Latest Posts



Jun 24

Partex and Singapore’s Experimental Drug Development Centre collaborate to bring forward an innovative approach for early drug discovery and development

Frankfurt, Germany, 3rd June 2024, 9am CET Partex, a leading provider of AI-driven solutions in the pharmaceutical industry, is thrilled...
Read More


Apr 24

Partex Partners with Lupin to Revolutionize Drug Discovery through AI-Driven Asset Search and Evaluation

Frankfurt, Germany, 23 April 2024 – Partex, a leading provider of AI-driven solutions in the pharmaceutical industry, is thrilled to...
Read More

Innoplexus wins Horizon Interactive Gold Award for Curia App

Read More

Impact of AI and Digitalization on R&D in Biopharmaceutical Industry

Data are not the new gold – but the ability to put them together in a relevant and analyzable way is

Innoplexus has contributed to big data and Artificial Intelligence (AI) projects in pharma R&D. In this paper we share our point of view on the impact of AI enabled technologies on R&D that can be summarized in one brief statement: “Data are not the new gold – but the ability to put them together in a relevant and analyzable way is.” The pharma industry, as one important player in the field of healthcare, is beginning to embrace this revolution that affects all aspects of biopharmaceutical research. One way to look at the dynamic development in medicine is to compare clinical trials which aim to assess the risk of i.e. cardiovascular disease over time.

The Framingham Heart study started in 1948 initiated by the National Heart, Lung and Blood Institute and Boston University, on 5’209 subjects; all residents in the city of Framingham Massachusetts. After more than 70 years, the study is now in its third generation and has embraced many newly available technologies – among them genetic testing. Over 1’000 papers have been published on the results of this study and a good part of the body of epidemiological and genetic evidence in cardiovascular disease stems from this longitudinal trial.

The All of Us Research Program started in 2017 funded by the US National Institute of Health. Its mission is to help “enable a new era of medicine in which researchers, health care providers, and patients work together to develop individualized care”. The trial is not limited to one disease state but tries to capture all the health conditions and their interdependencies.

Eric Topol, a leading cardiologist summarized the key study features in this table. It describes well the opportunities and challenges medicine faces since the arrival of big data and AI technologies.


From what we see as relevant and practical capabilities of big data and AI there will be at least 5 major impacts on pharma R&D:

1. Integrate all internal and external data into one consistent well linked and searchable platform

The ability to manage and integrate data generated at all levels of the value chain from molecule discovery to real-world use, from clinical trial design to regulatory dossiers, from reimbursement applications to adverse event reporting is widely recognized as one key success factor for the industry.

While companies take different approaches to the implementation (from “big bang” reinventions of their clinical IT systems to gradual changes in their data infrastructure) all of them believe that the foundation to compete in the new reality of “personalized medicine” is to understand data broader deeper and most importantly much quicker.

Fail fast was always the mantra of pharmaceutical R&D. Based on the broader, deeper and almost real time updated data generated by comprehensive data platforms, AI technologies now allow to fail – and to succeed – faster than ever before.

2. Fail faster and more transparently by employing modelling techniques in drug discovery and portfolio decision support

In 2003, the first whole human genome sequencing had cost roughly $3 billion. Only 3 years later the cost for one individual decreased to $300,000 and it took days to generate the information. Now the costs are approx. 1000$ or less and it takes an hour to decode an individual’s DNA.

With the acceleration of the understanding of patho-mechanisms on a receptor and genetic level comes an explosion of available data that can only be turned into information and finally actionable insight with the support of AI enabled modelling and decision support technologies. Such technologies will be key in supporting the definition and monitoring of work flows which today often lack transparency and well-defined decision points. It would turn these often not well structured and coordinated processes into continuously monitored modi operandi that are very likely to generate new analytical techniques and ways of working.

Big data/ AI enabled work processes allow to move from static processes to more fluid and adaptable working procedures. They allow to monitor and adjust the development course according to the most recent insights generated from systems that integrate all new findings in “real time”.

3. Make clinical development more efficient

Even though electronic data records are standard since long and rolling regulatory submissions are accepted for 20 years, clinical development still has significant untapped “efficiency reserves”. Discovery and clinical trials making up the largest share of costs to get a drug to commercialization, efficacy gains in these areas would have a direct impact measurable not only in time to market but also in real savings.

In our experience the low hanging fruit that can immediately be harvested through AI technologies are

  • Dynamic sample size estimations to adapt clinical trial protocols real time
  • Real time monitoring of trial centers and automated re-recruitment in new or better performing cites
  • Unified data platforms to allow automated real time data entry QA

Another field of AI enabled efficiency gains is the area of drug repurposing. Using all the wealth of data on a given molecule that is available internally and externally has the potential to identify untapped development opportunities of molecules that have already gone through the preclinical phase successfully.

Not only would an AI-enabled systematic evaluation of promising alternative clinical usages for specific molecules to reduce the research and discovery time. Through a comprehensive big data analysis of all relevant information, new clinical development pathways become simpler to identify and clinical protocols easier to establish.

4. Get to new levels of risk management

There are more Adverse Event Reports on Twitter than there are at the FDA. The discussion for guidance from regulatory authorities how the industry should capture these reports is currently ongoing and it is likely to lead to new levels of social media monitoring.

Also, in the “traditional” AER business”, pharma companies and service providers successfully deploy AI-based tools to detect rare patterns. Weak signals from i.e. one geography or one clinical setting can initiate relevant searches for potentially undiscovered rare drug interactions or adverse events.

But even beyond the AE discussion, pharma needs to gain a better understanding of what the scientific and patient community perceives as their reality of a therapeutic choice.

The relevance of such information goes beyond the room of AE reporting and scientific communication as it also affects perceptions in the area of policy-making as well as access and reimbursement. Early signals of how a specific intervention is perceived (also in comparison to another) by patients and their advocacy groups in their blogs, or by scientific society forums allow to detect and track valuable information also for R&D, clinical and commercial pharma departments way beyond the risk management aspect.

5. Tap into the wealth of information from real-world evidence to establish clinical and economic value propositions

Already in early R&D the question of the health economic value propositions of a new chemical entity is part of the drug development decision making. The dilemma of controlled clinical trials demanded by regulators not delivering data on the “real world” setting of medical practice is well understood but rarely solved. Yet not only payers require “real world evidence data” to establish pricing and reimbursement levels.

There are several models to deliver the required evidence: some pharmaceutical companies partner with health care providers and insurers to be able to provide economic information based on actual claims and patient data. Others offer to go at risk in “pay for performance” agreements.

What they all have in common: they require adequate and integrated big data/AI solutions that allow determining individual patient outcomes measured in clinical and financial terms.


Big data and AI technologies enable the pharma industry to harvest the fast-growing wealth of health data and to turn it into actionable insights for better, individualized health outcomes and enhanced financial results.


AI-based analysis and decision support will allow pharma R&D to produce solutions more effectively and to deliver on the promise of “personalized medicine”. This requires AI based

  • Integrated data platforms capturing the wealth of internal and external data in an easy to search way
  • Modelling apps and technologies for more effective drug discovery and faster decision making
  • Clinical research support solutions to automate (i.e. QA-) processes harvest efficiency reserves
  • Risk Management technologies detecting relevant AE and clinical and social media communication signals
  • Real world evidence information platforms to describe the full value potential of a pharmaceutical solution

Pharma blog banner

Featured Blogs