Urate Oxidase in Prevention and Treatment of Hyperuricemia

Home , Allopurinol, Urate oxidase

Urate oxidase in prevention and treatment of hyperuricemia associated with lymphoid malignancies C-H Pui1,2, MV Relling1,2, F Lascombes3, PL Harrison1, A Struxiano3, J-M Mondesir3, RC Ribeiro1,2, JT Sandlund1,2, GK Rivera1,2, WE Evans1,2 and HH Mahmoud1,2

1St Jude Children’s Research Hospital and 2the University of Tennessee, Memphis, College of Medicine, Memphis, TN, USA; and 3Sanoﬁ Winthrop Inc, Paris, France

Standard prophylaxis and treatment of malignancy-associated ily excretable metabolite with 5- to 10-fold higher solubility hyperuricemia in the USA has been allopurinol with vigorous than uric acid – offers a potentially useful alternative for the hydration, urinary alkalinization and osmotic diuresis. Urate 14 oxidase, the enzyme that converts uric acid to allantoin (a read- prophylaxis and treatment of hyperuricemia. However, the ily excreted metabolite that has 5- to 10-fold higher solubility non-recombinant product (Uricozyme; Sanofi Winthrop, than uric acid), is an alternative therapy; however, few pub- Paris, France), purified from cultures of Aspergillus flavus,is lished findings support this practice. Between February 1994 available only in France and Italy and only a few published and December 1996, we administered non-recombinant urate findings can be marshaled to support routine administration oxidase (Uricozyme) to 126 children with newly diagnosed non- of this agent for the prevention and therapy of hyperurice- B cell acute lymphoblastic leukemia (ALL) during the first 5 15–18 days of chemotherapy with methotrexate, 6-mercaptopurine or mia. both. Their blood levels of uric acid and other indicators of In December 1991, we began pharmacodynamic studies of tumor lysis were measured at diagnosis and during treatment methotrexate and mercaptopurine given before conventional and then compared with findings in 129 similarly treated his- remission induction therapy in patients with newly diagnosed torical controls who had received allopurinol to control hyper- ALL,19–21 with allopurinol used initially to prevent or treat hyp- uricemia. Clinical responses to urate oxidase were also determ- eruricemia.19–21 Because of confounding effects of allopurinol ined in eight patients with newly diagnosed B cell ALL or 22 advanced-stage non-Hodgkin lymphoma. Patients treated with on purine biosynthesis, we replaced it with urate oxidase in urate oxidase had rapid and significantly greater decreases in February 1994. This substitution allowed us to assess the effi- their blood uric acid levels than did the historical controls cacy and toxicity of urate oxidase in a large group of similarly (median maximal level during treatment, 2.3 vs 3.9 mg/dl, treated cancer patients. The results suggest a substantial P Ͻ 0.001). They also had lower creatinine (0.6 vs 0.7 mg/dl, advantage from the use of urate oxidase in the management Ͻ ؍ P 0.01) and blood urea nitrogen (11 vs 24 mg/dl, P 0.001) of chemotherapy-induced hyperuricemia. levels. Similar findings were made in the eight cases of B cell ALL or non-Hodgkin lymphoma. None of the patients required dialysis for acute renal failure. Six (4.5%) of the 134 children given urate oxidase had allergic reactions, manifested primarily Materials and methods by urticaria, bronchospasm and hypoxemia. Thus, non-recombinant urate oxidase is a more effective uricolytic agent than Patients allopurinol but is associated with acute hypersensitivity reactions, even in patients without a history of allergy. Keywords: acute lymphoblastic leukemia; urate oxidase; uric acid; Of 176 patients with non-B cell ALL, five with B cell ALL and hyperuricemia; tumor lysis syndrome six with stage III or IV B cell NHL who were consecutively admitted to St Jude Children’s Research Hospital from Febru- ary 1994 to December 1996, 126, five and three, respectively, Introduction were treated with non-recombinant urate oxidase (Uricozyme, provided free of charge by Sanofi Winthrop). The enzyme was Hyperuricemia, a well-recognized complication of lymphoid not given to 43 patients because of a history of allergy (atopic malignant diseases and their treatment, has been associated allergy, glucose-6-phosphate dehydrogenase deficiency and with high morbidity rates and delays in the delivery of chemo- pregnancy are listed as contraindications in the Uricozyme therapy, especially in patients with B cell or T cell acute lym- package insert), six because they received the new recombi- phoblastic leukemia (ALL) and advanced-stage B cell non- nant urate oxidase (which recently became available for clini- Hodgkin lymphoma (NHL).1–6 The standard prophylaxis or cal testing in a multicenter clinial trial), and four because they treatment of hyperuricemia in patients with neoplastic dis- did not receive methotrexate or 6-mercaptopurine as prein- orders consists of allopurinol, vigorous hydration, urinary alk- duction therapy. Children with newly diagnosed non-B cell alinization and osmotic diuresis.2–5,7 By inhibiting the enzyme ALL who were consecutively treated with allopurinol between xanthine oxidase, hence uric acid formation, allopurinol ther- December 1991 and February 1994 served as historical con- apy reduces the renal load of uric acid but increases the renal trols. Informed consent was obtained from the parents or legal excretion of the uric acid precursors hypoxanthine and xan- guardians of all patients. thine.8 Unlike hypoxanthine, xanthine is actually less soluble than uric acid in urine.9 In fact, occasional case reports of xanthine nephropathy and calculi suggest that allopurinol Chemotherapy treatment is detrimental to some patients.10–13 Urate oxidase, by converting uric acid to allantoin – a read- Patients with non-B cell ALL receiving non-recombinant urate oxidase and historical controls were the subjects of two con- secutive Total Therapy studies (XIIIA and XIIIB).19–21 Twenty- Correspondence: C-H Pui, St Jude Children’s Research Hospital, 332 nine patients treated with urate oxidase and all 129 historical N Lauderdale, Memphis, TN 38105, USA; Fax: 901 521 9005 controls were enrolled in the first study, XIIIA, which stratified Received 9 July 1997; accepted 6 August 1997 and randomized patients to receive either high-dose metho- Urate oxidase for hyperuricemia C-H Pui et al 1814 trexate (1 g/m2 intravenously over 24 h) or fractionated lower- immunophenotypes was analyzed by a two-tailed Fisher’s dose methotrexate (30 mg/m2 orally every 6 h for six doses) exact test. Seven patients with ALL who had adverse reactions as up-front window therapy (before conventional combination with the first dose of urate oxidase treatment and did not remission induction). Ninety-seven patients treated with urate receive subsequent doses, were excluded from the analysis. oxidase were enrolled in the second study, XIIIB, which strati- All comparisons were controlled at a type I error rate of fied and randomized patients to receive up-front 6-mercapto- ␣ = 0.05. All analyses were conducted using SAS Release purine (1 g/m2 intravenously over 6 h) alone, or high-dose 6.11. methotrexate or fractionated lower-dose methotrexate (at the same dosages as those used in the preceding trial), followed at 24 h by a 6-mercaptopurine infusion. Approximatley 96 h Results and 72 h after the start of methotrexate and 6-mercaptopurine treatment, respectively, chemotherapy with prednisone, The groups of patients given urate oxidase or allopurinol had vincristine and daunorubicin was begun. Hence, during the similar presenting ages (median, 5.6 vs 5.7 years), leukocyte first 96 h of treatment, patients in study XIIIB received more counts (11.7 vs 13.8 × 109/l), and blood levels of lactic chemotherapy than those in study XIIIA. dehydrogenase (1243 vs 957 U/l), uric acid (4.3 vs 4.3 mg/dl), Chemotherapy for B cell ALL or NHL was based on the LMB phosphorous (4.9 vs 4.7 mg/dl), BUN (8.0 vs 8.0 mg/dl), creat- 89 protocol,23 with minor modifications. Hence, all patients inine (0.5 vs 0.5 mg/dl) and calcium (9.4 vs 9.4 mg/dl). received a cytoreductive phase during the first week that Frequencies of the T cell immunophenotype were likewise included one dose of cyclophosphamide (300 mg/m2), one similar (14.3 vs 15.5%). dose of vincristine (2 mg/m2), 7 days of prednisone (60 mg/m2 During treatment, maximal levels of uric acid in patients per day) and intrathecal treatment (one dose with methotrex- receiving urate oxidase were significantly lower than those in ate and hydrocortisone for stage III NHL or three doses with the allopurinol group (median, 2.3 vs 3.9 mg/dl, P Ͻ 0.001, methotrexate, cytarabine and hydrocortisone for stage IV NHL Table 1), despite comparable degrees of leukemic cell lysis or B cell ALL). (data not shown). The decrease in uric acid levels was not only more pronounced but also more rapid in the urate oxidase group than in the allopurinol group (Figure 1). In the Supportive care urate oxidase group, the proportions of patients with uric acid levels below 1 mg/dl after 1 day and 2 days of treatment were

