TWN Info Service on Health Issues (Nov25/04)
28 November 2025
Third World Network

Overlooked Potential Users of the PABS System and Implications for Benefit-Sharing

New Delhi and London, Chetali Rao and Sangeeta Shashikant, 28 November – The Secretariat of the World Health Organization (WHO) circulated to delegations an information document titled Potential participants in/users of the PABS System beyond ‘participating manufacturers’, which overlooks a wide range of actors who use pathogen-related digital sequence information to develop commercial products and generate revenue.

The information document also classifies developers of VTDs as a category of users beyond “participating manufacturers”.   

The WHO Secretariat prepared the document at the request of some countries during the informal session of the Intergovernmental Working Group (IGWG) that met in Geneva on 6-10 October. The IGWG is tasked with negotiating an Annex to the Pandemic Agreement, on a Pathogen Access and Benefit Sharing (PABS) System.

The document circulated before the third session of IGWG lists developers of VTDs, laboratories and universities/research institutes/public health agencies as potential users of pathogen material (PABS Material) and sequences generated from the material (PABS Sequence Information) that are not “participating manufacturers”.

However, the actual ecosystem of actors who gain benefits (monetary or otherwise) is much broader. As discussed below, it includes gene synthesis/oligo providers, AI platforms, bioinformatics and cloud analytic firms, and animal health product developers. These are some of the regular users of sequence information generated from pathogens with pandemic potential infecting humans, and they derive substantial benefits from having access to such sequence information.  Once the PABS System is in place, the requirements of the PABS system, including benefit-sharing obligations, should logically also equally apply to such actors.

Further, is it correct to classify developers of VTDs as falling outside the definition of “participating manufacturers”? And what would be the consequences of such a classification? Excluding developers from this definition surely risks undermining implementation of Article 12.6(a) of the Pandemic Agreement, which requires provision of the target of 20% of real-time production of vaccines, therapeutics and diagnostics, during a pandemic emergency to the WHO, as well as other benefit-sharing obligations that are still under negotiation.

Major Free-Riders: Profiting from Sequence Information, Where is the Benefit Sharing?

Gene Synthesis: The global gene synthesis market size was estimated at USD 3.22 billion in 2024 and is anticipated to reach around USD 27.63 billion by 2034, expanding at a Compound Annual Growth Rate (CAGR) of 23.97% from 2024 to 2034, driven by leading providers like Twist Bioscience, Integrated DNA Technologies and Genscript. These platforms use AI to design and customise DNA constructs, including those linked to sequences of pathogens with pandemic potential.

For example, Twist Bioscience uses genetic sequences from databases like GenBank and others to prepare synthetic DNA and RNA that match the sequences derived from the databases. This way, they create commercial laboratory products such as synthetic controls (artificial viral or gene samples) for the detection and analysis of viral pathogens.  In the 3^rd quarter of 2025 (3QFY25), Twist Bioscience reported a record revenue of USD 96.1 million and is expecting revenue in the range of USD 374.0 million to 376.0 million for the full fiscal year 2025. The company’s revenue for the fiscal year 2024 was reported as USD 313.0 million, with synthetic genes (31%) and next-generation sequencing (53%) comprising 84% of total revenues.

AI-driven platforms:  In addition to DNA/RNA/protein sequences data, companies like Benevolent AI, Recursion Pharmaceuticals and Exscientia use a broad set of other data (like phenotypic data, omics data, chemical, structural and pharmacological data) from public and proprietary databases to improve AI models for drug discovery, training models, target validation and hypothesis generation.

For example, Benevolent AI is an AI drug discovery company which extracts genetic information about viruses – such as names of the viruses, sequences and structures of their proteins from databases – and feeds them into a graph that links all the information together. Their AI system then connects all the viral details to human proteins, biological pathways, scientific findings and spots how viruses interact with human bodies or which steps are the most important for the viruses to survive. This allows AI to search for existing drugs that can block these pathways or for new drugs. For example, Benevolent AI used its AI platform to hypothesise baricitinib for COVID-19 treatment in 2020. Its revenue in 2023 was £7.3 million, while it was reported to be £ 4.87 million in 2024.

Another company, Exscientia, which is now a part of Recursion Pharmaceuticals, uses AI and automation drug discovery platforms to screen and advance therapeutics, including those for infectious diseases. It collects gene and protein sequences from sequence databases and then uses this information to figure out which genes and proteins are linked to a particular disease and would make good drug targets. Exscientia’s AI platform then designs possible new molecules to interact with these targets. In 2021, the company entered into a USD 70 million collaboration with the Bill and Melinda Gates Foundation to develop broad-spectrum anti-viral therapeutics, including against coronavirus and other viruses with pandemic potential. Exscientia reported revenues of USD 25.6 million in 2023 and USD 34.7 million in 2022.

