The GMO banana Europe may never get to eat

Bananas, biotechnology and Europe’s evolving GMO rules meet in one unusually revealing story about a lost variety, a spreading disease and the question of whether regulation can keep pace with science.

In the mid twentieth century, the world nearly lost its favourite banana.

At the time, the Gros Michel variety dominated global exports to Europe and North America. It had a thick peel, travelled well and was widely remembered for its richer flavour compared with today’s Cavendish bananas. Then Panama disease, a soil borne fungal disease, caused by Fusarium oxysporum f. sp. cubense, spread through plantations in the 1950s and made large-scale production commercially unviable.

The memory of Gros Michel still survives in a surprisingly ordinary place: banana sweets. The familiar artificial banana flavour is strongly linked to isoamyl acetate, a compound found naturally in bananas. Gros Michel contained more of it than Cavendish, which helps explain why banana candy can taste like an exaggerated memory of a banana rather than the fruit most people buy today. The story is often simplified too far, but the chemistry points to a real sensory difference between the old export banana and the modern one.

The monoculture bargain

The banana industry survived only by switching to the Cavendish variety. Cavendish was resistant to the specific disease strain that had devastated Gros Michel, which kept the export system running. It also made global banana production even more genetically uniform.

That uniformity reduced resilience and created the conditions for the problem we are facing today. Cavendish bought time. It did not solve the underlying weakness of relying on one dominant commercial variety.

History is repeating itself. A more aggressive form of the same disease, Panama disease Tropical Race 4 or TR4, has been spreading across banana growing regions worldwide. Cavendish has little natural defence against it. This time, however, science offers tools that were not available in the 1960s.

The tools Gros Michel never had

Modern biotechnology makes it possible to protect bananas against disease and, in principle, even to bring back lost varieties such as Gros Michel, but with improved resistance. That does not mean one technology or one company will “save the banana”. It means the industry now has several routes out of the genetic corner it entered after Gros Michel collapsed.

Different scientific approaches are being explored.

In Australia, researchers at Queensland University of Technology have developed QCAV-4, a genetically modified Cavendish banana resistant to Panama disease TR4. Food Standards Australia New Zealand approved food derived from QCAV-4 in 2024, describing it as the first whole GM fruit it had assessed and approved, and the first GM banana approved in the world. The resistance is conferred by a gene from a wild banana relative, Musa acuminata ssp. malaccensis, which enables the plant to detect the infecting fungus and trigger a defence response.

Tropic Biosciences has taken a different approach. Rather than introducing genes from other banana species, the company uses gene editing to modify or regulate the banana’s own genes. By using CRISPR technology and its proprietary GEiGS platform, Tropic finetunes endogenous genetic pathways to improve resistance to diseases such as TR4 and Black Sigatoka, longer shelf life and reduced food waste. No foreign DNA is added. Under many regulatory frameworks, these bananas are therefore considered non transgenic.

Taken together, these developments mark a turning point for banana production. Where the collapse of Gros Michel once forced the industry into a genetic dead end, biotechnology now offers more than one possible path forward. The Cavendish can be protected. Genetic diversity can be restored. And perhaps one day, a banana with the taste people remember might return, this time with the resilience it previously lacked.

Europe’s banana test?

The key question is whether European consumers will ever see these bananas on supermarket shelves.

For years, the answer would have been no. Europe’s GMO regulation has been among the strictest in the world. In 2018, the Court of Justice of the European Union ruled that organisms obtained by mutagenesis are GMOs within the meaning of the GMO Directive, with limited exemptions for techniques with a long safety record. In practice, this pulled gene edited crops into the existing GMO framework. A precisely edited banana would be treated in the same way as a transgenic plant with a long, expensive and uncertain approval process. This made market access in the EU practically impossible for gene edited crops.

This situation may now be changing. In July 2023, the European Commission presented a proposal to modernise GMO regulation for plants produced using new genomic techniques. In April 2026, the Council of the EU adopted the text, although formal adoption by the European Parliament was still required before the new framework could enter into force. The proposal was built on the idea that genetically modified organisms are not all alike. A banana carrying small genetic changes that could also arise through conventional breeding is inherently less risky than a plant containing a foreign gene from an unrelated species.

