Innovation and collaboration: key catalysts for scientific break-throughs
During my time at Illumina I had a realisation: restricted access to data, and an industry-wide aversion to collaboration, was delaying and in some instances preventing the discovery of treatments and cures for cancer, as well as consistently hampering progress and innovation across biomedical research.
Even publicly available data for research, which is growing at an exponential rate, is very fragmented. So I was determined to do something about it and support researchers in searching for, and using, human genomic data.
I started a company and launched a freemium-access platform to support researchers across academia and industry, and in recent years there have been many other initiatives in both of those sectors to make sense, and the best utilisation, of the growing amounts of data in biomedical research, which is fantastic to see.
In a time of diminished success rates and ever more expensive drug-discovery pipelines we now need pharma, biotech and drug discovery researchers to truly embrace innovative data-driven tools and collaborative practices. Deliberate adoption of the data-driven order of the day will be essential for these solutions to succeed and provide genuine catalysts for scientific break-through towards better outcomes for patients.
One of the sparks for better and faster drug discovery will come from the combination of artificial Intelligence (AI) with new – or more widely available – types of data.
Artificial Intelligence and drug discovery
With the potential for AI to speed up research and drug discovery, operators need enough data to input to the AI in the first instance. Hugh Harvey, clinical artificial intelligence and predictive analysis researcher at King’s College London, puts it this way: “Clinical artificial intelligence has been ‘happening’ since the 1970s and the digital revolution – when it finally arrives – will come, slowly, surely and powerfully. The reason for the apparent snail’s pace is simple; feeding any artificial intelligence with enough knowledge to deal with the infinity of nuances, intricacies and quirks of the human body (and all that may weaken it) requires oceans of data.”
One company driving this forward is Benevolent AI. Its AI-driven system analyses biomedical research data, including papers and abstracts, to uncover relationships and links that already exist but just haven’t been discovered yet, because humans do not have the processing power to accomplish this level of analysis. Previously undiscovered treatments will be identified from the untapped wealth of historical research and scientific papers already in existence.
One of the platforms we have created – to broker cancer models – will become more and more useful as contract research organisations (CROs) provide better descriptions and more data characterisation of their products and services, enabling drug discovery researchers to do a comprehensive search to find the right model in the data landscape. Similarly, organisations like Benevolent AI, who are relying on in-silico drug discovery, require a constant input of new and specific data to uncover new targets for drug development.
We also see data collaborations emerge across the industry through initiatives such as the OpenTargets consortium, which uses genomic data to bring together both large pharmaceutical companies and academic institutions with the aim of accelerating pre-competitive drug discovery and target validation.
To succeed, players from across the drug discovery pipeline will need to be willing to share their data, and to ensure it is appropriately categorised, for everyone’s benefit.
Genomic data and cancer research
By determining the genetic characteristics of an individual’s cancer, clinicians may be able to decide whether a particular treatment may be effective. However, for this ‘future of cancer treatment’ to become properly established, pharma and biotech companies need access to more individuals’ genomic data. This is fast becoming the last hope of many patients, yet, while direct-to-consumer testing is becoming ever-more popular in the West, there has historically been no single repository or platform for researchers to locate or access human genomic data.
The late Dame Tessa Jowell drew attention to the Eliminate Cancer Initiative’s new Universal Cancer Databank project, which has been established for this very reason; she became one of an increasing number of individuals to donate their personal genomic data for research purposes.
Establishing better systems to enable and encourage sharing of data in this way, within responsible and ethical frameworks and with appropriate consents in place, is key. The more researchers support and draw on these repositories, using the data to discover and refine treatments, the more individuals and public healthcare systems will be motivated to donate genomic data.
Going full circle, this in turn supports pharma and biotech R&D teams by establishing a greater resource from which to draw.
Clinical trial data
As a willingness to share data from individuals increases, the tools to enable sharing of data related to clinical trials need to be developed, too.
There is the obvious benefit of accessing results from trials which succeed, but data from trials which fail is also valuable and could help pharma and biotech companies to make more accurate decisions on development strategies for different disease areas.
Several collaborations, including the not-for-profit OpenTrials.net, are aiming to fill in these missing pieces of the clinical trials puzzle, but there is still a significant way to go before all trials are reported equally and publicly.
In the meantime, with healthcare and pharmaceutical trials increasingly using technology such as wearables and mobile apps to track patients’ health throughout the duration of a trial, much larger amounts of data are being generated, all of which need to be analysed to maximise the benefits of continuous monitoring.
Platforms that enable researchers to visualise the volume of data created during trials are essential in the big data era, and Clarivate’s Cortellis already brings together data from across the drug and device innovation lifecycle in this way, giving valuable insights to CROs.
For all SMEs and start-ups that invest in tackling sluggish pipelines to drug discovery, the reality is that innovation can only deliver so much. Pharma and biotech companies need to play their part too, by embracing innovation, and collaboration, for the benefit of all involved with, and waiting for, the discovery of new drugs.
Fiona Nielsen is CEO and founder of Repositive