PFIC = Progressive familial intrahepatic cholestasis

PFIC1 = Progressive familial intrahepatic cholestasis type 1 = Byler disease = FIC1 deficiency

PFIC2 = Progressive familial intrahepatic cholestasis type 2 = Byler syndrome = BSEP deficiency

PFIC3 = Progressive familial intrahepatic cholestasis type 3 = MDR3 deficiency

Progressive familial intrahepatic cholestasis (PFIC) refers to heterogeneous group of autosomal recessive liver disorders of childhood in which cholestasis of hepatocellular origin often presents in the neonatal period or first year of life and leads to death from liver failure at ages usually ranging from infancy to adolescence 12. Recent molecular and genetic studies have allowed the identification of genes responsible for three types of PFIC and have shown that they were related to mutations in hepatocellular transport system genes involved in bile formation (Figure 1).

In our experience, PFIC represents 10 to 15% of causes of cholestasis in children and 10 to 15% of liver transplantation indications in children. PFIC1 and PFIC2 represent 2/3 of cases of PFIC, and PFIC3 1/3 of cases 3. The true incidence of PFIC is not precisely known, but PFIC is considered a rare disease with an estimated incidence of 1/50,000 to 1/100,000 births. All types of PFIC exist worldwide. Both sexes seem to be equally affected.

Main characteristics of PFIC 1–3 are summarized in Table 1.

Cholestasis is a major clinical sign in all PFIC forms.

It usually appears in the first months of life in patients with PFIC1, and is characterized by recurrent episodes of jaundice, which become permanent later in the course of the disease. In case of PFIC2, the initial presentation and the evolution seem to be more severe, with permanent jaundice from the first months of life and rapid appearance of liver failure within the first years of life. Early hepatocellular carcinoma (before one year of age) may complicate the course of PFIC2. Severe pruritus is usually observed in PFIC1 and PFIC2 patients. Phenotypic findings in PFIC1 and PFIC2 are similar, but some slight phenotypic difference have been identified 145678. Extrahepatic features that have been reported in PFIC1 patients (persistent short stature, sensorineural deafness, watery diarrhea, pancreatitis, elevated sweat electrolyte concentration, liver steatosis) 9, have not been described in PFIC2. In PFIC3, clinical signs of cholestasis are noted within the first year of life in about one third of patients and rarely in the neonatal period, in contrast to PFIC1 and 2. PFIC3 may also manifest later in infancy, in childhood or even in young adulthood. Gastrointestinal bleeding due to portal hypertension and cirrhosis is the presenting symptom in adolescent or young adult patients. Pruritus is usually mild. Evolution is characterized by chronic icteric or anicteric cholestasis, portal hypertension and liver failure. In half of the patient, liver transplantation is required at a mean age of 7.5 years. No liver tumor has yet been reported in association with PFIC3 10.

Patients with PFIC1 and PFIC2 have normal serum gamma-glutamyltransferase (GGT) activity, normal serum cholesterol level and very high serum bile acid concentration. Although PFIC1 and PFIC2 share similar laboratory findings, PFIC2 patients have higher transaminase and alpha-fetoprotein serum levels at diagnosis than those with PFIC1 1456789. Patients with PFIC3 have a persistent high serum GGT activity, normal serum cholesterol level and moderately raised concentrations of serum primary bile salts.

In PFIC1 patients, liver histology is characterized by canalicular cholestasis and the absence of true ductular proliferation with only periportal biliary metaplasia of hepatocytes. In PFIC2 patients, the same signs are seen but the liver architecture is more perturbed, with more pronounced lobular and portal fibrosis and inflammation. Hepatocellular necrosis and giant cell transformation is also much more pronounced in PFIC2 than in PFIC1 and may persist with time. These differences between PFIC1 and PFIC2 likely reflect the severe lobular injury present in PFIC2 (Figure 2A–B) 189.

In PFIC3, liver histology, obtained at the time of diagnosis (Figure 2C), shows portal fibrosis and true ductular proliferation with mixed inflammatory infiltrate. In a few instances, cholestasis is present in the lobule and in some ductules containing bile plugs. Slight giant transformation of hepatocytes can be observed. Cytokeratin immunostaining confirms the strong ductular proliferation within the portal tract. At a later stage (Figure 2D), there is extensive portal fibrosis and a typical picture of biliary cirrhosis. Interlobular bile ducts are seen in most portal tracts and there is neither periductal fibrosis nor biliary epithelium injury 11.

