Imaging Biomarker Shows Promise for Management of Chronic Liver Disease

A recent Radiology study shows the potential use of ¹⁸F fluorocholine PET/CT as a non-invasive imaging biomarker for chronic liver disease.

While the modality is already increasingly recognized as useful in detecting some cancers, the study demonstrated a significant association between ¹⁸F fluorocholine uptake and necro-inflammatory and fibrotic changes typically found in patients with liver disease.

“Chronic liver disease is a major cause of mortality worldwide, and over 90 percent of liver cancers are associated with it, so we felt it was worthwhile to pursue the topic of molecular imaging in the disease,” said the study’s lead author, Sandi Alexander Kwee, MD, PhD, an associate professor at the University of Hawaii Cancer Center and John A. Burns School of Medicine and program director for PET research at Hamamatsu/Queen’s PET Imaging Center in Honolulu, HI.

Dr. Kwee and his team were conducting a clinical trial evaluating ¹⁸F fluorocholine PET/CT for liver cancer when they recognized a chance to look beyond its utility in cancer assessment.

“We collected tumor samples from patients after surgery, and these samples contained the liver tumor and adjacent liver tissues, giving us the opportunity for radiologic-pathologic correspondence to study chronic liver disease along with liver cancer,” he said.

The team performed preoperative ¹⁸ F fluorocholine PET/CT on 48 patients with Child-Pugh class A or B disease who had resectable liver tumors. The relatively small population of the island created certain challenges with the study, Dr. Kwee said.

“The accrual of subjects for this trial took approximately five years to complete,” Dr. Kwee said. “A fair amount of effort was required to maintain reasonable consistency in imaging, biospecimen collection, etc. over the study period to minimize potential biases.”

The researchers obtained mean liver standardized uptake value (SUVmean) measurements from the PET images to assess liver choline metabolism. They also applied several histologic indexes to non-tumor liver tissue from the specimens to evaluate the presence of inflammation and fibrosis as well as other features common to chronic liver disease.

The researchers found significant correlation between liver SUVmean and Knodell histologic activity index (HAI), aspartate aminotransferase level and alanine aminotransferase level. They did not find significant correlation between albumin-bilirubin score, fibrosis-4 index, Model for End Stage Liver Disease score, albumin level, total bilirubin level or age.

Multi-Institutional Study Needed

The researchers analyzed the differences between individually scored components of the HAI as ordinal classifiers of necro-inflammation and fibrosis and identified significant variance in liver SUVmean among groups stratified by portal inflammation, piecemeal necrosis and fibrosis. Significant differences in liver SUVmean values were also noted between Metavir fibrosis stage F0 and higher, including F1, which indicates early fibrosis.

“As a somewhat skeptical researcher, I was surprised that the measurements obtained with PET showed reasonably good accuracy for discriminating early liver fibrosis, since detecting mild liver disease is often problematic for existing methods,” Dr. Kwee said. “But because our study comprised a very narrow patient group, these results need to be validated across a broader cohort of patients.”

He suggested a rigorous, multi-institutional study be conducted to assess the generalizability of their findings. “Emerging treatments and approaches that could potentially reverse liver fibrosis and optimize liver transplant selection may justify the development of imaging techniques to anatomically map and quantitatively track liver dysfunction,” Dr. Kwee said.

However, he acknowledged that cost-effectiveness needs to be proven with any expansion of clinical use of PET.

For non-oncological applications such as measuring organ function, he noted that a conventional PET/CT or PET/MRI design may not be necessary. Instead, more cost-efficient implementation of PET could be pursued without the CT or MRI component.

“With modern, sensitive PET detectors and a PET tracer with rapid biodistribution kinetics such as 18F fluorocholine, radiation exposure is further reduced while speeding up the exam,” Dr. Kwee said. “I hope to see novel scanner development being informed by the potential for broader clinical applications of PET.”