New therapeutics in primary biliary cirrhosis: will there ever be light?


  • Pietro Invernizzi,

    1. Liver Unit and Center for Autoimmune Liver Diseases, Humanitas Clinical and Research Center, Rozzano, MI, Italy
    2. Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis, Davis, CA, USA
    Search for more papers by this author
  • M. Eric Gershwin

    1. Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis, Davis, CA, USA
    Search for more papers by this author

Despite the fact that primary biliary cirrhosis (PBC) is the most common autoimmune liver disease and considered the model for all other organ-specific autoimmune diseases, there remains an enormous gap between our knowledge at the research bench and newer therapeutics for patients. In fact, we know more about the basic biology of PBC than virtually any other autoimmune disease, including rigorous definitions of the signature antimitochondrial autoantibodies, quality large scale epidemiological surveys, definition of autoreactive CD4 and CD8 responses, mapping of the immunodominant epitopes and a number of genome-wide association studies [1-8]. There are also several animal models of PBC, including models that develop fibrosis [9, 10]. However, there has not been a new drug approved for PBC for more than 20 years and indeed researchers and physicians have not reported any significant success in treating either the animal models or patients with any of the newer biologics; the one possible exception is the use of CTLA4 Ig [11-13], which appears successful in murine studies but has not yet been applied to patients. These data, on the basic autoimmunology of PBC, can be contrasted with similar data on our knowledge of autoimmunity in other human diseases characterized by a breach of tolerance [14-18], yet many of those diseases have had considerably more success in developing specific biological therapies. Interestingly, although ursodeoxycholic acid (UDCA) remains the only drug and is standard of care in the treatment of PBC, there are a significant number of patients who do not respond and indeed there is datum suggesting that its efficacy may be more cosmetic than hepatologists would like to believe. In any case, there is a great need for new therapies and in this respect we should note that beyond UDCA, there are no recognized second-line treatments available.

There is a long history on the use of fibrates in human disease, including agents such as bezafibrate (BZ) and fenofibrate (FF); both are fibric-acid derivatives used as lipid-lowering drugs to prevent cardiovascular pathology. Indeed, the effects of fibrates on cardiovascular morbidity appear secondary to its effectiveness in reducing LDL and triglycerides and improving HDL levels. In this respect, the effects are more dramatical in patients suffering with the metabolical syndrome. In addition, there is also data that demonstrate that BZ may delay progression to diabetes in patients with impaired glucose tolerance and also may slow progression in those patients with insulin resistance. Interestingly, the mechanism of action by which fibrates alter lipid profiles is not fully defined, although it is known that fibrates are ligands of the nuclear peroxisome proliferator-activated receptor (PPAR), primarily the PPAR-α but also the other two isoforms of PPAR, i.e. PPAR-δ and PPAR-γ [19]. While BZ is an agonist of the three isoforms, FF acts selectively on PPAR-α and along with gemfibrozil, is now approved by the US Food and Drug Administration for the treatment of hyperlipidemia. We should note that in addition to altering lipid metabolism, fibrates exert anti-inflammatory effects on multiple lineages of cells and also functions as a pregnane X receptor (PXR) agonist [20]. Indeed, there is increasing evidence that suggests that fibrates have a multifactorial mechanism of action that clearly requires more functional studies if it is to be fully understood.

There is a long history on the potential use of BZ for reduction in alkaline phosphatase (ALP). In fact, the effectiveness of BZ on the reduction of ALP as well as immunoglobulin M (IgM) was first reported by Iwasaki et al. [21] in 1999, alone or in combination with UDCA. Since that time, there have been a number of case reports and pilot clinical studies that substantially confirmed the original data in PBC; several studies also demonstrated the efficacy of other fibrates such as FF [20-38] (Table 1). These studies consistently report that both BZ and FF reduce ALP, gamma-glutamyl transpeptidase (GGT), alanine aminotransferase (ALT) and IgM levels. An additional observation of these studies, as expected, is that these agents also produce a reduction in triglycerides and cholesterol and therefore improve lipid profiles; this is particularly important in patients with PBC because of their hyperlipidemical status. Interestingly, fibrates also are reported to be beneficial on the common symptoms of fatigue and pruritus, without significant side effects. Despite these latter comments, we should note that the vast majority of these studies contain small numbers of patients, include short treatment times, are retrospective, do not include histology, have not focused on quality of life and survival and have not always adequately defined the dose and duration of concurrent use of UDCA.

