Article Text

Download PDFPDF

A novel twist in polycystic liver disease
  1. Wybrich R Cnossen1,
  2. Joost PH Drenth1
  1. Department of Gastroenterology and Hepatology, Nijmegen, The Netherlands
  1. Correspondence to Professor Joost PH Drenth, Department of Gastroenterology and Hepatology, Radboudumc, P.O. Box 9101, code 455, Nijmegen 6500 HB, The Netherlands; Joostphdrenth{at}

Statistics from

Liver cysts are a frequent finding on radiological imaging. Prevalence rates depend on the technique, but with CT or MRI, simple hepatic cysts are seen in approximately one-fifth of the population.1 Usually no more than one or two small (<3 cm) cysts are present. These cysts are innocuous and do not require immediate clinical attention. The situation is different when there are multiple and/or large liver cysts such as in polycystic liver disease (PLD). Though PLD is infrequent, patients are seen in every gastroenterology practice. Many patients give a positive family history of PLD. The majority of PLD patients will also have multiple renal cysts and suffer from autosomal dominant polycystic kidney disease (ADPKD). This is a progressive disorder that leads to end-stage renal disease. By contrast, some PLD patients do not have that renal phenotype and they are affected by isolated autosomal dominant polycystic liver disease (PCLD). While the genetic background of ADPKD and PCLD is dissimilar, there is great analogy in their phenotypical behaviour. PLD is characterised by an age-dependent increase of number and size of the liver cysts.1 Within this context, the physiological function of the liver is usually unaffected, but because of the unrelenting growth, significant hepatomegaly develops over time. This is associated with onset of symptoms, such as abdominal pain, early satiety, and dyspnoea which causes a compromised quality of life.2 Thus far, the primary endpoint of treatment is reduction of liver cyst volume as it is thought that this improves symptoms. Reduction of cyst burden can be achieved mechanically. There are a number of surgical and radiological interventions available that may be effective in selected cases.3 However, in patients with advanced PLD, these procedures are less effective or are prone to complications, hence, the need for medical treatment options. In order to better understand how we can target cysts, it is paramount to consider the genesis of these lesions. There are three elements that are required for cystogenesis. First, the right set of mutated genes; second, an appropriate scaffold and last adequate fluid production to fill the cyst.4 There has been impressive progress in the understanding of which genetic abnormalities are implicated in PLDs, but development of our understanding how a cyst develops and is maintained, has been less expeditious. This is, in most part, due to a lack of appropriate cell models for PLD. The discovery of an animal model (the PCK rat) that recapitulates many aspects of PLD has fuelled the field, and this signals another key development.5 Urribarri et al have isolated normal and cyst-derived cholangiocytes and were able to keep these cells over a number of passages while retaining the original features of the cell. They went on to elucidate the role of matrix metalloproteinases (MMPs) in hepatic cystogenesis. Why were they so interested in MMPs? PLD probably arises through defective development and maturation of the intrahepatic biliary tree resulting from defective dedifferentiation of cholangiocytes, and inappropriate secretion of cytokines. MMPs are expressed in the ductal plate (the bile duct nidus) and are required to initiate expansion of the liver during development. Under normal physiological conditions the biological activity of MMPs is low in most tissues. A key observation Urribarri et al made was the finding of persistent MMPs hyperactivity in cultured human and PCK rat cholangiocytes. This was associated with increased MMP gene expression and presence of endogenous MMP inhibitors. They went on to discover that cyst fluid components drive MMP hyperactivity. Specifically IL-6 and IL-8, and 17β-estradiol independently stimulated MMP-activity, probably through an autocrine/paracrine mechanism. They then directed their attention to human tissue in order to assess which MMP is most relevant to PLD. They found that MMP-3 was grossly overexpressed in cyst tissue samples from PLD patients and PCK rats. Final proof of the pivotal role of the MMPs in PLD came from experiments with marimastat, a MMP inhibitor. Oral administration of marimastat in a clinically relevant dose significantly curtailed liver cyst formation in young PCK rats. Importantly, no harmful/toxic effects of marimastat were reported. The findings of Urribarri et al are a remarkable deviation from the path researchers so far have been travelling. Earlier studies have focused on calcium signalling, cAMP production and MAPK/ERK signalling in cyst cholangiocytes.4 Indeed, cholangiocytes overproduce cAMP which can be repressed by somatostatin analogues (such as lanreotide or octreotide). Reducing cAMP diminishes cell proliferation, diminishes fluid hypersecretion and inhibits hepatorenal cystogenesis in experimental models. 6 As a corollary, clinical trials have demonstrated the efficacy in PLD as they consistently lower liver volumes in PLD. 7 The study reported here follows a new avenue as it focuses on the role of MMPs in PLD. This is the first evidence that suggests that pharmacological inhibition of MMP can be used for therapeutical purposes in PLDs. The authors used marimastat which specifically binds to the catalytic domain of MMPs and particularly inhibits MMP-1, MMP-2, MMP-3, MMP-7, MMP-9 and MMP-14. Apart from their role in PLDs, MMPs have been implicated in cancer, but also to contribute to the pathophysiology of vascular diseases such as atherosclerosis. Marimastat covalently binds to the zinc ion at the active site of MMPs and inhibits extracellular matrix degradation, angiogenesis, tumour growth and invasion, and metastasis. These data suggest that marimastat may be used as a therapeutical option in PLD, but before doing so we must examine its pharmacological characteristics. Urribarri et al report that an 8-week course was tolerated by the PCK rats. This finding is consistent with data in healthy volunteers where 1 week administration was uneventful apart from small, reversible increases in liver transaminases. 8 However, clinical trials with marimastat in cancer patients have been stopped because of musculoskeletal toxicity which affected >90% of patients. These side effects consisted of arthralgias and myalgias and improved following dose reduction or stopping of the drug.9 While the dose used in the PCK rats equals that used in clinical trials, musculoskeletal toxicity requiring dose modification only became apparent after prolonged use (mean of 133 days).9 As indicated, most PLD patients have concomitant polycystic kidneys. It would be ideal for a drug to target renal and liver phenotype. Unfortunately, marimastat did not affect renal cystogenesis in their PCK rats. A possible explanation for the absence of extrahepatic systemic effects is the high first-pass effect of oral marimastat in rodents. This leads to low systemic levels of marimastat. Support for this line of thinking comes from observations with another MMP inhibitor batimastat. Intraperitoneal injections with batimastat resulted in a significant reduction of cyst number and kidney weight in a mouse model for ADPKD.10 Altogether, these data provide a fresh look on the pathogenesis of PLDs and provides us, clinicians, with a potentially useful therapeutical tool.



  • Contributors WRC and JPHD contributed to the choice and development of the concepts expressed in this paper. JPHD drafted the paper, WRC worked on subsequent versions.

  • Competing interests WRC is supported by a grant of the Institute for Genetic and Metabolic Disease (IGMD) of the Radboudumc.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles