Khazal Paradis, MD
Division of Pediatric GastroenterologyNutrition, Department of Pediatrics, Hôpital Sainte-Justine, University of Montréal, Montréal, Que
Clin Invest Med 1996; 19 (5): 311-6.
[résumé]
Tyrosinemia, a genetic disorder of the liver and kidneys, is caused by reduced activity of fumarylacetoacetate hydrolase (FAH), the final enzyme in the degradation of tyrosine. The consequent presence of succinylacetone in urine or blood is pathognomonic of tyrosinemia and is used as a confirmatory test in the Quebec neonatal screening program. Due to a complex founder effect, the province of Quebec has an unusually high prevalence of tyrosinemia, particularly in the SaguenayLac Saint-Jean region (where the prevalence is 1 in 1850). Tyrosinemia has several different clinical presentations, ranging from acute liver failure with severe coagulopathy early in life, to slowly progressing cirrhosis with multiple nodules and variable renal dysfunction, to normal liver function with renal failure. Hepatocarcinoma has been found in approximately one third of cases. FAH complementary DNA has been cloned and mapped to chromosome 15q23-q25. The mutation observed in Quebec is a splice mutation at intron 12. This mutation is common and has been observed in other areas of the world as well, although more than 20 mutations causing tyrosinemia have now been described. Liver transplantation remains the definitive treatment. The author's team has carried out 28 liver transplantations (including 2 combined liverkidney transplantations) in 25 children. The overall survival rate has been 92%; two children died as a result of primary nonfunction. The primary indications for transplantation were hepatic nodules (in 14 cases), neurological crises (6) and hepatic (3) or renal failure (2). An abnormal glomerular filtration rate (GFR) of less than 80 mL/min per 1.73 m2 was documented before transplantation in 54% of the cases. The rate normalized after liver transplantation in most patients, with rapid improvement in tubular function. However, patients with a severely low rate (less than 55 mL/min per 1.73 m2) before transplantation still had borderline renal function and poor growth after the transplantion, despite normal liver function. Therefore, for children with a consistently low GFR, careful consideration should be given to performing a combined liverkidney transplantation, and a renal biopsy should form part of the pretransplantation evaluation.
La tyrosinémie est une maladie génétique du foie et des reins. Elle est causée par une activité réduite de la dernière enzyme impliquée dans la dégradation de la tyrosine, l'hydrolase fumarile acéto-acétate (HFA). Il en résulte la présence de succinile acétone dans l'urine ou le sang et cette caractéristique pathognomonique de la tyrosinémie sert de test diagnostique dans le programme de dépistage néo-natal du Québec. À cause d'un effet fondateur complexe, la Province de Québec a une prévalence élevée de tyrosinémie particulièrement dans la région du SaguenayLac Saint-Jean où la prévalence est de 1:1850. Les modes de présentation clinique de la tyrosinémie sont variés, et incluent une insuffisance hépatique précoce avec troubles graves de la coagulation, une cirrhose lentement progressive avec nodules multiples et troubles rénaux variables, et une insuffisance rénale avec fonction hépatique normale. Le cancer hépato-cellulaire survient chez environ un tiers des sujets. L'ADNc de la HFA a été cloné et localisé au chromosome 15q23-q-25. La mutation observée au Québec est une mutation d'épissage à l'intron 12. Cette mutation est fréquente et a été observée ailleurs dans le monde; plus de 20 mutations distinctes causant la tyrosinémie ont été identifiées. La transplantation hépatique est le meilleur traitement. L'équipe de l'auteur a procédé à 28 transplantations hépatiques chez 25 enfants, y compris deux transplantations hépato-rénales combinées. Le taux de survie global est de 92 %; deux sujets sont décédés à la suite du non-fonctionnement du greffon. Les indications de transplantation étaient : nodules hépatiques (n = 14), crises neurologiques (n = 6), et insuffisance hépatique (n = 3) ou rénale (n = 2). Un taux de filtration glomérulaire (TFG) inférieur à 80 mL/min par 1.73 m2 a été observé avant la transplantation chez 54 % des sujets. Le TFG était normal après la transplantation hépatique chez la plupart des sujets, avec une amélioration rapide de la fonction tubulaire. Par contre, les sujets avec un TFG gravement abaissé (moins de 55 mL/min par 1.73 m2) avant la transplantation avaient une fonction rénale limitée et une faible croissance après la transplantation, en dépit d'une fonction hépatique normale. Dans le cas d'enfants avec un TFG gravement abaissé, il faut donc envisager particulièrement la greffe hépato-rénale combinée et pratiquer une biopsie rénale dans l'évaluation pré-transplantation.
This article surveys the numerous contributions of clinicians and researchers around the world to the understanding of the pathophysiology of tyrosinemia and to advances in therapy for the condition as well as the experience of the Montreal Pediatric Liver Transplant Program.
A recent international survey, involving primarily centres where neonatal screening is not available, proposed that the disorder be classified according to the patient's age at presentation of symptoms (less than 2 months, 2 to 6 months and more than 6 months), since this appeared to have a significant effect on prognosis.[6] This classification may be less useful with the advent of new therapies, as discussed below.
