• Sustainability
  • DE&I
  • Pandemic
  • Finance
  • Legal
  • Technology
  • Regulatory
  • Global
  • Pricing
  • Strategy
  • R&D/Clinical Trials
  • Opinion
  • Executive Roundtable
  • Sales & Marketing
  • Executive Profiles
  • Leadership
  • Market Access
  • Patient Engagement
  • Supply Chain
  • Industry Trends

The Data Challenge in Rare Diseases and Complex Therapies

Pharmaceutical ExecutivePharmaceutical Executive-07-01-2021
Volume 41
Issue 7

Closing the gaps in complex therapy data remains a hurdle, but “getting clever” with AI can offer a path forward.

The panelists at the “Data: The Cornerstone of Success” roundtable at Veeva’s North American Commercial and Medical Summit last month were in agreement that COVID-19 has not just created dramatic shifts in patient and healthcare provider (HCP) behaviors but is also changing the industry’s relationship with data. For Eric Solis, director and lead data scientist at Takeda, COVID has exposed the need for access to broader data sets. “We’ve been bringing in information outside of our therapeutic areas to understand how much the new trends we’re seeing can be applied to our markets,” he said. COVID has served as an impetus “to challenge our data providers and ourselves to shift our ability to consume and analyze the data.” Saket Malhotra, Ipsen’s head of data, digital, and commercial IT, noted that, before COVID, companies were already trusting of data and were shifting toward a data-driven or insights-driven culture. However, he explained, “now we have accelerated the data, digital, and technology investments to focus on customer and patient-centric capabilities and experiences and we’re bringing in more out-of-the-box innovative thinking to drive the business’ strategic objectives—that is the shift post-COVID.”

In complex therapies and rare diseases, however, the post-COVID shift in attitudes and access to data has not been realized with the same sense of optimism. Shekhar Sattiraju, executive director and head of commercial operations at Chiasma, told Pharm Exec that while the data are getting “bigger and better,” processing speeds are doubling every few months, and analytical approaches are improving, there is still work to be done to improve lives of patients with rare diseases, and a key step is to help them get diagnosed. “If we consider the big picture, there are 7,000 rare diseases and this equates to about 25 to 30 million patients. The unfortunate part is that among these, there are millions who are still looking for a diagnosis,” observed Sattiraju. “So, the critical question for us is, how do we maximize our ability to find those patients, and help them connect to an appropriate diagnosis? From my perspective, we are making headway through advanced analytics, AI, and machine learning; we are still in an early phase of this journey and have much more progress to make when it comes to getting the most from the data.”

Solis agreed that in complicated therapy areas “there are a ton of different factors that are influencing the underlying behavior that we’re trying to affect.” He told Pharm Exec, “The segments and nuances, and types of patients and physicians that we’re interested in don’t fall out of the data readily. You try to infer a diagnosis, a segment, a subset, a severity through longitudinal views on patients and physicians, but you run into the problem of the data having gaps.” He pointed to the “Venn diagram problem:” If two data sets are combined, certain patients or events occur in one but not the other and there’s an intersection that offers a more complete picture. “But as we add more context and nuance by bringing in other data sources, those gaps tend to multiply,” he said. “We need analytics approaches that deal with those gaps to get better insight and visibility.”

Solis, however, is excited about how AI and machine learning can be leveraged “not necessarily to get to the answers, but to ask questions that can be followed up with more data to identify where we need to focus our efforts.” The “fun part” is “creatively employing data science to get as much out of data as possible.” He explained, “You can get clever in the way you engineer features and build structure into your analytical approaches.” Solis’ team has been working with vendors to identify key gaps in the data and build in additional data sources. “We’ve been partnering closely with Veeva to leverage what we’re calling crowdsourcing. That is, understanding anonymized, large industry trends, but at the individual prescriber levels, so that we can start to disentangle these individual shifts in HCP preferences. This is important if we want to get to a more advanced, predictive, next best action, or a really tailored approach.”

By leveraging insights from broader behaviors across markets and therapeutic areas—especially the “idiosyncratic shifts” fueled by COVID—Solis’ team aims to “supercharge” their ability to understand customers. This approach is “a potential game changer for rare diseases and more nuanced therapy areas.”

Julian Upton is Pharm Exec’s European and Online Editor. He can be reached at jupton@mjhlifesciences.com.

Related Videos