That is the right scientific instinct. The method used to produce a genetic change should not be the only decisive factor. What matters is the safety profile of the final product: the trait, the exposure, the intended use and the biological context. While the proposal moves in this direction, this product-based logic is not applied consistently throughout the framework.

A narrow opening in EU GMO rules

The EU Commission proposes a two-tier system. It creates two categories for plants produced by new genomic techniques.

The first category covers plants whose genetic changes could also arise naturally or through conventional breeding. These changes are limited in number, and any genetic material must come from the plant’s own gene pool. Such plants would no longer be subject to full GMO authorisation. Instead, developers would go through a verification process to confirm that the criteria are met.

Category 1 plants would not require a GMO label. From a scientific perspective, this direction makes some sense. It recognises that a small, targeted change within a plant’s own gene pool does not necessarily carry the same risk profile as introducing genetic material from a more distant source. The approach still comes with boundaries. Certain traits, such as herbicide tolerance and production of known insecticidal substances, are excluded and organic farming remains off limits.

The second category includes gene edited plants with more complex changes. These would still be regulated as GMOs, although some procedures would be streamlined. Member States would also retain the right to prohibit cultivation of these crops at national level.

This distinction is scientifically imperfect. Safety is determined by the characteristics of the final plant, its intended use and the resulting exposure, not simply by the method used or the number of modifications. Risk must always be assessed case-by-case, and small changes can, in principle, result in greater risk than transferring genetic material from a distant species. Still, as a practical step towards a more proportionate system, the proposal is meaningful.

The motivation behind the proposal is clear. The existing GMO legislation dates back to 2001 and does not reflect the development of precise gene editing tools. At the same time, other regions including the United States, Canada and Brazil have already updated their regulatory frameworks. The European Union risks falling behind in agricultural innovation at a time when gene edited crops are increasingly being presented as part of the response to climate change, food security and the need to reduce chemical pesticide use.

Permission is not acceptance

Optimism should still be tempered. A more proportionate framework does not make market access automatic. Much will depend on how the possible new rules are implemented in practice. Qualification as a Category 1 plant will need to be demonstrated. Borderline cases may lead to uncertainty or delay. The regulatory framework may also evolve over time through delegated acts. Even if formally adopted, most provisions are expected to apply only after a 24-month transition period, meaning the new framework is unlikely to offer practical relief before around mid-2028, offering no immediate relief for projects already in development.

Practical and political challenges remain. Member States can still restrict cultivation of Category 2 crops, creating the risk of uneven acceptance across the EU. Public perception also matters. European consumers have long been cautious about biotechnology in food. Even without GMO labelling, retailers may impose their own standards.

Concerns around traceability are already being raised, especially by organic and non-GMO sectors. Detecting small, targeted genetic changes can be technically difficult or even impossible, when the same change could also occur naturally or through conventional breeding. This complicates segregation but does not in itself relate to safety. It is primarily a question of protecting business interests and consumer choice.

Intellectual property is another unresolved issue. The proposal does not ban patents on gene edited plants. Instead, it introduces transparency requirements. Patents remain a concern, particularly for small and medium-sized breeders, who may face high licensing costs or legal uncertainty. At the same time, patents play an important role in incentivising investment in plant breeding. Striking the right balance will be critical.

Plants get reform. Microbes wait.

It is also important to note what the proposal does not address.

Genetically modified microorganisms and ingredients produced using precision fermentation remain outside the scope of reform. The Commission argues that scientific knowledge in this area is more limited, a claim that is difficult to reconcile with the long history of safe use of such organisms in food production. As a result, regulatory barriers for microbial and fermentation-derived technologies remain unchanged. This matters for Europe’s wider bioeconomy. Plant biotechnology may now receive a more modern framework, while other biotechnology platforms continue to face slow and uneven regulatory pathways.

The risk of regulating yesterday’s science

In this context, the banana serves as a useful illustration. A simple story about a disease-resistant or non-browning fruit leads to a broader discussion about how innovation and regulation interact. The Commission’s proposal signals a willingness to adapt, acknowledging scientific advances while maintaining a precautionary framework. It accepts, at least partly, that all genetic technologies should not be treated as though they carry the same risk.

The direction of travel is positive, but the path forward is narrow and uncertain. Europe now has a chance to move closer to a truly product-based, science-driven approach to GMO regulation, where consumer choice is built on safety and evidence rather than inherited anxiety about the breeding method used.