Paired with years of medical history and proteomic measurements, genetic data can inform multiple points along the path from research to care, from deciding which drug targets deserve further investigation to improving clinical development and indicating people who could benefit from expedited intervention.
Bringing that information together across millions of people is ultimately about making human genetics more useful in the real world—connecting what is written in DNA with how disease actually develops and turning that knowledge into something on which researchers and clinicians can act.
Connecting genetics with patient care
Genetic insights are also being used to identify people at particularly high risk of developing certain diseases, allowing surveillance—and, where necessary, treatment—to begin earlier. Baras points to pancreatic and ovarian cancer, from which relatively few people face substantially elevated danger. “Population-level testing just isn’t feasible, but if you can narrow it down to tens of thousands rather than hundreds of millions using genetic markers, something like an annual abdominal MRI might be practical.” For those patients, more intensive monitoring could mean detecting the condition at an earlier stage, when the chances of successful treatment are considerably better.
In the years since the discovery, RGC has mapped more than 100 protective genetic insights linked to numerous diseases, with around 50 insights progressing into Regeneron drug programs. Pinpointing these rare variants across millions of people is a “needle in a haystack” exercise, according to Baras, though advances in technology are helping refine the process. RGC’s research teams work alongside AI models trained on expansive de-identified genetic and health datasets, sifting through large volumes of information for unusual patterns and flagging those that warrant closer expert scrutiny.
One of those searches recently produced a promising result in Huntington’s disease. In June 2026, RGC scientists found people carrying the Huntington’s variant who stayed symptom-free two decades past when the disease should have appeared and some who never developed the disease at all. In other words, the same “incurable” variant being present from birth may not be a life sentence for everyone who has it.
Hear from Dr. Aris Baras
on how “superhumanism”
is driving drug discovery
and development programs.
With access to a large and detailed body of de-identified human data, RGC scientists can distinguish which genetic associations are likely to offer a credible route to new medicines. Some of the most promising signals in this process don’t come from sick patients but from so-called “superhumans”—individuals carrying protective variants that lower their risk of certain diseases. Studied carefully, these rare occurrences indicate what happens when a particular gene is altered, giving drug developers a model to imitate.
In 2006, researchers using data from the Dallas Heart Study identified participants carrying variants that disrupted PCSK9, a gene involved in regulating cholesterol. The selected group had unusually low levels of harmful cholesterol and, crucially, an almost 90% lower risk of coronary heart disease. Regeneron Pharmaceuticals (Regeneron), the parent company of RGC, leveraged these genetic insights to inform the development of treatment targeting PCSK9, translating naturally occurring genetic protection into a therapeutic approach.
What ‘superhumans’ can teach us
TriNetX, a global health-data company and one of RGC’s collaborators, has a network covering more than 14,200 clinical sites and 309 million patients. This scale is matched by what Jeff Brown, TriNetX’s chief scientific officer, describes as “data depth and generalizability”—richer patient data, including clinical notes, biomarkers, vital signs, and laboratory findings, along with a population broad enough to reflect different ages, health profiles, and disease states.
Finding reliable relationships across that data requires very large populations. An uncommon genetic variant may appear in only a tiny fraction of people, so millions of individuals are needed to identify enough carriers and compare what happens to them over time. To date, RGC has paired de-identified genetic data with de-identified health records from around 3 million participants, already one of the largest such datasets ever assembled. RGC is now on the path to 10 million, with the ultimate goal of reaching 50 million—enough data to unlock discoveries not yet possible today. Baras’s team is collaborating with health systems, academic centers, and research institutions to make this a reality, expanding both the number of people represented and the insights available about them.
Genetic data and patient histories together provide a much fuller picture, though they still leave gaps. This is where proteomics, the large-scale study of proteins in the body, comes in. While DNA is largely fixed from birth, the proteins circulating in blood change over time as an individual responds to illness, treatment, and other physiological processes. Linking those measurements with long-term patient and genetic information gives a detailed account of how inherited differences are expressed in the body and how they relate to disease.
People’s genetic information, their DNA, can indicate whether they carry variants associated with a higher or lower risk of particular conditions. But genetics alone cannot show whether someone developed the disease, when symptoms appeared, or how their well-being changed over time—details that must be drawn from medical records.
Building a fuller picture of human health
Linked with other biological information and years of medical history, those findings can also strengthen research, reveal more about how illness progresses, and help uncover health risks earlier.
“The key is having enough biological data,” says Dr. Aris Baras, senior vice president and head of Regeneron Genetics Center (RGC), which has spent more than a decade building large de-identified genetic and health datasets. “The more of it you can gather, the better you can understand which genes are driving disease and which genetic changes might protect against it.”
For dozens of incurable inherited disorders, scientists have long known the underlying genetic causes. Huntington’s disease is a good example: The variant responsible for the neurodegenerative illness was discovered decades ago, yet there remains no therapy that can halt its progression. The same is true for many similar conditions, in which an understanding of the cause hasn’t translated into effective treatments.
Drawing on genetic and de-identified health data from millions of people, Regeneron Genetics Center® is finding novel drug targets and new ways to spot disease risk earlier.
Hear from Tim Walsh on how KPMG’s AI-powered tool is helping companies plan around tariffs.
Part of the difficulty is predicting how a medicine will behave in people prior to clinical trials. Preclinical studies and animal models are essential, but neither can fully reproduce human biology. Addressing that shortfall requires better human data. By studying how naturally occurring genetic differences relate to health and illness, researchers can identify drug mechanisms that are 2.6 times more likely to succeed in clinical development.
