Liver cell transplantation was performed in a child with urea cycle disorder poorly equilibrated by conventional therapy as a bridge to transplantation. A 14-month-old boy with ornithine transcarbamylase (OTC) deficiency received 0.24 billion viable cryopreserved cells/kg over 16 weeks. Tacrolimus and steroids were given as immunosuppressive treatment while the patient was kept on the pre-cell transplant therapy. Mean blood ammonia level decreased significantly following the seven first infusions, while urea levels started to increase from undetectable values. After those seven infusions, an ammonium peak up to 263 μg/dL, clinically well tolerated, was observed. Interestingly, blood urea levels increased continuously to reach 25 mg/dL, after the last three infusions. Eventually, he benefited from elective orthotopic liver transplantation (OLT) and the post-surgical course was uneventful. We conclude that use of cryopreserved cells allowed short- to medium-term metabolic control and urea synthesis in this male OTC-deficient patient while waiting for OLT.
The urea cycle is the final and crucial pathway for the metabolism of waste nitrogen in humans. Hence, an alteration of this pathway causes an accumulation of ammonia, which has been described to cause several deleterious effects of the brain, for instance neurological damage and cognitive deficits. Urea-genesis is liver-based and hepatocyte is the single cell able to express the ornithine transcarbamylase (OTC) gene, and by this way fully eliminate ammonia. Male patients with OTC deficiency, which is a severe form of urea cycle disorder presenting from birth, are at high risk of severe brain damage due to hyperammonemia. Management by diet protein restriction and use of ammonium scavengers are not always sufficient to control the disease and stabilize the patient (1), and orthotopic liver transplantation (OLT) becomes then the ultimate treatment to avoid repeated encephalopathic episodes and irreversible brain damage (2).
Rapid transplantation is limited by the organ shortage and long waiting time, during which irreversible brain damage may occur and jeopardize the long-term psychomotor development.
Liver cell transplantation (LCT) has been shown to restore partial metabolic activity in various animal models of inborn errors of metabolism (3,4). Because human trials have already reported short or medium metabolic effects in different inborn metabolic diseases, such as Crigler–Najjar disease (5), Glycogen storage type 1a disease (6) and Refsum disease (7), this alternative therapy has been proposed as a bridge during his waiting time for OLT. Moreover, in LCT, cells can be made available without competing with whole organs, by using resected segments of reduced livers or even segment four obtained from split livers (8). One additional advantage of LCT is the possibility to cryopreserve cells that are available for infusions when required.
In a previously reported OTC patient, Horslen et al. have recently demonstrated that using fresh and cryopreserved cells, isolated from 10 different livers, only a short-term (11 days) metabolic stabilization has been observed with a normal protein intake during this short period. The child was subsequently transplanted after this LCT trial (9).
In the current study, we show that LCT with cryopreserved cells can be successful to stabilize an OTC-deficient patient for as long as 6 months, providing metabolic control and urea synthesis while waiting for OLT.
Patient and Methods
A 14-month-old boy (10 kg) with a urea cycle disorder was referred to our unit for liver transplantation. He had experienced numerous attacks of hyperammonemia since birth, when he presented with acute encephalopathy and ammonium level of 548 μg/dL. He had positive orotic aciduria and a liver biopsy confirmed OTC deficiency (enzymatic activity assay). The child had impaired psychomotor development and was maintained on a restricted protein intake (1 g/kg/day), sodium benzoate therapy (250 mg/kg/day) and arginine/citrulline supplementation. He was hospital bound most of the time because of recurrent episodes of encephalopathy and hyperammonemia.
His parents were not considered as living donor candidates, and the boy was registered in the Eurotransplant waiting list. In view of his unstable condition, and following parental information and institution review board approval, a program of LCT was proposed in order to stabilize the child while waiting, as a bridge to transplantation. No contra-indication to LCT was found, and left to right cardiac shunt was amongst others excluded.
A port-a-cath device was inserted surgically in the portal system according to the technique described elsewhere (10) and allowing repetitive infusions. The last injection was made after direct portal puncture during surgery to remove the port-a-cath, which was no more in the portal system. During the ninth cell injections with port-a-cath, an ultrasound was performed before and after each infusion to control the catheter position. Moreover, contrast Doppler ultrasound was applied in order to follow flow direction before infusions, while the portal pressure was monitored before and after infusions. Central venous pressure, blood pressure, pulse rate, temperature and oxygen saturation were also monitored during infusions.
Cell isolation procedure has been reported earlier (7). Briefly, hepatocytes were isolated at the liver cell transplantation center at the St. Luc Hospital, Université Catholique de Louvain using the classical two-step perfusion method under good manufactory practice guidelines. In this case, 5.4 billion hepatocytes with a viability of 89% estimated by the trypan blue exclusion (TBE) method (plating: 42%) were isolated from one ABO compatible 19-year-old boy reduced-liver (no fatty infiltration) recovered postmortem. Serologic tests for hepatitis B, hepatitis C, HIV and CMV were negative.
