The rise of techbio and its IP needs: IP strategies for data-driven innovation
In this article we will consider the way the techbio sector is developing and the strategies for protecting arising intellectual property, which may differ from strategies used in the traditional biotechnology and technology sectors.
Techbio is defined as the interface between biotech and tech, and focusses on using cutting-edge techniques from both sectors to drive innovation. During recent years there has been a surge of interest in this evolving sector. Broadly speaking, the techbio sector can be divided into two main areas. The first of these is the use of data to drive traditional innovation, and is largely driven by existing biotechnology, pharmaceutical and life science companies. We will focus on this area in this article, and will examine the arising IP needs compared to traditional approaches to innovation in the biotech space. The second area is the development of new platforms for driving innovation, and is largely driven by tech companies. We will explore this area further in a later article since the IP challenges of this part of the sector are related but distinct from those facing biotech companies.
Traditionally, innovation in the biotech and life sciences sectors has relied heavily on wet data. Whilst this approach provides a robust system, it places a heavy burden upon the early stages of research: a time when funds may be scarce and uncertainty levels are high. Using data-driven solutions may allow companies to focus resources upon projects having a greater chance of success, driving pipelines forward in a cost effective manner.
The traditional approach
Life sciences and biotech companies have long placed a heavy emphasis on the importance of wet data. In vitro studies are commonly followed by testing in an animal model, and the process culminates in an expensive and lengthy clinical trial. There are many advantages to this approach, including a deep understanding of the activity of a candidate molecule and an acknowledgment of the relevant safety considerations. However, this wet experiment focussed approach requires a large investment of both time and money at an early stage of development when the outcome, and even the aims, of a project can be far from clear.
Take, for example, the development of a small molecule pharmaceutical. Initial experiments are likely to be devised on the basis of an understanding from the literature of the causes of a particular disorder or the workings of a particular pathway. From this premise a library of small molecule candidates is chosen for initial screening, probably based upon structural similarity to a component of a pathway thought to be involved in a particular disorder. These initial wet experiments are likely to be performed in vitro, with a large proportion of the tested compounds found to be inactive.
It is only after extensive in vitro testing that the most promising candidate compounds are likely to move to testing in an animal model. Animal models can be extremely useful research tools. However, by definition they are based upon the biology of an organism which is not human, which is itself a challenge for researchers looking to devise a pharmaceutical for human use. Further, it is known that many disorders do not have an adequate animal model, hampering the development of treatments for these disorders.
Finally, once data in an appropriate animal model has indicated a reasonable chance of success for a candidate compound, clinical trials are required in order to demonstrate a reasonable toxicology profile in a healthy population, as well as a suitable therapeutic efficacy. This is a lengthy and expensive process in itself, but it also comes at the end of a process which has already taken many years, a huge amount of investment and has seen a large number of candidate compounds fall by the wayside.
There will always be a role for wet experiments and clinical trials in the biotech and life sciences sectors, but what if these expensive stages of testing could be focussed upon candidate compounds known to have a greater chance of success? This is where data-driven solutions in the techbio space can play a pivotal role.
The role of data-driven solutions
Techbio solutions offer a data-driven way in which to focus research upon candidate compounds having an increased chance of success. Taking the small molecule pharmaceutical example introduced above, data-driven approaches can reduce, or remove, the need for initial wet experiments. For example, a machine learning model trained based on a library of known chemical structures labelled with known therapeutic effects can be used to predict which other chemical structures are candidate compounds for the treatment of a certain medical condition.
Selecting an appropriate pathway through which a particular disorder can be treated is a challenging but vital initial stage in the traditional approach to pharmaceutical development. Performing this step manually, using wet experiments, requires an in depth knowledge of the relevant field, but also an element of good fortune to select a premise which has the potential to yield relevant candidate compounds. Using machine learning approaches to analyse the relevant data can reduce the need for good fortune, allowing the assimilation of a much larger data set and the arrival at a premise that is a more accurate reflection of the clinical situation and therefore more likely to succeed.
The use of data-driven solutions within a biotech process does not need to end once a relevant premise or pathway has been established. Rather, computer modelling can be used to determine the candidate compounds most likely to interact at an appropriate point in the selected pathway. This has a greater chance of success than basing decisions on the structural similarity of a candidate compound to a component of a pathway alone because it allows additional parameters such as steric interactions and affinity to be accurately modelled.
Taken together, these and other techbio approaches to pharmaceutical innovation can reduce the risks associated with early stage drug development, reducing upfront costs and allowing companies to take viable candidates into the clinic at a fraction of the cost of candidates arrived at through traditional approaches alone, for which the candidate atrophy rate will have been much higher.
Within the traditional approach to pharmaceutical development there are a number of possibilities for arising IP, including patent protection, know-how and trade secrets.
Highly prized candidate compounds are almost always patent protected and these patents, and associated Supplementary Protection Certificates (SPCs), can be extremely valuable. Primary patent protection is likely to focus upon the structure of a candidate compound, which may be defined chemically or through the nucleic acid or amino acid sequence of a biologic in a composition of matter patent. Follow on patents are also available for novel and inventive formulations, second generation molecules, methods of treatment, and dosage regimens, amongst other things.
Ancillary inventions may relate to proprietary assays and laboratory techniques involved in the selection of candidate compounds, but these do not form the core assets of a biotech company and are often protected as trade secrets or kept as know-how rather than being the subject of patent protection. An evolved IP strategy will include preferred options for protecting this innovation whilst focussing costs upon the core assets of the company.
It is likely that the IP position for companies using techbio solutions within traditional methodologies will be similar to a traditional IP strategy approach, with patent protection sought for candidate compounds and follow on inventions, and trade secrets and know-how used to provide additional protection for associated innovations.
Taking the pharmaceutical development example introduced earlier, small molecule candidates selected using data-driven solutions will initially be protected under a composition of matter patent, with additional patents available for novel and inventive formulations, second generation molecules, methods of treatment, and dosage regimens, irrespective of whether these innovations were arrived at using data-driven solutions or traditional wet experimental techniques. As for companies developing candidate compounds using traditional approaches, the primary focus, and therefore value, surrounding this area of the techbio sector resides in the compounds themselves.
Methodologies surrounding the generation of candidate compounds are likely to be of secondary importance to biotech companies as they look to assimilate data-driven solutions into their standard experimental toolkit. These will often therefore be protected as know-how or confidential information, at least in the first instance. Biotech companies relying on data-driven approaches should also take care to protect their data sets, which may have taken considerable investment to develop and can be of significant commercial value. Although unregistered IP rights such as database rights may be available in certain countries, it is also advisable to implement IT security measures for protecting access to the data, and review contractual provisions in contracts with employees, contractors, commercial partners and customers restricting use and dissemination of such data sets.
In contrast, techbio companies developing new platforms for driving innovation will have such methodologies at the core of their business and will increasingly look to protect these platforms per se rather than merely the products thereof. Protection for the data processing platforms may also be of interest for companies developing diagnostic tools, for example, a machine learning model which processes biomarkers or genetic sequence data from a patient to generate a prediction of whether the patient suffers from a particular health condition. We will focus on these aspects of the techbio sector in a subsequent article.
Related webinar: biotech case law
Our most recent biotech webinar is now available to view.Read more