets/pdf_file/0008/190655/e96854.pdf

Analyzing the debate around writing and constructing layperson summaries of clinical trial results-and proposing a “reader-centered” approach.

Considerations on Layperson Trial Summaries in the EU

 The regulation will apply six months after the European Commission will publish the notice in the 

 that the EU portal and EU database are fully functional.

This communication was originally expected on May 28, 2016 (Article 99), but, while writing this paper (January2017), no indication from EMA was released.

Since 2015, the debate on the items listed in Annex V and how they should be disclosed to average-literacy people has highlighted their potentially critical aspects.

 In 2016, a consultation document on layperson summary, released by the Medical Research Council Technology Return of Results Toolkit, specified the structure of each item and sub-item (see Table 1). The Regulation (Article 39) states that the trial sponsor should guarantee the clarity of layperson summaries and their proper submission to EU database; penalties for breach of transparency provisions will be proposed.

	Therefore, while projecting the hierarchy of a layperson summary, it may be useful to keep into account the motivation and the cognitive profile of the expected reader.

	Therefore, in our opinion, the investigational medicinal products used in clinical trials and the overall results should have the maximum evidence and should be discussed as the first topic. Then, general information on study procedures, patient populations, and adverse effects should be described in detail to locate the results in their proper context. More technical information required to guarantee the transparency may be summarized at the end in a few sentences (see Figure 1). The proposed hierarchy agrees on the regulatory requirements, as described in Annex V, with readers’ interest and expectations.

	In our proposed hierarchy (right column of Fig. 1), we have deleted point 8 (“Comments on the outcome of the clinical trial”). The reason for this deletion is that, in our opinion, it is useless to perform comments related to a drug’s development based on a single clinical trial. We agree with some authors’ opinions 

	On the contrary, we believe that some comments on the particular result of the single clinical trial, written with clarification purposes, could be included in the overall results, so as to keep the document as readable as possible.

 A recent European Health Literacy Survey

 shows that almost half (29% to 62%) of the population involved in the analysis had inadequate or problematic health literacy, with great differences among countries (see Figure 2).

 are currently available, and a third (the OECD’s Program for the International Assessment of Adult Competencies [PIAAC]) is in progress.

The Summary of Clinical Trial Results refers to the IALS that identifies five levels of proficiency ranging from level 1 (lowest) to 5 (highest) and points out that communications written for the general public should use simple everyday language to ensure ease of reading and understanding.

	The readability of texts can be formally determined using different metric tools currently available in many European languages (see Table 2). For a layperson summary written in English, the Flesch Reading Ease Score should be used: the higher the test score, the easier the text is to read; anything scoring 70 and above is considered easy to read.

Sponsors are encouraged to consider including an English version if the trial did not include the UK.

 Table 3 below mentions the writing style indications to maximize the readability of the layperson summaries.

	According to our experience, reporting all available identifiers, such as the compound code when it comes to early trials or the international non-proprietary name and the trade names for later stages studies, is not a priority for the reader. These data do not represent the principal interest of the layperson reader, and they could even become a confounding factor, especially when fixed combinations and/or free combinations of molecules are at stake. Following these observations, we suggest using only generic names of the study drugs, avoiding incomprehensible acronyms or trade names that can vary in different countries.

Description of adverse reactions and their frequency 

		Were related or not to the study drug and if there were deaths in the trial.

			1. Regulation (EU) n. 536/2014 of the European Parliament and of the Council of April 16, 2014, on clinical trials on medicinal products for human use, and repealing Directive 2001/20/EC. Off. J of the European Union L 158, 27.5.2014, p.1. Available online: 

			2. EMA policy on publication of clinical data for medicinal products for human use (Policy 0070). Available online: 

			3. Sroka-Saidi K, Boggetti B, Schindler TM. Transferring regulation into practice: The challenges of the new layperson summary of clinical trial results. EMWA. 2015;24(1):24–27. Available online: 

