Elena Bonfiglioli (General Manager, WW Healthcare, Global Pharma and Life Sciences, Microsoft) and Tamara Elias (SVP, Strategy and Business Incubation, Nuance) speak with Delphine Zurkiya (Senior Partner, McKinsey & Company) about bringing data & AI, talent, and expertise together in the biopharmaceutical industry to help patients live longer, healthier lives. They focus on two of the questions posed to the industry in a recent report by McKinsey prior to the recent announcement of GenAI and GPT Foundation Models.1 Their conversation reflects their own views and should not be assumed as any professional (including legal) advice.
Delphine: Our research has shown that there is large untapped potential for digital solutions along the life sciences value chain. How can the tech industry play a larger role in driving patient outcomes and realize the full potential of innovative therapies?
Elena: The pharma industry is exploring many exciting frontiers with healthcare providers, data scientists, and other ecosystem partners. Top of mind for me is helping the pharma industry improve upon the success rate for new therapy development, with only around 12 percent of drugs in clinical trials making it through to regulatory approval today.2
Industry leaders are increasingly using advanced analytics to accelerate drug discovery and development—for example, use of large biomedical datasets and real-world evidence (RWE) for in-silico modeling of human biology—and we are now seeing emerging use cases in foundation models and generative AI.3 Some of these models, like PubMedBERT, could empower biologists in various scenarios of scientific discovery by helping them mine and generate biomedical text.4 Other foundational models encode molecule representations and have the potential to fundamentally change drug discovery not only by predicting which molecules (drug candidates) bind better to certain proteins (disease targets), but also now generating new molecules that can then be tested in the lab. There are many other advances in AI that can help across the care continuum too, such as Text Analytics for Health, which extracts and labels relevant medical information from unstructured texts such as doctor’s notes and electronic health records. This can then be fed back into the discovery and development process to improve treatments.
In addition to AI, the tech industry can help by ensuring all this data can be stored, matched, and available to be analyzed by as many researchers and companies as possible, and done in a way that upholds principles of privacy and security. Terra, a secure biomedical research platform co-developed by the Broad Institute of MIT and Harvard and Verily, provides this capability.
Tamara: Biopharma creates molecules that can save patient lives, but this is just one part of the journey. We need to find the right patients for therapy, monitor those patients while on therapy, and monitor post-therapy to watch for any recurrence. This calls for a continuous journey and not a point solution.
We still see too many non-adherent patients taking less than 80 percent of their prescribed medicine,5 or worse, patients who have delayed medical care all together. Remote patient monitoring technology can really help here. Devices like weight scales, pulse oximeters, blood glucose meters, blood pressure monitors, and wearables can offer clinical decision support by enabling providers with continuous data to better care for patients.
There is also a significant amount of data being captured through medical images across the care continuum. At Nuance, we see AI increasingly being used to support healthcare clinicians with their clinical decision-making (for example with Precision Imaging Network). Research and development (R&D) researchers can use all of this data to improve therapeutic interventions—using AI for biomarker discovery and patient identification and monitoring in clinical trials—and ultimately accelerate time to market.
No company can do this alone. Working with partners who have the data, understand digital touchpoints, and can deploy AI models, we can help patients get what they need at every stage of the journey to achieve the best possible outcomes.
Delphine: This partnership and ecosystem theme is critical. How can biopharma companies rely more on ecosystem partners, create more flexible and resilient operating models, and overcome their preference for owning capabilities and capacity outright?
Elena: Many of the advances in science and technology go beyond the core capabilities of a pharmaceutical company. We’ll see transformational change as AI capabilities are embedded in solutions, which requires collaboration across the ecosystem. Tech players can provide secure and scalable infrastructure, such as cloud, AI, and machine learning toolboxes. Academic institutions, government bodies, and healthcare providers can provide large, curated datasets to supplement those of pharma companies.
McKinsey’s recent research1 confirms my own experience that large talent gaps in data science and engineering can be barriers to innovation. Embedding these new technologies in core capabilities and collaboration tools across an organization will democratize the use of AI, widening the user base beyond data scientists and engineers and promoting innovation. What we believe is going to be the real change is when these enhanced capabilities are embedded in the solutions of many, not in the hands of only a few.
At Microsoft, we value collaboration with our partners to provide the connecting glue so that pharma companies can focus on the science. For example, through our partnership with SOPHiA GENETICS, we are providing secure and scalable cloud infrastructure, coupled with the SOPHiA DDM™ Platform, enabling multimodal data-driven care across a network of more than 750 connected healthcare institutions. So far, the SOPHiA DDM™ Platform has supported the analysis of more than 1.2 million genomic profiles. With each incremental profile, SOPHiA GENETICS’ algorithms become more robust, benefiting patient outcomes.
All of the partners in our ecosystem are pushing the boundaries in different ways. The ability to collaborate and build on these capabilities will be a key driver of change in the years to come.
Tamara: To win in a modern ecosystem, pharma companies will need a different approach to capability building. Two century-old companies come to mind—one medtech that believed it needed to create its own digital capabilities to protect its intellectual property, contrasted to a pharma company that joined a partnership ecosystem to augment digital and AI tooling beyond its core capabilities. The better model is to co-create new solutions to ensure that patients achieve the right outcomes with the molecules biopharma has created and is continuing to create.
Tech players are investing more than pharma companies in tech-related healthcare deals. This underscores what Elena described—to drive innovation, there needs to be a symbiotic relationship between players in the ecosystem: tech players who provide the infrastructure and the intelligence toolkit, pharma companies who push the boundaries in developing innovative therapies, and providers and payors who engage with patients and provide the data to advance science across patient populations, care settings, and patient journeys.
Too often, real-world data is siloed in individual databases. It is often said that there are medicines waiting to be discovered in the rainforest. Well, many innovative therapies are waiting to be discovered in data that we can access today. By leveraging this real-world evidence along with the capabilities of our ecosystem partners—medication adherence players, AI developers, and data companies who are experts at “making the molecule”—allows biopharma to leverage the best of external expertise.
Figuring out what we can do together is relatively easy; deciding how to work together is harder. But we’re truly excited about the advances we’ll make together.
Life Sciences Solutions
Discover how the Microsoft Cloud is helping life sciences organizations accelerate innovation.
Note: Microsoft and McKinsey & Company share ownership and publication rights of this blog post.