Competition Law and Genomic Ecosystem Concentration Risks

1. Introduction

The genomic ecosystem is no longer confined to laboratories performing DNA sequencing. It increasingly consists of an interconnected chain involving:

• DNA sequencing and genotyping platforms;
• genomic databases and biobanks;
• direct-to-consumer genetic-testing services;
• clinical genomic testing;
• cancer and rare-disease diagnostics;
• bioinformatics and interpretation software;
• cloud storage and computational infrastructure;
• pharmaceutical and biotechnology research;
• AI models trained on genomic and clinical datasets;
• genetic counselling and precision-medicine platforms; and
• pharmaceutical licensing and drug-discovery networks.

This creates a distinctive competition-law problem: concentration may occur not merely through market share, but through control over data, infrastructure, interoperability, algorithms, standards and access to downstream markets.

The most important modern example is Illumina/GRAIL, where competition authorities examined whether control of a critical upstream DNA-sequencing technology could be combined with a downstream cancer-detection business to disadvantage competing diagnostic developers. The FTC ultimately required divestiture, while the Fifth Circuit found substantial evidence supporting the Commission's competitive concerns but remanded on the treatment of rebuttal evidence; Illumina subsequently announced the divestiture.

2. Meaning of Genomic Ecosystem Concentration

Genomic concentration can occur at several levels.

A. Horizontal concentration

Two genomic companies operating at the same level may merge, for example:

sequencing company + sequencing company

or:

genomic testing platform + competing genomic testing platform.

This may eliminate an existing competitor or a future competitor.

B. Vertical concentration

A company controlling an upstream input acquires a downstream genomic application:

sequencing platform → genomic test → clinical diagnosis.

The concern is that the upstream company could favour its own downstream service.

C. Data concentration

A company may accumulate:

• DNA sequences;
• phenotypic information;
• family relationships;
• medical records;
• longitudinal health information;
• disease outcomes;
• population-level genomic information.

Even where the data itself is not sold, control over a sufficiently valuable dataset can create a competitive advantage.

D. Ecosystem concentration

The most complex situation is:

Sequencing + genomic database + cloud + AI + diagnostics + pharmaceutical research

under one corporate ecosystem.

Such integration can produce competitive advantages that are difficult to measure through traditional market-share analysis.

3. Why Genomic Data Can Be a Competitive Asset

Genomic data has several characteristics that make it potentially important under competition law.

3.1 Scale

Large genomic datasets can improve:

• variant identification;
• disease association studies;
• clinical prediction;
• drug discovery;
• population genetics;
• AI training.

3.2 Variety

A genomic ecosystem may combine genetic information with:

• age;
• sex;
• clinical history;
• medication;
• imaging;
• laboratory results;
• disease progression;
• treatment response.

The combination may be more commercially valuable than any individual dataset.

3.3 Velocity

Repeated clinical observations can generate longitudinal datasets.

3.4 Uniqueness

A dataset may contain information that cannot easily be recreated because:

• patients are difficult to recruit;
• genomic sequencing is expensive;
• historical clinical records are difficult to reproduce;
• longitudinal outcomes take years to develop.

3.5 Feedback effects

More users → more genomic data → better algorithms → better products → more users → still more data.

This resembles the network effects considered in digital-platform competition.

4. Principal Competition Risks

A. Acquisition of nascent competitors

A large sequencing or genomic-data company could acquire a smaller company before it becomes a serious competitive threat.

This creates the classic nascent-competition or killer-acquisition concern.

The relevant question is not merely:

"Does the target have substantial market share today?"

but also:

"Could the target become an important competitive constraint tomorrow?"

B. Data foreclosure

A dominant genomic company might restrict competitors' access to:

• genomic datasets;
• API access;
• research databases;
• phenotype information;
• variant databases;
• clinical outcome datasets.

If rivals cannot replicate the dataset economically, access restrictions could increase barriers to entry.

C. Preferential treatment

An integrated company could provide its own downstream products with:

• better sequencing prices;
• faster processing;
• privileged access to datasets;
• earlier access to new technologies;
• preferential API access;
• better computational resources.

The competitor might technically remain able to operate, but its competitive position could deteriorate.

D. Interoperability restrictions

Genomic ecosystems depend upon interoperability.

Examples include:

• sequencing-machine compatibility;
• laboratory information systems;
• genomic-data formats;
• clinical databases;
• APIs;
• cloud platforms;
• electronic health-record systems.

A dominant platform could potentially make interoperability more difficult for rivals.

E. Algorithmic advantage

A company possessing both genomic data and AI infrastructure may continuously improve its algorithms.

