A substantial proportion of pediatric liver-transplant patients with normal liver tests have varying degrees of subclinical chronic allograft injury, according to new findings.

"This study gives us possible markers that may help predict the long-term outcomes of children who have had liver transplants and identified their role in immunosuppression," Dr. Vera F. Hupertz, medical director of pediatric hepatology and transplantation at Cleveland Clinic Children's in Cleveland, Ohio, told Reuters Health by email. She was not involved in the research.

For the study, Dr. Sandy Feng of University of California, San Francisco, and her colleagues used data collected for iWITH, an immunosuppression-withdrawal trial that included 157 stable long-term recipients of liver transplants from living or deceased donors (45% boys).

Participants were <18 years old and underwent transplants for non-viral and non-autoimmune liver disease when they were no more than six years old. They were at least four years from surgery (mean, 8.9 years), with official liver-test results for more than four years post-transplant.

The children had alanine aminotransferase and gamma-glutamyltransferase <50 IU/L and were stable on calcineurin inhibitor monotherapy without rejection during the past two years. They underwent liver biopsies at 12 North American pediatric liver-transplant centers between 2012 and 2014.

The researchers used unsupervised cluster analysis of histologic features to sort biopsies into three clusters. They also conducted transcriptional and cytometric analyses of liver tissue samples and a systems-biology study that incorporated clinical, serologic, histologic and transcriptional data.

Alanine aminotransferase

The mean level of alanine aminotransferase in study participants was 27.6 U/L, and that of gamma-glutamyl transferase was 17.4 U/L, the team reports in Gastroenterology, online August 22.

Interface activity characterized cluster 1 (n=34), whereas periportal or perivenular fibrosis without interface activity was characteristic of group 2 (n=45) and cluster 3 lacked both factors (n=78).

The team found a module of genes whose expression correlated with levels of alanine aminotransferase, class II donor-specific antibody, portal inflammation, interface activity, perivenular inflammation, portal and perivenular fibrosis, and cluster assignment.

This module was enriched with genes that regulate T-cell mediated rejection (TCMR) of liver and other transplanted organs. Functional pathway analysis found over-representation of TCMR gene sets in the cluster 1 subgroup only.

Dr. Hupertz said, "decisions on changing immunosuppression, particularly lowering or withdrawing it, should not be based on liver enzymes, but rather on what is happening at the microscopic tissue level."

"This study supports being very careful with immunosuppression or withdrawal without getting a tissue sample," she said, adding that it should probably not be lowered if there's evidence of inflammation or fibrosis.

Dr. Douglas Mogul, medical director of pediatric liver transplantation at Johns Hopkins Medicine in Baltimore, Maryland, said the mainstays of current patient management and liver-health assessment are blood tests or imaging.