1275
Comparison of oral iron chelator L1 and desferrioxamine in iron-loaded patients
The efficacy of the oral iron chelator 1,2-dimethyl3-hydroxypyrid-4-one (L1) was compared with that of subcutaneous desferrioxamine in 26 patients with transfusional iron overload. Immediately after red-cell transfusion, 20 patients to randomised receive either were a 12 h desferrioxamine (50 mg/kg daily as subcutaneous infusion), or L1 (50 mg/kg daily by mouth). Patients were evaluated during treatment with the other drug after transfusion the next month. Mean (SD) daily urinary iron excretion was lower during L1 than during desferrioxamine (12·3 [6·7] vs 18·2 [15·3] mg/day). In 5 patients the dose of L1 was raised from 50 to 75 mg/kg daily; mean urinary iron excretion rose from 13·8 (7·0) mg/day to 26·7 (17·8) mg/day, comparable with that during desferrioxamine (24·9 [24·3] mg/day). Faecal iron excretion rose slightly over baseline in 6 patients studied during L1 administration (from 8·5 [0·9] mg/day to 12·2 [0·9] mg/day). Pharmacokinetic studies showed an elimination half-life for L1 of 117-237 min. Studies in dogs and in volunteers showed no absorption of the L1-iron complex, excluding a contribution of absorption of intraluminal complexes of L1 and food iron to urinary iron excretion. Further animal toxicity testing is needed before L1 can be studied in a broader group of patients.
Introduction The institution of regular programmes of red-cell transfusion has improved the quality of life in patients with thalassaemia major, steroid-unresponsive DiamondBlackfan anaemia, and sickle-cell disease. The penalty of this success has been the emergence of transfusional iron overload.1 For these patients, nightly subcutaneous infusion of desferrioxamine has been the only chelation treatment available, but it has the disadvantages of a cumbersome, parenteral mode of administration, which results in reduced compliance during adolescence, serious toxicity with intensive use2 and high cost. To improve compliance in transfused patients, and to provide greater access to treatment, a safe, inexpensive, orally effective chelator is
patients with transfusional iron overload in whom desferrioxamine treatment was failing. 26
Patients and methods transfusion-dependent patients with a mean age of 22 years (range 8-49 years) were admitted to hospital for supervised adminstration of L 1. 24 patients had homozygous thalassaemia, and 2 had Diamond-Blackfan anaemia. 21 patients were considered to be at risk of accelerated iron loading: 6 had been poorly compliant with nightly subcutaneous desferrioxamine for at least 2 years (defined as less than 80% of the prescribed amount of drug taken), and 6 had refused subcutaneous desferrioxamine completely. 5 patients had desferrioxamine-induced neurotoxicity2 which limited the dose that could be given. 10 other patients who were eligible for the study refused to take part. All patients had evidence of moderate to severe iron overload (see table). Liver dysfunction was present in 17 patients. 9 patients had insulin-dependent diabetes mellitus and 1 patient showed evidence of hypothyroidism and hypoparathyroidism. Flat responses to gonadorelin stimulation testing were present in 13 patients. 8 patients required digoxin and diuretics for iron-related cardiac dysfunction. The mean transfusional iron exposure (iron transfused since the initial transfusion per kg of present body weight) was 2-11 (0-3) g/kg. Mean total desferrioxamine taken (estimated since start of chelation therapy per kg present body weight) was 89-4 (31-9) 26
g/kg. L1has been used orally in human studies in the UK,3,4 but it is available commercially. On the basis of animal studies by other investigators,5,6 we applied to the Health Protection Branch, Health and Welfare Canada, for permission to produce and administer Ll to patients with severe transfusional iron overload in whom desferrioxamine is not a therapeutic option. In September, 1988, permission was obtained for a pilot study (file no 9427-H1117-41C, HPB, Ottawa, Canada). The study was approved by our hospital’s human subject review committee. Written informed consent was obtained from each patient, or from a parent before entry to the not
study. Ll was synthesised by direct reaction in aqueous solution of methylamine and maltolIt was purified by four recrystallisations from water. The purity was established by melting point, elemental analysis, 400 MHz proton nuclear magnetic resonance, infrared and mass spectroscopy, and thin-layer chromatography. Ll and ferric chloride (3-05/1) were dissolved in water to yield a solution of pH
0-5, which
was
then neutralised with three moles of sodium
hydroxide per mole of iron. The sample was lyophilised to produce an orange powder containing (Ll)3Fe of 98% purity. 300 mg capsules of Llwere produced by the department of pharmacy at the hospital for Sick Children, Toronto.