Cryopreservation and thawing
Isolated cells were frozen in an University of Wisconsin solution with 5 mM Hepes, 20 mg/L dexamethasone, 40 ∥IU/L insulin, 1 M glucose, 3.75% human albumin and 10% DMSO by controlled rate freezing as demonstrated by a consensus on liver cells cryopreservation (11). After cryopreservation in liquid nitrogen, cells were rapidly thawed in a 37°C water bath and washed twice in a stable solution of plasmatic proteins (85% of albumin) containing bicarbonate (0.84 mg/L), glucose (2.5 g/L) and 10 IU/L heparin. After two centrifugations at 1200 rpm and 4°C, cells were resuspended in the same solution at the desired concentration for injection. Viability of the thawed cells was estimated by TBE method and additional functional tests including intracellular adenosine triphosphate assay and tetrazolium salts reduction test (succinate deshydrogenase activity) were also performed. The infused cell suspensions were constituted by ∼95% of albumin positive cells (with hepatocyte morphology) as demonstrated by immunocytochemical studies. In primary cultures, these cells were also able to produce albumin and to synthesize urea as demonstrated by biochemical assays (data not shown).
The child received 3.5 billion cryopreserved cells (2.4 billion viable cryopreserved cells). The median viability as determined by TBE method was 70% (range: 50–84%). The cells were given in 10 successive infusions over 16 weeks. Each infused cell suspension contained 30 to a maximum of 100 × 106 cells/kg body weight, to reach 5–10% of the total liver cells number (5,6,12). A first series of seven infusions (71% of total infused cells) was performed during 1 month when he arrived in our unit. Three last infusions were performed in 15 days while he came back to hospital for follow-up 2.5 months later.
He was maintained on the same dietary and medical management as before. Nevertheless, the controlled diet was probably not followed during the period between the two steps of infusions. Immune suppression included steroids given at a dose of 1 mg/kg progressively tapered to reach 0.25 mg/kg at 1 month and tacrolimus (Prograft, Fujisawa, Berlin, Germany) to reach a through level of 6–8 ng/mL. Before each cell infusion, intravenous hydrocortisone (10 mg/kg) was administered. The patient was maintained under the same immunosuppressive and metabolic treatment until a post-mortem graft became available 6 months after the first infusion and 2 months after the last one. The post-OLT immunosuppressive treatment included tacrolimus monotherapy according to our immunosuppressive OLT protocol. To evaluate immunization against infused cells, donor-HLA-antibodies were measured 2 months after the end of the infusions just before OLT. The patient had orthotopic transplant 6 months after the first infusion.
Due to the deleterious effect of hyperammonemia episodes, plasma ammonia was followed on a regular basis before and after infusions. Blood urea nitrogen was monitored as a final product of the urea cycle activity. Additional ammonia measures were done whenever the patient was unwell.
The engraftment of the transplanted cells was not followed on liver biopsies to avoid unnecessary invasive exams, as this therapy was proposed as a bridge to liver transplantation.
The explanted liver was not analyzed for cell engraftment due to lack of informed consent.
The patient was evaluated using the early cognitive development rating scales (EEDCP, 2000).
The data were analyzed using the Student's t- test (paired).
The infusions were perfectly tolerated without any acute adverse events. Clinical parameters remained within normal range and no important changes in liver function tests were observed (Table 1). Pre- and post-cell-transplant Doppler ultrasound of the portal system and hepatic artery neither show any change of portal pressure nor blood flow.
Table 1. Ammonia, AST (normal AST laboratory values: 14–63 IU/L) and ALT levels (normal ALT laboratory values: 6–33 Iu/L) were followed during the pre-transplant period, first series of infusions, post first series period, second series of infusions and the last period, free of infusions, while waiting OLT. Data are presented as mean ± SD. Slight increase in AST and ALT levels was recorded during the two infusions periods with a maximum of 151 and 130 IU/L, respectively. These values returned to control levels 2 days after the last infusion of each series
Post 1st series
Post 2nd series
Ammonia levels (μg/dL)
118.0 ± 11.7
71.4 ± 7.7
110.9 ± 39.3
82.7 ± 27.5
138.6 ± 4.9
23.5 ± 2.6 (n = 11)
76.0 ± 13.8 (n = 24)
54.6 ± 34.4 (n = 11)
82.8 ± 32.6 (n = 7)
38.5 ± 20.3 (n = 12)
36.8 ± 2.7 (n = 11)
57.7 ± 11.8 (n = 24)
40.5 ± 11.9 (n = 11)
59.7 ± 31.6 (n = 7)
39.3 ± 7.9 (n = 12)
Before LCT, several episodes of hyperammonemia occurred reaching 548 μg/dL at birth (7 days of live) and 422 μg/dL 5 months later. The patient had recurrent episodes of severe psychomotor decompensation, while under diet and treatment. During first month, the patient received a first series of seven infusions with a total of 1.7 billion viable cells. During this period, analysis of blood ammonia (μg/dL) showed a significant decrease in its mean level 71.4 ± 7.7 (median: 73; range: 38–131) (n = 32) as compared to the mean value before LCT 118.0 ± 11.7 (p = 0.0048) (median: 96; range: 25–548) (n = 58).
