Deloitte predicts R&D productivity slump: Is AI the answer to profitable drug discovery?

Producing life saving or life changing drugs requires significant time and resources to be invested, in the hopes of recouping high returns if a drug makes it to market. However, a recent report by Deloitte shows that projected returns on investment into pharmaceutical R&D have fallen to 1.9%. This is the lowest figure in nine years, with projections steadily dropping since reaching 10.1% in 2010.

Increasing costs for development and low returns have created a difficult predicament for the pharmaceutical industry. In 2018, the cost to bring an asset to market, including the cost of failure, increased to a record high of $2,168m, almost double the $1,188m recorded in 2010. Meanwhile, forecast peak sales per asset have more than halved since 2010 to only $407m.

Speaking to these results, Colin Terry, consulting partner for European life sciences R&D at Deloitte, said “the results signal a time for substantive change in the pharmaceutical industry. Despite the launch of many successful products, growing development costs and regulatory constraints are making it more difficult than ever for companies to redeem their R&D investment”. He continued that cutting R&D cycle time and costs will be vital for success. His firm recommendation is that “the industry act now to embrace new technologies and seek out talent with the right skill sets to challenge the status quo”.

These stark results may be a catalyst for the industry to move away from their tried and trusted methods towards new ways of operating. As the report makes clear, “the industry has yet to unlock the full potential of truly breakthrough R&D capabilities”. To do so will require a complete digital transformation focused on maximizing R&D productivity.

RELATED: Register now for Smart Labs Digital to discover how to increase R&D efficiency with AI and new technology

Using artificial and augmented intelligence to maximize R&D returns

There are a number of ways technology can be utilized across research and development to improve productivity and reduce costs and cycle times.

Using robotic process automation could help to automate repetitive tasks so they can be done faster, cheaper and more accurately. Not only is this cost-effective, it also allows technical resource to be focused on more engaging, high-value tasks, increasing productivity.

Natural language processing and machine learning can be used to help researchers comb through increasingly high levels of data. A new life sciences paper is uploaded every 30 seconds, this means that by the end of the day nearly 3000 new pieces of scientific literature exist. There are also increasing numbers of biomedical, clinical trial and biomarker data being released. Within this data could be the potential for a new highly profitable drug; however it would be impossible for one person to accurately comb through the information. Using a technological solution allows for this excess of information to be sorted, categorized and reviewed with important links and connections found. It can also narrow the research focus for a new drug saving valuable time.

Drug repurposing can also be a potential gold mine for cost reductions and profitability. A large amount is already known about the drug or a late stage candidate, so finding a secondary use for the formulation by cross referencing it against disease defects can be a strong strategy.

Read More: How Alphabet's DeepMind AI Algorithm solved the protein-folding problem