The main innovation trends within the life sciences are: processes and technologies augmented by artificial intelligence (AI); cell and gene therapy (CGT); and clustered regularly interspaced short palindromic repeats (CRISPR-) Cas technologies.
The coronavirus disease (COVID-19) pandemic has shown that shorter development timelines are possible for life sciences innovations.
Successful technology transfer in the field of life sciences is a complex challenge requiring the input of many parties: scientists and/or inventors, the technology transfer office/department professionals, patent attorneys, consultants, etc. The objective of this report is to provide a reference document for all essential stakeholders involved in patenting innovation within the life sciences sector, particularly within the technical fields of biotechnology and pharmaceuticals. Given the broad target audience, this document seeks to identify and address key themes applicable to all.
The document can also be considered as a tool to assist innovation stakeholders in steering research and technology development towards patentable outcomes, and to support business decisions that will be made as patent prosecution progresses. Moreover, the document emphasizes the pre-grant patent practice, specifically the characterization of potentially patentable research outcomes and inventions and their use as a basis for the patent application filing, and the patent prosecution that follows.
While parts of this report provides a global perspective, country- or region-specific patenting diagnostic approaches are examined for Europe, Japan and the United States of America.
To support the comparative analysis presented in this report, the study also incorporated insights from patent attorneys and technology transfer professionals in Europe, Japan and the United States of America. These perspectives were gathered through a series of expert interviews designed to complement the desk study with practical, day-to-day experiences from professionals working in these jurisdictions.
The interviews were conducted using two structured questionnaires tailored respectively to patent attorneys and to technology transfer professionals. The information collected informed the analysis presented throughout.
To encourage open and candid contributions, participants were assured that their responses would remain anonymous. Consequently, individual responses and underlying data are not disclosed and the insights are presented in aggregated or synthesized form. In total, the survey received 28 responses: 20 from life sciences patent attorneys and 8 from technology transfer professionals.
Trends within innovation in life sciences
The innovation opportunities within the life sciences field are limitless. The following three main trends that we identified through market research and literature searches are outlined below.
AI-enhanced processes and/or technology
Both AI and machine learning (ML) have become increasingly prevalent terms in the vocabulary of innovators across a range of industry sectors. AI and ML are closely related and connected terms, with ML being a subset of AI.
AI refers to the capability conferred to a computer system that allows it to mimic cognitive functions including learning and problem-solving, utilizing both mathematics and logic to process new information and make decisions. ML is the application of AI, implementing mathematical models of data that enable a computer to learn without the requirement of direct instruction; the computer therefore uses experience to learn and improve. In other words, AI allows the system to “mimic the thinking process” and perform tasks independently, and ML confers the ability to develop this intelligence.
AI for clinical trials
AI offers those within the life sciences sector wide and varied opportunities. Currently, an area of drug development that has begun to benefit from AI and ML is clinical trial design, operation and analysis. AI technologies such as ML can provide benefits from the very beginning of the clinical trial process, with data gathered during pre-clinical studies being analyzed to estimate the lowest clinically effective dosage to be used as part of the wider dosing regimen.
AI also offers the ability to optimize patient recruitment and thereby increase the success of a clinical trial. Clinical trials have specific recruitment requirements with inclusion and exclusion criteria that patients must meet. Identifying suitable patients poses a significant challenge and is the major cause of delays to trials; recruitment can take up to one-third of trial duration, and 86 percent of trials fail to meet the established recruitment timeline.
It is hoped that AI, when used to examine the vast data available in electronic medical records (EMR), may offer a solution to optimizing patient recruitment. The United States Food and Drug Administration (FDA) has identified the following criteria that AI could adopt:
reducing population heterogeneity
choosing patients most likely to have a measurable clinical endpoint
identifying a population more capable of responding to the treatment.
AI can be used to automatically analyze EMR and identify patients that meet the eligibility criteria of various clinical trials, recommending these matches to both patients and sponsors. A pilot of such a system was successfully conducted by Mayo Clinic, IBM Watson Health and Novartis.
The monitoring of ongoing clinical trials is an essential aspect of quality control mandated by regulatory bodies to ensure both the safety of participants and the integrity of the trial, and is often outsourced to specialist contract research organizations. This is time-consuming and costly, with trial site monitoring (including management of collected data and performance reviews) representing one of the top three costs across all study phases (an estimated 9–14 percent of total costs).