Cloud Analytics Firms: The global bio-informatics market was valued at USD 16.66 billion in 2024 and is expected to exceed USD 52 billion by 2034, growing at a CAGR of 12.05% from 2025 to 2034. With the expansion of digital technology and bioinformatics, technology platforms have mushroomed, many of them dependent on access to genetic sequences.

Many software platforms freely and anonymously access genomic data from the International Nucleotide Sequence Database Collaboration (INSDC) and similar public databases. These platforms range from open source tools to commercial (paid private) platforms, and they use the retrieved sequences for many purposes. By applying advanced, complex computational algorithms and sometimes manual reviews, these platforms transform raw genomic data into a curated format. Some commercial databases or software platforms require users to pay a licensing fee for access or use. In this way, these databases or software platforms obtain sequences from free sources like  INSDC, but then generate revenue by offering specialised analysis, data curation or enhanced functionality from the use of sequences that have been freely accessed.

An example is BioNumerics, a commercial bioinformatics platform developed by Applied Maths (a BioMérieux company). It provides a unified platform for storing and analysing many types of biological data, including sequences, fingerprints and phenotypic information. One important application is sequence-based typing of pathogens and viruses, which is particularly useful in outbreak settings for rapid identification and characterisation of viral strains. In practice, BioNumerics is used to import sequence or isolated data of pathogens, compare them across labs, regions or countries and generate analysis or outputs related to surveillance and outbreak investigations. The company monetises this through software licenses, service contracts, even though much of the underlying sequence data may originate from public databases. Thus, BioNumerics forms a part of the broader value chain built on accessing and using sequence information.

There are many companies like BioNumerics that use pathogen data to generate commercial products and/or services which are then sold or licensed to developers/manufacturers.

Cloud platforms: Cloud platforms range from general-purpose giants such as Amazon Web Services and Google Cloud Platform, to specialised workbenches like Terra or Software-as-a-service platforms, which provide the necessary infrastructure for scientists to upload, securely store and analyse vast amounts of genetic sequence data efficiently. Cloud bioinformatics platforms utilise genetic information by hosting, processing and transforming raw sequence data into valuable scientific products like comparative reports, novel discoveries, and clinical support tools. Many of these platforms are commercialised through subscription fees, licensing, pay-per-user models or service contracts.

Animal health product developers/manufacturers: Genomic sequencing of viral pathogens infecting humans is also being leveraged in the animal industry to develop vaccines and therapeutics for animal health. For example, major animal companies like Zoetis (USD 9.2 billion), Elanco Animal Health (USD 4.22 billion), and Merck Sharp & Dohme (USD 5.55 billion) also use human influenza strains to develop vaccines and diagnostics for livestock and pets, an approach that helps to control cross-species outbreaks and zoonotic threats. For example, Zoetis has a long history of using influenza strains, both from avian and mammalian sources, to develop vaccines for poultry and cattle.  

While influenza does not seem to be within the scope of the PABS system, the example provides an insight into the use of human pathogens for the development of vaccines, diagnostics and other products to control animal diseases. 

A glimpse of how sequences from pathogens infecting humans have been used by animal health companies can be seen from the patent filings of these companies, which list the uses of sequence information of human pathogens for vaccine manufacturing. For example, patent application WO2006115843 titled “Nipah Virus vaccines” describes creating animal vaccines against the Nipah virus by putting the Nipah “G” and “F” surface protein into pox virus vectors and showing that they protect pigs. Nipah virus F and G genes tied to GenBank Accession no. NC_002728. This accession, NC_002728, is the Nipah/Malaysian human isolate reference genome. Thus, the Nipah virus genome isolated from human cerebrospinal fluid can be used to produce vaccines for pigs.

Multiple patents have been filed by animal health product developers, including Merial Inc. (now part of Boehringer Ingelheim Animal Health USA Inc.) and Merck, covering various aspects of vaccine production, diagnostics, and related technologies to improve animal health outcomes.

Although the animal sector represents a multi-billion dollar industry, with a strong focus on swine vaccines and diagnostics, numerous companies and research institutes are engaged in the development of vaccines targeting emerging zoonotic threats such as Nipah and Zika virus. These efforts frequently utilise sequence data originally derived from pathogens infecting humans. Many of these vaccine candidates remain in clinical stages of development and have not yet reached commercialisation.  