PFIC should be suspected in children with a clinical history of cholestasis of unknown origin after exclusion of other main causes of cholestasis, (e.g. biliary atresia, Alagille syndrome, alpha1antitrypsine deficiency, cystic fibrosis, sclerosing cholangitis and extrahepatic bile duct obstruction) 11. High serum bile acid concentration excludes primary disorders of bile acid synthesis 58. Patients with PFIC1 and PFIC2 have normal serum GGT activity, while patients with PFIC3 have high serum GGT activity. PFIC3 patients can also be distinguished from patients with the other types of PFIC (PFIC1 and PFIC2) in that they present rarely with cholestatic jaundice at the neonatal period, but rather later in infancy, in childhood or in young adulthood. A schematic approach to the diagnosis of PFIC is proposed in Figure 3. This combined clinical, biochemical, radiological and histological approach associated with liver immunostaining and biliary lipid analysis, should help to select PFIC candidates in whom a molecular diagnosis can be proposed.

Several other autosomal recessive PFIC-like diseases are known. Liver diseases resembling PFIC with normal GGT activity have been identified as inborn errors in primary bile acid synthesis and represent distinct disorders 58. Other liver diseases resembling PFIC with normal serum GGT activity are recognized. Familial Amish hypercholanemia represents a PFIC-like disorder not due to a primary defect of the transport system involved in bile formation, but to a tight junction protein defect combined with a defect of primary bile acid conjugation 6364. Cholestasis is due to impaired transport of unconjugated bile acids into bile and to bile leakage into plasma through abnormal canalicular tight junctions increasing paracellular permeability. Another category of progressive cholestastic liver disease of childhood could be due to abnormal villin expression, leading to loss of structural integrity of canalicular microvilli impairing biliary secretion system function 65. A case of hypercholanemia due to mutations in the m-epoxide hydrolase gene has also been reported (EPHX1) 6667. Arthrogryposis-renal dysfunction cholestasis syndrome is a complex disease due to mutation of VPS33B involved in intracellular trafficking and targeting of apical proteins. The gene defect results in a loss of apical protein expression in the liver and kidneys 68. From a liver point of view, it represents another PFIC-like disorder with normal GGT activity. In cases of high serum GGT activity, sclerosing cholangitis must be excluded 5969. North American Indian Childhood Cirrhosis due to a defect in the cirrhin gene is characterized by high GGT activity 70 as is Aagenes syndrome, a lymphedema cholestasis syndrome of unknown cause 71.

Genotyping should be used to confirm the diagnosis of PFIC in affected children. Heterozygosity of parents for the defects found in affected patients confirmed the recessive inheritance of the disease. This understanding has already allowed prenatal diagnosis. Antenatal diagnosis of PFIC requires clinical and biochemical expertise 56.

Therapy with UDCA should be considered in the initial therapeutic management of children with all types of PFIC 72. UDCA therapy may be effective in some patients, especially those with PFIC3 with missense mutations who have less severe disease in comparison to children with a mutation leading to a truncated protein 11.

Some patients with PFIC1 or PFIC2 may also benefit from surgical biliary diversion 7374. Nasobiliary drainage may help to select potential responders to biliary diversion 75. So far, clear genotype-phenotype correlation data are missing and remain to be defined in order to identify those PFIC1-2 patients who could benefit from UDCA or biliary diversion 360. Preliminary data suggest that PFIC2 patients with p.D482G or p.E297G mutations may respond well to biliary diversion 60.

If these therapies fail, liver transplantation represents the only alternative 76. In our experience, extrahepatic features, such as diarrhea, liver steatosis and short stature sometimes associated with PFIC1, do not improve or may be aggravated after successful biliary diversion or liver transplantation 9. Chronic diarrhea may become intractable when biliary bile salt secretion is restored after liver transplantation 6920. It is often associated with severe liver steatosis and/or steatohepatitis that may lead to cirrhosis with time and to indication to retransplantation. Liver steatosis and diarrhea may recur after retransplantation (E. Jacquemin, personal data). Diarrhea might be favorably managed by bile adsorptive resin treatment 920.

In the future, therapies such as cell, gene or specific targeted pharmacological therapies (i.e. FXR inducers, chaperone drugs), might represent an alternative therapy for all types of PFIC 2760777879. In PFIC2, it is uncertain whether hepatocyte transplantation or gene therapy with modified hepatocytes represents a good therapeutic approach. Indeed, with this approach, it may be a risk to leave premalignant liver cells in place, especially in patients with severe biallelic ABCB11 mutations 80.

