1 Introduction

Hepatorenal tyrosinemia or tyrosinemia type I (HT1) is a severe autosomal recessive disorder caused by the deficiency of fumarylacetoacetate hydrolase (FAH), the final enzyme in the degradation of tyrosine. This leads to accumulation of proximal metabolites including maleylacetoacetate (MAA) and fumarylacetoacetate (FAA), which are converted to toxic byproducts, succinylacetoacetate (SAA) and succinylacetone (SA) (Mitchell et al. 2001).

The worldwide incidence of HT1 is estimated to be 1:100,000–1:125,000 and a high number of cases have been described in regions such as Quebec (Canada), Scandinavia and some parts of Turkey (see Aktuglu-Zeylek and Gigden, this book Chap. 15) and India (Mayorandan et al. 2014). In Quebec, the carrier rate is about 1/60, reaching 1/20 in the Saguenay-Lac Saint-Jean area (1/1,846 live births) where the founder mutation IVS12+ 5G>A is particularly prevalent (De Braekeleer and Larochelle 1990; Grompe et al. 1994). Universal newborn screening for HT1 has been performed in Quebec since 1970 (Grenier et al. 1982).

In this chapter, we summarize the liver features of HT1 patients before the era of 2-(2-N-4-trifluoromethylbenzoyl)-1,3-cyclohexanedione (NTBC), including clinical, biological, histo-pathological and imaging findings, focusing on the Quebec experience. Then we describe the current hepatic situation of patients treated in the Quebec NTBC Study and our recommendations for liver surveillance and management of the Quebec cohort. Early data about the course of HT1 patients before the availability of NTBC have been published elsewhere (Larochelle et al. 1973; Paradis et al. 1990). Data about liver transplantation in Quebec are reported in another chapter of this volume (Alvarez and Mitchell, Chap. 5).

2 Clinical Presentation

In children who are not detected by newborn screening and who do not receive treatment with NTBC, the clinical course of HT1 may be highly variable. The spectrum includes acute (or sub-acute) and chronic forms (Larochelle et al. 1967; van Spronsen et al. 1994). The acute form is associated with rapid disease progression, characterized by severe liver disease in the first 6 months of life. Liver synthetic function is particularly severely affected. Symptoms include bleeding abnormalities, hypoglycemia, ascites, oedema, vomiting, irritability and jaundice. Laboratory findings show abnormally prolonged prothrombin (PT) and activated partial thromboplastin time, increased international normalized ratio (INR), and decrease of coagulation factors (factors II, V, VII, IX, X, XI and XII) as well as elevation of serum aminotransferases, gamma glutamyl transpeptidase. Since the defect in hepatocytes is primarily synthetic rather than cholestatic, hyperbilirubinemia tends to occur late, when liver failure is already established. Therefore, lack of hyperbilirubinemia is not infrequent and does not exclude HT1 in a patient with liver failure. Conversely, a patient who presents with marked hyperbilirubinemia and relative preservation of synthetic function is unlikely to have HT1.

The chronic form classically affects patients older than 1 year of age. Fibrosis and cirrhosis develop, but episodes of acute liver failure can occur in the context of infections and other stresses. Some patients have renal tubular dysfunction with a Fanconi syndrome of variable severity resulting in phosphaturia with hypophosphatemia and sometimes rickets, glycosuria, generalized aminoaciduria, hypercalciuria, hyperchloremic metabolic acidosis and hyperuricosuria/hypouricemia). Cardiomyopathy is described but in our experience is rarely of clinical importance. Episodes of acute hepatic porphyria-like neurologic crises can occur (Mitchell et al. 1990). Not all patients fall neatly into these two groups. In practice, there is a spectrum of disease severity. Some patients have an intermediate, “subacute” course. Rare patients have a prolonged course, in which cirrhosis and renal failure develop insidiously in early adulthood.

A key consideration is the high risk of hepatocellular carcinoma (HCC) as a mid- or long-term complication of HT1 (Paradis et al. 1994; van Spronsen et al. 2005; Koelink et al. 2006). The presence of a liver nodule in patients with tyrosinemia is an indication for urgent evaluation and is considered to be cancerous until proof to the contrary. Liver parenchymal changes of some HT1 patients, including fibrosis and steatosis, may make difficult the diagnosis of HCC on imaging techniques.