AI tools for the monitoring of trial site performance and data quality may offer a framework to increase efficiency by predicting and detecting risks in real-time, consistent with the lean philosophy. By applying AI tools, an advanced form of risk-based monitoring may be implemented; improving risk prediction as a result of real-time analysis of clinical data being combined with predictive analytics (based upon available data from similar trials) can transform the ability of a sponsor to protect patients, reduce trial duration and lower costs.
Additionally, AI could assist in the monitoring of clinical trial participants. For successful progression through a clinical trial, it is essential that patients adhere to the trial procedures and rules throughout, and that data-points used to assess the impact of the therapy are collected both reliably and efficiently. Patient dropout, experienced by 85 percent of clinical trials, is an area of concern; in affected trials, patient dropout averages about 30 percent. Most of these dropouts are the result of a lack of adherence to the trial protocol, and have a significant impact upon costs associated with the trial. A linear increase in non-adherence rates results in an exponential increase in additional patients required to conserve the essential statistical power of the results.
Reducing the adherence burden and making endpoint detection more efficient are essential to minimizing dropout. It is hoped that AI techniques, in combination with wearable technology, can offer an approach that enables mobile, real-time and personalized patient monitoring systems. Underlying deep-learning models can analyze and periodically update the measurement data tailored to the patient, and can be adaptive to changes in disease expression and patient behavior. AI can also assign a dropout risk to each patient by detecting a lack of adherence to the study protocol, alerting the trial sponsor who can reach out to the patient to assist them with protocol adherence.
AI for drug development
Advances in the high-throughput screening of compounds have significantly enhanced the discovery of potential new drugs, but the process is still laborious and time-consuming with low levels of success. It is essential that progress is made in enhancing the chances of success in the early stages of drug development, as this will enable therapeutics to be developed both faster and cheaper. AI techniques also have the ability to discover new compound candidates.
The investment bank Morgan Stanley has estimated that a 20–40 percent reduction in preclinical development costs may allow for the additional development of four to eight novel molecules across a subset of United States (US) biotech companies. This would represent a potential 15 percent increase in approved novel therapies, generating additional sources of revenue and benefiting a wider patient population.
It is hypothesized that these AI drug development platforms could realize significant revenue growth and investment within the biopharmaceutical sector. Indeed, in the period 2015–2020, equity funding of AI within the healthcare sector rapidly grew worldwide.
When combined with technological advances that have allowed for the collection of large data sources combining genomic data, patient medical records and advancements in medical imaging in projects such as United Kingdom (UK) Biobank, it is hoped that AI platforms will be able to utilize these data to identify possible therapeutic targets in the drug development phase. AI may also be able to assist in drug design by predicting the three-dimensional (3D) structure of receptors and drug–receptor interactions. This is referred to as de novo drug design (DNDD), a computational approach that results in the generation of novel molecular structures.
Advantages of DNDD include the ability to examine a broader chemical space, designing compounds that are likely to constitute novel intellectual property (IP) and identifying novel therapies in a shorter period than traditionally achieved, optimizing drug design while reducing cost and time to market.
AI as an inventor
AI can clearly advance research within the life sciences. However, what of the debate surrounding AI as an inventor? Could AI be listed as the inventor on patent filings? It is suspected that patent filings were secured on AI-generated inventions as long ago as during the 1980s; in such instances, it is further speculated that filers were advised to list a human as the primary inventor.
In August 2019, the World Intellectual Property Organization (WIPO) announced two Patent Cooperation Treaty (PCT) filings for AI-generated inventions in which no human was an inventor. These filings listed the AI named Device for the Autonomous Bootstrapping of Unified Sentience (DABUS), the inventor and the owner of the AI, as the patent applicant and prospective owner of any granted patents. Both the European Patent Office (EPO) and UK Intellectual Property Office (UKIPO) had evaluated the patent filings, and found that they met with the prerequisites of patentability. However, both applications were rejected by UKIPO as DABUS is not a person and therefore not eligible to be listed as an inventor under UK patent law. Appeals were dismissed by both the High Court and Court of Appeal, with the Supreme Court hearing the case in March 2023.
Because there is no substantive requirement within the Patents Act that an inventor must be human but only the “actual deviser” of the invention, representatives of Dr Stephen Thaler (the creator of DABUS who submitted the patent filings) argued that specifying “no person” as the inventor of AI-generated inventions would be in keeping with the purposive interpretation of the Patents Act.
Other countries in which the patents in question were filed have also debated the issue of AI as an inventor, with all jurisdictions that operate an examination process refusing to grant a patent.
CGT
Healthcare is currently undergoing a revolution in our approach to the treatment of disease: personalized medicine represents a shift in our ability to tailor treatment to individual needs. Using next-generation sequencing, we can now sequence an entire human genome for a cost of less than USD 1,000, and our ability to extrapolate meaningful data and conclusions from this information is ever increasing as a result of advancements in our AI and ML capabilities and projects (e.g., UK Biobank and 100,000 Genomes).
These findings allow us to better understand the physiology of disease, revealing novel therapeutic targets for a new generation of advanced therapy medicinal products (ATMPs), also known as CGTs, that promise to prove effective against rare diseases such as retinitis pigmentosa and some blood cancers such as leukaemia and lymphoma, that have proved to be difficult, if not impossible, to treat.
Definitions and regulations of ATMPs and CGTs vary by jurisdiction. Within the United States of America and the European Union (EU), the FDA and the European Medicines Agency (EMA) regulate ATMPs and CGTs under the framework governing biological products.
The first ATMP to be approved by the EMA was ChondroCelect in 2009 (a tissue-engineered product targeted at cartilage defects), and the first to be approved by the FDA was PROVENGE™ in 2010 (somatic cell therapy for prostate cancer). As of January 2023, the EMA has approved 25 ATMPs (including those that have subsequently had market access withdrawn). Eleven of these approvals occurred from 2020, highlighting the increasing numbers of ATMPs successfully making it to market. As of July 2025, the FDA has approved 45 CGTs. In both markets the majority of approved therapies are gene therapies, defined here as one containing or consisting of recombinant nucleic acid. Gene therapies are often intended to be administered for the purpose of introducing functional genes into cells to replace missing or defective genes to correct a genetic disorder. However, with advances in precision gene editing, it is likely that this definition will continue to evolve along with the regulatory provisions that govern their development and use.
Immunotherapies such as chimeric antigen receptor T (CAR-T) cells are responsible for a significant amount of the activity within the field of gene therapy. It may not seem obvious to consider CAR-T therapies as gene therapies; however, they are considered as such because the T cells are modified with a recombinant nucleic acid encoding a chimeric antigen receptor. It is the CAR that redirects the action of the T cell through binding to target antigens, resulting in the activation of the T cell. Since their initial development in 1989, multiple generations of CAR design have improved CAR-T therapies.
CAR-T therapies can be either autologous (obtained from the same individual) or allogeneic (sourced from a donor) by design. The benefits of having an off-the-shelf allogeneic therapy is that it may (i) reduce the cost and (ii) improve access to the therapy as a result of economies of scale and not requiring specialized local expertise. Researchers have also considered alternative lymphocytes to T cells such as natural killer cells (NK), resulting in CAR-NK therapies. It is hoped that these may provide a truly allogeneic therapy as they do not have to be donor matched.
CAR-T immunotherapies have had significant success in the treatment of B cell leukemia and lymphoma but have shown limited therapeutic effect in solid tumors and hematological malignancies. CAR-T therapies are limited in their ability to infiltrate solid tumors, suffer from antigen escape and are highly susceptible to environmental conditions within the tumor microenvironment.
The ATMP industry is in clear need of IP guidance and support, including professional IP services that can clearly articulate to innovators what can be protected, the value of protection and the effective exploitation of IP assets related to all aspects of the ATMP industry.
CRISPR/Cas
One of the most significant discoveries in modern times, CRISPR and their associated Cas proteins have been evolved by prokaryotes to provide an innate immune mechanism to combat attack from bacteriophages. The CRISPR-Cas mechanism of immunity incorporates short fragments of DNA excised from the invading bacteriophage within the host genome. Post-transcription, it is these fragments that guide protein complexes to foreign DNA sequences targeted for degradation.
