5

Privacy, Ownership, and Accessibility Considerations in the United States

The fourth session of the workshop explored legal and ethical considerations related to access, use, and analysis of pathogen genomics data in public health applications. Richard Wilder, senior scholar at the Georgetown University O’Neill Institute for National and Global Health Law, moderated the panel discussion. Alexandra Phelan, associate professor at Johns Hopkins Bloomberg School of Public Health, outlined considerations regarding sovereignty over genetic resources and efforts to ensure access to and share benefits from these resources. Claudia Emerson, founding director of the Institute on Ethics and Policy for Innovation at McMaster University, highlighted privacy implications and ethical considerations related to genome sequencing in public health surveillance. Benjamin Berkman, head of ethics of genetics and emerging technologies at the National Institutes of Health (NIH) Department of Bioethics, discussed the NIH data-sharing policy and ethical issues associated with potential harms, sample repurposing, and data accessibility in pathogen genomics. Bronwyn MacInnis, director of pathogen genomic surveillance at the Broad Institute, described barriers to data accessibility and initiatives to expand data-analysis capability and access.

INTERNATIONAL GOVERNANCE OF PATHOGEN GENOMICS DATA

Phelan discussed sovereignty over genetic resources, genetic resource digital sequence information (DSI), and multilateral access and benefit sharing (ABS) for pandemic pathogen samples and sequences. She described how during the COVID-19 pandemic, inequities—particularly those related

to vaccine access—profoundly shifted perception on collective health security within the public and within the Global North scientific community. The Global South has long held the view that health security involves fair distribution of health resources and a foundation of trust underlying global outbreak and response, and she maintained that this understanding is spreading in the Global North.

Sovereignty over Genetic Resources

Sovereignty is a founding principle of international law that became increasingly relevant during the decolonization movements of the 1950s and 1960s, said Phelan. Given that sovereignty involves the right of a state to rule itself, develop its own laws, and determine how resources within its territory are used, the extraction of resources was a particularly forceful feature of colonialism that denied sovereignty, self-determination, and economic prosperity, she explained. The 1992 Convention on Biological Diversity expressly recognizes states’ preexisting sovereign rights over genetic resources within their territories and defines genetic resources to include any material of microbial origin containing functional units of heredity.¹ The objectives of the Convention on Biological Diversity extend beyond conservation of biological diversity to sustainable use and fair and equitable sharing of benefits that arise from the use of genetic resources. The 2010 Nagoya Protocol governs stipulations on ABS.² Currently, ABS via the Nagoya Protocol is largely carried out on a bilateral basis; the country of origin for a genetic resource is permitted to require prior informed consent before a recipient takes that resource, and such transfer is performed on mutually agreed-upon terms, often within a material transfer agreement. Terms under the Nagoya Protocol may include equitable sharing of benefits such as research funding, joint ownership of intellectual property, corporation contributions, scientific collaboration, technology transfer, capacity building, and publication acknowledgment. Phelan noted that in some circumstances, this bilateral model would be ineffective. Thus, the Nagoya Protocol allows for multilateral ABS regimes to enable distribution of specific genetic resources based on agreed-upon values such as public health, risk, or need. For example, the World Health Organization

___________________

¹ Convention on Biological Diversity, United Nations, New York, is available on https://treaties.un.org/doc/Treaties/1992/06/19920605%2008-44%20PM/Ch_XXVII_08p.pdf (accessed October 8, 2024).

² Nagoya Protocol on Access to Genetic Resources and the Fair and Equitable Sharing of Benefits Arising from Their Utilization, United Nations, New York, is available on https://treaties.un.org/doc/Treaties/2010/11/20101127%2002-08%20PM/XXVII-8-b-Corr-Original.pdf (accessed October 8, 2024).

(WHO) Pandemic Influenza Preparedness Framework applies to physical samples of human influenza with pandemic potential, but does not apply to genetic DSI (WHO, 2011).

Genetic Resource Digital Sequence Information

Technological advancements in pathogen sequencing and the development of medical countermeasures from sequence data have reduced the need for physical samples, Phelan explained. She described this dynamic as holding potential to bypass ABS obligations. To address this, some countries have included DSI of genetic resources within the scope of domestic ABS laws; however, these laws are typically limited to bilateral sharing. A series of efforts is under way to incorporate sequence data into the scope of multilateral ABS approaches. The December 2022 Convention on Biological Diversity Conference of the Parties established an agreement stipulating that benefits of the use of DSI on genetic resources should be shared equitably, that access should be open and not hinder research and innovation, and that a distinctive approach is required but should not necessarily be legally binding (United Nations Conference of the Parties, 2022). Phelan highlighted ongoing work that explores whether commercial use of DSI should only be subject to benefit sharing for funds, the types of not-for-profit use that should be governed, and whether benefit sharing should be voluntary or compulsory for specific triggers.