			4. Gillow, C. Layperson summaries of clinical trial results: Useful resources in the vacuum of regulatory guidance, 

			5. Brauburger K, New European clinical Trial Regulation: The Requirement for lay summaries and its impact on medical communicators, 

http://www.amwa.org/files/2015v30n2_online_RegulatoryInsights.pdf

 Annual Meeting, 2014; pp. 6-7. Available online: 

			7. Gross A, Lay Summary. Writing for Lay Audiences - Questions & Answers, pp.6-8. Available online:  http://webcache.googleusercontent.com/search?q=cache:nVbf5uaUj2sJ:www.clinesco.co.uk/   

			8. Summary of Clinical Trial Results for Laypersons. Available online: 

			9. Bierer B, Introduction to Harvard MRCT and the Return of Results Guidance Document, in Proceedings Harvard Multi-Regional Clinical Trials Center (MRCT), 2014, pp. 5-6. Available online: 

			11. Health literacy - The solid facts, World Health Organization, 2013. Available online: 

			12. Adult Literacy and Lifeskills Survey (ALL). Available online: 

Outlining considerations on layperson summary writing of clinical trial results in Europe, and proposing a "reader-centered" approach to constructing these summaries.


Layperson Summary in the EU: Analysis of the Debate and Personal Remarks

	The European journey toward clinical trial transparency (see Figure 1) started in 2003 when the European Medicines Agency (EMA) first published the European public assessment reports. Since 2012, the EMA behavior on trial transparency was reactive, meaning that trial information was available upon request. In 2012, a more proactive measure was proposed for clinical trial transparency, and a long process of consultation with stakeholders, including clinicians, patients’ associations and pharmaceutical companies began.

 In 2014, a new regulation on clinical trials (EU Reg. n.536/2014)

 was published and in October 2014, the EMA policy 0070

 reinforced the concept of clinical trial transparency. In addition, in 2012, the EU Pharmacovigilance Regulation was published.

	According to the European Treaty, a regulation shall have a general application and is directly applicable to all member states. Therefore, each of them should implement the EU Reg. n.536/2014 that applies to all clinical studies conducted in Europe, except for non-interventional studies. This new legislation replaces the EU Clinical Trials Directive (EUCTD), which was approved in 2001 and implemented in 2004.

 The EU reg. n.536/2014 established a novel definition of a clinical study. As stated in article 2 of the Regulation, a clinical study is any investigation about humans with the objective of ascertaining the safety and efficacy of those medicinal products, intended to:

		Discover or verify the clinical, pharmacological or other pharmacodynamic effects of one or more medicinal products

		Identify any adverse reactions to one or more medicinal products

		Study the absorption, distribution, metabolism and excretion of one or more medicinal products (Regulation 536/2014 art.2/1)

	To ensure a sufficient level of transparency in clinical trials, the EU reg. n. 536/2014 states that “the EU database should contain all relevant information as regards the clinical trial submitted through the EU portal. The EU database should be publicly accessible, and data should be in an easily searchable format, with related data and documents linked together by the EU trial number and with hyperlinks, e.g., linking together the summary, the layperson's summary, the protocol and the clinical study report of one clinical trial, as well as linking to data from other clinical trials which used the same investigational medicinal product. All clinical trials should be registered in the EU database before being started. As a rule, the start and end dates of the recruitment of subjects should also be published in the EU database.”
	
	Based on this legal context, on Oct. 2, 2014, EMA released the policy 0070 on the publication of clinical data for medicinal products for human use,

 as a complementary tool ahead of the implementation of the EU Reg. n. 536/2014, effective since May 2016. In order to fully accomplish the EU reg n. 536/2014 requirements to guarantee clinical trial transparency, biopharmaceutical companies should start implementing an internal procedure to upload information of clinical trials to the EudraCT platform and provide the layperson summaries (see Figure 2).


	
	Policy 0070 on publication of clinical data for medicinal products for human use is the result of an extensive EMA initiative, which started in 2012 and involved more than 1,000 stakeholders among academic, industry representatives, patients’ associations, editors and lawyers. EMA was committed to continuously extending its approach to clinical trials data transparency, with the aim to make medicine development more efficient.
	