required. A cohort of our transfusion-dependent, iron-overloaded patients have not achieved net negative iron balance, either because substantial neurotoxic effects limit the amount of desferrioxamine they can take or because of poor compliance with the regimen of subcutaneous desferrioxamine. They are thus at high risk of organ toxicity and death from iron overload. Trials of the iron chelator 1,2-dimethyl-3hydroxypyrid-4-one (LI) in iron-loaded patients in the United Kingdom have shown short-term efficacy in a small number of patients. 3,4 We report our experience with Ll in
ADDRESSES: Divisions of Haematology-Oncology (N. F. Olivieri, MD, Prof M H Freedman, MD) and Clinical Pharmacology (G. Koren, MD, C. Hermann, MD, Y. Bentur, MD, D. Chung, BSc, J. Klein, MSc) and Departments of Paediatrics (N. F. Olivieri, G. Koren, Prof M. H. Freedman), Clinical Biochemistry (P. St. Louis, PhD), and Biochemistry (Prof D. M. Templeton, MD), Hospital for Sick Children, Toronto, and Departments of Chemistry (Prof R. A. McClelland, PhD) and Clinical Biochemistry (Prof D. M Templeton), University of Toronto, Toronto, Canada. Correspondence to Dr N. F. Olivieri, Hospital for Sick Children, Division of Haematology-Oncology, 555 University Avenue, Toronto, Canada M5G 1X8.
1276
CLINICAL AND LABORATORY DATA
For the randomised crossover comparison of Ll and desferrioxamine 20 patients were admitted after their monthly transfusions. Patients who took desferrioxamine at home were told to discontinue infusions of the drug 72 h before transfusion. A controlled-iron diet (9-2 [1’6] mg/day) was started on admission. Dietary contents were adjusted to individual preferences by a dietitian, and the iron content was calculated from patient diaries reviewed daily with each patient by one of us. The next day, urine collection for baseline iron measurement was begun, with all urine passed into acid-washed plastic containers under nursing supervision. Blood samples were taken at baseline and on days 1, 3, and 5 for cell counts, differential, serum electrolytes, blood urea nitrogen and creatinine, total protein, albumin, aspartate and alanine aminotransferases, calcium, phosphorus, magnesium, copper, and zinc. Desferrioxamine concentration was measured in the first serum sample. 24 h after admission, each patient began treatment as randomised with either L1 or desferrioxamine, each at a dose of 50 mg/kg daily. L was given in divided doses at 0800 h, 1600 h, and 2400 h, at least 1 h before and after ingestion of food or liquids. Desferrioxamine was infused over 12 h. Each drug was given on days 2, 3, and 4, and all urine was collected in 24 h volumes and analysed for iron. All patients received a diet with sufficient vitamin C, and were given an extra 100 mg ascorbic acid per day during administration of the chelators. Patients were discharged on the morning of day 6. After the next blood transfusion 3-4 weeks later, each patient was readmitted and received the other drug under the same protocol; equal numbers received Ll first and desferrioxamine first. Each patient therefore served as his or her own control in this phase of the study. The protocol ensured that each patient was studied at a similar haematocrit during both treatments, which is necessary in patients with thalassaemia because urinary and faecal iron excretion can vary with haemoglobin level.8 5 patients who took part in the crossover study were readmitted to hospital immediately after a later transfusion for administration of Ll at a dose of 75 mg/kg daily with the same protocol, to determine the effect on urinary iron excretion of a higher dose of Ll. 6 patients were admitted to evaluate total iron excretion (faecal and urinary) on Ll (75 mg/kg daily). Throughout this study, patients were maintained on the standard iron diet. On admission, each patient received 10 uCi chromium-51-labelled chromic chloride orally in 5 ml distilled water. The stool radioactivity was
