It’s no industry secret that the resources and costs associated with drug discovery, development, and efficacy are astronomical.
According to a recent study from Tufts Center for the Study of Drug Development, one new therapy – from discovery through Phase III clinical trials – costs approximately $2.9 billion to bring to market, with an average development time span of 10 years. Further, a drug substance in Phase I Discovery has only an 8 percent chance of being approved and brought to market. According to Drug Discovery Trends annual survey of top 50 revenue producing pharmaceutical products, most of these drugs were released to the market in the last five years, and they earned an average of $7.6 Bn in annual sales.
It’s vital to recognize the inherent costs and challenges in drug development to truly appreciate the potential transformative power of data science and data collaboration in the field of pharmaceuticals.
Using data science to speed product development not only reduces development costs and increases success probability, but every month that a new drug therapy is delivered faster to the market has a potential $436 million revenue, based on the 2023 top 50 average.
The Covid-19 pandemic showcased a glimpse into what is possible when pharmaceutical companies collaborate to accelerate the development of new drug therapies. Within mere months, the global community transitioned from lacking effective treatments for Covid-19 to having multiple vaccines and therapies at their disposal. This swift progress was largely due to the unprecedented levels of data sharing and collaboration. Both industry competitors in drug substance research and governmental regulatory agencies played pivotal roles, with the latter assessing the efficacy, safety, and quality of these new therapies through notably condensed clinical trial timelines.
Undoubtedly, the billions of dollars in grants and pre-purchased immunizations extended by governments played an instrumental role in mitigating the risks for the pharmaceutical industry, enabling them to rapidly introduce Covid-19 drug therapies to the world. While discussions on financial incentives like grants or tax relief – designed to promote pre-competitive collaboration and data sharing among pharmaceutical companies without dampening competition – are noteworthy, they are more apt for political discourse and are not the centerpiece for this discussion. We must focus instead on the repeatable lesson learned: using data science and collaboration to advance medical science.
Although pre-competitive collaboration in the pharmaceutical industry is not a new term, the open exchange of models and data from studies, experiments, and patient outcomes enabled researchers to focus on promising substances faster, gain “buy-in” from governments, and shorten the time to delivery of successful Covid-19 vaccines and treatments. Although the rapid pace of Covid-19 drug therapy development resulted from more ad hoc data sharing, a long-term data discovery and data sharing initiative to expedite therapies for cancers, Alzheimer’s, and other chronic conditions necessitates two key industry shifts. First, companies must establish data platforms that enable them to effectively and quickly discover data that is relevant to a target disease. Second, the pharmaceutical industry, as a whole, requires a change in culture and mindset to improve the level of secure, efficient data sharing under pre-competitive collaboration agreements while maintaining a competitive advantage. Having the right data platform will be pivotal and significantly ease this essential industry culture shift.
The data platform needed to enable this enhanced data discovery and collaboration is underpinned by three vital components:
- Industry standard ontologies: The platform first requires industry standard ontologies that describe and enrich information across various data silos. An example of such an effort is the Pistoia Alliance Identification of Medicinal Products ontology, which encodes the International Organization for Standardization (ISO) 11238 and related standards. Such ontologies are developed using pre-competitive collaboration agreements across pharmaceutical companies. These ontologies enable the data platform to label data uniformly, thus linking information stored in diverse formats across different silos and even organizations.
- Semantic alignment: Leveraging the ontologies, the second component of the data platform is semantic alignment. This involves using artificial intelligence and machine learning algorithms to automate the labeling of data in disparate formats across various sources with industry standard terminology defined in the ontologies. A clear advantage of semantic alignment is that it accelerates data discovery. For instance, prior effective treatments for SARS became a key basis for government investment in Covid-19 research. Concurrently, the platform prioritizes trust, security, and accessibility with technology features that control and authenticate access to each piece of data.
- Blockchain technology: The final component is a non-reputable blockchain technology that ensures complete data credibility. This includes providing mathematical evidence verifying the legitimacy of every piece of shared or persisted data. With time stamps for every unit of data and stringent security measures, the data can effectively “defend itself” across domains. This ensures that researchers can collaborate with peers and regulatory officials throughout the drug development process with full confidence in the data’s integrity.
The future of pharmaceutical innovation hinges on enhanced data discovery and collaboration. Through data sharing and collective effort, the pharmaceutical industry can accelerate the development of new therapies. This not only elevates the quality of life for global citizens but also boosts the financial performance of drug portfolios for individual companies.
Photo: manop1984, Getty Images