	Moreover, EMA intended to foster public scrutiny to clinical study information by the scientific community and to develop knowledge in the interest of public health, protecting and nurturing the latter, while promoting a better-informed use of medicines. Indeed, the policy states “A high degree of transparency will take regulatory decision-making one step closer to EU citizens, and promote the better-informed use of medicines. (…) The access to clinical data will benefit public health in the future.”

	The scope of the EMA Policy 0070 relates to proactively sharing study-level and patient–level clinical data, i.e., clinical reports (clinical overviews, clinical summaries, clinical study reports, protocol and protocol amendments, sample case report forms, documentation of statistical methods) and individual patient data (IPD), submitted under the centralized marketing authorization procedure after Jan. 1, 2015.


	
	The process included two sequential phases: in the first one, only the clinical reports supporting centralized marketing authorization application (MAA) submitted only after Jan. 1, 2015, and-starting from July 1, 2015-clinical reports supporting the request for line extensions followed the same rules. In the second phase, after finding the most appropriate way to make them available in compliance with privacy and data protection laws, the IPD will also be provided.


	
	To protect the privacy of patients, EMA issued guidance

 to the pharmaceutical industry on anonymization procedures of clinical reports. The direction aims at assisting companies by recommending methods, techniques and processes that could be applied to clinical reports in order to achieve an adequate anonymization, while retaining a maximum of scientifically useful information on medicinal products for the benefit of the public. Indeed, the same data can be adequately anonymized in different ways, depending on the context of the data release.
	
	In the case of public data release, the risk of re-identification should be slight, whereas, for non-public data sharing, a higher risk could be acceptable.
	
	Due to an active and rapid IT research in this field, many anonymization techniques are currently available to marketing authorization holders (MAHs)/applicants. The legislation is not prescriptive about the techniques to be used by data controllers. According to the Article 29 Working Party Opinion, examples of techniques that could apply to clinical reports are masking, randomization and generalization.


	
	When anonymizing structured data, the main goal is preventing the linkage of sensitive datasets to an identified dataset (or similar background knowledge about individuals). While it is obvious to remove the so called “identifiers,” the attributes that may be used for linkage are defined as “quasi-identifiers,” which are not identifiers, per se, but may be combined to generate a linkage. In addition, these data cannot be removed because they may be required for the analysis. Assuming that directly identifying information has been already eliminated from the dataset, the k-anonymity privacy model should be implemented to enhance data protection. A dataset is k-anonymous if, regarding the quasi-identifiers, each data item cannot be distinguished from at least k-1 other data items. This property can be used to define equivalence classes of indistinguishable entries.

 To fully accomplish the generalization procedure, entries from equivalence classes that violate the privacy model (i.e., outliers) are automatically replaced with semantic-free placeholders, while the percentage of suppressed records is kept under a given threshold.

 Available IT tools can calculate utility measures to guarantee the minimal loss of information after k-anonymization and aggregation procedures.
	
	

Redaction of commercially confidential information


	
	To allow pharmaceutical companies to protect any information contained in the clinical reports submitted to the EMA-which is not in the public domain or publicly available and where disclosure may undermine the legitimate economic interest of the applicant/MAH-EMA issued an External guidance on the identification and redaction of commercially confidential information (CCI) in clinical reports for the purpose of publication in accordance with EMA Policy 0070.


	
	For healthcare companies, the guidance is a working tool and a reference document that aims at supporting them in the procedure of redaction of CCIs and in the preparation of their justifications. EMA will scrutinize the justification for the redaction to assess whether the definition of CCI applies.

In 2014, EU Regulation n.536/2014 represented a significant step toward transparency of clinical trials in Europe. However, despite the EMA’s policy 0070, which reinforced this concept, several aspects remain to be determined or evaluated, leaving room for additional requirements and local interpretation.

E. Ornago

Articles

Anonymization and Redaction of Clinical Trials According to the EU Regulation

Considerations on Layperson Trial Summaries in the EU

Layperson Summary in the EU: Analysis of the Debate and Personal Remarks

Transparency: The EU Prospective