Multilateral Access and Benefits Sharing for Pandemic Pathogens and Sequences

Parallel processes are under way that could influence whether a multilateral DSI instrument would apply to pathogen genomic sequence data, said Phelan. WHO member state negotiations for a pandemic agreement began in December 2021 and have centered on the proposed creation of a Pathogen Access and Benefits Sharing system, an issue she described as contentious. To date, member states have agreed to but not yet ratified that states have sovereign rights over biological resources; a public health need exists for the rapid and timely sharing of DSI for pathogens with pandemic potential; rapid, timely, fair, and equitable sharing of benefits from pathogen samples and sequences is appropriate; a Pathogen Access and Benefits Sharing system should be established and governed by a new instrument negotiated in parallel with or subsequent to the pandemic treaty; and legal certainty and research and innovation are important. Issues that have yet to be determined include the definition of a pathogen with pandemic potential, the definition of sequence information, the process for accessing samples and DSI, the degree to which benefits will be shared, the use of legally

binding contracts, and potential monetary contributions from industry.³ She highlighted that the target date for adoption of a finalized treaty is December 2024 or May 2025. Phelan underscored that inequitable access to diagnostics, vaccines, and therapeutics renders the world less safe. Moreover, most countries are reluctant to share samples and sequences with an international system that does not ensure access to countermeasures. These issues need to be resolved before the next pandemic, Phelan emphasized.

PRIVACY AND ETHICAL CONSIDERATIONS RELATED TO HUMAN GENOME SEQUENCES IN PUBLIC HEALTH DATASETS

Emerson explored privacy implications, ethical considerations, and the mitigation of ethical risks involved in public health surveillance activities. She clarified that her discussion of privacy concerns related to pathogen genomic surveillance assumes circumstances involving unconsented data, the presence of human genomic information in samples collected from wastewater surveillance and other environmental samples, and the potential presence of other health information—such as data on communicable and noncommunicable diseases, substance use, demographic data, and certain types of metadata—in collected samples.

Privacy Implications

Emerson said that privacy implications in the context of public health extend beyond theoretical risk. Although privacy breaches are uncommon, samples may include identifiable information within genomic sequences or in conjunction with other data. Potential identification of individuals or communities could create stigma, cause discrimination, and perpetuate existing inequities. This potential also raises questions of fairness in terms of how data are acted upon; for example, spread of infection could be used to justify restrictions on individuals or specific communities that result in unintended consequences. She raised the ethical question of whether linking and analyzing a person’s unconsented data is inherently harmful in invading privacy and autonomy, regardless of whether other harms result from the data use. In determining the nature and scope of obligations regarding data use, public health actors must strike a balance between privacy protection and the duty to act on data and share them further, as well as being responsive to findings. Observing that public interests typically prevail—particularly given the rarity of privacy breaches—Emerson emphasized the

___________________

³ Additional agreements related to digital sequence information on genetic resources were reached at the 2024 COP 16, decision 16/2. See https://www.cbd.int/conferences/2024/cop-16/documents (accessed March 2, 2025).

importance of ensuring that benefits outweigh assumed risks. To illustrate, she presented a case study of wastewater surveillance in Wisconsin that raised privacy concerns (see Box 5-1).

Ethical Considerations

Emerson outlined ethical considerations in public health surveillance activities, noting that WHO recognizes surveillance as a duty that governments owe their constituents, thereby bringing issues beyond privacy into play (WHO, 2017). These considerations include privacy risks to individuals and communities, fairness in distribution of benefits and burdens justice in remediating or responding to harms, participation of historically marginalized populations, responsiveness to community health needs, and ethical data management that covers data sharing. Furthermore, ethical considerations extend to awareness of the complex sociocultural environment and sources of trust, or lack thereof, in public health. She remarked that equitable partnerships contribute to equitable participation in the research enterprise as well as in capacity strengthening and attribution, particularly when partners from low- and middle-income countries (LMICs)

are involved. Equitable participation should also inform how resources are allocated. Emerson added that response time is often overlooked as an ethical issue, but it offers an important justification for trade-offs inherent in public health that may impinge on privacy or other freedoms. Emerson highlighted management of expectations as a core component of managing ethical risks in public health. This entails engaging with partners and communities to foster relationships, clarifying the obligations of various actors to maintain confidentiality, and translating benefits to the public and partners. She suggested that confidentiality may be a more useful concept to public health than privacy, because the data are seen, accessed, used, and shared. Moreover, clear communication to the public about benefits and risks helps to satisfy a principle of reciprocity. Emerson continued that the values that underpin public health should also anchor genomic surveillance programs: promotion of the common good; respect for persons and communities; solidarity; equity; and justice in global partnerships.