Serum alphafetoprotein (AFP) is increased in virtually all patients at the time of diagnosis. When NTBC treatment is started during the first month of life, AFP level typically normalizes over 12 to maximum 24 months of life. In non-NTBC treated patients AFP levels tend to improve with age and diet treatment, but usually do not normalize (Larochelle et al. 1967; van Spronsen et al. 1994).

HT1 patients are followed with repeated assay of serum AFP. This test should not be used alone as the only marker of HCC because it lacks specificity and sensitivity. However it is useful clinically when integrated with clinical and imaging considerations. Serum AFP can fluctuate, but a steady increase of AFP levels raises the suspicion of a HCC. Serum AFP levels are elevated in most cases of HCC but also can be high in patients with regenerative nodules (such as in the recuperation phase from severe liver crises in non-NTBC treated patients), fatty nodules and if NTBC treatment is stopped. Importantly, in some HT1 patients, HCC can occur with normal or only slightly elevated serum AFP levels (Paradis et al. 1990; van Ginkel et al. 2015).

In the Québec NTBC Study, we use an integrated clinical approach that includes (1) careful consideration of the clinical history (time of starting NTBC treatment, rapidity of AFP response, (2) adherence to treatment as seen by serial plasma NTBC and plasma and urine SA levels), (3) monthly surveillance of serum AFP levels and serial imaging with careful comparison to previous imaging studies. No single parameter suffices to diagnose or to exclude HCC.

Abdominal imaging is an essential element. Abdominal ultrasound (US) and magnetic resonance (MR) imaging are the modalities of choice for serial evaluation of liver, kidneys and pancreas in HT1 patients, preferably by the same radiologist using detailed comparison with previous studies. The liver screening is important to exclude liver nodules. US permit to detect small nodules sometimes difficult to see on MR. Nodules are hyper or hypoechoic in a heterogeneous (when macronodular cirrhosis is present) or normal parenchyma on US. It is impossible to make the difference between regenerative nodules, dysplasic nodules or HCC. However, MR imaging with the hepatocellular contrast allows a better characterization of lesions. Several publications suggest the Gd-BOPTA contrast (gadobenate dimeglumine, MultiHance®, Bracco Imaging SpA, Milano, Italy) and more recently Gd-EOB-DTPA (gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid [or gadoxetic acid: Primovist®/Evovist®]) as improved tool to detect small HCC nodules (Grazioli et al. 2010; Granito et al. 2013; Haradome et al. 2011). The MR dynamic acquisition is performed with the intravenous injection of 0.1 cc/kg = 0.025 mmol/ml of Primovist® with a pre-MR scanning followed by 25 seconds (sec), 70 sec, 3 minutes (min), 5 min, 20 min and 60 min post-injection scanning. It permits to increase the specificity of lesions. Most of the HCC nodules present with a portal washout followed by a hypointense lesion with a capsule and a hypointense lesion on delayed phase (Fig. 6.1).

Fig. 6.1
figure 1

MR imaging results suggesting HCC in a 16-year old boy with HT1presenting with AFP increase. Gd-EOB-DTPA-enhanced 3D T1–weighted images obtained at (a) precontrast (b) arterial phase 20 sec and (c) at 60 sec (portovenous phase). On segment VIII of the liver, an hypointense lesion was seen in pre-contrast image with a significant arterial phase enhancement with wash-out on the porto-venous phase compared to the liver parenchyma

Full size image

In a recent meta-analysis, up to 400 articles related to agents contributing to the characterization of HCC nodules have been reviewed by Liu et al. (2013). Authors have also studied more than 200 publications focusing on agents to diagnose liver metastases from various origins. Altogether, with Primovist® as contrast agent in the diagnosis of HCC, the sensibility is around 91% (false negative rate may be related to a hyperintense aspect of HCC in ~7% of cases) and the specificity reaches 95% (false positive rate may be related to a hypo-intense aspect of other lesion like cholangiocarcinoma).

Possible extrahepatic findings on abdominal imaging include nephromegaly, nephrocalcinosis and renal cortical hyperechogenicity. The pancreas can occasionally be hyperechogenic but this is not typically associated with clinical signs.