After the first series of infusions, when he left hospital, transient hyperammonemia episodes, clinically well tolerated, up to 263 μg/dL (2 months after the seventh infusion) were detected biochemically, related to decreased compliance to treatment and diet or to possible rejection of the cells (mean ammonia for this period: 110.9 ± 39.3) (median: 101.5; range: 43–263) (n = 13). A second series of three infusions was then made in 15 days. Mean ammonia levels during this period decreased to 82.7 ± 27.5 with no statistical significance (p = 0.514) (median: 80; range: 41–132) (n = 6). Later, respectively, 5 days and 2 months after the last infusion, two hyperammonemia episodes occurred up to 338μg/dL during intercurrent infection, being well tolerated clinically (mean ammonia levels of this period: 138.6 ± 4.9) (median: 110; range: 46–338) (n = 24). During these two episodes, the diet was reduced to 0.5 g/kg until ammonia levels returned again to normal values. The rest of the treatment was not modified.
In parallel, we evaluate the levels of urea as a probable indicator of metabolically active engrafted cells. Urea starts to become detectable 1 month after the first infusion, at the end of the first infusion series (two peaks up to 13 mg/dL). These significant urea levels were maintained up to 2 months before decreasing to reach undetectable values. Importantly, urea levels increased again after the last infusions and reached the highest value of 25 mg/dL. The post-LCT average was 12.3 ± 0.6 (n = 40) as compared to undetectable pre-LCT values (n = 12) (p = 0.0008) (Figure 1).
From the end of the first infusion series, the child psychomotor development markedly improved, according to EEDCP scales, with acquisition of sitting position and speech, which had disappeared earlier. No psychomotor decompensation occurred during post-LCT hyperammonemia episodes unlike pre-LCT episodes.
Following LCT and before the planned OLT, there was no evidence of sensitization to donor HLA's as shown by negative panel reactive antibodies. Six months after first infusion, when a post-mortem graft was available, he was liver transplanted. Following OLT, ammonia and urea levels returned within the normal range, and the post-transplant course was uneventful.
While waiting for OLT, this patient received safely 0.24 billion viable cryopreserved cells/kg. Cells were infused over a period of 16 weeks in two different series. Single donor origin aimed to avoid immune sensitization in view of the planned OLT.
We demonstrate that cryopreserved cells can be used for elective LCT, and exert metabolic control. Indeed, the first series of infusion (1 month—71% of all infused cells) led to the control of ammonia levels, which remained within normal ranges. This immediate and transient result confirms the data obtained by Horslen et al. (9) using a larger number of cells from different donors as well as prior observations from our center showing that metabolic control can be obtained immediately following infusion of a large cell mass (7).
Our patient required, however, a second series of infusion when ammonia levels increased again. This secondary loss of effect was probably due to non-permanent engraftment or rejection of a part of the infused cells. Indeed, animal studies have shown immunogenicity of infused murine hepatocytes (13). The increased ammonia levels may also be explained by a decreased compliance to diet following the first infusions, as the patient was not hospital bound in this period. Nevertheless, the port-a-cath access allowed us to plan additional infusions, leading again to normalization of ammonia levels during the infusion period.
Whereas the ammonia levels can be influenced by several parameters, urea synthesis is a more significant indicator of metabolically active engrafted cells. Urea levels increased progressively in blood to reach the highest detected value 5 months after the first infusion. This result indicates that donor liver cells were engrafted and had become metabolically active. This allowed to stabilize the patient further and to protect him against encephalopathic episodes while waiting for OLT.
Previous animal studies have demonstrated that the efficiency of cryopreserved cells can be comparable to that of freshly isolated cells. Fresh or cryopreserved cells resulted in similar decrease in urinary uric acid excretion in dalmatian dogs (14). Liver repopulation with cryopreserved/thawed hepatocytes was demonstrated in mice model (15,16). The cryopreserved cells have also the ability to retain hepatic function and to respond to liver injury as it was demonstrated in immunodeficient mice (17). In humans, Baccarani et al. described use of cryopreserved cells in a bioartificial liver as a source of human liver cells to bridge a patient, affected by fulminant hepatic failure, to emergency liver transplantation (18). Nevertheless, exclusive use of cryopreserved cells was to our knowledge not yet reported in LCT.
We conclude that cryopreserved/thawed hepatocytes can be safely infused to achieve metabolic control. We showed immediate efficacy in ammonia levels and medium-term efficacy in urea synthesis as well as improved psychomotor development. LCT should be considered as a bridge to transplantation for OTC deficiency patients to avoid neurological decompensation.
The authors thank the UCL pediatric radiology unit for the medical support, Magda Janssen, Annick Bourgois, Céline Caillau and Khuu Ngoc Dung for their collaboration. This work was supported by a grant Télévie for liver cells cryopreservation and by grant WALEO. LCT program is supported by a grant from DGTRE, Région Wallonne, WALEO/HEPATERA.