Realizing the potential of AI in pharma R&D

The pharmaceutical industry is under constant pressure to deliver safe and effective treatments — and quickly. This urgency requires companies to embrace innovative solutions that can streamline the traditionally lengthy process. Increasingly, artificial intelligence (AI) in pharma R&D is addressing many of these challenges.

Recent research from McKinsey notes that pharma companies were already using AI in many instances before the public awakening to genAI. Even still, the potential of generative AI is immense — $60 billion to $110 billion annually in new economic value for the industry. Expect AI-powered technology to have profound effects on drug discovery and development, speed to market and other aspects of R&D. 

Learn how companies use AI in lowering the cost of pharma R&D, including six use cases, and what lies on the horizon.

What is the role of AI in pharma R&D?

The pharmaceutical industry has traditionally been cautious in adopting new technologies. This hesitance is understandable, given the life-changing nature of pharma products, not to mention tough regulations and the substantial financial risks involved in drug development. Companies can spend over a decade to get drugs to market at a cost of billions of dollars. While technology drives pharma innovation, companies must be careful about which tech they invest in. 

That said, industry leaders recognize the vast potential of AI, including generative AI, for pharma R&D. These technologies are transforming how new medications are discovered, developed, tested and brought to market.

Regulators, including the U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA) have acknowledged the increasing use of AI algorithms and machine learning in drug development. The FDA is in the early days of creating a risk-based regulatory framework to accommodate AI’s rapid advancements.

Meanwhile, the EMA and Heads of Medicines Agencies have launched an AI work plan for maximizing AI’s benefits in the pharma industry. This multi-year plan will focus on various policy areas related to AI, as well as “the fundamental role of experimentation in accelerating learning and gaining new insights.” 

3 benefits of using AI for research and development

Pharma companies are always looking to reduce the timeline and cost of new drugs and treatments without sacrificing outcomes. AI and related technologies can contribute to each stage of R&D by streamlining processes, reducing costs and spurring innovation through cutting–edge products. Here are a few of the ways AI can improve pharma R&D today.

Redefining longstanding workflows

AI can redefine and improve pharma R&D workflows in two key ways: replacing manual, low-level cognitive tasks and giving statistical insights into what’s working and what could be improved. These optimized workflows streamline drug development processes and makes them more responsive to changes in research direction or market needs.  

Removing knowledge silos 

Knowledge silos are created when people or groups fail to share information or skill sets across the enterprise. This is often intentional — a byproduct of job specialization, isolated teams, poor communication or clunky systems that inhibit sharing of data. These silos become especially taxing when colleagues with similar responsibilities and experience can’t learn from each other’s findings. They might even work on the same team, yet they lack visibility into other areas of the organization. Research is delayed, inhibited or even duplicated because of such silos.

To break down these silos, you need knowledge-sharing and communication tools that leverage the power of AI and make information easy to find and access. These solutions give teams greater access to all available resources, including subject matter experts.

Reducing costs

AI can help companies identify inefficiencies, reduce redundancies and optimize processes, reducing waste (and costs) in each case. 

Pharmaceutical companies can also use AI for R&D cost-savings by improving decision-making efficiency and reducing redundant efforts. AI-powered knowledge systems, meanwhile, enable access to vast data sets and expert insights, reducing the time and resources typically spent on rediscovering existing knowledge or duplicating previous efforts.

6 ways R&D teams are using AI in pharma

The power of modern AI can help organizations find information faster, improve their operations and unlock innovation. 

The potential uses for AI in the pharmaceutical and life sciences industries are limited only by imagination and the outer edges of technology. AI and its healthcare applications will continue to evolve, but you can already realize many potential benefits. Here are some ways AI is used in pharma today and will continue to evolve new capabilities.

Target identification 

Target identification is crucial in drug development, particularly for complex diseases like stroma-rich cancers, which include colorectal and pancreatic cancers. Companies like Phenomic AI and Boehringer Ingelheim are using AI for more precise target identification of these cancers. 