The competitive cycle can become:

More patients → more data → better model → better diagnosis → more patients.

Competitors with smaller datasets may face increasing difficulty catching up.

5. Relevant Competition-Law Framework

Several doctrines can be applied.

5.1 Merger control

Competition authorities can examine whether a concentration may:

• eliminate actual competition;
• eliminate potential competition;
• increase barriers to entry;
• facilitate foreclosure;
• combine strategically important datasets;
• reduce innovation.

5.2 Abuse of dominance

A dominant genomic platform may face scrutiny for:

• discriminatory access;
• refusal to supply;
• tying;
• self-preferencing;
• exclusionary rebates;
• interoperability restrictions;
• discriminatory API access.

5.3 Essential-facility principles

Where access to a genomic infrastructure or dataset satisfies the applicable legal requirements, refusal of access can potentially raise essential-facility or refusal-to-deal issues.

The threshold remains demanding, however.

5.4 Vertical foreclosure

Authorities may examine whether integration allows a company to:

1. restrict an important input;
2. raise rivals' costs;
3. disadvantage downstream competitors; and
4. ultimately reduce competition or innovation.

5.5 Innovation competition

Genomic markets are heavily dependent on research and development.

Therefore, competition may be harmed even before consumers experience a conventional price increase.

6. Important Case Laws

1. Illumina, Inc. v. GRAIL, Inc. / FTC

Most directly relevant case

Illumina, a major next-generation DNA sequencing company, acquired GRAIL, which was developing a multi-cancer early-detection test using DNA sequencing.

The FTC argued that Illumina controlled an important input for competing multi-cancer early-detection tests and could have incentives to disadvantage GRAIL's rivals.

The FTC's administrative proceeding ultimately resulted in an order requiring divestiture. The Fifth Circuit subsequently held that substantial evidence supported the Commission's determination concerning the competitive threat, although it remanded because of the legal standard applied to part of Illumina's rebuttal evidence. Illumina then announced that it would divest GRAIL.

Principle

The case demonstrates that:

Vertical integration involving a critical genomic technology can create competition concerns even where the merging parties are not conventional horizontal competitors.

Relevance to genomic ecosystems

A similar analysis could arise where:

genomic database + genomic testing platform

or:

sequencing infrastructure + AI diagnostic platform

are combined.

2. Illumina/Pacific Biosciences

The FTC challenged Illumina's proposed $1.2 billion acquisition of Pacific Biosciences, alleging that the acquisition could eliminate PacBio as a nascent competitive threat in next-generation DNA sequencing.

The parties abandoned the transaction in January 2020.

Principle

The case illustrates the importance of nascent competition.

Competition law may intervene even where the acquired company is not currently dominant if it represents a meaningful future competitive constraint.

Genomic relevance

This is particularly important because genomic technologies evolve rapidly. A smaller sequencing technology may become an important alternative after technological development.

3. Google/Fitbit

The European Commission's review of Google's acquisition of Fitbit is highly relevant by analogy because it involved the accumulation of health-related data.

The Commission examined whether combining Fitbit's data with Google's existing datasets could strengthen Google's competitive position, including in advertising and digital-health contexts. The Commission ultimately approved the transaction subject to commitments, including restrictions concerning the use of certain Fitbit health data and interoperability.

The Commission assessed datasets using characteristics including:

• volume;
• variety;
• velocity; and
• value.

Principle

Data concentration can constitute a competition concern even when the merging companies are not straightforward horizontal competitors.

Genomic relevance

The same methodology can be relevant to:

genomic data + clinical data + pharmaceutical research data.

A genomic dataset may become substantially more valuable when combined with longitudinal clinical information.

4. Apple/Shazam

The European Commission's review of Apple's acquisition of Shazam examined whether combining datasets could strengthen Apple's position in related digital markets.

The Commission considered whether the data increment could increase barriers to entry and expansion and examined the possibility that access to Shazam's data could affect competing services.

Principle

The competitive significance of data depends not simply on the quantity of data but on:

• uniqueness;
• usefulness;
• substitutability;
• ability of rivals to obtain comparable data;
• ability to combine it with other datasets.

Genomic relevance

This is important because two genomic databases of similar size may have radically different competitive significance.

A database containing rare-disease genetic information and long-term clinical outcomes may be substantially harder to reproduce than a general population dataset.

5. IMS Health GmbH & Co. OHG v NDC Health GmbH & Co. KG

C-418/01

This landmark EU competition case concerned access to a copyrighted "brick structure" used for pharmaceutical sales-data analysis.

The Court considered when refusal by a dominant undertaking to license or provide access to a commercially important resource can constitute abuse of dominance.

