We thank Walker and colleagues for their comments on our article showing that the extent of the biochemical response to ursodeoxycholic acid (UDCA) has a major impact on the prognosis of primary biliary cirrhosis (PBC), thus confirming previous data indicating that not only the fall in bilirubin levels but also in liver enzymes may have prognostic values in patients treated for PBC.1, 2 In this context, Walker and colleagues highlight the possible role of fibrates as new therapeutic agents for PBC. Indeed, they found in their cohort from Northeast England that the combination of fenofibrate and UDCA was able to markedly decrease both alkaline phosphatase and immunoglobulin M levels in patients with PBC with an incomplete response to UDCA, thus confirming recently published results from Iwasaki et al. in Japan.3 First, Iwasaki and colleagues showed that bezafibrate (400 mg/day) was as effective as UDCA (600 mg/day) in reducing liver enzymes. Second, they showed that the combination of bezafibrate and UDCA produces a higher reduction in alkaline phosphatase levels than does UDCA (600 mg/day) alone, but there were no significant differences in other biochemistries between the two therapeutic regimens. Fibrates derivatives have a 40-year history in the management of dyslipidemia. Their use led to the discovery of peroxisome proliferator-activated receptor (PPAR) genes. This class of drugs is generally well-tolerated; however, safety issues have arisen from their use. Their parsimonious use in Europe for the treatment of PBC came initially from the reports of Schaffner4 and Summerfield et al.,5 who showed that clofibrate could induce paradoxical hypercholesterolemia, intrahepatic, and choledocolithiasis in patients with PBC. There were further reports of cases of acute and chronic liver injuries described under the forms of acute cholestatic, autoimmune, chronic liver diseases and biliary tract diseases as well as acute pancreatitis mostly associated with gallstones.6-8 At the same time, occurrence of severe photosensitivity, cutaneous rash, rhabdomyolysis, polymyositis, and renal failure, in addition to the suspected possible role in human liver carcinogenesis, which was demonstrated in rodents, discouraged hepatologists from prescribing fibrates even in the case of dyslipidemia in their patients with PBC.
We have a little experience of the use of fibrates for dyslipidemia in PBC. Following the reports from Japan, the review of these cases indeed suggests that fibrates may potentiate the effects of UDCA given at a daily dose of more than 13 mg/kg body weight on alkaline phosphatase and immunoglobulin M levels. We did not find the side-effects mentioned above with the combined therapy. Fibrates are PPAR alpha receptor activators, which are known to exert an array of effects on a myriad of genes including those involved in hepatic lipid metabolism, suppression of bile acid synthesis, increased biliary secretion of phospholipids and cholesterol, inflammation, and defense against oxidative stress, all of these being processes which possibly act in the pathobiology of PBC. As for UDCA over the last 20 years, it remains to be demonstrated that the effects of fibrates will not only be “cosmetic” but really effective in terms of progression toward end-stage disease. Lessons from the UDCA story will certainly be useful to achieve this goal.