Most of medicine is built to react. You feel a symptom, you see a doctor, a test confirms what’s already wrong, and treatment starts. The system works. It just works late. For chronic disease, the gap between the moment your biology starts to change and the moment anyone puts a name to it can be enormous. On average, more than seven years pass between the onset of a chronic disease and a formal diagnosis.1
Seven years is a long stretch to be sick without knowing it. It also happens to be the window when the disease could often have been slowed, redirected, or headed off altogether. So a real question sits underneath all of this: could we see these diseases coming while there’s still time to do something? More and more, the answer is showing up in the epigenome.
Reacting versus predicting
Reactive medicine confirms disease after it announces itself. The tools are familiar and often very good, but by design they look for damage that has already happened. Spirometry diagnoses COPD once lung function has measurably dropped. A fasting glucose test flags diabetes once blood sugar regulation has already broken down.
Predictive medicine runs on a different timeline. Instead of confirming disease, it detects risk, ideally years ahead while the biology is changing but symptoms haven’t arrived. The payoff goes past earlier treatment. Step in early enough and the disease may never fully take hold.
The sticking point has always been finding a signal that shifts early enough to matter and before irreparable harm has been done. We believe epigenetics is the missing piece to do this.
Risk shows up in the epigenome before it shows up in you
Chronic diseases rarely start with symptoms. They start with biology: quiet, coordinated changes in how genes are being switched on and off, well before anything registers on a routine test or in how you feel.
Because the epigenome is the body’s active control layer, responsive to metabolism, inflammation, environment, and stress, those changes often turn up in DNA methylation patterns early in the process. The information has been there the whole time. We just couldn’t read it at scale or make sense of it.
This is what makes methylation a good foundation for prediction as a live readout of processes that are underway rather than a snapshot of the damage already done. Read a million of those signals from one blood sample and you’re looking at a disease’s earliest fingerprints, sometimes years before a standard test would catch them.
Top chronic diseases, billions of people, found too late
Chronic diseases cause roughly three-quarters of deaths worldwide,2 and more than four billion people live with at least one.3 Take four of the most common.
Cardiovascular disease is the leading cause of death in the United States, responsible for hundreds of thousands of deaths a year.4 Worldwide, it kills someone roughly every two seconds.5
Type 2 diabetes affects well over a hundred million Americans once you count prediabetes, and a large share of them have no idea.6
MASLD, the current name for fatty liver disease, affects an estimated 100 million U.S. adults, and the great majority go undiagnosed.7 It tends to stay silent until it’s advanced.
COPD affects tens of millions and is badly underdiagnosed, partly because the standard test usually isn’t ordered until symptoms are already serious.8
The common thread is delay. The biology shifts quietly for years, huge populations go undiagnosed, and earlier detection could change how each story ends.
What finding it early actually buys you
Catching disease sooner isn’t just reassuring. It changes what’s medically possible. Earlier in the process there are simply more moves available, like lifestyle changes that still work and drugs that are far more effective before the damage sets in.
It also shifts the math. Chronic and mental-health conditions account for an estimated $4.5 trillion in annual U.S. healthcare spending, much of it piled up in late-stage treatment and complications.9 Moving even part of that effort upstream, toward prediction and prevention, is one of the larger opportunities anywhere in healthcare.
That premise is the whole idea behind the Infinite Epigenetics approach: easily accessible blood tests that read for about a million epigenetic signals, interpreted by a biological AI model, built to estimate risk for several chronic diseases from a single sample. In internal validation across its four lead programs, COPD, type 2 diabetes, cardiovascular disease, and MASLD, the models have shown a strong ability to separate people who have a condition from people who don’t.
Where this fits in clinical workflows
Epigenetic testing complements traditional lab panels, and the combination of these two test types is where the real clinical value lives. A risk estimate derived from DNA methylation is not a diagnosis. It’s a reason to look harder, talk to a clinician, and in many cases run a confirmatory test. The performance numbers coming out of research cohorts are encouraging, the field is advancing rapidly, and rigorous same-population validation and regulatory review are part of a maturing science.
What’s new about epigenetic testing isn’t certainty. It’s timing. For the first time we have a biological signal rich enough, and tools strong enough, to go looking for many chronic diseases years before they would otherwise turn up in standard labs.
For most of its history, medicine has detected disease only once symptoms arise. Epigenetics offers a different option: reading the body’s earliest warning signs from a few drops of blood, while there’s still room to change the outcome. Reactive care will always have its place. We are passionate about making the future of chronic disease – for patients, clinicians, and the systems that pay for it – look a lot more like prediction.
This article is for general education and isn’t medical or investment advice. Infinite Epigenetics is building a biological AI platform meant to predict, detect, and help prevent chronic disease from a single blood sample. Its operating companies, TruDiagnostic and Tally Health, have together collected more than 120,000 epigenetic samples and supported over 50 peer-reviewed publications.
Sources
- Gopalan et al., “Prevalence and Predictors of Delayed Clinical Diagnosis of Type 2 Diabetes,” 2019; Manikat et al., “Peri-Complication Diagnosis of NAFLD,” 2025; Larsson et al., “Impact of COPD Timing on Clinical and Economic Outcomes,” 2019.
- WHO, “Noncommunicable Diseases,” 2025.
- Boers et al., “Global Burden of COPD Through 2050,” 2023; Global Burden of Cardiovascular Diseases and Risks 2023 Collaborators, “Global, Regional, and National Burden of Cardiovascular Diseases and Risk Factors in 204 Countries and Territories,” 2025; Younossi et al., “The Global Epidemiology of NASH,” 2023; International Diabetes Federation, “IDF Diabetes Atlas,” 2025.
- CDC, “Heart Disease Facts,” 2024.
- WHO, “Cardiovascular Diseases,” 2026.
- CDC, “National Diabetes Statistics Report,” 2026; CDC, “Diabetes in the US,” 2026.
- Le et al., “Estimated Burden of MASLD in US Adults,” 2025; Kaiser Permanente, “Many Adults May Be Unaware That They Have Liver Disease,” 2025; Unalp-Arida and Ruhl, “Prevalence of MASLD and Fibrosis Defined by Liver Elastography,” 2025.
- COPD Foundation, “COPD Prevalence, Disease Burden Varies Significantly by State,” 2025; American Lung Association, “COPD in Your State,” 2026; Ho et al., “Under- and Over-diagnosis of COPD,” 2019. Lamprecht B, et al. “Determinants of Underdiagnosis of COPD in National and International Surveys.” Chest. 2015.
- CDC, “Fast Facts: Health and Economic Costs of Chronic Conditions,” 2026; McKinsey Health Institute, “The Health of Nations: Stronger Health, Stronger Economies,” 2026.