Personalised medicine – or precision medicine – customises therapies for individual patients based on their genetic makeup, lifestyle and environment.
Treatment often focuses on targeting specific molecules or pathways affecting a patient’s disease. This approach is a significant departure from the traditional one-size-fits-all model of drug development.
More than 300 personalised medicines are currently available in the US market, with roughly 47% targeting rare diseases, 40% addressing cancer and 13% covering other conditions. Nine of the top 10 most expensive personalised medicines are gene therapies targeting the root cause of disorders, rather than just managing symptoms.
Patents in personalised medicine
Patenting in this emerging sphere is extremely complex. Some elements of personalised medicine and the stages in creating treatments are patentable, while others are not. There are also significant differences between regions.
“While drugs themselves are patentable as innovative therapies, methods of diagnosing diseases are generally not patentable. But the landscape is varied and companies need to tailor their patent strategies for different jurisdictions,” says Rouse Deputy CEO, Linda Chang.
For example, the EU and China both restrict patents on diagnostic and treatment methods, while the US does not disallow these.
The EU and China both allow patenting of DNA sequencing, although with restrictions. China requires that inventors demonstrate how their findings are applied in the therapeutic context. In the US, natural phenomena such as genetic material cannot be patented due to legal principles preventing patents on ‘products of nature.’
As the genetic data sets and treatment processes used in personalised medicine may not always be patentable, companies operating in this sphere often rely on trade secrets to protect intellectual property.
“Keeping certain elements as trade secrets can protect innovation from competitors while avoiding the public disclosure that patents require,” notes Chang.
Trade Marks
Personalised medicine trade marks are typically registered in Class 5 (pharmaceuticals) and Class 44 (medical services).
In China, it’s advisable to file applications early and to have backup names ready. This is because there is a high rejection rate in Class 5. The use of local language names is highly recommended in order to avoid products picking up an unapproved nickname which often gets out of the control of brand owners.
“The interplay between patents and trademark portfolios is crucial to observe. Once a drug’s patent expires, trade marks help to maintain the brand’s competitive edge. But brand owners must guard against so-called ‘genericide,’ where a trademark becomes so widely used that it loses its legal protection. This was the case with Aspirin, for example,” says Chang.
Data Privacy
The patient journey for personalised medicine is long and complex. It includes risk assessment, screening, disease identification through gene sequencing, and the formulation of a personalised treatment plan. There is also administration and monitoring of the treatment, as well as ongoing supportive care (often for a lifetime).
Each stage of the journey generates sensitive personal data, making privacy a critical concern. While an organization who collects and keeps the personal information owns them, patients typically have the rights to modify even delete their personal data.
“There is an obligation to ensure data security, particularly when sharing data across borders or with third parties. It’s mandatory to comply with regional regulations such as the EU's GDPR, HIPAA in the United States, and China’s PIPL,” emphasises Chang.
In the EU, genetic data, health information and biometric data are classified as sensitive and subject to strict protection under GDPR. Patients must provide explicit consent for data collection and be informed about how their data is used.
Under HIPAA in the United States, data such as biometric identifiers and medical records are protected. But there are no specific federal regulations regarding the transportation of this data.
China’s PIPL also includes strict rules about handling and storing biometric and medical health data, requiring biometric information and personal identifiers to be stored separately.
Copyright
If databases of genetic information are creative in their selection or presentation, these can be protected under copyright law in many jurisdictions. Factual data itself is not copyrightable
In the EU, for example, the Database Directive provides copyright protection for original databases that reflect the author’s intellectual creation. Medical institutions thus need to ensure that their databases demonstrate originality, either in how data is selected or how it’s arranged.
“Raw data is generally not protected by IP laws. But when raw data is cleaned and structured, it forms datasets that can hold commercial value and can be subject to IP protection,” says Chang.
Regulatory Issues and Ethics
Personalised medicine presents unique regulatory challenges. Genetic therapies typically require more sophisticated clinical trials, stricter data analysis and ongoing safety monitoring, as these therapies permanently alter a patient's genes.
In some cases, companion diagnostic tools – used to determine whether a patient is suitable for a personalised treatment – must undergo a joint approval process with the medicine itself.
Ethics reviews for personalised medicine are also often stricter, due to the involvement of genetic manipulation and patient-specific data. There is an emphasis on informed consent and transparent communication between medical providers and patients.
AI-generated IP in personalised medicine
Artificial intelligence and machine learning are instrumental in the field of personalised medicine, aiding drug discovery, disease diagnosis, drafting treatment plans and more. However, AI cannot typically be deemed an inventor.
“With regard to the inventorship and ownership of IP, the broad consensus view is that AI cannot be the inventor. This is because an inventor must be a natural person under the current legal framework,” says Chang.
“But what if an AI system is to create a novel idea without human intervention? Can the owner of the AI system then be the owner of a patent, for example? This remains to be seen in the future,” she concludes.
=====