counted to measure excretion of 5’Cr. After 95% of the label had been recovered in the stool (3-5 days in all patients), all urine and stools were saved in acid-washed containers for the next 72 h, in 24 h and 72 h collections, respectively. When these collections were completed, the patients were given the first dose of Ll followed 3 h later by a second dose of 10 p Ci 5’CrCl3. Administration of Lthen continued every 8 h for the remainder of the study. When 95% of the second dose of 51Cr had been recovered in stool, three 72 h collections of all stool and daily collections of urine were analysed for iron. Use of the 5’Cr label showed that, at the start of the baseline faecal iron collection, all iron consumed before the institution of iron restriction had passed out of the gastrointestinal tract. Similarly, the collections for faecal iron on Lwere started only after stool formed before the first ingestion of L1 had been excreted.9 Studies were done in dogs and in 2 volunteers to determine whether absorption of intraluminal dietary iron from the gut is facilitated by administration of Ll. After an overnight fast, 4 mongrel dogs were anaesthetised with 10 mg/kg pentobarbitone and mechanically ventilated. The jugular vein, femoral artery, and both ureters were cannulated, and a duodenal tube was introduced. After stabilisation for 1 h, urine was collected for 20 min and a blood sample was drawn at the midpoint of the collection. In 2 dogs 60 mg/kg elemental iron (as a 25% aqueous solution of ferrous sulphate) was given through the duodenal tube, which was then flushed with 20 ml distilled water. Blood and urine samples were collected for 6 h. The other 2 dogs received the equivalent of 60 mg/kg iron as (Ll)3Fe suspended in 25 ml 1% methylcellulose and the duodenal tube was flushed with 20 ml 1 % methylcellulose. 2 healthy adult volunteers ingested the (Ll)3Fe complex, prepared in vitro as described above, equivalent to 60 mg elemental iron. Urine was then collected for 48 h and analysed for iron. In 4 patients who took part in the iron balance study, the pharmacokinetics of Lwere studied after the first dose of the drug. Samples (2 ml) were drawn through an indwelling venous catheter placed in the antecubital fossa just before Ll ingestion and then every 15 min to 1 h, every 30 min to 3 h, and every 1 h to 6 h. Serum was immediately separated and stored at - 20°C for analysis within 2 months. The elimination half-life of Ll was calculated by least squares regression of the decay phase of serum concentrations. Urinary iron was determined by direct atomic absorption spectrometry of a diluted sample with a Varian Spectra AA-10 atomic absorption spectrophotometer (Varian Techtron, Australia). The analyser was calibrated with standard iron solutions prepared from BDH analytic grade iron standard. Stool was processed for iron analysis’O by shaking with an equal volume of water in the tared plastic collection containers, and triplicate aliquots of 1-3 g were digested by gentle heating in 10 ml concentrated nitric acid (GR; Merck, Darmstadt) . After volume reduction, the procedure was repeated three times with 10 ml nitric acid/hydrogen peroxide in equal proportions. The final concentrate was diluted to 25 ml with water. Iron was measured by flame atomic absorption spectrometry (Perkin-Elmer 306 spectrometer). Method blanks were prepared with each batch of samples, and measurements were shown to be free from matrix effects by standard addition. Desferrioxamine was measured at baseline by
iron excretion during treatment with L1 and desferrioxamine (DFO) in 5 patients who took two doses of L1.