[In the above-mentioned cases, sequences may have been accessed via the INSDC sequence databases, especially GenBank (and possibly others as well). INSDC consists of: GenBank based at the National Center for Biotechnology Information, United States; DNA Databanks of Japan, based at the National Institute of Genetics, Japan, and the European Nucleotide Archive based at the European Bioinformatics Institute, European Molecular Biology Laboratory, United Kingdom. These sequence databases mirror each other’s sequence information.

A key concern with such databases is that they allow anonymous access and full dataset downloads without requiring users to accept terms or comply with access and benefit-sharing (ABS) obligations. Anonymous access to sequence information makes it impossible to effectively operationalise the PABS System and is antithetical to the principle of legal certainty in ABS terms and conditions mandated by Article 12.5(b) of the Pandemic Agreement.]

Can VTD Developers Credibly Be Classified as Non-Manufacturers?

The WHO Secretariat’s information document also incorrectly identifies developers of VTDs as users of the PABS System beyond “participating manufacturers”. 

Presumably, it follows from the first IGWG Bureau’s draft PABS Annex text that incorporates an extremely narrow definition of “participating manufacturer” i.e. “Participating Manufacturer” means a public or private entity, for profit or not-for-profit, that manufactures vaccines, therapeutics and/or diagnostics, including by means of licensing agreements, and that has signed a legally binding contract with WHO regarding that entity’s participation in the PABS System”.

The approach of the Secretariat and Bureau is a significant departure from the definition agreed in the Pandemic Influenza Preparedness Framework (PIP Framework) which defines “Influenza vaccine, diagnostic and pharmaceutical manufacturers” as “public or private entities including academic institutions, government owned or government subsidized entities, nonprofit organizations or commercial entities that develop and/or produce human influenza vaccines and other products derived from or using H5N1 or other influenza viruses of human pandemic potential.”.

[The PIP Framework is an access and benefit-sharing system for influenza viruses with pandemic potential, developed through intergovernmental negotiations and adopted by the World Health Assembly in 2011.]

The Secretariat’s document clearly fails to recognise the inherently blurred line between a developer and a manufacturer of VTDs. Exclusion of product developers from the definition of “participating manufacturer” could adversely impact benefit-sharing, such as commitments to deliver a certain percentage of VTDs to the WHO.

Development of VTDs is not limited to the conceptual or design stage. It may entail some degree of manufacturing activity, including pilot-scale or clinical-grade production (for clinical trials). This raises a fundamental ambiguity: would an entity whose primary role is product development – but which necessarily conducts very small-scale manufacturing – fall within, or outside, the definition of a “participating manufacturer”? The same entities can also, at any point, commence full-scale manufacturing on their own or through partnerships or contract manufacturing organisations, including during a public health emergency.

If VTD developers are excluded from the definition of “participating manufacturers”, then entities signing contracts with the WHO as “developers” would not be subject to the obligations imposed on manufacturers, such as providing access to VTDs or other mandated benefits (to be negotiated), including providing other manufacturers with non-exclusive production licenses. Yet, during a public health emergency of international concern (PHEIC) or a pandemic, these same developers could rapidly scale up manufacturing to capitalise on increased demand. However, these developers would remain free of any commitments to supply WHO under the PABS System, because the benefit-sharing obligations apply to a “participating manufacturer” and not to a “developer” (under the Bureau’s proposed definition).

Product developers also usually outsource production using contract manufacturing organisations (CMOs). In such a scenario, the product developer remains the holder of all rights and is, in effect, the legal manufacturer. For example, Moderna outsourced commercial manufacturing of its COVID-19 vaccine to Lonza in 2020.

In another case, Novavax entered the COVID-19 response primarily as a product developer of NVX-CoV2373 vaccine, with very limited pre-existing in-house commercial manufacturing capacity. Early in the pandemic, it relied on public funding agreements with CEPI (Coalition for Epidemic Preparedness Innovations) and the US government to support clinical development and large-scale manufacturing, while having no large-scale manufacturing facility of its own. To produce doses, it instead relied on a network of CMOs and local producers, including SK Bioscience, Takeda and FUJIFILM Diosynth Biotechnologies, to make the vaccine.

As demand surged Novavax rapidly scaled up global supply through this network and signed purchase agreements, such as with Gavi/COVAX as a contract supplier, while retaining control over the underlying intellectual property, technology and product strategy. This illustrates how a company can move from a pure developer role into a manufacturer and commercial supplier without necessarily being subject, from the outset, to the same obligations that a formally designated manufacturer might face under the PABS system, if the system’s obligations only apply to narrow notions of who “manufactures”.

Developers are actually an integral part of the ecosystem of production and would typically be the legal owner by owning intellectual property rights over the VTDs being developed.

Thus, it would be unfortunately short-sighted to consider developers of VTDs as users of PABS system beyond “participating manufacturers”.