Novel oral treatment of Gaucher s disease with N-butyldeoxynojirimycin (OGT 918) to decrease substrate biosynthesis

Transcription

1 Articles Novel oral treatment of Gaucher s disease with N-butyldeoxynojirimycin (OGT 918) to decrease substrate biosynthesis Timothy Cox, Robin Lachmann, Carla Hollak, Johannes Aerts, Sonja van Weely, Martin Hrebícek, Frances Platt, Terry Butters, Raymond Dwek, Chris Moyses, Irene Gow, Deborah Elstein, Ari Zimran Summary Background Current treatment for Gaucher s disease involves administration of intravenous glucocerebrosidase to degrade glucocerebroside stored in lysosomes. Lowering the rate of biosynthesis of glucocerebroside should decrease accummulation of this substrate. We investigated the safety and efficacy of OGT 918 (N-butyldeoxynojirimycin), an inhibitor of glucosyltransferase, as a novel oral treatment for nonneuronopathic Gaucher s disease. Methods We recruited, into a 1-year open-label study, 28 adults (seven with previous splenectomies) from four national Gaucher s referral clinics, who were unable or unwilling to receive enzyme treatment. We measured liver and spleen volume by computed tomography or magnetic resonance imaging at baseline and at months 6 and 12, and biochemical and haematological variables monthly, including chitotriosidase activity (a sensitive marker of Gaucher s disease activity). Patients were started on 1 mg oral OGT 918 three times daily. Findings Baseline liver volumes were times normal and spleen volumes times normal. At 12 months, mean liver and spleen volumes were significantly lowered by 12% (95% CI ) and 19% ( ), respectively (each p< 1). Haematological variables improved slightly. Mean organ volume and blood counts improved continually between 6 months and 12 months of treatment. Mean chitotriosidase concentrations fell by 16 4% over 12 months (p< 1). Six patients withdrew because of gastrointestinal complaints (two), personal reasons (two), or severe preexisting disease (two). The most frequent adverse effect was diarrhoea, which occurred in 79% of patients shortly after the start of treatment. Interpretation Decrease of substrate formation by OGT 918 improves key clinical features of non-neuronopathic Gaucher s disease. The strategy justifies further trials in this and other glycosphingolipid storage disorders. Lancet 2; 355: Department of Medicine, University of Cambridge, Addenbrooke s Hospital, Cambridge CB2 2QQ, UK (Prof T Cox FRCP, R Lachmann MRCP); Academic Medical Centre, University of Amsterdam, Amsterdam, Netherlands (C Hollak MD, Prof J Aerts PhD, S van Weely PhD); Institute of Inherited Metabolic Disorders, Prague, Czech Republic (M Hrebícek MD); Oxford Glycobiology Institute, Department of Biochemistry, University of Oxford, Oxford, UK (F Platt PhD, T Butters PhD, Prof R Dwek FRS); Oxford GlycoSciences, Abingdon Science Park, Oxon, UK (C Moyses MRCP, I Gow BSc); Shaare Zedek Medical Centre, Jerusalem, Israel (D Elstein PhD, A Zimran MD) Correspondence to: Prof Timothy Cox Introduction Gaucher s disease is the most common glycosphingolipid lysosomal storage disorder and the first to be treated by enzyme replacement. Inherited deficiency of the lysosomal enzyme, glucocerebrosidase, leads to the accumulation of glucocerebroside, mainly in the cells of the mononuclear phagocyte system. 1 The clinical features of the disorder are dominated by progressive hepatosplenomegaly, hypersplenism, skeletal lesions, and, in the rare neuronopathic forms, neurological disease. 