Fig 1-Urinary
1277
Fig 2-Correlation of mean daily iron excretion induced by L1 and desferrioxamine (each at 50 mg/kg daily) in 20 patients.
high-performance liquid chromatography (HPLC)l1 to ensure that washout of the drug from previous treatment was complete. The detection limit was 1 (ig/ml. Lwas analysed by an HPLC method with ultraviolet detection (to be published elsewhere); recovery was 81 (6-5)% and the detection limit was 1 ug/ml. The coefficient of variation at concentrations measured in our patients was less than 5%. The quantitative measures of iron excretion before and during chelation treatment were compared by the Student’s t test. Correlation between variables was tested by linear least squares regression analysis. Values are expressed as mean and SD.
Results
fell from 126-2 (153) g/1 on the morning before the first dose of Ll, to 106 0 (12-0) g/I on the morning of the day of discharge. Faecal iron excretion during treatment accounted for 15 % (range 0-28%) of the total iron excretion. In all 4 dogs used in the studies with the (Ll)3Fe complex blood pressure was stable throughout the experiment. After intraduodenal administration of iron as ferrous sulphate, plasma iron rose from 10-3 (2-5) mg/1 to 42-8 (3-9) mg/1 at 2 h and gradually fell to 295 (28) mg/16 h after administration. Intraduodenal administration of the (Ll)3Fe complex did not cause plasma iron to rise and it fell from 11.4 (1-4) mg/1 to 4-2 (03) mg/1 at 6 h. Measurement of urinary iron excretion for 48 h after ingestion of the (Ll)3Fe complex by 2 volunteers showed no urinary excretion of the (Ll)3Fe complex; no iron could be detected in the urine and the red complex itself was absent. In all 4 patients who took part in the pharmacokinetic study, Ll was detectable at concentrations above 3 mg/1 in the first sample (15 min). Elimination half-life varied between 117 and 237 min (mean 157 min). Peak serum concentrations occurred 45-90 min after ingestion of Ll, and varied between 7-2 mg/1 and 21 mg/1.
Discussion the difficulties associated with its long-term administration to transfusion-dependent patients, desferrioxamine remains the mainstay of iron parenteral chelation treatment.Because of the mode of administration, patient compliance becomes erratic in children as young as 10 years old and it declines further during the teenage years, at the time when transfusional iron accumulation is accelerated.12 There is an increasing body of evidence that iron chelation reduces the morbidity and mortality associated with iron overload. Regular desferrioxamine treatment removes hepatic iron and prevents hepatic fibrosis in iron-loaded patients." Survival, linear growth, and sexual maturation may also be improved.14,lS Although prevention of cardiac disease in patients who start treatment after the age of 10 years has been shown in small studies,16 reversal of cardiac dysfunction has been less convincingly demonstrated. In some cases, clinical improvement and unexpectedly long survival have followed intensification of treatment in patients with iron-related cardiac dysfunction, but in other patients, progressive cardiac deterioration has not been changed.l’ The key to success in chelation treatment seems to be prevention of accumulation of iron in the myocardium, with its attendant cardiac dysfunction.! The poor compliance with desferrioxamine treatment of many patients with thalassaemia major is, therefore, a major obstacle to prevention of iron-related morbidity and
Despite
In the 20 patients who took part in the crossover study mean
urinary iron excretion was 12-3 (6-7) mg/day during Ll treatment and 18.2 (15-3) mg/day during desferrioxamine.
Increasing the dose of Ll to 75 mg/kg daily in 5 patients raised the urinary iron excretion from 13-8 (7-0) mg/day (on 50 mg/kg Ll daily) to 26-7 (17-8) mg/day (p < 0-05); this was comparable with the urinary excretion of 24-9 (24-3) mg/day during desferrioxamine treatment in these 5 patients (fig 1). There was a highly significant correlation between daily urinary iron excretion during Ll (50 mg/kg daily) and during desferrioxamine (fig 2). Urinary iron excretion was significantly correlated with serum ferritin during both treatment periods (r=0-45, p
Comparison of oral iron chelator L1 and desferrioxamine in iron-loaded patients
The efficacy of the oral iron chelator 1,2-dimethyl3-hydroxypyrid-4-one (L1) was compared with that of subcutaneous desferrioxamine in 26 patients with transfusional iron overload. Immediately after red-cell transfusion, 20 patients to randomised receive either were a 12 h desferrioxamine (50 mg/kg daily as subcutaneous infusion), or L1 (50 mg/kg daily by mouth). Patients were evaluated during treatment with the other drug after transfusion the next month. Mean (SD) daily urinary iron excretion was lower during L1 than during desferrioxamine (12·3 [6·7] vs 18·2 [15·3] mg/day). In 5 patients the dose of L1 was raised from 50 to 75 mg/kg daily; mean urinary iron excretion rose from 13·8 (7·0) mg/day to 26·7 (17·8) mg/day, comparable with that during desferrioxamine (24·9 [24·3] mg/day). Faecal iron excretion rose slightly over baseline in 6 patients studied during L1 administration (from 8·5 [0·9] mg/day to 12·2 [0·9] mg/day). Pharmacokinetic studies showed an elimination half-life for L1 of 117-237 min. Studies in dogs and in volunteers showed no absorption of the L1-iron complex, excluding a contribution of absorption of intraluminal complexes of L1 and food iron to urinary iron excretion. Further animal toxicity testing is needed before L1 can be studied in a broader group of patients.
Introduction The institution of regular programmes of red-cell transfusion has improved the quality of life in patients with thalassaemia major, steroid-unresponsive DiamondBlackfan anaemia, and sickle-cell disease. The penalty of this success has been the emergence of transfusional iron overload.1 For these patients, nightly subcutaneous infusion of desferrioxamine has been the only chelation treatment available, but it has the disadvantages of a cumbersome, parenteral mode of administration, which results in reduced compliance during adolescence, serious toxicity with intensive use2 and high cost. To improve compliance in transfused patients, and to provide greater access to treatment, a safe, inexpensive, orally effective chelator is
patients with transfusional iron overload in whom desferrioxamine treatment was failing. 26
Patients and methods transfusion-dependent patients with a mean age of 22 years (range 8-49 years) were admitted to hospital for supervised adminstration of L 1. 24 patients had homozygous thalassaemia, and 2 had Diamond-Blackfan anaemia. 21 patients were considered to be at risk of accelerated iron loading: 6 had been poorly compliant with nightly subcutaneous desferrioxamine for at least 2 years (defined as less than 80% of the prescribed amount of drug taken), and 6 had refused subcutaneous desferrioxamine completely. 5 patients had desferrioxamine-induced neurotoxicity2 which limited the dose that could be given. 10 other patients who were eligible for the study refused to take part. All patients had evidence of moderate to severe iron overload (see table). Liver dysfunction was present in 17 patients. 9 patients had insulin-dependent diabetes mellitus and 1 patient showed evidence of hypothyroidism and hypoparathyroidism. Flat responses to gonadorelin stimulation testing were present in 13 patients. 8 patients required digoxin and diuretics for iron-related cardiac dysfunction. The mean transfusional iron exposure (iron transfused since the initial transfusion per kg of present body weight) was 2-11 (0-3) g/kg. Mean total desferrioxamine taken (estimated since start of chelation therapy per kg present body weight) was 89-4 (31-9) 26
g/kg. L1has been used orally in human studies in the UK,3,4 but it is available commercially. On the basis of animal studies by other investigators,5,6 we applied to the Health Protection Branch, Health and Welfare Canada, for permission to produce and administer Ll to patients with severe transfusional iron overload in whom desferrioxamine is not a therapeutic option. In September, 1988, permission was obtained for a pilot study (file no 9427-H1117-41C, HPB, Ottawa, Canada). The study was approved by our hospital’s human subject review committee. Written informed consent was obtained from each patient, or from a parent before entry to the not
study. Ll was synthesised by direct reaction in aqueous solution of methylamine and maltolIt was purified by four recrystallisations from water. The purity was established by melting point, elemental analysis, 400 MHz proton nuclear magnetic resonance, infrared and mass spectroscopy, and thin-layer chromatography. Ll and ferric chloride (3-05/1) were dissolved in water to yield a solution of pH
0-5, which
was
then neutralised with three moles of sodium
hydroxide per mole of iron. The sample was lyophilised to produce an orange powder containing (Ll)3Fe of 98% purity. 300 mg capsules of Llwere produced by the department of pharmacy at the hospital for Sick Children, Toronto.