ETHICAL ISSUES IN PATHOGEN GENOMICS

Berkman discussed ethical issues in pathogen genomics related to potential harms, sample repurposing, and data accessibility. He emphasized that pathogen genomics differs from human-subject research because it involves deidentified samples that do not require oversight from institutional review boards. However, there are lessons to be gleaned from ethical issues that have arisen in human genetics work that could benefit the field of pathogen genomics, particularly as the ability to link clinical and pathogen genomic data expands. To illustrate, he cited the example of lawsuits filed by parents arguing that they did not provide adequate consent for use of their newborns’ blood spots in public health research, resulting in several states dropping the samples from research. Similarly, controversy arose early in the COVID-19 pandemic regarding the repurposing of samples collected for other pathogens toward SARS-CoV-2 research. Other examples include reidentification of individuals, the Henrietta Lacks case,⁴ and downstream commercial uses flowing from the use of deidentified samples.

___________________

⁴ Henrietta Lacks was a Black patient with cancer from whom cancerous cells were taken during a biopsy in 1951 and subsequently used, propagated, and commercialized without consent for research purposes. The HeLa cell line has been used widely in cancer, immunology, and infectious disease research, enabling numerous discoveries. Research using these cells, including familial medical records and genomic sequence data, were published without obtaining consent from her or from her family after her death. In 2023 her family reached a settlement with a science and technology company that profited from the use of the HeLa cell line without family consent.

NIH’s Policy on Genomic Data Sharing

NIH policy facilitates sharing of large-scale genomic data in all NIH-funded research to enable data aggregation for analyzing complex questions, Berkman explained. Applying to both human and nonhuman genomic data, the policy requires investigators to deposit data in a repository with access that may be controlled or open to the public. Institutional certification stipulates appropriate secondary uses of submitted data to define the breadth of sample and data usage.

Implementation of NIH’s genomic data-sharing policy has given rise to numerous challenges, said Berkman. For instance, the policy includes an expectation that explicit consent be obtained for all human samples and sequencing, as well as the encouragement to obtain consent for future secondary use and broad sharing of genomic and phenotypic data. This may raise reluctance to give consent in identifiable populations with privacy interests or individuals with rare diseases who could be easily reidentified. A further complication is that the policy only applies to samples collected after its adoption on January 25, 2015, noted Berkman. For the deposit of samples predating the policy, the established standard is that data were obtained with consent language that is not inconsistent with sharing (i.e., it does not explicitly state that data will not be shared). This poses the challenge of determining what language is or is not consistent with sharing. Additionally, the policy stipulates that each investigator articulate a plan for sharing data but does not specify the review process, criteria, or persons responsible for doing so. The policy does not address data repository design and long-term maintenance, and Berkman emphasized the funding and coordination required to make resources available in perpetuity. Similarly, a centralized resource or support for data deposition has yet to be established. The policy requires substantial datasets to be deposited in a large-scale repository without defining “substantial.” Berkman stated that exemptions from the requirement to share are supposed to be in place—such as in cases where a population is potentially identifiable or faces stigma—but reaching group consensus on a process for granting exemptions has proved challenging.

General Ethical Issues in Pathogen Genomics

Outlining general ethical issues in pathogen genomics, Berkman noted potential psychosocial, economic, and nonwelfare harms. Psychosocial harms include anxiety, embarrassment, and stigma. Workplace or insurance discrimination constitute economic harms. Nonwelfare harms refer to violations, such as autonomy infringement, that are free from tangible harm. He contended that individual harms are very rare in pathogen genomics, particularly

psychological harms. Studies have indicated that the psychosocial reactions individuals may have upon learning undesirable information about themselves tend to be of relatively brief duration, he noted. Berkman maintained that group harms are more relevant to pathogen genomics in that psychosocial, economic, and nonwelfare harms may affect a population simultaneously. For instance, identifying a group as particularly susceptible to a specific pathogen or as the source of an emerging outbreak may produce negative consequences to that population. Deidentification of populations at the community level could mitigate such effects, but removal of data in the deidentification process could decrease the value of the dataset, said Berkman.