3 Pathology

The histo-pathologic changes are variable depending on acute or chronic presentation mode (Dehner et al. 1989; Watanabe et al. 1983; Demers et al. 2003). In the early form, macroscopically, the liver will be enlarged and frequently pale. Usually, histo-pathologic changes include usually micronodular cirrhosis, fibrotic septa, bile duct proliferation within portal tracts and steatosis. The normal hepatic architecture is substituted by pseudoacinar or nodular formations around a central duct with bile plugging. Accumulation of iron in hepatocytes or Kuppfer cells can occur, as can giant cell transformation. In the chronic form of HT1, the liver is nodular, coarse and enlarged and histological analysis shows micro and macronodular cirrhosis, variable amount of steatosis, fibrotic septa and mild lymphoplasmatic infiltrate. Ductular proliferation, cholestasis or intralobular inflammation is less marked. Large or small type liver cell dysplasia may be present and is considered to be a premalignant lesion. Ultra-structural features include fatty inclusions into the hepatocytes with dilatation of endoplasmic reticulum and variable changes in the mitochondria.

HCC lesion size may be variable but, usually, small HCC measures less than 2 cm on diameter and macroscopically, it can be indistinguishable from macroregenerative nodules, in particular when liver cirrhosis occurs. A fibrous capsule and an approximately nodular aspect with irregular borders are usually present. Some green areas may alternate with yellow parts corresponding respectively to bile staining and fat accumulation into tumor cells. The diagnosis is confirmed microscopically. HCC measuring less than 2 cm of diameter is well-differentiated and is composed of thin and multiple strands of hepatocytes. Some trabecular and acinar patterns are often mixed at the microscopic analyze. Tumor cells are assembled is cords of variable thickness separated by sinusoidal blood spaces. Initially, lesions have a thin trabecular pattern but with clonal expansion inducing the dedifferentiation, trabeculae become thicker. Mallory hyaline and intracytoplasmic proteinaceous accumulations are visible as well as loss of reticulin fibers (Washington and Harris 2010) (Fig. 6.2).

Fig. 6.2
figure 2

(a) Macroscopic appearance of a liver from an adolescent who presented with hepatomegaly and renal Fanconi syndrome and received NTBC treatment from the age of 11 months. He had with a nodular lesion located in segment VIII, measuring around 15 × 15 mm. (b, c) Microscopically, the lesion is multinodular, well limited by a thin fibrotic capsule. The lesion is highly acinar (continuous arrow), with substantial cell proliferation. Multiple hepatocytes strands are present (dotted arrow) (Hematoxylin and eosin, ×200 (b) and ×400 (c))

Full size image

4 Quebec Cohort

In this section, the clinical and imaging features of the liver and the hepatological management of Quebec patients with HT1, is described.

Currently, 88 patients (43 females) are individually followed by a pediatric hepatologist at the coordinating center. Each patient is examined at least once a year during a interdisciplinary clinical visit, as described in the Quebec NTBC Study protocol. Patients have liver US every 6 months and abdominal MR imaging every 12 months. Recently, an annual fibroscan has been added. The usefulness of this examination of liver stiffness in HT1, and its optimal frequency, will be evaluated over the next years. In stable patients, AFP and alanine aminotransferase are measured every 3 months and albumin, INR and PT are assessed every 6 months.

Among entire cohort, 85 patients had a neonatal screening for HT1. Currently the median age is 13.9 years (range (r): 41 days (d) to 42.4 years (y)). The first use of NTBC occurred on February 1994 in a 5 year-old girl. NTBC treatment was started during the first month of life in 67 of 88 patients (median age: 25 days, range: 2 days to 21.7 years). Currently, the median duration is 13.8 years. At the evaluation in spring 2016, 81/88 patients (92%) had normal levels of alanine aminotransferase (<35 U/L). In the seven remaining patients (including five males in whom median BMI was 31.2, range, 26–50.1) with increased ALT (mean level 49.7 U/L, range 36–76 U/L), liver steatosis was confirmed on abdominal ultrasound. No other cause was found for the elevated level of AFP in these patients.