required. A cohort of our transfusion-dependent, iron-overloaded patients have not achieved net negative iron balance, either because substantial neurotoxic effects limit the amount of desferrioxamine they can take or because of poor compliance with the regimen of subcutaneous desferrioxamine. They are thus at high risk of organ toxicity and death from iron overload. Trials of the iron chelator 1,2-dimethyl-3hydroxypyrid-4-one (LI) in iron-loaded patients in the United Kingdom have shown short-term efficacy in a small number of patients. 3,4 We report our experience with Ll in
ADDRESSES: Divisions of Haematology-Oncology (N. F. Olivieri, MD, Prof M H Freedman, MD) and Clinical Pharmacology (G. Koren, MD, C. Hermann, MD, Y. Bentur, MD, D. Chung, BSc, J. Klein, MSc) and Departments of Paediatrics (N. F. Olivieri, G. Koren, Prof M. H. Freedman), Clinical Biochemistry (P. St. Louis, PhD), and Biochemistry (Prof D. M. Templeton, MD), Hospital for Sick Children, Toronto, and Departments of Chemistry (Prof R. A. McClelland, PhD) and Clinical Biochemistry (Prof D. M Templeton), University of Toronto, Toronto, Canada. Correspondence to Dr N. F. Olivieri, Hospital for Sick Children, Division of Haematology-Oncology, 555 University Avenue, Toronto, Canada M5G 1X8.
1276
CLINICAL AND LABORATORY DATA
For the randomised crossover comparison of Ll and desferrioxamine 20 patients were admitted after their monthly transfusions. Patients who took desferrioxamine at home were told to discontinue infusions of the drug 72 h before transfusion. A controlled-iron diet (9-2 [1’6] mg/day) was started on admission. Dietary contents were adjusted to individual preferences by a dietitian, and the iron content was calculated from patient diaries reviewed daily with each patient by one of us. The next day, urine collection for baseline iron measurement was begun, with all urine passed into acid-washed plastic containers under nursing supervision. Blood samples were taken at baseline and on days 1, 3, and 5 for cell counts, differential, serum electrolytes, blood urea nitrogen and creatinine, total protein, albumin, aspartate and alanine aminotransferases, calcium, phosphorus, magnesium, copper, and zinc. Desferrioxamine concentration was measured in the first serum sample. 24 h after admission, each patient began treatment as randomised with either L1 or desferrioxamine, each at a dose of 50 mg/kg daily. L was given in divided doses at 0800 h, 1600 h, and 2400 h, at least 1 h before and after ingestion of food or liquids. Desferrioxamine was infused over 12 h. Each drug was given on days 2, 3, and 4, and all urine was collected in 24 h volumes and analysed for iron. All patients received a diet with sufficient vitamin C, and were given an extra 100 mg ascorbic acid per day during administration of the chelators. Patients were discharged on the morning of day 6. After the next blood transfusion 3-4 weeks later, each patient was readmitted and received the other drug under the same protocol; equal numbers received Ll first and desferrioxamine first. Each patient therefore served as his or her own control in this phase of the study. The protocol ensured that each patient was studied at a similar haematocrit during both treatments, which is necessary in patients with thalassaemia because urinary and faecal iron excretion can vary with haemoglobin level.8 5 patients who took part in the crossover study were readmitted to hospital immediately after a later transfusion for administration of Ll at a dose of 75 mg/kg daily with the same protocol, to determine the effect on urinary iron excretion of a higher dose of Ll. 6 patients were admitted to evaluate total iron excretion (faecal and urinary) on Ll (75 mg/kg daily). Throughout this study, patients were maintained on the standard iron diet. On admission, each patient received 10 uCi chromium-51-labelled chromic chloride orally in 5 ml distilled water. The stool radioactivity was
