The same disease, opposite outcomes. Something genetic had to explain it.
At the University of Illinois Chicago, a decade-long inquiry into why some diabetes patients go blind while others do not has yielded both a specific genetic culprit and a broader method for understanding how genes and environment conspire to produce disease. Ophthalmologist Michael Grassi and his collaborators identified the FLCN gene as a driver of diabetic retinopathy risk, using a novel combination of genomic and transcriptomic analysis drawn from one of medicine's most consequential long-term diabetes trials. The finding matters not only for the millions living with diabetic retinopathy, but as a replicable template for untangling other diseases where heredity and circumstance are deeply intertwined.
- Diabetic retinopathy strikes unpredictably — a teenager with perfect glucose control can go blind while a veteran with decades of poor management keeps his sight, and that inexplicable gap has driven researchers to search for genetic answers.
- Previous approaches to the disease's genetic complexity kept hitting dead ends, because so many genes and environmental factors interact in ways that standard single-method analyses could not resolve.
- By pairing genomics with transcriptomics — asking not just what variants exist but which genes actually switch on under high-glucose conditions — Grassi's team found a way to cut through the noise and isolate FLCN as a genuine risk driver.
- Validation across independent cohorts and the use of Mendelian Randomization strengthened the causal case, moving the finding from correlation toward something closer to mechanism.
- The current standard of care — laser surgery and monthly eye injections — only manages damage already done, and this genetic roadmap points toward prevention and earlier intervention before vision is lost.
Michael Grassi was a young retina specialist when a paradox stopped him cold: a nineteen-year-old with carefully controlled diabetes went blind, while a Vietnam veteran with decades of poor blood sugar management kept his sight. The same disease, the same organ, opposite outcomes. Something genetic had to explain it, and Grassi spent the next decade trying to find what.
Diabetic retinopathy — damage to the light-sensitive tissue at the back of the eye caused by chronic high blood sugar — is one of the leading causes of vision loss in working-age adults. Yet its progression varies enormously from person to person, shaped by factors that go well beyond glucose control. Working with Barbara Stranger at Northwestern University, Grassi eventually found an approach that worked: combining genomics with transcriptomics in a way that had not been tried before for this question. The team examined how genes behaved in cells from type 1 diabetes patients — some with retinopathy, some without — when exposed to high glucose. A distinct set of genes responded differently depending on retinopathy status. They then tested whether genetic markers tied to those genes correlated with retinopathy risk in larger populations, and finally confirmed that increasing the activity of one gene in particular — folliculin, or FLCN — could actually cause retinopathy to develop.
The study, published in eLife, drew on cell lines from the Diabetes Control and Complications Trial, a landmark study that had carefully documented retinopathy severity across thousands of participants over decades. That clinical precision allowed the researchers to correlate genetic patterns with real outcomes, and they validated their findings across independent cohorts using Mendelian Randomization to sharpen the causal inference.
What gives the work its broader significance is not only the identification of FLCN, but the method itself — a reproducible template for studying other diseases where genetics and environment interact in poorly understood ways. Today's treatments for diabetic retinopathy are reactive: laser surgery and injections every four weeks to manage damage already underway. A clearer picture of genetic vulnerability could open the door to prevention, earlier intervention, and new screening strategies. For Grassi, a decade of pursuit has yielded not just an answer, but a better way to ask the question.
Michael Grassi was a young retina specialist when he encountered a puzzle that would occupy the next decade of his career. A nineteen-year-old patient with meticulously controlled diabetes went blind. A Vietnam veteran with decades of poorly managed blood sugar kept his sight. The same disease, the same organ, opposite outcomes. Something genetic had to explain it.
Grassi, now an associate professor of ophthalmology at the University of Illinois Chicago, set out to find what. Diabetic retinopathy—damage to the light-sensitive tissue at the back of the eye caused by high blood sugar—affects millions of people with diabetes. It is one of the leading causes of vision loss in working-age adults. Yet not everyone with diabetes develops it, and not everyone who does loses their sight at the same rate or to the same degree. The disease is heterogeneous, shaped by factors beyond glucose control alone. For years, Grassi pursued the genetic architecture underlying this variation, trying different approaches, hitting dead ends.