Both the repurposing of collected samples and the open data movement entail ethical questions, Berkman remarked. Early in the COVID-19 outbreak, scientists used samples collected for Seattle Flu Study surveillance of influenza to determine whether SARS-CoV-2 was emerging, despite somewhat strict data-use limitations on the samples. Although the IRB endorsed repurposing, this use of samples generated resistance at the federal and state levels. This raises the question of whether public health emergencies justify relaxing protections for human-subject research (Berkman et al., 2021). Berkman also discussed ethical considerations of data democratization via the open data movement that advocates for free, open access to and use of datasets, particularly those created via public funds. He posed the question of whether nontraditional researchers—citizen scientists, or persons not affiliated with a research institution or who do not have the training typically associated with researchers—should be allowed access to individual-level data. Expanded access could enable more discoveries, answer novel scientific questions, and enhance perspectives. Nevertheless, the risks of possible reidentification, individual privacy concerns, and group harms remain. Individuals lacking training and without institutional oversight may veer from norms the research profession has established to mitigate these risks. Berkman underscored that open data access also carries the potential for data misuse, particularly among those seeking to use high-reputation data to justify ideologically driven views and misinformation. He stated that the ethics of research with human subjects is an imperfect fit with the field of pathogen genomics, but it may yield lessons regarding data-sharing policies in the face of potential group harms, data repurposing, and data access to nontraditional researchers.

DEMOCRATIZATION EFFORTS AND INTEGRATION OF ALTERNATE DATA SYSTEMS

MacInnis discussed efforts to democratize pathogen genomics, or expand access to genomics tools and technologies, through the lens of who is using and benefiting from the data. She emphasized the need to expand

access to pathogen genomics tools and technologies into lower-resource environments worldwide. This need was evident during the COVID-19 pandemic, when there were substantial international discrepancies in public availability of SARS-CoV-2 sequences in public repositories (Chen et al., 2022). MacInnis called for breaking down silos and engaging a broader range of stakeholders in the pathogen genomics enterprise. Across all settings, she noted that practitioners who may not specialize in the computational skills to analyze these data (e.g., epidemiologists, clinicians, frontline health care workers, biologists) can contribute insights that extract value from the data. She highlighted current barriers to the democratization of pathogen genomics, including cost; perceived value of the use of genomic data in relation to the high cost of generating them; perceived risks; limitations in technical capacity; expertise silos; and issues related to ownership, contribution, and attribution. Barriers to procurement of reagents and technical support in LMICs continue to hamper research efforts in the Global South, she added (Mukhwana et al., 2024).

Fostering Engagement in Pathogen Genomics Surveillance

Several efforts to foster broadened engagement in the pathogen genomic surveillance space are under way, said MacInnis. The Centers for Disease Control and Prevention’s (CDC’s) Pathogen Genomics Centers of Excellence (PGCoE) network seeks to expand and deepen collaboration between U.S. public health agencies and academic institutions. PGCoE objectives include exploring, advancing, and understanding new technologies to maximize applications to public health and providing training on these technologies to increase surge capacity. Moreover, PGCoEs are establishing partnership agreements and regulatory documentation such as IRBs, material transfer agreements, and data-use agreements. She highlighted the value of building generalizable and universal templates before a health crisis to facilitate barrier navigation during emergency contexts. Similarly, cross-sector trust developed via PGCoE collaboration fosters efficient cooperation during crises, she added. MacInnis highlighted the PGCoE National Education Program to integrate workforce development. Taking a holistic approach, the program extends beyond technical and analytical training to create cross-disciplinary, de-siloed learning pathways, she said. Moreover, the program is seeking to generalize genomic principles and connect genomic data to decision making.

Democratization will make engaging the public an important element of pathogen genomic work, said MacInnis. To illustrate, she described how the lack of IRB oversight of SARS-CoV-2 sampling resulted in the public being unaware of their contributions to COVID-19 genomic sequencing efforts (Bendix, 2021). She asserted that as the public develops situational

awareness about genomic sequencing, questions will arise regarding access to such information.

Building Capacity for Data Handling and Analysis

The Broad Institute created the Terra data platform to enable genomic data management analysis in the cloud, thereby making it more accessible. MacInnis emphasized that the cloud-based platform facilitates secure, scalable collaboration with partners in other parts of the world. Furthermore, the platform is designed to be accessible to users with little to no coding ability. The platform uses open-source, containerized workflows based on highly vetted community standards. These features enable users without formal training in genomics to access standardized, reproducible, and interoperable bioinformatic analyses, she added.

