- Top of page
- Methods and materials
Methotrexate (MTX) is the most commonly prescribed systemic therapy for severe psoriasis. MTX, a folate antagonist, competitively inhibits dihydrofolate reductase, an enzyme necessary for methionine, purine and thymidylate synthesis and, ultimately, DNA synthesis. MTX possesses potent anti-inflammatory effects on T-cell mediated immune responses as it inhibits proliferation or induces apoptosis in activated T-cells and blocks the abnormal rapid epidermal cell proliferation, both responsible for the characteristic skin lesions in psoriasis.1 Long-term, weekly low-dose methotrexate therapy is associated with some serious adverse reactions as myelosuppression, interstitial pneumonitis and hepatoxicity. The pathogenesis of MTX-induced hepatic damage is poorly understood, but intra hepatocellular accumulation of a polyglutamated metabolite of MTX might be responsible for liver toxic effects. Several studies have suggested that a pre-existent liver disease, overweight, diabetes mellitus (DM) or alcohol use carries an increased risk of hepatotoxicity.2
Early studies in MTX treated psoriatic patients reported a very high prevalence of hepatotoxicity with fibrosis occurring in up to 50% and cirrhosis in up to 20%.3–5 Concerns about liver injury in psoriasis patients has led to dermatologic guidelines that stipulate monitoring periodically after every 1500 mg cumulative dose, in order to identify liver injury by means of serial liver biopsies. Liver biopsy is considered as the gold standard method in the assessment of histological changes,6 but its application is complicated by some important limitations. A liver biopsy is an invasive procedure and carries inheritent side effects such as pain and localized bleeding.7, 8 However, recent literature suggests that MTX might be significantly less hepatotoxic and reports that liver fibrosis and cirrhosis are notably less prevalent than previously assumed.9, 10 These data suggest that reconsideration of our current monitoring strategies for patients on MTX may be warranted. Recent literature proposes that procollagen III aminopeptide (PIIINP) measurements may be of additional value in some patients.11, 12
In the present study, we reviewed the clinical details and liver biopsies of 125 patients on long-term, low-dose MTX therapy for psoriasis seen in a large tertiary referral centre from 1976–2005. Our aim was to establish the prevalence of MTX-induced liver injury and to delineate potential contributing factors.
- Top of page
- Methods and materials
One of our aims was to establish the prevalence of MTX-induced liver injury because the large variation in previous literature.2, 9, 10 In this substantial cohort of 278 liver biopsies taken over 30 years MTX-induced liver fibrosis (≥Roenigk grade 3a) was only seen in 15% (19 patients) of our population. Progression to a higher stage of liver injury mostly occurred at a cumulative dose-range between 1500 and 6000 mg, which translates to at least 2 years of MTX treatment at 15 mg weekly. Progressive development of liver damage was infrequent above 6000 mg MTX (after about 8 years with 15 mg weekly) when patients have proven to tolerate high cumulative dosages MTX without previously obtaining pathological hepatic differences. In contrast to our data, others found that the cumulative probability of developing advanced hepatic fibrosis doubled after 5000 mg.9, 10
We carefully monitored our patients for MTX induced toxicity by regular liver biopsies. As such our practice mirrors the international guidelines and as such the data can be viewed as a field study for implementation of these guidelines.13, 14 As a consequence, it is unlikely that we missed patients who were treated with MTX and actually developed MTX-hepatotoxicity.
Most long-term observational studies in rheumatoid arthritis indicate that toxicities, rather than lack of efficacy, are the most common cause of discontinuing MTX.15 This appears to be in contrast with the situation in psoriasis. We and others found that low-dose, long-term MTX treatment seems to be relatively safe and effective, and that undesired side effects lead to discontinuation in a minority.16
Another aim of our study was to delineate potential contributing factors to the development of MTX-induced liver injury. Risk factors play an important role in the development of liver fibrosis. At any given cumulative MTX dosage, patients with obesity and or DM had a significantly worse liver histology compared with patients without these risk factors. The presence of fatty infiltration (steatosis) of the liver is highly prevalent among obese and or diabetic patients, and probably plays a role. This reminds us of the situation in non-alcoholic steatohepatitis (NASH) where the presence of steatosis is associated with hepatitis with necroinflammation and pericellular fibrosis.17 Indeed, a recent case series showed that the presence of NASH in psoriatic patients contributes to the hepatotoxicity of MTX.18 These data suggest that histological monitoring of MTX toxicity could be tailored to obese patients with or without DM. Other researchers have emphasized the role of alcohol consumption as a risk factor for MTX-induced fibrosis.10 Surprisingly, we could not confirm this finding in our study and is possible that the association is indeed absent in our cohort. On the other hand, the retrospective nature of our study carries inherent limitations, which may hamper the interpretation of the data on alcohol use. The retrospective nature of our study could have resulted in the under-reporting of risk factors. But even so, we found a significant correlation between the percentage of patients with liver disease and the presence of any risk factors. On the other hand, we cannot rule out the possibility of entry bias, e.g. that physicians do not consider MTX as a therapeutic option for psoriatic patients with a history of alcohol (ab)use.
We detected a correlation between grades of histological change and γ-GT. In 40% of patients with Roenigk >1, γ-GT was significantly elevated. In contrast, ASAT, ALAT, and AP serum concentrations were not elevated beyond the normal range for any of the Roenigk grades, which fits with other literature data.10, 19 One of the clinical consequences is that as a whole, normal liver enzymes do not exclude progression of liver injury. Conversely, elevated liver enzymes do not necessarily correlate with MTX induced hepatotoxicity. In this respect, liver enzymes serve as a poor predictor for MTX induced liver injury. One exception may be γ-GT levels, as our data are consistent with the notion that they may serve as a marker for MTX hepatotoxicity. Regular PIIINP measurements might be of additional value in monitoring these patients.11, 12
In our study, patients with advanced MTX-induced fibrosis and cirrhosis (Roenigk ≥ 3a) ran a benign course. Liver histology did not deteriorate (and sometimes even improved) under ongoing MTX therapy. Our relatively large series that documented a large number of biopsies over a 20-year period confirms those of two small studies.9, 10
In conclusion, MTX-induced fibrosis occurs in a minority of patients. Liver histology deterioration is seen mostly at cumulative MTX dosages between 1500 and 6000 mg, which call for close monitoring at this stage. Our study confirms that normal liver function tests do not guarantee normal liver histology. Furthermore, risk factors play an important role in the development of liver injury and they should be taken into account before the decision is made to start MTX. Obese patients with or without DM should be monitored closely by regular liver biopsies. From our studies it is difficult to indicate a correct interval, but it is probably more often than stated in the guidelines.