The quiet revolution transforming the future of drug development

Ray Chohan co-founder and vice president of new ventures at innovation intelligence expert Patsnap looks at technology’s ability to expedite the analysis of data in pharma and how it can help bring about new therapies for patients. 

The central role of AI and machine learning in drug and vaccine development is now widely recognised within the pharma industry. What’s less well known is technology’s ability to expedite the time taken to gather valuable data and explore both patent and intellectual property rights relating to particular molecular structures.  

Prior to the outbreak of Covid-19, the time taken to develop a new drug or vaccine could take upwards of a decade. Granted, peer reviews, the roll-out of field testing and international approvals were key barriers to speeding up the process, but what is not widely appreciated is the quiet revolution that has taken place in the area of early-stage drug research - specifically where patents and IP are concerned. Until the past few years, this was typically a very manual process that took highly trained researchers months to undertake. 

This front-end development phase of drug development can be hugely time consuming, requiring many hours sifting through patent applications to see if a particular molecular structure has already been registered. Manually searching out relevant research and peer reviewed papers to see whether previous studies provide credence to current investigations is another exercise that can soak up weeks of time. And of course, the added possibility of human error – such as misfiling of new found data – can have a very significant impact on the early stages of progress. In short – this administrative behemoth was crying out to be automated. We simply needed vision and technology to fix it.

Covid acted as a catalyst – driving pharmaceutical businesses to adopt new processes and methods that were previously not on the boardroom agenda. And the impact of this new way of thinking has been significant. While this first stage of research used to take between six to nine months to achieve, a combination of artificial intelligence (AI) and software analytics can bring this time down significantly. 

Working in life sciences since 2014, we have developed a business model based on machine learning that can interrogate medical documentation from all over the globe – including molecular structures, registered patents, research papers and technical documents. From this massive amount of data, we enable deep analysis of a particular chemical area. 

Just a couple of years ago, the early phases of molecular investigation would entail many hours of research, sifting through published data, field trials and research documents looking for relevant information. Unearthing a clinical study that supported your initial findings was like finding a needle in a haystack. Very few pharma businesses are now using manpower to undertake these vital first steps in drug development. Instead, most are using proprietary software such as ours that collates data from thousands of sources and presents it as usable management information. It’s this scale of automation that is the prime driving force behind the new revolution in drug development. 

While the pharma sector is now largely on board with the adoption of this type of analytical technology, it has only been in the past year or so – since the outbreak of Covid-19 – that the medium cap and large corporations have all put technology investment at the top of their budget lists. Indeed, according to data from the Pistoia Alliance (a global life sciences member organisation) AI, machine learning and Blockchain are all top priority areas for investment over the next year. This trend is echoed by the fact that earlier this year Patsnap raised $300m for future development of its AI-driven R&D and IP analytics capability. 

The next decade is all about open collaboration. In our work with pharma businesses, we have seen how our platform has created a bridge between ‘drug development’ and ‘business development’ departments. What was often previously a chasm has now been filled in – using valuable information to create a shared dialogue. In our experience, poor internal communication has been one of the significant brakes on faster drug development – and it is hugely encouraging to see new technologies like ours enabling genuine collaboration between bench-level departments, marketing consultants and c-suite managers. 

Looking ahead to the next decade, we see pharma and biotech companies acting more like software companies did back in the nineties. We are already witnessing that shift, which is being driven by new technology adoption and the emergence of platforms.  

Big pharma corporations are eyeing up the benefits of a more collaborative, open style which will see the opening up of IP access rights, thus allowing smaller enterprises to use their platforms and knowledge to develop and deliver new drug therapies – in much the same way that the various Covid-19 vaccines were developed. 

This model will happen in biotech and pharma over the coming years. We are entering a new era, the likes of which we have never seen. Coronavirus has been one of the world’s most serious pandemics – however it has been a catalyst to change that could help bring about tailored drugs, personalised medicine and reduced side effects to people across the globe. 

And in some part, it is the ability of machine learning that has contributed to this exciting prospect. Technology’s ability to grind though millions of pages of data, in hours not months, is helping to realise the dream of effective vaccines and truly bespoke drug therapies.