This AI-driven approach allows researchers to use digital screening and experimental validation for improved evaluations of these specific cancers, hopefully unlocking better drug targets. 

Virtual AI screening is a powerful approach for comparing chemical structures against targets, predicting binding likelihood and identifying targets for further testing. All of this can be done at a greater scale and speed than previous computational or manual efforts.

Drug discovery 

AI can improve the rigor of drug discovery efforts by introducing structured, data-driven inquiries. And AI can do this at scale by leveraging deep learning and chemical libraries to predict and optimize molecule interactions.

For example, Atomwise is using AI to streamline the small-molecule drug discovery process and discover unique chemical structures that could lead to breakthrough medicines. Recently, Atomwise used AI to develop a TYK2 inhibitor‌ candidate that could help treat immune-mediated inflammatory diseases such as psoriasis and inflammatory bowel disease. The company’s AI-powered platform can sort through over 15 quadrillion synthesizable compounds, meaning that countless other drug candidates could be forthcoming.

Drug design

AI technologies are reshaping and accelerating ‌traditional drug design methodologies. For example, biotech startup Cradle is using generative AI to speed up protein design and optimization. The company has numerous industry partners and is pursuing more than 10 R&D projects that focus generative AI capabilities on protein modalities. 

Separately, university researchers are using AI tools to speed up work on Parkinson’s disease. An AI platform quickly scanned a library of chemical compounds and identified a handful of promising compounds for further study.

Knowledge management

Roche has harnessed AI to help employees quickly access critical information across its global network and connect with subject matter experts for verification and further insight. This AI-driven approach has transformed Roche’s communication and innovation. The changes aren’t just about processes; Roche is creating an interconnected culture of knowledge sharing that improves business results and patient outcomes.

Since adopting Starmind in 2020, Roche has seen substantial growth in platform engagement, with user numbers swelling from 1,000 to nearly 9,000 by 2023. The platform’s AI capabilities have helped save around 91,000 hours previously spent searching for information, proving a significant increase in operational efficiency. Roche continues to expand its AI capabilities by integrating with systems like Google Cloud Search and expanding multi-language support. 

Clinical trial documentation 

Clinical trial documentation is an important but time-consuming process that can delay drug development. Generative AI tools are being used to automatically create complex documents such as clinical study reports, patient narratives, and summary clinical safety documents. Employees spend less time drafting, reviewing and approving these essential components.

For example, Yseo uses pre-trained large language models specifically developed for biopharma use. These AI tools automatically generate clinical documentation, creating over 10,000 reports in 2023 and saving thousands of hours of manual labor. The company hopes to automate other aspects of document processing, including FDA approvals. 

Manufacturing 

AI can help pharma manufacturers like Amgen improve operational efficiency and accelerate production of essential medicines. The company partnered with Amazon Web Services to increase the throughput and reliability of medicine production.

Amgen's new manufacturing facility will leverage Amazon SageMaker and other machine learning technologies to analyze manufacturing data points in real time. Use cases include enabling predictive maintenance for shop floor equipment, improving ergonomic safety for workers and reducing the need for human intervention in ‌packaging lines.

Get started with enterprise AI for pharma R&D

AI in pharma R&D is transforming the entire life cycle, from drug discovery and design to clinical trials and manufacturing. Companies already use AI for lowering the cost of pharma R&D, optimizing workflows and reducing the time to market.  

Starmind’s AI-powered Expertise Directory can help pharma R&D teams optimize their operations, especially in global enterprises. The platform improves communication and knowledge sharing across the organization, enabling employees to quickly locate information and connect with internal experts. Starmind’s solution helps break down R&D silos and accelerate innovation. 

Curious how Starmind can help your R&D teams streamline drug discovery, accelerate clinical trials and eliminate duplication of efforts? Watch how Roche uses Starmind for more efficient and effective R&D processes.