Clinical Benefit of Intermediate-Dose of Steroid In Chronic Graft-Versus-Host-Disease of the Liver Presented with the Elevation of Serum Aminotransferases

Chronic graft-versus-host disease (GVHD) of the liver is generally known as cholestatic jaundice. Although there are many aetiologies of elevated hepatic enzyme after hematopoietic stem cell transplantation (HSCT), such as iron overload and hepatitis B and/or C, acute hepatic enzyme elevation has been recently considered as a manifestation of chronic liver GVHD. Here we compare the incidence and the therapeutic response of chronic liver GVHD between two phenotypes; hepatic enzyme elevation type and hyperbilirubinemia type.

We reviewed the data of 402 patients who underwent allogeneic HSCT at our institute from 1995 to 2009, and searched the cases that satisfied the following criteria: the serum aminotransferase (aspartate aminotransferase (AST) and/or alanine transaminase (ALT)) levels were over 3 times above their normal ranges while the bilirubin level remained within the double of its normal range (hepatic enzyme elevation type), or the bilirubin level was over twice above its normal range while the serum aminotransferase levels remained within 3 times of their normal ranges (hyperbilirubinemia type). We interpreted that the treatment was successful when AST, ALT and bilirubin satisfied neither of the criteria above.

Of the 402 patients, 76 patients (18.9%) had chronic GVHD of the liver (hepatic enzyme elevation type in 65 (16.2%) cases and hyperbilirubinemia type in 11 (2.7%) cases). One-year cumulative incidence of chronic liver GVHD were 16.3±1.9% in hepatic enzyme elevation type, and 3.0±0.9% in hyperbilirubinemia type. Among the patients with hepatic enzyme elevation type, median time of the onset was at 161 days (range, 60–510 days) after transplantation. Median peaks of serum AST, ALT and bilirubin were 260 IU/L (74-3,492 IU/L), 422 IU/L (89-2,236 U/L), and 1.7 mg/dL (0.5-38.4 mg/dL), respectively. Thirty patients had recently ceased all immunosuppressive drugs and 35 patients had been tapering in line with standard clinical practice. Liver biopsies were undergone in 10 patients and showed damaged and degenerative interlobular bile ducts, one of characteristic features of GVHD. Other features, such as lobular hepatitis and many acidophilic bodies scattered throughout the lobules, were also detected. Regarding viral infection, CMV immunohistochemistry and EBER in situ hybridization were negative. Several treatments were performed for 65 patients; increased dosage of calcineurin inhibitor (CI; Cyclosporine or Tacrolimus) for 25 patients, prednisolone (PSL) alone for 18 patients, CI plus PSL for 20 patients, and only two patients were taken a wait-and-see approach. Next we evaluated the therapeutic response in 60 patients, since five died soon after onset. Median time until treatment success were 31 days (3-85 days) and 63 days (7-259 days) in patients with or without PSL, respectively (n=35 vs 25, p<0.0001). Further, we analyzed the dose-effect of PSL. Median time until treatment success were 25 days (3-85 days) in patients received 1 mg/kg or more PSL (n=25), and 34.5 days (19-72 days) in patients received under 1 mg/kg PSL (n=10) (p=0.62). We also compared the difference of response to steroid between hepatic enzyme elevation type and hyperbilirubinemia type. Three of 38 (7.9%) patients resulted in treatment failure in hepatic type, while four of eight (50%) patients in hyperbilirubinemia type (P=0.012).

In our study, hepatic enzyme elevation type is more frequent than hyperbilirubinemia type in chronic liver GVHD. Intermediate-dose steroid can prevent progressive hepatocellular injury and bile duct destruction in hepatic enzyme elevation type, resulting in good outcome compared with in hyperbilirubinemia type.

No relevant conflicts of interest to declare.