2,3 Macrophagetargeted human glucocerebrosidase is currently supplied to a few thousand patients, in whom it reduces organomegaly, improves hypersplenism, and frequently lessens bone pain. 4,5 Treatment is required for life and involves repeated intravenous infusions that are inconvenient for some patients; intravenous administration by indwelling catheters may be required, which carries a risk of infection, especially in patients who have undergone splenectomy. There are estimated to be 2 3 cases worldwide, 3 and, therefore, only a fraction of patients are receiving treatment. In practice, the costs of replacement enzymes for such rare disorders and concerns about the ability of the enzyme to penetrate the central nervous system restrict the usefulness of this approach. An oral treatment that uses a small molecule has the potential to improve access to treatment and lessen the burden of treating these disorders with parenteral agents. Inhibition of substrate formation is a new approach to the glycosphingolipidoses 6 and was first advocated by Radin and colleagues 7 for the treatment of non-neuronopathic Gaucher s disease. The strategy seeks to lower the formation of glycosphingolipids to rates at which the residual enzyme activity in a given patient can catabolise stored and incoming lysosomal substrate. 8 The iminosugar, N- butyldeoxynojirimycin (OGT 918) is an inhibitor of the ceramide-specific glucosyltransferase that initiates the glycosphingolipid biosynthetic pathway and catalyses the formation of glucocerebroside. Previous clinical trials have explored the use of OGT 918 at doses of up to 3 g daily as an -glucosidase-1 inhibitor in an attempt to decrease viral burden in patients with HIV-1 infection, in whom it was well tolerated. 9,1 In an in-vitro model of Gaucher s disease, the compound prevented the lysosomal storage of glucocerebroside, and, therefore, it was a candidate for clinical use in glycosphingolipid disorders. 6 Furthermore, oral administration of OGT 918 in gene-knockout mice that serve as a model of human G M2 gangliosidosis delays the onset of disease and prolongs the life of affected animals. 11,12 These and other studies of animals lacking key enzymes in the pathway of ganglioside biosynthesis generated by targeted gene disruption 13 show that deficiency of glycolipids may be compatible with normal growth to maturity. THE LANCET Vol 355 April 29, For personal use only. Not to be reproduced without permission of The Lancet.

2 Patient Sex Age at entry Genotype* Liver Spleen Avascular Previous enzyme into trial (years) (O/E)* (O/E)* necrosis therapy 1 F 62 N37S/L444P M 64 N37S/NK 1 5 SPX + 3 M 69 N37S/N37S M 49 N37S/L444P M 54 N37S/L444P M 52 N37S/L444P F 28 L444P/NK 2 4 SPX F 44 N37S/L444P F 43 N37S/NK 1 8 SPX + 1 F 39 R463C/NK 2 7 SPX + 11 F 35 N37S/IVS M 34 N37S/Rec( ) M 35 N37S/N37S M 22 N37S/N37S F 22 N37S/V394L F 58 N37S/N37S M 31 N37S/L444P F 32 N37S/L444P 2 SPX F 5 N37S/N37S M 51 N37S/NK F 38 N37S/N37S NK NK F 59 N37S/N37S 1 5 SPX + 23 M 62 N37S/IVS M 47 N37S/N37S F 43 N37S/N37S M 48 N37S/L444P 1 8 SPX + 27 M 3 N37S/NK F 3 N37S/N37S 2 9 SPX=previous splenectomy; NK=not identified from test panel at that centre. *O/E=observed/expected; for liver, organ volume expected is 2 14% of bodyweight, 16 for spleen, organ volume expected is 2% of bodyweight. 16 Patient had several pathological fractures. Table 1: Baseline characteristics of patients Decrease in formation of substrate therefore represents a potential approach for Gaucher s disease and other glycosphingolipid lysosomal disorders for which no specific treatment is available. We did a 1-year open-label study to investigate the efficacy and safety of OGT 918 for nonneuronopathic Gaucher s disease. Patients and methods The trial was approved by the local independent ethics committee of each participating centre and the appropriate national regulatory agencies. Patients From March to December, 1998, we enrolled adult patients aged years, who had non-neuronopathic Gaucher s disease (confirmed by leucocyte acid -glucosidase assay, mutation analysis of the glucocerebrosidase gene, or both) who were unable or unwilling to be treated with intravenous enzyme infusions from Gaucher s clinics in Cambridge, UK, Amsterdam, Prague, and Jerusalem. Patients were required to have measurable enlargement of the liver and spleen; we included splenectomised patients if their liver constituted more than 2 5% of bodyweight. We enrolled patients with an intact spleen if they had haemoglobin concentrations of less than 11 5 g/dl or platelet counts of less than /L. We excluded patients who had received enzyme therapy less than 3 months before enrolment, a history of lactose intolerance, or clinically severe diarrhoea; pregnant or breastfeeding women; and fertile individuals who did not agree to use adequate contraception. Before starting the trial, each patient gave written informed consent. Treatment and assessment Laboratory studies and studies in animals suggest that a steadystate plasma concentration of 1 2 g/ml (5 1 mol/l) of OGT 918 would be sufficient to inhibit partially the synthesis of glycosphingolipids. 