Hot on the Trail of a Noninvasive Test for Pediatric NAFLD

As a gastroenterologist/hepatologist at Cleveland Clinic Children’s, Naim Alkhouri, MD, sees pediatric patients with nonalcoholic fatty liver disease (NAFLD) on a daily basis. Challenged to diagnose the presence and severity of the disease using inadequate, invasive tests developed for adults, he is dedicated to finding noninvasive testing methods for this most common form of liver disease in children.

How current tools fall short

“Liver biopsy, the current gold standard for diagnosing NAFLD, is impractical for screening purposes,” he says. Dr. Alkhouri follows American Academy of Pediatrics recommendations to use liver enzyme testing to screen overweight children who have risk factors for NAFLD. But the tests lack sensitivity, and he feels the cutoff value is too high.

“It’s a misconception that if a child has NAFLD, they’ll have elevated fatty liver enzymes,” he explains. “They can have normal alanine transaminase (ALT) but have nonalcoholic steatohepatitis (NASH) or fibrosis. Many children with borderline values are missed.”

Even after the diagnosis of NAFLD has been made, there is no good noninvasive method of determining its severity.

For these reasons, Dr. Alkhouri has been working with colleagues in Italy who share his passion to develop noninvasive methods of identifying the disease at all its stages.

Pinpointing NASH within the NAFLD population

Some 10 to 15 percent of children with NAFLD develop NASH, a more aggressive form of the disease that may progress to fibrosis and, ultimately, cirrhosis.

A differential diagnosis is desirable, because patients with NASH may benefit from treatment with vitamin E or, potentially in the future, one of two investigational agents, obeticholic acid or elafibranor (GFT505). After puberty, bariatric surgery may help resolve the disease.

Although development of cirrhosis is uncommon — a recent review of the United Network for Organ Sharing (UNOS) database by Cleveland Clinic researchers found 330 young adults who received liver transplants for NAFLD-related cirrhosis — its unpredictability requires that all patients with NASH be watched. “Some of these patients were younger than 18,” says Dr. Alkhouri, who led the database review.

Since widespread use of biopsy to identify the small percentage of patients with NASH is unrealistic, Dr. Alkhouri and colleagues developed a potential screening tool that uses multiple variables to identify NASH in pediatric patients with NAFLD.

As published in Digestive and Liver Disease in late 2014, this tool evolved through a study of all patients with biopsy-proven NASH (N = 302) seen in Rome’s Bambino Gesù Children’s Hospital from 2003 through 2009. After reviewing all anthropometric, clinical and laboratory characteristics, the Italian researchers and Dr. Alkhouri developed a risk-prediction nomogram based on cholesterol, total bilirubin and waist circumference percentile. They found that a cutoff score of less than 40 percent rules out NASH with a 73.2 percent positive predictive value and a 70.7 percent negative predictive value.

The research team is currently evaluating these variables to ensure they are not specific to this cohort. Until external validation can be performed, Dr. Alkhouri is unsure whether the nomogram will be applicable to a wider population. “We do the calculations now, but I wouldn’t make major decisions based on these indices until the study can be done in an American cohort,” he says.

Breath testing: An encouraging possibility

Dr. Alkhouri is even more excited about the potential of using breath testing in children with NAFLD. In a pilot study published in 2015 involving 61 patients, he found that NAFLD produced a distinctive “breath print” that could be identified using selective ion flow tube mass spectrometry (see image at top of post).

He is now investigating whether the technology can distinguish between NASH and NAFLD. With only 10 percent of his more than 400 patients with NALFD having undergone a biopsy, validating the findings will take time. Yet he is encouraged by the potential of obtaining a quick, immediate diagnosis. “We could develop a smartphone app that would work like a Breathalyzer,” he says.

Fine-tuning a fibrosis scoring system for kids

Dr. Alkhouri recently conducted a thorough review of algorithms and fibrosis scores used to categorize the severity of NAFLD in adults, and found none to be applicable to children.

In 2009, Dr. Alkhouri’s colleagues at Bambino Gesù Children’s Hospital in Rome developed a fibrosis scoring system they called the pediatric NAFLD fibrosis index (PNFI). It proved useful for identifying mild (stage 1) fibrosis but was later found inadequate for predicting significant fibrosis.

So the team went back to work, this time with Dr. Alkhouri, examining 20 possible factors and combinations in 242 pediatric patients with biopsy-proven NAFLD. They found that a model including ALT, alkaline phosphatase, platelet count and gamma-glutamyl transferase could predict advanced fibrosis with a high degree of sensitivity and specificity. This model, published in PLoS One in 2014, is called the pediatric NAFLD fibrosis score (PNFS).

The PNFS provides a more specific cutoff value that helps determine which patients should undergo liver biopsy to confirm advanced fibrosis.

“Children who do not meet this cutoff do not need aggressive monitoring for liver-related morbidities during childhood,” Dr. Alkhouri explains, “but they should be encouraged to maintain a healthy lifestyle that includes weight loss, if appropriate.”

Homing in on a solution

A realist with a dream, Dr. Alkhouri plows on, impelled by a growing number of young patients diagnosed with NAFLD and confident that noninvasive testing methods are within reach.

He hopes that establishing an Ohio-based research network will increase the number of children with biopsy-proven disease who can be used for comparison, thereby accelerating the speed of research.

He understands that the overarching issue hampering the diagnosis of NAFLD in children is lack of understanding about the disease’s natural history. “Everything we know is extrapolated from adults,” he notes. “How — or if — the disease will progress in an 8-year-old isn’t known.”

He hopes the development of noninvasive tests will lead to answers. “There’s a huge unmet need,” he says.