DISCUSSION

The Role of International Agreements

Wilder asked about the relevance of the Nagoya Protocol, given that international rights of sovereignty that provide the impetus for questions regarding ABS are already in place. Phelan replied that the creation of the Nagoya Protocol enabled countries—particularly those in the Global South—to mobilize arguments around sovereignty and genetic resources. Stating that the global health space has not addressed equitable access to vaccines, diagnostics, and therapeutics with adequate speed, she commented on the lack of structures that systematically address injustices and inequities between nations. The Convention on Biological Diversity forum became a receptive space for arguments, with treaty negotiations that resulted in its objectives expanding beyond biological conservation to encompass equitable sharing of benefits. Phelan emphasized that states are the main actors and promulgators of cooperation across the world, but transnational corporations are now profoundly shifting global interactions, and international law is a mechanism to influence these dynamics. She added that the Pandemic Influenza Preparedness Framework engages industry through legally binding contracts, an approach she described as unusual and innovative.

Private Sector Proprietary Data

An attendee asked about progressing a larger data-sharing initiative within a capitalist context, noting that many institutions send their samples to private companies that return project-specific results while maintain-

ing closed and proprietary data. Noting legitimate restrictions on science, Phelan pointed to citing practices as an example. She contended that governments and regulatory agencies are appropriate parties for determining the extent to which intellectual property protections apply to genomics data. Berkman commented that most large-scale data collections receive some level of public funding, which somewhat mitigates this issue. He described a push-pull dynamic that allows sufficient intellectual property protections to incentivize innovation, while simultaneously generating adequate value from large-scale data-sharing efforts to motivate participation from private entities. Emerson noted that the public can be reluctant to share data with the private sector, which could be addressed by enhancing the public’s understanding about the role of those data in innovating novel interventions and products.

Ethical Evaluation in Pathogen Genomics Investigations

An attendee asked whether ethical evaluation resources are available to genomic epidemiology experts using pathogen genomics in investigations of Legionella, multi-drug-resistant organisms, measles virus, and other pathogens causing outbreaks, particularly in marginalized communities. She remarked on the need to balance the benefit of engaging with marginalized communities to determine risk with the potential for group harms. Berkman underscored the value of bioethicist consultation and stated that human genomics resources offer helpful frameworks to consider for pathogen genomics applications. He added that any major hospital system or academic medical center has an ethics consultation service, IRB committee, or an ethics committee. Institutions partnering with health systems could look to these structures for bioethics consultations. Wilder highlighted that a potential resource in the Gates Foundation, WHO, and other research funders that require applied-ethics due diligence. A participant commented that the field would benefit from more formalized structures to provide ethics guidance to public health entities.

Expedited Protocol Considerations

A participant remarked on the challenge of writing an IRB protocol during an outbreak, which could be mitigated by developing a pathogen-agnostic IRB for data sharing and research engagement. Berkman replied that the regulations governing human-subject research focus on individual biomedical harms and benefits, thus are not perfectly suited to public health research. The U.S. Department of Health and Human Services has developed a public health emergency IRB that features some regulatory flexibility of human-subject protection rules during an emergency. Berk-

man contended that in some cases, following regulations could impair the rapid generation of data that is vital for addressing an emerging pandemic, thereby constituting a strong ethical argument for flexibility. He suggested developing a body devoted to reviewing public health protocols that allows some regulatory flexibility to facilitate urgent research in the face of an emerging pandemic or disease.

Another participant noted the reemergence of pre-pandemic regulatory hurdles and asked how to facilitate data sharing between public health emergencies in that context. She added that the value of nontraditional ABS components, beyond vaccines and diagnostics, should be publicized to encourage open data sharing. Phelan replied that although vaccines, diagnostics, and therapeutics are tangible benefits that most people have experienced, open data in and of itself is a benefit that translates into capacity building and technology transfer. However, these nontangible benefits are difficult to specify within the terms of a treaty. Berkman noted the duplication of effort that arises when each research group creates individual data collection and sharing protocols for IRBs. He suggested that establishing a broad, centralized IRB protocol applicable to numerous small projects could expedite the approval process during interim periods, without needing to relax regulatory rules designed to protect human subjects. Emerson commented that pathogen-agnostic umbrella protocols could expedite the approval process, noting that WHO has attempted to develop the type of fast-track protocols created during the COVID-19 pandemic but has been unable to overcome coordination and capacity challenges. She added that the African Vaccine Regulatory Forum has successfully fast-tracked the clinical trial ethical review, emphasizing that this effort shortened the review process to 10 days by establishing agreements that avoid separate approvals for all components of multisite studies. MacInnis suggested that a forum could work to develop universal regulatory documents that address ethics elements.