counted to measure excretion of 5’Cr. After 95% of the label had been recovered in the stool (3-5 days in all patients), all urine and stools were saved in acid-washed containers for the next 72 h, in 24 h and 72 h collections, respectively. When these collections were completed, the patients were given the first dose of Ll followed 3 h later by a second dose of 10 p Ci 5’CrCl3. Administration of Lthen continued every 8 h for the remainder of the study. When 95% of the second dose of 51Cr had been recovered in stool, three 72 h collections of all stool and daily collections of urine were analysed for iron. Use of the 5’Cr label showed that, at the start of the baseline faecal iron collection, all iron consumed before the institution of iron restriction had passed out of the gastrointestinal tract. Similarly, the collections for faecal iron on Lwere started only after stool formed before the first ingestion of L1 had been excreted.9 Studies were done in dogs and in 2 volunteers to determine whether absorption of intraluminal dietary iron from the gut is facilitated by administration of Ll. After an overnight fast, 4 mongrel dogs were anaesthetised with 10 mg/kg pentobarbitone and mechanically ventilated. The jugular vein, femoral artery, and both ureters were cannulated, and a duodenal tube was introduced. After stabilisation for 1 h, urine was collected for 20 min and a blood sample was drawn at the midpoint of the collection. In 2 dogs 60 mg/kg elemental iron (as a 25% aqueous solution of ferrous sulphate) was given through the duodenal tube, which was then flushed with 20 ml distilled water. Blood and urine samples were collected for 6 h. The other 2 dogs received the equivalent of 60 mg/kg iron as (Ll)3Fe suspended in 25 ml 1% methylcellulose and the duodenal tube was flushed with 20 ml 1 % methylcellulose. 2 healthy adult volunteers ingested the (Ll)3Fe complex, prepared in vitro as described above, equivalent to 60 mg elemental iron. Urine was then collected for 48 h and analysed for iron. In 4 patients who took part in the iron balance study, the pharmacokinetics of Lwere studied after the first dose of the drug. Samples (2 ml) were drawn through an indwelling venous catheter placed in the antecubital fossa just before Ll ingestion and then every 15 min to 1 h, every 30 min to 3 h, and every 1 h to 6 h. Serum was immediately separated and stored at - 20°C for analysis within 2 months. The elimination half-life of Ll was calculated by least squares regression of the decay phase of serum concentrations. Urinary iron was determined by direct atomic absorption spectrometry of a diluted sample with a Varian Spectra AA-10 atomic absorption spectrophotometer (Varian Techtron, Australia). The analyser was calibrated with standard iron solutions prepared from BDH analytic grade iron standard. Stool was processed for iron analysis’O by shaking with an equal volume of water in the tared plastic collection containers, and triplicate aliquots of 1-3 g were digested by gentle heating in 10 ml concentrated nitric acid (GR; Merck, Darmstadt) . After volume reduction, the procedure was repeated three times with 10 ml nitric acid/hydrogen peroxide in equal proportions. The final concentrate was diluted to 25 ml with water. Iron was measured by flame atomic absorption spectrometry (Perkin-Elmer 306 spectrometer). Method blanks were prepared with each batch of samples, and measurements were shown to be free from matrix effects by standard addition. Desferrioxamine was measured at baseline by
iron excretion during treatment with L1 and desferrioxamine (DFO) in 5 patients who took two doses of L1.