Table 1. Clinical studies with fibrate (bezafibrate and fenofibrate) in primary biliary cirrhosis
AuthorCountry (year)FibrateFibrate Dose (mg/day)UDCA Dose (mg/day)Patients Fibrate (n), UDCA + fibrate (n), UDCA (n)SafetyEfficacyRef
  1. BZ, bezafibrate; FF, fenofibrate; UDCA, ursodeoxycholic acid; ALP, Alkaline phosphatase; GGT, gamma-glutamyl transferase; IgM, Immunoglobulin M; IgG, Immunoglobulin G; AMA: anti-mitochondrial antibodies; TG, tryglycerides.

Iwasaki et al.Japan (1999)BZ40011, /, /NoneReduced ALP, IgM [21]
Kurihara et al.Japan (2000)BZ400600/, 11, 11NoneReduced ALP, GGT [37]
Miyaguchi et al.Japan (2000)BZ400600/, 11, /Not reportedReduced ALP, ALT, IgG, IgM [22]
Nakai et al.Japan (2000)BZ400 /, 13, 11NoneReduced ALP, GGT, IgM [23]
Ohmoto et al.Japan (2001)BZ400600/, 11, 6Not reportedReduced fibrosis markers [24]
Ohira et al.Japan (2002)FF150/200600–900/, 7, 7NoneReduced ALP, GGT, IGM, symptoms [25]
Yano et al.Japan (2002)BZ4009001, 1, /NoneReduced GGT, ALT, IgM [26]
Kanda et al.Japan (2003)BZ400600/, 11, 11PolydipsiaReduced ALP, symptoms [27]
Dohmen et al.Japan (2004)FF100/150600/, 9, 9NoneReduced ALP, GGT, IgM, AMA [28]
Itakura et al.Japan (2004)BZ400600/, 8, 8Not reportedReduced ALP, GGT, TG [29]
Kita et al.Japan (2006)BZ400600/, 22, 17NoneReduced ALP, GGT, ALT, IgM [30]
Ohomoto et al.Japan (1996)BZ400600/, 10, 10Not reportedReduced ALP, GGT, ALT, IgM, symptoms [31]
Iwasaki et al.Japan (1998)BZ400600/, 32, 35Abdominal painReduced ALP, GGT, IgM, TG, cholesterol [32]
Walker et al.UK (2009)FF134–200Not reported/, 16, /Not reportedReduction ALP, IgM [33]
Hazzan et al.Israel (2010)BZ400900–1500/, 8, 8Not reportedReduced ALP, GGT [34]
Levy et al.USA (2011)FF16013–15 mg/kg/, 20, 20HeartburnReduced ALP, IgM, AST, TG [35]
Takeuchi et al.Japan (2011)BZ40012–15 mg/kg/, 15, 22NoneReduced ALP, IgM, TG, cholesterol [36]
Han et al.China (2012)FF20013–15 mg/kg/, 22, /NoneReduced ALP, GGT, ALT, TG cholesterol [38]
Honda et al.Japan (2013)BZ400600/, 19, /Not reportedReduced ALP, GGT, IgM, TG, cholesterol [20]
Lens et al.Spain (2013)BZ40013–16 mg/kg/, 30, /NoneReduced ALP, GGT, IgM, TG, cholesterol [39]

In the current issue of Liver International, Lens et al. [39] fill a considerable gap in our knowledge on fibrates and improve on the quantity and quality of data presented in the multiple studies cited in Table 1. In particular, the data reported confirmed a significant decrease in cholestasis and serum transaminases, as well as a beneficial effect on pruritus during their one-year therapeutical trial with BZ. In fact, a major strength of this study is the three-month BZ discontinuation period, during which liver function tests significantly increased along with recurrence of pruritus; these then improved after restarting BZ. Overall, the data reported suggest that fibrates are a potential second-line treatment of PBC when used in combination with UDCA in patients with an inadequate UDCA response. We do urge prospective large cohort studies to evaluate the effect of fibrates not only on biochemical markers, but also on histology, quality of life, need for transplantation and survival. In fact, it will be particularly critical to combine this data with well defined subsets of patients with PBC with various disease features, including staging, duration of disease, gender, concurrent sicca, fatigue and pruritus in order to fully define which patients are most fit to respond. PBC is relatively uncommon and such studies will require a multiteam and multicentrical effort to produce results of high quality that our patients need and deserve.


Conflict of interest: The authors do not have any disclosures to report.