6,11,12 Treatment with OGT 918 was started at a dose of 1 mg given orally three times daily. The target plasma concentrations for OGT 918 were up to 2 g/ml (1 mol/l). Our protocol allowed for dose adjustment throughout the trial based on peak or trough plasma concentrations, tolerability, and organ-volume response after 6 months of treatment. The dose could be decreased to 1 mg once daily or increased by 1 mg three times daily each month; a maximum dose of 3 mg three times daily was permitted. We did clinical examinations of all patients and measured liver and spleen volumes by spiral computed axial tomography or magnetic resonance imaging at baseline, 6 months, and 12 months. Haematological and routine biochemical variables were monitored monthly, as well as chitotriosidase activity, an enzyme marker that is released from activated macrophages and is raised in Gaucher s disease, 14,15 which makes it a sensitive marker of disease activity. Our primary purpose was to decrease the rate of glycolipid biosynthesis and organ volumes. For the first five patients recruited, we recorded a full kinetic profile of plasma OGT 918 concentrations after the first dose of a single 1 mg capsule, and after 1 month of treatment. For all patients, peak and trough plasma concentrations of OGT 918 were measured 2 4 weeks after starting treatment and at months 3, 6, and 12, as well as after dose increases. In these patients we also measured, throughout the study, concentrations of glycolipid (G M1 ganglioside) in circulating leucocytes, which are the main source of stored glycolipid in this disorder. 6 Analysis of N-linked oligosaccharides was also done to investigate the effects of OGT 918 on -glucosidases. At the end of the 12-month study, patients who benefited from OGT 918 treatment (defined by a reduction in organomegaly or improved symptoms) were able to continue treatment in an extended-use protocol. In these patients, safety data and biochemical and haematological measurements are collected every 3 months, and organ volumes are measured every 6 months. Statistical analysis Given that Gaucher s disease is a rare genetic disorder in which numbers of patients and specialist sites are limited for clinical research, the planned sample size of 24 patients was judged practical to implement. We included all treated patients in the analysis population and all patients with data in the statistical comparisons and calculation of descriptive statistics at each time point. Within-patient comparisons were done on the percentage change from baseline for liver and spleen volumes at 6 months and 12 months, and G M1 and chitotriosidase values at 12 months; absolute haemoglobin and platelet values were compared at 12 months. To limit variability, we used the means of values for screening and day 1, months 5 and 6, and months 11 and 12 to represent baseline, month 6, and month 12, respectively, for 1482 THE LANCET Vol 355 April 29, 2

3 haemoglobin, platelet counts, and chitotriosidase (if data were available). We used two-sided paired t tests for all comparisons with corresponding 95% CI calculated to show precision. Results Study population Seven (25%) of the 28 patients had undergone previous splenectomy and six (21%) had previously received enzyme therapy (table 1). If we assumed that healthy livers and spleens comprise 2 14% and 2% of bodyweight, respectively, 16 liver enlargement ranged from 1 1 to 2 7 times normal and splenic enlargement from 5 1 to 24 8 times normal at baseline. Safety Six patients withdrew during the study two because of gastrointestinal complaints (patients 2 and 13 at weeks 4 and 5, respectively), one