Your next routine checkup could reveal whether you are headed toward cirrhosis or liver cancer, and the warning might come from three blood measurements your doctor already takes.
Researchers at Karolinska Institutet have developed a prediction tool that uses standard liver enzyme tests to forecast the 10 year risk of severe liver disease with remarkable accuracy. The model, called CORE (Cirrhosis Outcome Risk Estimator), outperformed existing screening methods in a study tracking nearly half a million people over three decades.
The timing matters. Liver cirrhosis ranks as the 11th leading cause of death globally, while liver cancer (predominantly hepatocellular carcinoma) stands as the sixth most common cancer and fourth most common cause of cancer death. Yet these conditions often develop silently, showing no symptoms until complications force patients into emergency rooms.
What Makes CORE Different
Most people getting annual physicals already have their blood tested for liver enzymes: AST, ALT, and GGT. CORE takes those three measurements, adds the patient’s age and sex, and calculates their likelihood of developing serious liver disease within a decade. The researchers designed it specifically for primary care settings, where most patients first seek help but doctors lack specialized tools for catching liver problems early.
This is an important step towards being able to offer early screening for liver disease in primary care.
The study, published in The BMJ on September 29, analyzed health records from 480,651 Stockholm residents who underwent checkups between 1985 and 1996. Over the following 30 years, about 1.5 percent developed severe liver disease: cirrhosis, liver cancer, or conditions requiring transplantation. The long follow up period proved crucial, since liver disease typically progresses across decades rather than years.
CORE achieved an 88 percent accuracy rate in distinguishing between people who would or would not develop major liver problems. The current recommended screening tool, called FIB-4, managed only 79 percent. At every level of sensitivity, CORE demonstrated better specificity than FIB-4, meaning it caught more true cases while triggering fewer false alarms.
The researchers tested CORE on two additional populations, one in Finland and another in the UK, and found consistent performance. The model also costs less than FIB-4 to run: about $5.70 versus $12.30 at the Karolinska University Laboratory, since all three enzyme measurements can be taken from a single blood tube rather than requiring separate samples.
The Bigger Picture
Metabolic dysfunction associated steatotic liver disease, the condition formerly known as fatty liver disease, now affects an estimated 38 percent of people worldwide. It is particularly common among those with type 2 diabetes or obesity. Many of these patients remain unaware their livers are struggling until the damage becomes irreversible.
Primary care has not had the tools to detect the risk of severe liver disease in time.
Drug treatments for preventing liver disease progression are becoming available, including resmetirom for patients with fibrotic steatohepatitis. But these therapies only help if doctors identify at risk patients before cirrhosis develops. Early detection also enables surveillance for complications like esophageal varices and enables lifestyle interventions (alcohol abstinence for alcohol related liver disease, weight loss for metabolic conditions) that can slow or halt progression.
The study had limitations worth noting. The training data came from a relatively old cohort, with the most recent lab values from 1996. Some variables had high rates of missing data, particularly platelet counts and bilirubin levels. The validation datasets showed higher rates of liver disease than the training population, possibly reflecting different risk profiles or improved disease detection in more recent years.
The model also needs testing in populations outside Western Europe and among groups at especially high risk, such as people with confirmed fatty liver disease or type 2 diabetes. The researchers acknowledge that integrating CORE into electronic health record systems, where it could automatically calculate risk from routine lab orders, will prove essential for widespread adoption.
A web calculator is already available at www.core-model.com for doctors and nurses. But the bigger question remains: at what risk threshold should primary care physicians refer patients to specialists or order additional tests? The researchers suggest a CORE score of 0.4 percent roughly corresponds to the established FIB-4 cutoff of 1.30, while 5 percent matches FIB-4’s 2.67 threshold. Health economic studies will need to determine what level of risk justifies the costs of further evaluation.
For now, CORE represents something simpler and potentially more valuable than another diagnostic tool: a way to have an informed conversation about liver health before symptoms appear. Whether that conversation leads to lifestyle changes, specialist referrals, or simply closer monitoring, it beats the alternative of waiting for cirrhosis to announce itself.
The BMJ: 10.1136/bmj-2024-083182
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.