Principle

The case is important for the relationship between:

• intellectual property;
• proprietary information structures;
• market access; and
• Article 102-type refusal-to-deal analysis.

Genomic relevance

A genomic database may involve intellectual-property rights, database rights, contractual rights and confidential information.

But ownership alone does not automatically create an obligation to provide access. The stringent conditions governing refusal-to-supply cases remain important.

6. Bronner v Mediaprint

Case C-7/97

Bronner concerned access to a dominant newspaper company's home-delivery infrastructure.

The case established a demanding framework for determining when refusal by a dominant undertaking to provide access to infrastructure may amount to abuse.

Principle

Traditional essential-facility/refusal-to-deal doctrine requires careful examination of factors such as:

• indispensability;
• elimination of effective competition;
• feasibility of alternative facilities; and
• justification.

Genomic relevance

Suppose a company controls a genomic database or sequencing infrastructure.

The critical question would not simply be:

"Is this database valuable?"

It would be:

"Is access indispensable under the applicable legal test, and does refusal eliminate effective competition?"

7. Alphabet/Enel X — Android Auto

C-233/23, 25 February 2025

This recent CJEU judgment is especially significant for genomic ecosystems because it concerns interoperability.

Google had refused to make Android Auto interoperable with Enel X's JuicePass application. The Court held that refusal by a dominant platform to ensure interoperability can constitute abuse even where the platform is not indispensable to the applicant's business, where the platform was designed to permit third-party use and the refusal can hinder competition.

Principle

Modern platform competition cannot always be analysed exclusively through the traditional "essential facility" concept.

Genomic relevance

Consider a dominant genomic platform that controls:

genomic-data API + interpretation platform + clinical software.

If it permits some third parties to connect but denies interoperability to a competing genomic service, competition authorities may examine whether that conduct prevents or delays competitive development.

7. Comparative Case-Law Matrix

8. Genomic Ecosystem "Data Moat"

A particularly important future competition issue is the emergence of a genomic data moat.

Imagine:

Company A

→ owns sequencing machines
→ operates testing laboratories
→ collects genomic data
→ connects genomic data with medical records
→ develops AI interpretation tools
→ sells diagnostic services
→ partners with pharmaceutical companies.

Each activity reinforces the others.

The result can be a feedback loop:

More sequencing → more data → better algorithms → better diagnostics → more customers → more sequencing → more data.

Competition law must therefore consider the ecosystem effect, rather than examining each market in isolation.

9. Genomic Database Concentration

A merger between two genomic databases could generate several forms of competitive harm.

A. Loss of independent datasets

Two independently controlled datasets become one.

B. Increased entry barriers

New entrants may be unable to reproduce the combined dataset.

C. Reduced bargaining power

Hospitals, researchers and laboratories may have fewer alternative data providers.

D. Research foreclosure

A dominant company could restrict access to academic or commercial researchers.

E. Pharmaceutical foreclosure

If a genomic-data platform also operates pharmaceutical research services, it might preferentially use its own database and limit competitors' access.

F. AI advantage

The database could become training infrastructure for genomic AI.

This introduces a second layer of concentration:

data concentration → AI concentration → diagnostic concentration.

10. Genomic Ecosystem and Self-Preferencing

Self-preferencing could take several forms.

For example, a genomic platform might rank:

1. its own genetic test;
2. its own laboratory;
3. its own interpretation service;

above independent alternatives.

The issue becomes particularly serious if the platform controls the interface through which doctors, hospitals or patients access genomic services.

The relevant competition question is whether the conduct distorts competition on the merits, rather than simply whether the company prefers its own products.

11. Tying and Bundling

A dominant genomic company could potentially bundle:

sequencing + interpretation software

or:

genomic testing + cloud storage

or:

genomic database access + AI tools.

Competition concerns may arise if customers cannot realistically purchase the components separately and rivals cannot compete for one component without obtaining the other.

However, bundling is not automatically unlawful; the competitive effects, market power, foreclosure and objective justifications must be examined.

12. Interoperability as a Competition Issue

The genomic ecosystem increasingly depends upon interoperable systems.

A competition authority could therefore examine:

• API restrictions;
• proprietary genomic formats;
• laboratory software compatibility;
• database portability;
• cloud interoperability;
• restrictions on third-party analytical tools;
• discriminatory access to sequencing platforms.

The Alphabet/Enel X judgment demonstrates the increasing importance of interoperability in modern Article 102 analysis. The CJEU specifically recognised that a refusal to provide interoperability can raise competition concerns even when the platform is not strictly indispensable in the traditional sense.