because of concern about preexisting pulmonary hypertension (patient 18 at 4 months), one for religious reasons (patient 22 on day 2), one to have a child (patient 14 after 6-month assessment), and one because of an unrelated serious adverse event (patient 1, hepatic venous thrombosis 3 days after starting OGT 918). In the latter patient, primary hepatocellular carcinoma was later diagnosed and she died within a few months. Retrospectively, raised serum -fetoprotein concentrations were discovered that predated entry into the study. Patient 7 required hospital admission for a painful bone episode; she had experienced previous episodes of avascular necrosis and was not withdrawn from the study. The most frequent adverse effect was diarrhoea, which was reported in 79% of patients shortly after OGT 918 treatment was started. This effect improved spontaneously within several weeks or responded to treatment with loperamide or codeine phosphate. Diarrhoea recurred transiently in two of the three patients whose dose was increased after 6 months. Two patients (patients 1 and 5) have been withdrawn from the continuing extended-use protocol because of paraesthesiae. Electromyography confirmed peripheral neuropathy of the sensory axonal type; patient 1 has improved during follow-up. Pharmacokinetics In the first five recruited patients, after the first dose of 1 mg OGT 918, the time of maximum plasma concentration and half-life were 2 5 h and 6 3 h, Plasma OGT 918 ( g/ml) Peak Trough Time (months) Figure 1: Peak and trough plasma concentrations of OGT 918 after chronic dosing at 1 mg three times daily in 15 patients Patients 3, 9, 11, 15 17, 19 21, Percentage change from baseline Liver volume Spleen volume Screening Month 6 Study visit Month 12 Figure 2: Percentage changes from baseline in liver and spleen volume at 6 months and 12 months respectively, with a maximum concentration of 86 g/ml. Steady-state plasma concentrations of OGT 918 were achieved after 4 6 weeks of treatment (figure 1). The study protocol allowed dose titration from 1 mg to 3 mg three times daily, but the maximum administered dose was 2 mg three times daily in three patients; four patients doses were lowered to 1 mg once or twice daily. Efficacy Primary-endpoint data were available for 23 patients after 6 months and for 22 patients after 12 months. The anticipated effect of OGT 918 to lower the rate of glycolipid biosynthesis was confirmed by decline in cellsurface G M1 expression in leucocytes taken from the subgroup of five patients. Mean leucocyte G M1 values fell by 38 5% (95% CI , p= 6) over 12 months in these patients. Analysis of plasma glycoproteins showed no notable increase in detectable N-linked glucosylated oligosaccharide during treatment (< 1% of total N-linked oligosaccharide), which shows that there was no major inhibitory effect on N-glycan-processing -glucosidases at the concentrations of OGT 918 used. After 6 months and 12 months, organomegaly decreased significantly (figure 2, table 2). At 6 months, the mean decrease in liver volume from baseline was 7% ( , p< 1) and for spleen volume 15% ( , p< 1); at 12 months the decrease from baseline was 12% ( , p< 1) and 19% ( , p< 1). There was little change in mean haemoglobin concentrations or platelet counts after the first 6 months of treatment. After 12 months, the mean increase in THE LANCET Vol 355 April 29,

4 Patient Liver volume (L) Spleen volume (L) Haemoglobin (g/dl) Platelet count ( 1 9 /L) Chitotriosidase ( mol h 1 ml 1 )* months months months months months months months months months months months months months months months SPX SPX SPX SPX SPX SPX SPX SPX SPX NK NK SPX SPX SPX SPX Dots show results missing for patients who withdrew. SPX=previous splenectomy; NK=not identified from test panel at that centre. *Platelet count <1 1 9 /L before enrolment. Patient chitotriosidase deficient. Dose increase after 6 months of treatment. Chitotriosidase reference range mol h 1 ml 1. Table 2: Organ volume, haemoglobin, platelet count, and chitotriosidase values before and during OGT 918 treatment haemoglobin was 26 g/dl ( 5 to 57, p= 95) and platelet count /L ( , p= 14; table 2). Nine of the 22 patients who completed the study, had anaemia at baseline (haemoglobin <11 5 g/dl). Of these, five had a haemoglobin increase of more than 5 g/dl at month patients could be assessed at month 12 for platelet response (platelets < /L at baseline). Of these, four had an increase of more than /L platelets. There was a slow decrease in plasma chitotriosidase activity that continued throughout the study, up to 16 4% ( ) at month 12 (p< 1, table 2). Among the 18 patients in the extended study, there is evidence of continued improvement in organ volume and haematological variables. Discussion As well as assessing use of OGT 918 for Gaucher s disease, this study provided an opportunity to test the hypothesis that reducing the rate of the biosynthesis of substrate would improve the effects of an enzyme deficiency. 7,8 Our findings supported this principle, since glycolipid synthesis was decreased, organomegaly was significantly reduced, blood counts improved, and chitotriosidase activity was significantly lowered. 15,17 We chose non-neuronopathic Gaucher s disease for initial investigation because improvement of the clinical features of this disease has been shown after enzyme replacement. 1,4,5,18 The trial duration was 1 year because in non-neuronopathic Gaucher s disease the cellular target to decrease formation of substrate is an indirect one: the macrophages that are the principal focus of the disorder in the bone marrow, spleen, and liver become engorged with glycolipid, mainly because of phagocytosis of other blood cells rather than the accumulation of endogenous glycolipid-breakdown products generated in situ. In view of this indirect mechanism of storage, 3,8 we expected that measures to lower substrate biosynthesis would improve the manifestations of disease gradually. A decline in the expression of leucocyte cell-surface G M1 confirmed the ability of OGT 918 to lower the rate of glycolipid synthesis in vivo, which is a key observation, since circulating leucocytes are the main source of the stored glycolipid. 19,2 Although treatment to decrease substrate formation partially had been thought a possibility for Gaucher s disease and related glycosphingolipidoses for some years, 6,7 a well tolerated agent of the appropriate selectivity was needed. Before we did this trial, OGT 918 was known to have no serious unwanted effects in patients who had HIV- 1 infection. 9,1 The antiviral action of the iminosugar depends on inhibition of glucosidases in the endoplasmic reticulum, whereas its preferential effects on glycosphingolipid biosynthesis are seen at lower plasma concentrations. Limited antiviral activity was seen at plasma concentrations that exceeded those in this study (doses of 1 mg three times daily). Apart from gastrointestinal disturbances, the drug was well tolerated 9,1 and, therefore, was suitable for investigation in patients who had glycosphingolipidoses. Despite the clear improvement in spleen and liver volumes, as well as the progressive decline in chitotriosidase activity, haemoglobin concentrations and platelet counts did not increase significantly in all patients. No clear explanation is apparent for the more pronounced effects of OGT 918 on visceromegaly than on blood counts. However, the extent of reduction in liver and spleen volumes and plasma chitotriosidase activities compares favourably with the effects of current intravenous low-dose enzyme replacement therapy The length of the study did not permit informative analysis of any effect on bone disease. We have shown that the iminosugar OGT 918 has clinical potential, based on its dose-related selectivity towards the glucosyltransferase enzyme involved in glycolipid biosynthesis, to improve the outcome for patients 1484 THE LANCET Vol 355 April 29, 2