The CRISPR locus is preceded by an adenine- and thymine-rich sequence and flanked by genes encoding the Cas proteins. CRISPR-Cas systems can be divided into two primary classes, further divided into six types and again into various subtypes, with the classification determined by the action of the Cas proteins in the cleavage of foreign DNA. The effector module comprises a multi-protein complex in class 1 (types I, III and IV) systems, as opposed to a single effector protein in class 2 (types II, V and VI) systems.
The mechanism of CRISPR-Cas defense can be broken down into three stages: adaptation, crRNA (CRISPR RNA) processing and interference.
In adaptation, a short DNA sequence is extracted from the genome of the intruding bacteriophage. This sequence is then incorporated into a genomic locus of the host referred to as CRISPR, consisting of short repeat elements interspaced by unique sequences referred to as spacers. The spacers are the extracted short DNA sequences originating from the invading bacteriophage. The mechanism of integration of spacers differs between the different types.
The next stage, crRNA processing or biogenesis, details the transcription of the CRISPR locus generating a crRNA precursor (pre-crRNA) that undergoes processing to produce mature crRNA, each consisting of a spacer sequence flanked by a partial repeat sequence.
In the final interference or targeting stage, this crRNA subunit is used to identify target DNA sequences with which it forms an R-loop as it hybridizes. This is followed by degradation of the target.
Since the discovery of the mechanism of CRISPR-Cas, modifications of the system as well as identification of a broader spectrum of Cas proteins have positioned CRISPR-Cas as a fundamental tool of modern biological research. Beginning with its utilization as a gene-editing tool, application areas have broadened to encompass gene regulation, epigenetic editing, chromatin engineering and live-cell chromatin imaging. Future directions include utilization of CRISPR-Cas systems in gene therapy for the benefit of human health.
Other notable life sciences trends
Other identified life sciences trends include an increase in the use of wearable devices,
Life sciences patent attorneys and technology transfer professionals who participated in our survey reported a perceived increase in life sciences patent filings, ranked from highest to lowest as follows: cell and gene therapies, AI-driven innovations, vaccine technologies, diagnostic biomarkers, monoclonal antibodies, and cell-free technologies.
The effects of COVID-19 on life sciences
The COVID-19 pandemic was a testing time globally, with lockdowns imposed on society, movement restricted and global industry severely impacted. In this section we consider the effects of the pandemic in the field of life sciences and its lasting impacts on the industry.
In a survey circulated among life scientists based in Canada, France, Germany, Italy, Spain, Turkey, the United Kingdom of Great Britain and Northern Ireland, and the United States of America, 881 responses were collected to gauge the impact of COVID-19 upon their daily lives. Of these responses, 62 percent identified as experimentalists, 34 percent as computational biologists and 4 percent as administrative support personnel. The responses showed that 77 percent reported a complete shutdown of their institute.
The effect of COVID-19 was not limited to researchers, but also severely impacted the supply chains that support them. The response from regulatory bodies such as the National Institutes of Health was to create the Accelerating COVID-19 Therapeutic Interventions and Vaccines program to increase collaboration between actors within the life science industry. The FDA and the EMA both collaborated for the purpose of sharing clinical trial data to support the rapid approval of therapies and industry bodies. This led to interesting debates on IP surrounding the innovations and vaccine designs; indeed, many companies sought patent protection although they were primarily incentivized in the development by large contracts for the sale of vaccines. Others, such as Oxford University, developed a vaccine that they then patented and licensed to AstraZeneca on the condition that it would be broadly licensed and sold at cost during the pandemic.
A different perspective on the COVID-19 pandemic is that it demonstrated that shorter timelines for bringing vaccine and drug candidates to market are possible.
The effects of COVID-19 upon the life sciences industry were varied. Some benefited, especially those who were able to pivot their research and activity towards development of vaccines, testing kits, medical devices and personal protective equipment (PPE). Others were severely hampered in their development of therapeutics for indications separate from COVID-19, with some attempting to repurpose existing drug candidates.
In the aftermath of the pandemic, patenting trends observed by both lifes ciences patent attorneys and technology transfer professionals who participated in our survey indicate a significant impact of COVID-19. Specifically, 80 percent of life sciences patent attorneys rated the impact at 7 or higher (on a scale where 10 represents a very high impact and 1 represents no impact), while 88 percent of technology transfer professionals also rated the impact at 7 or above on the same scale.