Typically, newborns with tyrosinemia have extremely high levels of a-fetoprotein (AFP); it has been suggested that this level reflects liver injury even in utero.[7] This speculation has been supported by the observation of advanced macro- or micronodular cirrhosis in very young infants.[8] In most children, AFP levels slowly decline during the first year of life but never return spontaneously to normal levels, even on strict dietary therapy. Approximately one third of infants affected present with symptoms of liver failure and coagulopathy by 2 months of age.[6] Older children may present with severe coagulopathy and no other signs of decreased synthetic function.[8]
Renal involvement in tyrosinemia is extremely variable; tubulopathy is detected in more than 80% of patients, with half of the children having clinical rickets despite therapy.[9] On histologic examination, 25% have a variable degree of glomerulosclerosis and 50% have mild to moderate interstitial nephritis and fibrosis. Approximately one third of patients have ultrasonographic evidence of nephrocalcinosis. Glomerular function has been found to be abnormal (less than 80 mL/min per 1.73 m2) in 50% of patients at the time of transplant evaluation at our institution (Fig. 1) and appears to deteriorate with time. The process appears to be at least partially reversible through liver transplantation (Fig. 1).
One of the most troubling aspects in the management of tyrosinemia is the very high incidence of hepatocarcinoma.[10] A careful examination of 25 livers resected in our program revealed two cases of hepatocarcinoma and six cases of dysplasia, for a combined incidence of 32%. All of the livers were cirrhotic, with 75% having mixed macro- and micronodular cirrhosis. Nodules were either regenerative, fatty or carcinomatous. A blind retrospective radiologic analysis of screening ultrasonographic and computed-tomography scans of these livers could not detect which patients had hepatocarcinoma, and the inability to detect this has been described by others.[11] Nevertheless, all of the livers with carcinomatous nodules, identified after transplantation, had also been correctly identified as having nodules before transplantation. The use of both imaging techniques as complementary screening tools is recommended, since nodules may be detected by one technique and not the other. Preliminary use of magnetic resonance imaging at our institution has not provided further power to distinguish benign nodules from malignant ones.
In 1977, Lindblad, Lindstedt and Steen[12] showed that tyrosinemia is due to a deficiency in fumarylacetoacetate hydrolase (FAH), the last enzyme in the tyrosine catabolic pathway (Fig. 2). The deficiency leads to an accumulation of the metabolites fumarylacetoacetate and maleylacetoacetate, which together form succinylacetone. Detection of succinylacetone in urine, plasma or amniotic fluid is considered pathognomonic of tyrosinemia, since succinylacetone is not found in any other condition. Succinylacetone is a potent inhibitor of alanine dehydrogenase, a key enzyme in the heme synthetic pathway (Fig. 2).
It has been suggested that the renal damage in tyrosinemia is secondary to the toxic effects of maleylacetoacetate.[13] The biochemical cause for the neurological crises experienced by many patients remains unclear: in view of the clinical similarities between these attacks and acute porphyric attacks, the attacks are postulated to be due to alterations in the heme biosynthetic pathway (Fig. 2) caused by the accumulation of succinylacetone.[14] The cause of the unusually high incidence of hepatocarcinoma in the disorder remains unclear.
The FAH gene has been mapped to the long arm of chromosome 15, and the predominant mutation in the Quebec population has been found to be a splice mutation.[15] This mutation is common in European and Middle-Eastern patients as well.[16] Several other mutations have also been described, with many patients having the compound heterozygote form. Normal FAH enzymatic activity has been detected in certain nodules from resected livers.[17] In-situ polymerase chain reaction analysis of these FAH-positive nodules has suggested some form of reverse-mutation process with reversion to the normal allele. This reversed mutation was not, however, detected in fibroblasts of affected patients but was noted in some nodules adjacent to hepatocarcinoma.[18]
After the initial dose of NTBC, serum levels of succinylacetone become undetectable in minutes or hours,[21] although the elevated AFP levels may take several months to normalize. The drug is administered as a tasteless powder in a divided dose of 0.5 to 2 mg/kg per day, while a diet with strict restriction of phenylalanine and tyrosine is maintained. The results of pharmacokinetic and dose-response studies are still pending. It remains unlikely that NTBC will prevent hepatocarcinoma in affected children; several children treated with NTBC were found to have dysplastic nodules at transplantation, and one child was found to have hepatocellular carcinoma after 3 years of therapy. It is too soon to determine whether institution of NTBC at birth prevents hepatocarcinoma or delays its onset. The transgenic mouse model of tyrosinemia developed by Grompe and colleagues[22] is the first clinically relevant animal model. Unfortunately, cancerous nodules develop in the mice despite NTBC treatment at birth or even during gestation. These studies suggest that NTBC may not prevent hepatocellular carcinoma and that physicians should continue to be vigilant to detect nodules in patients' livers.
Liver transplantation remains the only definitive therapy for tyrosinemia. As of 1995, the Montreal Pediatric Liver Transplant Program had performed transplantations for 25 children, two of which were combined liverkidney transplantations, with a survival rate of 92%. Three patients required retransplantation, two because of vanishing bile duct syndrome and one because of early hepatic artery thrombosis. In the two patients with hepatocarcinoma, the pretransplant metastatic work-up showed no metastasis; therefore, no chemotherapy was administered. Both patients are alive and well 2 and 6 years after transplantation, as has been reported previously by others.[23] Only 3 of the 25 patients continue to have very abnormal GFR values after transplantation, despite normalization of the tubulopathy and hypercalciuria within weeks of the transplant.[24] The two survivors with combined liverkidney transplants have done extremely well, with no evidence of liver or kidney rejection.
The author would like to thank Drs. Grant Mitchell, Frédéric Faucher, Jean Larochelle, Sylvianne Forget, Josée Dubois, Heidi Patriquin, Dikkens St-Vil, Andrée Rasquin and Pierre Russo for their assistance. The publication of this article was supported in part by a grant from the Fonds de la recherche en santé du Québec.