5 with Gaucher s disease. Partial inhibition of glycosphingolipid biosynthesis can be achieved in patients with Gaucher s disease, which shifts the substrate-product equilibrium in favour of lysosomal degradation and reduces the amount of accumulated substrate stored. The strategy depends on the presence of residual glucocerebrosidase activity but, as with enzyme-replacement treatment, prediction of the degree of response to OGT 918 from the glucocerebrosidase genotype has not been possible. The introduction of iminosugars that inhibit glycolipid formation offers particular promise for glycosphingolipid storage diseases for which enzyme treatment is unavailable or ineffective, or for which the clinical condition is beyond other means of correction. For glycolipid disorders such as Fabry disease and the G M2 gangliosidoses, in which direct lysosomal storage occurs as a result of failure to degrade glycolipids in situ and residual enzyme is present, decline in substrate formation may be beneficial. Peripheral neuropathy was seen in two patients in our study who underwent extended treatment; one patient after drug withdrawal. This effect could be unrelated to glycosphingolipid depletion and laboratory studies with other iminosugars, such as the galactose analogue, N- butyldeoxygalactonojirimycin, which is more selective for the glycosphingolipid pathway than OGT 918, will help to resolve this issue. 21,22 Further exploration of OGT 918 is warranted in patients with glycosphingolipid diseases. This compound could be used as monotherapy and has additional potential as a treatment combined with enzyme replacement. Combination therapy could lead to significantly lower doses of each drug being required to benefit patients with Gaucher s disease. Our data show that OGT 918 reverses many of the clinical signs and laboratory hallmarks of Gaucher s disease and may broaden the options for treating this family of metabolic diseases, including those with involvement of the nervous system. Contributors Timothy Cox was involved in the conception of the clinical trial and was principal investigator of the UK centre. Timothy Cox and Robin Lachmann, a clinical investigator who did the pharmacokinetic profile studies, were the principal investigators. Carla Hollak was principal investigator and physician in charge of patients at the Amsterdam Centre and did the pharmacokinetic profile studies. Johannes Aerts and Sonja van Weely contributed importantly to discussions as well as genotyping and enzymatic studies, including chitotriosidase measurements. Martin Hrebícek was principal investigator and physician in charge in Prague and contributed to several drafts of the paper. Frances Platt, Terry Butters, and Raymond Dwek identified the action of iminosugars on glycolipid biosynthesis and suggested their use in the treatment of the glycosphingolipidoses based on experiments done with cultured cells and animal studies, as well as contributing to the writing of the paper, assaying glycoconjugates, and advising on study conduct. Chris Moyses and Irene Gow contributed to protocol design, study-centre coordination, and trial logistics. Deborah Elstein was responsible for coordination of the study in Jerusalem, including collection of all primary data and Ari Zimran was responsible for overall conduct of the study in Jerusalem. Deborah Elstein and Ari Zimran also made individual contributions to the first and subsequent drafts of the paper. Acknowledgments We thank the patients who participated in this study; the study nurses Elizabeth Morris, Naomi Wright, Maaike Wiersma, and Tereza Paterová, as well as Adi Sasson, for their assistance in the daily conduct of the trial; Andy Garrett for statistical assessment; Joan Grantham and Lyn East for secretarial assistance; David Halsall and Marri Verhoek for genotype analysis; Deena Geudalia for technical assistance; Gabriele Reinkensmeier for G M1 analysis; and Howard Mellor for oligosaccharide analysis at the Oxford Glycobiology Institute. The Wellcome Trust provided the flow cytometer for G M1 analysis at Oxford University Biochemistry Department. RL is a Wellcome Trust Clinician Scientist. FP is a Lister Institute Fellow. CM and IG are employees and shareholders in Oxford GlycoSciences plc. RD, TB, and FP are shareholders in Oxford GlycoSciences, as is the University of Oxford. RD is currently a non-executive director of the company. RD, TB, and FP are in receipt of a grant from Oxford GlycoSciences to study glycolipid storage disorders. References 1 Beutler E, Grabowski GA. Glucocerebroside lipidoses: Gaucher disease. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic and molecular bases of inherited disease, 7th edn. New York: McGraw-Hill, 1995: Zimran A, Kay A, Gelbart T, et al. Gaucher disease: clinical, laboratory, radiologic, and genetic features of 53 patients. Medicine (Baltimore) 1992; 71: Cox TM, Schofield JP. Gaucher s disease: clinical features and natural history. Bailliere s Clin Haematol 1997; 1: Grabowski GA, Barton NW, Pastores G, et al. Enzyme therapy in nonneuronopathic Gaucher disease: comparative efficacy of mannoseterminated glucocerebrosidase from natural and recombinant sources. Ann Intern Med 1995; 122: Zimran A, Elstein D, Levy Lahad E, et al. Replacement therapy with imiglucerase for non-neuronopathic Gaucher s disease. Lancet 1995; 345: Platt FM, Neises GR, Dwek RA, Butters TD. N-butyldeoxynojirimycin is a novel inhibitor of glycolipid biosynthesis. J Biol Chem 1994; 269: Inokuchi J, Radin NS. Preparation of the active isomer of 1-phenyl-2- decanoylamino-3-morpholino-1-propanol inhibitor of murine glucocerebroside synthetase. J Lipid Chem 1987; 28: Conzelmann E, Sandhoff K. Partial enzyme deficiency: residual acitvities on the development of neurological disorders. Dev Neurosci ; 6: Tierney M, Pottage J, Kessler H, et al. The tolerability and pharmacokinetics of N-butyl-deoxynojirimycin in patients with advanced HIV disease (ACTG 1): the AIDS Clinical Trials Group (ACTG) of the National Institute of Allergy and Infectious Diseases. J Acquir Immune Defic Syndr Hum Retrovir 1995; 1: Fischl MA, Resnick L, Coombs, R, et al. The safety and efficacy of combination N-butyl-deoxynojirimycin (SC-48334) and zidovudine in patients with HIV-1 infection and 2 5 CD4 Cells/mm 3. J Acquir Immune Defic Syndr 1994; 7: Platt FM, Neises GR, Reinkensmeier G, et al. Prevention of lysosomal storage in Tay-Sachs mice treated with N-butyldeoxynojirimycin. Science 1997; 276: Jeyakumar M, Butters TD, Cortina-Borja M, et al. Delayed symptom onset and increased life expectancy in Sandhoff disease mice treated with N-butyldeoxynojirimycin. Proc Natl Acad Sci USA 1999; 96: Takamiya K, Yamamoto A, Furukawa K, et al. Mice with disrupted GM2/GD2 synthase gene lack complex gangliosides but exhibit only subtle defects in their nervous system. Proc Natl Acad Sci USA 1996; 93: Aerts JM, Hollak CE. Plasma and metabolic abnormalities in Gaucher s disease. Bailliere s Clin Haematol 1997; 1: Hollak CE, van Weely S, van Oers MH, Aerts JM. Marked elevation of plasma chitotriosidase activity: a novel hallmark of Gaucher disease. J Clin Invest 1994; 93: Elstein D, Abrahamov A, Hadas-Halpern I, et al. Low-dose frequency imiglucerase as a starting regimen of enzyme replacement therapy for patients with type I Gaucher disease. QJM 1998; 91: Guo Y, He W, Boer AM, et al. Elevated plasma chitotriosidase activity in various lysosomal storage disorders. J Inherit Metab Dis 1995; 18: Barton NW, Brady RO, Dambrosia JM, et al. Replacement therapy for inherited enzyme deficiency: macrophage-targeted glucocerebrosidase for Gaucher s disease. N Engl J Med 1991; 324: Brady RO. Glucocerebroside lipidoses: Gaucher s disease. In: Stanbury JB, Wyngaarden JB, Fredrickson DS, eds. The metabolic basis of inherited disease, 4th edn. New York: McGraw-Hill, 1978: Kattlove HE, Williams JC, Gaynor E, Spivack M, Bradley RM, Brady RO. Gaucher cells in chronic myelocytic leukemia: an acquired abnormality. Blood 1969; 33: Platt FM, Neises GR, Karlsson GB, Dwek RA, Butters TD. N-butyldeoxygalactonojirimycin inhibits glycolipid biosynthesis but does not affect N-linked oligosaccharide processing. J Biol Chem 1994; 269: Andersson U, Butters TD, Dwek RA, Platt FM. N-butyldeoxygalactonojirimycin: a more selective inhibitor of glycosphingolipid biosynthesis than N-butyldeoxynijirimycin, in vivo and in vitro. Biochem Pharmacol 2; 59: THE LANCET Vol 355 April 29,