Fig 1-Urinary
1277
Fig 2-Correlation of mean daily iron excretion induced by L1 and desferrioxamine (each at 50 mg/kg daily) in 20 patients.
high-performance liquid chromatography (HPLC)l1 to ensure that washout of the drug from previous treatment was complete. The detection limit was 1 (ig/ml. Lwas analysed by an HPLC method with ultraviolet detection (to be published elsewhere); recovery was 81 (6-5)% and the detection limit was 1 ug/ml. The coefficient of variation at concentrations measured in our patients was less than 5%. The quantitative measures of iron excretion before and during chelation treatment were compared by the Student’s t test. Correlation between variables was tested by linear least squares regression analysis. Values are expressed as mean and SD.
Results
fell from 126-2 (153) g/1 on the morning before the first dose of Ll, to 106 0 (12-0) g/I on the morning of the day of discharge. Faecal iron excretion during treatment accounted for 15 % (range 0-28%) of the total iron excretion. In all 4 dogs used in the studies with the (Ll)3Fe complex blood pressure was stable throughout the experiment. After intraduodenal administration of iron as ferrous sulphate, plasma iron rose from 10-3 (2-5) mg/1 to 42-8 (3-9) mg/1 at 2 h and gradually fell to 295 (28) mg/16 h after administration. Intraduodenal administration of the (Ll)3Fe complex did not cause plasma iron to rise and it fell from 11.4 (1-4) mg/1 to 4-2 (03) mg/1 at 6 h. Measurement of urinary iron excretion for 48 h after ingestion of the (Ll)3Fe complex by 2 volunteers showed no urinary excretion of the (Ll)3Fe complex; no iron could be detected in the urine and the red complex itself was absent. In all 4 patients who took part in the pharmacokinetic study, Ll was detectable at concentrations above 3 mg/1 in the first sample (15 min). Elimination half-life varied between 117 and 237 min (mean 157 min). Peak serum concentrations occurred 45-90 min after ingestion of Ll, and varied between 7-2 mg/1 and 21 mg/1.
Discussion the difficulties associated with its long-term administration to transfusion-dependent patients, desferrioxamine remains the mainstay of iron parenteral chelation treatment.Because of the mode of administration, patient compliance becomes erratic in children as young as 10 years old and it declines further during the teenage years, at the time when transfusional iron accumulation is accelerated.12 There is an increasing body of evidence that iron chelation reduces the morbidity and mortality associated with iron overload. Regular desferrioxamine treatment removes hepatic iron and prevents hepatic fibrosis in iron-loaded patients." Survival, linear growth, and sexual maturation may also be improved.14,lS Although prevention of cardiac disease in patients who start treatment after the age of 10 years has been shown in small studies,16 reversal of cardiac dysfunction has been less convincingly demonstrated. In some cases, clinical improvement and unexpectedly long survival have followed intensification of treatment in patients with iron-related cardiac dysfunction, but in other patients, progressive cardiac deterioration has not been changed.l’ The key to success in chelation treatment seems to be prevention of accumulation of iron in the myocardium, with its attendant cardiac dysfunction.! The poor compliance with desferrioxamine treatment of many patients with thalassaemia major is, therefore, a major obstacle to prevention of iron-related morbidity and
Despite
In the 20 patients who took part in the crossover study mean
urinary iron excretion was 12-3 (6-7) mg/day during Ll treatment and 18.2 (15-3) mg/day during desferrioxamine.
Increasing the dose of Ll to 75 mg/kg daily in 5 patients raised the urinary iron excretion from 13-8 (7-0) mg/day (on 50 mg/kg Ll daily) to 26-7 (17-8) mg/day (p < 0-05); this was comparable with the urinary excretion of 24-9 (24-3) mg/day during desferrioxamine treatment in these 5 patients (fig 1). There was a highly significant correlation between daily urinary iron excretion during Ll (50 mg/kg daily) and during desferrioxamine (fig 2). Urinary iron excretion was significantly correlated with serum ferritin during both treatment periods (r=0-45, p