Serum AFP (N<10 ng/ml after the age of 12–18 months) is measured every 3 months. Excluding patients younger than 18 months, two patients had chronically increased AFP levels. The first is a 22-year-old man who started NTBC at 5 months of age in whom AFP levels decreased to about 50 ng/ml, and have remained at this level for almost 15 years. Extensive evaluations have been performed in this patient. No sign of HCC is evident. The second patient is a 23-year-old man who started NTBC at 8 months of age. His AFP levels are normal but liver imaging is potentially compatible with HCC. He is currently under evaluation for a possible liver transplantation.

In our experience, no patient who received NTBC treatment before the age of 1 month has developed HCC. Conversely, even early-treated patients are considered to be at risk, and current plans are to follow them throughout their lives. Among the patients who began treatment at an older age, some have developed HCC and are described in the chapter about liver transplantation in HT1 in Quebec (Chap. 5).

Altogether, 14 liver biopsies have been performed in 12 patients over the 20 y follow-up period. The main indication was unexplained cytolysis and/or increased AFP levels with suspicion of liver dysplasia. The Metavir score (Bedossa and Poynard 1996) was normal (F0-F1) for all except two patients who had an abnormal Metavir score of (F3). One of these patients eventually developed HCC and underwent liver transplantation.

Among the nontransplanted patients in the Quebec cohort, 21 patients began NTBC treatment after the first month of life due for example to birth before the NTBC era or birth outside of Quebec in a region without screening, with delayed diagnosis. To date, all patients except the two young men discussed above, show normal AFP and imaging without suspicion of hepatic nodules,. Typically, serum SA levels are near to or within the control reference range (<24 nmol/L), discussed in the chapter dealing with clinical challenges in tyrosinemia (Chap. 19).

5 Management of Liver Complications

The management of liver disease is challenging during the acute phase of HT1. In our experience, in recent years, due to the effectiveness of generalized neonatal screening combined with rapid NTBC and diet treatment, acute liver crisis have not occurred. Historically, during liver crises, non-NTBC treated HT1 patients were hospitalized. Caloric intake was maximized by enteral or parenteral routes, in order to limit catabolism and the enhanced breakdown of phenylalanine and tyrosine to the toxic metabolites that accumulate in HT1. Intake of phenylalanine and tyrosine was restricted and subsequently increased as permitted by the clinical course. Management of infections, which are important precipitants of acute liver crises, was a major goal. The clinical description and biological findings in HT1 patients prior the era of NTBC are described elsewhere (Mitchell et al. 2001).

In screened, NTBC-treated HT1 patients, our clinical approach for chronic liver disease focuses mainly in the detection and follow-up of nodules in order to assess the indication for liver transplantation. It is difficult to be certain of the malignant potential of nodules, in particular in those with imaging signs compatible with HCC. For this reason, we still believe that any patient metabolically well controlled with therapeutic levels of NTBC, who presents with increase of serum AFP and with repeated imaging confirming such nodules, should be assessed for liver transplantation. In our experience, the major risk factor for the development of liver nodules and of HCC, is the absence of early diagnosis leading to a delayed NTBC treatment (after the first month of life). Over the last 22 years, no patient who was diagnosed following a positive neonatal screen and who was treated before 1 month of age, has required liver transplantation.

6 Future Perspectives

Research into new therapeutic avenues, including stem cells and gene therapy, is providing interesting preliminary results (Hickey et al. 2014, 2016). The main challenge of these approaches is to substitute all hepatocytes without leaving a single FAH-deficient cell, which is not yet possible. Research in genetically modified stem cells is discussed elsewhere in this volume. Although it is not currently applicable clinically, these techniques may play a role in the future for the management of HT1.

7 Conclusion

Since the publication of Lindstedt and Holme (Lindstedt et al. 1992), NTBC therapy has revolutionized the medical management of HT1 patients. For two decades, we moved from the symptomatic treatment of life threatening hepatic emergencies to the elective regular surveillance of patients who have a high quality of life and who function normally in society. In Quebec, and increasingly elsewhere, identification of HT1 by newborn screening permits treatment to be started before the development of clinical symptoms or of detectable organ damage. Of course, this treatment requires adherence to a special diet and to the prescription of NTBC. Adherence has proven difficult for some patients, particularly during adolescence, as discussed in the chapter in this volume devoted to future clinical challenges. The clinical course to date is promising and continued follow-up of the Québec NTBC study cohort will continue to provide answers about the long-term hepatic outcome of the medical treatment of HT1 patients.