Then, working with Barbara Stranger at Northwestern University, he found a method that worked. The team combined multiple analytical approaches in a way that had not been tried before for this particular question. They started by examining how genes behaved differently in cells from people with type 1 diabetes—some of whom had developed retinopathy and some who had not. They looked at what happened when those cells were exposed to high glucose levels. A distinct set of genes responded differently depending on whether the person had retinopathy. Next, they took genetic markers associated with these genes and tested whether they correlated with retinopathy risk in larger populations. Finally, they went further: they tested whether actually increasing the amount of one particular gene—folliculin, or FLCN—could cause retinopathy to develop. It could.
The work appeared in eLife under the title "Integration of genomics and transcriptomics predicts diabetic retinopathy susceptibility genes." The researchers had identified FLCN as a gene that increases retinopathy risk. But they had done something else too: they had created a template, a reproducible method that could be applied to other diseases where genetics and environment interact in ways we do not yet understand.
The study drew on cell lines from the Diabetes Control and Complications Trial, a landmark clinical study that had generated biological samples from thousands of people with type 1 diabetes over decades. Because the DCCT had carefully documented retinopathy severity in each participant, the researchers could correlate genetic patterns with precise clinical outcomes. They validated their findings across independent cohorts, using a statistical technique called Mendelian Randomization to strengthen the causal inference.
What makes this work significant is not just the gene itself, though that matters. It is the approach. Diabetic retinopathy has proven difficult to study precisely because it is so genetically complex. Many genes contribute. Many environmental factors matter. The interplay between them is not simple. By combining genomics—the study of genetic variation—with transcriptomics—the study of which genes are actually turned on or off in cells—Grassi and his team created a way to cut through that complexity. They could ask not just whether a genetic variant existed, but whether it actually changed how cells behaved under the conditions that trigger disease.
Today, the standard treatment for diabetic retinopathy is laser surgery to preserve central vision or injections into the eye every four weeks. These are interventions that manage the damage after it has begun. Understanding the genetic factors that make some people vulnerable could open a different path: prevention, or earlier intervention before vision is lost. It could lead to new drugs, new screening strategies, new ways of thinking about who is at highest risk and why. For Grassi, it represents a decade of pursuit finally yielding not just an answer, but a way to ask better questions.
Citas Notables
I encountered two individuals with disparate outcomes, a 19-year-old who had well-controlled diabetes for five years and went blind, and a Vietnam veteran, who had poorly controlled diabetes for over 30 years but had no vision problems.— Dr. Michael Grassi, associate professor of ophthalmology at UIC
It has been a challenge to study diabetic retinopathy because it is so heterogeneous. There are so many genetic factors that can contribute.— Dr. Michael Grassi
La Conversación del Hearth Otra perspectiva de la historia
Why did it take ten years to find this gene? Wasn't the technology already there?
The technology existed, but the question was harder than it looked. You can sequence someone's DNA and find thousands of variants. The challenge is figuring out which ones actually matter for disease. Grassi needed a way to connect genetic variation to what was actually happening inside cells.
So he combined two different types of data—genetics and gene expression.
Exactly. He looked at which genes were turned on or off in cells from people with retinopathy versus without it, then asked whether genetic variants predicted those differences. That's the bridge. It's not just "you have this variant," it's "you have this variant and it changes how your cells respond to high glucose."
The story mentions a nineteen-year-old who went blind and a veteran who didn't. Did finding FLCN explain that difference?
The study doesn't say it explains those specific cases. But it points toward why such differences exist. Some people carry genetic variants that make their retinas more vulnerable to glucose damage. Others don't. It's part of the answer to why the same disease hits different people so differently.
What happens next? Does knowing about FLCN change how doctors treat diabetic retinopathy?
Not immediately. Right now, treatment is still laser surgery or injections. But if you know which gene increases risk, you can develop drugs that target it. You can identify high-risk patients earlier. You can test whether blocking FLCN actually prevents vision loss. That's the real payoff—moving from managing damage to preventing it.
And the method itself—why does that matter beyond this one disease?
Because many diseases are like this. Genetic variation plus environment plus time equals disease, but not in everyone. The approach Grassi developed—combining genomics and transcriptomics, validating across cohorts—that's a template for understanding any disease where genetics and environment interact. It's a tool that can be used again.