Fractal Podcast: Suman Giri on Scaling AI Innovation in Pharma

Introduction

The pharmaceutical industry stands at the cusp of an artificial intelligence (AI) revolution. As the world grapples with unprecedented health challenges and escalating costs, pharma leaders are turning to AI and machine learning (ML) to accelerate the discovery of life-saving treatments, optimize clinical trials, and personalize patient engagement at scale.

At the forefront of this transformation is Suman Giri, Global Head of Data Science & Analytics at pharmaceutical giant Merck. In a captivating Fractal podcast episode hosted by Sagar Shah, Suman shares his insights on the current state and future potential of AI in pharma, the talent imperatives for building high-impact analytics organizations, and the criticality of responsible AI innovation in healthcare.

With a distinguished career at the intersection of data science and healthcare, spanning insurance, provider, and life sciences contexts, Suman brings a unique perspective on the challenges and opportunities ahead. His compelling vision for a more agile, data-driven, and patient-centric pharmaceutical industry offers a roadmap for unlocking the full power of AI to enhance and extend human life.

AI Applications Across the Pharma Value Chain

"We use data science in every stage of the product lifecycle, from research to manufacturing to commercialization," explains Suman. "It is absolutely crucial in better understanding the diseases we are treating, how prevalent they are, and how we can develop personalized medicine strategies."

In drug discovery, AI is enabling researchers to parse vast troves of biomedical data, identify novel disease targets, and predict molecular properties with unprecedented speed and accuracy. DeepMind‘s AlphaFold model, which has solved the protein folding problem at scale, is just the beginning. The next frontier lies in protein-ligand binding prediction, which could dramatically accelerate the identification of drug candidates.

Suman emphasizes, "In the next 5-10 years, AI-driven molecular design will become the norm rather than the exception. We will see a significant compression in drug discovery timelines, getting treatments to patients faster than ever before."

In the clinical development phase, ML is being leveraged to optimize trial design, patient recruitment, and monitoring. Novartis, for instance, has partnered with Trials.ai to use AI for intelligent patient-trial matching, achieving a 70% improvement in enrollment rates.

"There is immense potential to use AI to make clinical trials more targeted, more diverse, and more efficient," notes Suman. "We are just scratching the surface of what‘s possible in terms of personalizing the trial experience for each patient."

On the commercial side, Suman highlights a range of sophisticated data science applications at Merck, including:

• Demand forecasting: Using machine learning to predict product demand across geographies, considering factors like seasonality, competition, and market events.
• Marketing mix optimization: Analyzing the performance of different marketing channels and optimizing spend allocation based on predictive ROI models.
• Content compliance monitoring: Leveraging natural language processing to review and flag promotional materials for potential compliance risks.
• HCP engagement personalization: Developing data-driven segmentation and targeting models to tailor sales rep interactions with healthcare providers.
• Patient journey mapping: Applying advanced analytics to claims, EMR, and lab data to identify patients with rare diseases and surface opportunities for early intervention.

"By 2030, I believe the majority of pharma commercial decisions will be driven by AI," asserts Suman. "From dynamic pricing to omnichannel marketing to patient support programs, every aspect of the business will be powered by predictive analytics."

Building the Analytics Talent Pipeline

To realize the full potential of AI, pharmaceutical companies must attract, develop, and retain world-class analytics talent. However, with fierce competition from tech giants and startups, this is no easy feat.

"Our number one challenge is talent," admits Suman. "We are looking for that rare combination of deep technical expertise and domain knowledge in pharma and healthcare."

Merck‘s talent strategy centers on creating an environment where data scientists can work on meaningful problems, grow their skills, and see the tangible impact of their efforts. Key initiatives include:

• Academic partnerships: Collaborating with universities on capstone projects, internships, and curriculum development to build a strong entry-level talent pipeline.
• Rotational programs: Providing opportunities for data scientists to rotate across different therapeutic areas and functions, gaining exposure to the business.
• Dual career tracks: Offering both technical and managerial career paths to support the aspirations of individual contributors and people leaders.
• Capability building: Investing in training programs, hackathons, and external education to keep skills on the cutting edge.
• Community engagement: Partnering with organizations like Analytics Vidhya to build brand awareness and source diverse talent.

The analytics talent gap in pharma is significant and growing. A McKinsey report estimates that by 2025, the industry will face a shortfall of 30,000 to 40,000 data science and AI experts. Closing this gap will require a concerted effort across academia, government, and industry to develop the next generation of analytics leaders.

The Future of AI in Pharma

Looking ahead, Suman highlights several key trends that will shape the future of AI in pharma:

1. Quantum Leap in Drug Discovery

"In the next 5-10 years, quantum computing will start solving problems that are intractable today," predicts Suman. "We will see a revolution in our ability to simulate molecular interactions and predict drug efficacy and safety."

Quantum AI has the potential to break through the limitations of classical computing, enabling pharma companies to explore vast chemical spaces and accelerate the drug discovery process. Early adopters like Menten AI are already using quantum techniques to design novel protein drugs for cancer.

2. Decentralized Data Ecosystems

"I am a big believer in the Web 3.0 landscape," says Suman. "Decentralized data ecosystems, powered by blockchain and federated learning, will enable secure and privacy-preserving analytics at scale."

As concerns around data privacy and security intensify, pharma companies will need to embrace new paradigms for data sharing and collaboration. Web 3.0 technologies like Nebula Genomics and Shivom are pioneering decentralized models for patients to own and selectively share their genomic data for research.

3. Metaverse for Medicine

"In the next 5-10 years, we will see the first tangible use cases for the metaverse in pharma," envisions Suman. "From virtual clinical trials to immersive HCP education, the possibilities are endless."

The convergence of virtual reality, AI, and 5G is paving the way for a new era of remote and personalized healthcare. Companies like AppliedVR are using VR therapeutics to help patients manage chronic pain, while Medivis is leveraging holographic technology to assist surgeons in the operating room.

4. Responsible and Sustainable AI

"We have a moral obligation to ensure that the AI systems we build are transparent, fair, and accountable," emphasizes Suman. "Responsible and sustainable AI must be a non-negotiable part of every project we undertake."

As the stakes of AI in healthcare grow higher, so too does the imperative to embed ethics and responsibility into every facet of the development process. Pharma companies will need to put in place robust governance frameworks, bias detection methods, and explainability techniques to maintain trust and mitigate unintended consequences.

Conclusion

The pharmaceutical industry is on the cusp of an AI-powered renaissance that will transform every aspect of how medicines are discovered, developed, and delivered. As Suman Giri‘s insights make clear, realizing this potential will require bold leadership, strategic investments in talent and technology, and an unwavering commitment to responsible innovation.

The road ahead is not without challenges. Legacy data silos, regulatory hurdles, and skepticism about AI remain significant barriers to adoption. But for visionary leaders like Suman, the imperative to harness data science for good is too great to ignore.

"Every day, I am motivated by the opportunity to build AI systems that can help get the right treatment to the right patient at the right time," reflects Suman. "It is a privilege to work on some of the most important problems facing humanity today."

As the AI revolution in pharma accelerates, the potential for data science to improve and extend human life has never been greater. By staying at the forefront of this transformation, companies like Merck are poised to make a meaningful difference for patients around the world for generations to come.