13. Innovation Competition

Genomic competition is unusual because today's small company may become tomorrow's major technological competitor.

Therefore merger analysis should consider:

• R&D pipelines;
• patents;
• scientific personnel;
• clinical trials;
• research partnerships;
• genomic datasets;
• algorithmic capabilities;
• potential alternative sequencing technologies.

This was central to the concerns surrounding both Illumina/PacBio and Illumina/GRAIL.

14. Possible Competition-Law Remedies

Where concentration creates demonstrable competitive concerns, possible remedies may include:

Structural remedies

• divestiture;
• separation of competing businesses;
• prohibition of particular acquisitions.

Behavioural remedies

• non-discriminatory access;
• data-access commitments;
• interoperability requirements;
• API access;
• data-silo arrangements;
• licensing commitments;
• prohibition on self-preferencing.

Merger conditions

Authorities may require:

• independent governance;
• firewalls;
• restrictions on combining datasets;
• continued supply to rivals;
• transparent access conditions.

The Google/Fitbit transaction illustrates the use of data-silo and interoperability commitments, while Illumina/GRAIL ultimately illustrates the much stronger structural-remedy approach of divestiture.

15. Challenges for Competition Authorities

A. Defining the relevant market

The market may involve:

• sequencing;
• genetic testing;
• genomic interpretation;
• databases;
• cloud computing;
• AI diagnostics;
• pharmaceutical research.

Traditional market definitions may therefore understate ecosystem power.

B. Measuring data value

Data has no single market price.

Its competitive value may depend upon:

uniqueness × quality × scale × interoperability × analytical usefulness.

C. Privacy versus competition

Competition authorities must distinguish:

• legitimate privacy protections;
• legitimate security restrictions; and
• restrictions that unnecessarily exclude competitors.

D. Innovation is difficult to quantify

A merger may not immediately increase prices but could reduce:

• future R&D;
• scientific diversity;
• alternative technologies;
• diagnostic innovation.

E. Rapid technological change

Genomic technologies can change faster than conventional merger-control timelines.

16. Proposed Analytical Framework

A useful competition-law assessment of a genomic concentration can follow this sequence:

Step 1 — Identify the ecosystem

↓

Step 2 — Identify the relevant genomic assets

↓

Step 3 — Determine market position

↓

Step 4 — Assess data uniqueness and replicability

↓

Step 5 — Examine vertical relationships

↓

Step 6 — Assess foreclosure incentives

↓

Step 7 — Examine interoperability and API access

↓

Step 8 — Analyse nascent and potential competitors

↓

Step 9 — Assess innovation effects

↓

Step 10 — Examine efficiencies and legitimate justifications

↓

Step 11 — Consider behavioural or structural remedies

17. Key Legal Principles Emerging from the Cases

Six broad principles emerge.

Principle 1 — Control of infrastructure can confer ecosystem power

Illumina/GRAIL demonstrates the importance of control over an upstream genomic technology.

Principle 2 — Nascent competitors matter

Illumina/PacBio demonstrates that competition law can address the elimination of emerging competitive threats.

Principle 3 — Data can be a competition asset

Google/Fitbit and Apple/Shazam demonstrate that merger analysis can examine the competitive significance of accumulated datasets.

Principle 4 — Access to proprietary resources requires careful analysis

IMS Health demonstrates the interaction between intellectual property and access obligations.

Principle 5 — Essential-facility doctrine remains demanding

Bronner prevents competition law from becoming a general compulsory-sharing regime.

Principle 6 — Modern interoperability can extend beyond traditional essentiality

Alphabet/Enel X shows the growing importance of interoperability in digital-platform competition.

18. Conclusion

Genomic ecosystem concentration represents a multidimensional competition-law problem. The principal risk is not simply that one company obtains a large percentage of genomic tests. The greater concern can arise when a single undertaking progressively controls sequencing infrastructure, genomic databases, clinical information, AI capabilities, cloud infrastructure, diagnostic interfaces and pharmaceutical research relationships.

The strongest existing precedent is Illumina/GRAIL, because it demonstrates how competition authorities can analyse the combination of an upstream genomic technology with a downstream genomic diagnostic business. Illumina/PacBio adds the important dimension of nascent competition. Google/Fitbit and Apple/Shazam provide useful principles for analysing data accumulation, while IMS Health, Bronner, and Alphabet/Enel X provide frameworks for proprietary access, essential facilities and interoperability.

Accordingly, future genomic merger review is likely to require an assessment that goes beyond conventional market shares and asks:

Who controls the data, who controls the infrastructure, who controls interoperability, who controls the algorithms, and whether rivals can realistically reproduce those competitive assets?