In addition to the natural history and complications of PFIC1-3 (portal hypertension, liver failure, cirrhosis, hepatocellular carcinoma, extrahepatic manifestations), children with PFIC1-3 are theoretically at risk of developing further in the disease course, biliary stones, drug-induced cholestasis and ICP. Girls under UDCA therapy who reach adulthood with their native liver must not stop UDCA during pregnancy because of the risk of developing severe ICP as seen in a previously reported patient 53 who became pregnant (personal data Dr N. Ganne-Carrié, Hepatology Unit, Jean Verdier Hospital, Bondy, France).

Patients with BSEP deficiency, especially those with biallelic truncating mutations, are at considerable risk for hepatobiliary malignancy (15% may develop hepatocellular carcinoma or cholangiocarcinoma) [2580 and E. Jacquemin personal data]. These findings justify close monitoring of hepatocellular carcinoma (At least, serum alpha fetoprotein dosage every 6 months and liver ultrasonography every year) in PFIC2 patients from the first year of life.

In very rare PFIC patients (< 10%), only one mutated allele or no mutation is identified. This can be explained by mutations that may map to regulatory sequences of the genes. A gene involved in the transcription of PFIC genes (i.e. FXR) or in protein trafficking could also be involved 8182. It is also possible that other unidentified genes involved in bile formation may be responsible for the PFIC1-3 phenotypes. Furthermore, it may be hypothesized that combined heterozygous mutations for two genes (i.e. MDR3 and BSEP) lead to PFIC like phenotype 49. An interesting possibility is also that in a heterozygous state, the mutated protein may have a dominant negative effect on protein expression/function 83. Modifier genes and environmental influences could play a role in the expression of PFIC 58.

After liver transplantation the possible "recurrence of PFIC" on liver grafts due to alloimmunization of the recipient against the BSEP, MDR3 or FIC1 proteins of liver donor remains a theoretical matter of debate. It is hypothesized that PFIC patients with a severe mutation leading to absence of the gene product would be immunologically naive for the FIC1, BSEP or MDR3 gene products. In this context, alloimmunization could occur after liver transplantation. To our knowledge, this complication has not been reported so far after liver transplantation 7684 but we have observed a case of an PFIC2 male patient who experienced an unexplained severe bout of pure hepatocellular cholestasis resembling PFIC2 after cadaveric liver transplantation (E. Jacquemin, personal data). In addition, in case of parental living donor transplantation, it could be expected that the heterozygous status of the liver graft leads to a predisposition for developing lithiasis or cholestasis favored by immunosuppressive drugs 33 that may interfere with canalicular protein function, as we have seen in a PFIC2 patient (E. Jacquemin, personal data). Nevertheless, this complication is likely to be very rare, since it has not been reported in a broad series of living donor transplantation for PFIC 84.

As already mentioned above, studies performed in cell lines have determined the effects of some PFIC gene mutations (ATP8B1 171819, ABCB11 262829, ABCB4 414344). This should allow testing of some drugs that may restore, at least partially, the function or the targeting of a mutated protein 277879. Besides these cell studies, mouse models of PFIC1-3 have been created by genetic engineering 858687. Mainly, the PFIC1-2 models have demonstrated that the phenotypes of the mice were much less severe than in humans. Although the phenotypes were aggravated by supplementation with cholic acid, the main bile salt in humans, this was not sufficient to induce liver injury comparable to that seen in PFIC1-2 patients. The likely reason for this is that mice, unlike humans, may easily detoxify hydrophobic bile salts by hydroxylation, via the cytochrome P450 enzyme system, resulting in a less cytotoxic bile salt pool. Therefore, it would be interesting to generate new mouse models that closely resemble PFIC1-2. Recently, cytochrome P450 knockout mice have been generated 88. It is expected that crossing PFIC mouse models with mice with a disrupted liver cytochrome P450 oxidoreductase gene will generate mice sensitive to human bile salts. Such new mouse models could develop liver diseases that closely reproduce PFICs and will allow better experimental studies of the pathophysiology of PFICs and test the effects of treatments.

PFIC: Progressive familial intrahepatic cholestasis; GGT: gamma-glutamyltransferase; BSEP: bile salt export pump protein; UDCA: Ursodeoxycholic acid; ICP: intrahepatic cholestasis of pregnancy; BRIC: benign recurrent intrahepatic cholestasis; ASBT: apical sodium bile salt transporter; CFTR: cystic fibrosis transmembrane conductance regulator; NBFs: nucleotide binding folds; MDR3: Class III Multidrug Resistance; P-gp: P-glycoprotein; MRS: Magnetic Resonance Spectroscopy; FXR: farnesoid X receptor.

The authors declare that they have no competing interests.

The authors equally contributed to this review article. They read and approved the final version of the manuscript.