A Disturbing Trend in Drug Discovery

On average, it takes at least 10 years for a drug to enter the market, with 6-7 years spent in clinical trials. According to some estimates, the average price of development and approval of a new drug is $2.6 billion. This number reflects the high failure rate both in preclinical and clinical research. In parallel, R&D returns on investment among largest pharma companies are steadily decreasing, from reported 10.1% in 2010 to about 3% in 2017. In this context, any changes to improve the process of drug discovery would have a tremendous financial and societal impact in the industry whose combined annual revenues reach $446 billion in the US and over $1 trillion globally.

Drug Discovery and Development Timeline

Source: Pharmaceutical Research and Manufacturers of America

Promise of Artificial Intelligence

Artificial intelligence has shown great promise in accelerating and improving the process of drug discovery at every stage. In early discovery and preclinical research, AI has helped researchers to make sense of the vast data sets derived from chemical, genomic and various biological studies to come up with novel ideas for new drugs and therapies.

Within clinical research, the application of AI has the potential to enable better preparation and execution of clinical trials, contributing to overall reduction in R&D costs and failure rates.

Challenges Remain

Recent years have seen a growing number of drug discovery-oriented AI companies and an increasing number of alliances between big pharma and AI solutions providers. However, full embracement and implementation of AI-powered computational approaches to drug discovery requires addressing current challenges.

• Technological: Building suitable algorithms that reflect the full complexity of natural phenomena requires large volumes of high quality data with a high cost of generation.
• Cultural/social: Many researchers may be unwilling to integrate machine learning tools with traditional drug discovery approaches due to a skepticism toward AI. Some of it maybe deserved due to a failed promise of bioinformatics.

Drug Discovery focused AI in Chicago Area and the broader Midwest

rMark Bio

rMark Bio, a Matter Start Up, helps monitor R&D processes and the market in the post approval phase (collaboration with Takeda Medical Affairs group) to enable best-decision making and reduction of inefficiencies.  This exciting new firm was cofounded by Lev Becker of The University of Chicago.

AICure 

AICure, another Matter Start Up, improves compliance in clinical trials by using artificial intelligence to visually confirm medication ingestion. The clinically-validated platform works on smartphones to reduce risk and optimize patient behavior.

Qrative

Qrative, a start up cofounded by Mayo Clinic,  synthesizes knowledge from multiple biomedical sources (using nference technology). Discovers potential rare disease indications and subsets of patients who may respond favorably to an existing drug.

Antidote

Antidote makes sense of unorganized and unstructured data about clinical trials. Enrolls more patients in appropriate trials.

Scinote

Scinote uses AI to write a draft scientific manuscript based on provided data. Allows researchers to get a "head start" when writing a scientific manuscript to submit for publishing.

AbbVie

AbbVie, in addition to internal efforts to build machine learning tools, has announced the following collaborations:

• Atomwise uses AI to predict drug candidates by leveraging a convolutional neural network trained on huge amount of organic chemistry data. This allows researchers to generate novel drug candidates much faster than traditional methods alone.)
• Calico (a Google company) – details of this collaboration have not been disclosed. Calico’s Chief Computing Officer is Daphne Koller one of the leading experts in application of AI in biomedical sciences.
• Prof. Rick Stevens (University of Chicago and Argonne National Laboratory) - Works on novel strategies for large-scale computational approaches to drug screening.
• Prof. David Page Jr. (University of Wisconsin, Madison, Center for Predictive Computational Phenotyping) – monitoring of clinical trials, pharmacovigilance.
• Prof. Jian Peng (University of Illinois, Urbana-Champaign) – AI to accelerate drug discovery. Developing computational techniques for processing, integrating and analyzing massive datasets in genomics, systems biology and molecular biology. Includes integrative and etwork-based approaches for understanding molecular mechanisms of human diseases and accelerating drug development.
• Prof. Andrew Ferguson (University of Illinois, Urbana-Champaign) – conducting research using AI and data mining for vaccine design.

Sources

https://www2.deloitte.com/us/en/pages/life-sciences-and-health-care/articles/measuring-return-from-pharmaceutical-innovation.html

https://blog.benchsci.com/