All patients received hydration fluid with NaHCO3 to main- 50.9 and 46.8%, respectively, compared with 1% and 1.8% tain the urine pH у6.5 (to prevent methotrexate-induced in the allopurinol group (P Ͻ 0.001 for both comparisons). toxicities). Non-recombinant urate oxidase, diluted in 50 ml Moreover, the minimal uric acid levels among patients in the normal saline, was administered at 100 units/kg per day intra- urate oxidase group were significantly lower than correspond- venously over 30 min for 5 days, under the auspices of US ing values in the allopurinol group: median (range), 0.6 (0– FDA IND 43183 and 46051. Allopurinol was given at 4.1) vs 2.3 (0.6–6.3) mg/dl, P Ͻ 0.001. Patients given urate 300 mg/m2 per day orally for 5–13 days to historical controls. oxidase also had better renal function, as indicated by lower Oral phosphorous binder (aluminum hydroxide or calcium maximal levels of BUN (median, 11 vs 24 mg/dl, P Ͻ 0.001) carbonate) was given to individual patients as indicated. or creatinine (median, 0.6 vs 0.7 mg/dl, P = 0.01). Although patients with B cell ALL or advanced-stage B cell NHL lacked historical controls, their responses to urate oxi- Determination of uric acid levels dase were nonetheless impressive. Plasma uric acid levels decreased during antineoplastic chemotherapy in each of the Plasma collected in an EDTA tube was used to measure uric five patients with B cell ALL and in the three with stage III acid levels. To block enzymatic degradation of uric acid by NHL (Figure 2), dropping below 1 mg/dl after 4 days of treat- urate oxidase ex vivo, we maintained temperatures at 0° to ment. Renal function worsened transiently in a single case, in 4°C during sampling, transport to the laboratory, and prep- which the creatinine level increased from 1.4 mg/dl to aration for assay. The samples were assayed with the EKTA- 2.1 mg/dl on day 3, returning to normal (0.7 mg/dl) on day 6. CHEM 700 analyzer (Johnson & Johnson Clinical Diagnostics, None of the patients required dialysis for renal failure; how- Rochester, NY, USA), based on the oxidation of uric acid in ever, six (4.5%) of the 134 children with ALL or B cell NHL the presence of exogenous urate oxidase (from Candida utilis) who received urate oxidase developed allergic reactions. All to form allantoin and hydrogen peroxide. Hydrogen peroxide reactions (five grade 3 and one grade 2 according to the Com- then reacts with a leuco dye in the presence of peroxidase to mon Toxicity Criteria of the National Cancer Institute) produce a colored compound that is measured by reflectance occurred during infusion of the first dose of urate oxidase with spectophotometry. The lower limit of sensitivity was a median onset of 6 min (range 1–17 min). They were mani- 0.5 mg/dl. fested as bronchospasm and hypoxemia in five patients, urticaria in four, facial or lip swelling in two and hypotension or emesis in one each. All reactions resolved within 1–12 h Statistical analysis (median 1 h) after treatment with intravenous diphenhydramine and a glucocorticoid, subcutaneous epinephrine and Two-tailed Wilcoxon rank-sum tests were used to compare oxygen. distributions of blood levels of uric acid, blood urea nitrogen One additional patient, a 12-year-old black boy with T cell (BUN), creatinine, calcium, phosphorous, and lactic dehydro- ALL and a presenting leukocyte count of 410 × 109/l had an genase during the first 96 h of treatment, as well as presenting oxygen saturation of 90% without clinical symptoms during leukocyte counts, between the 119 patients with ALL who had routine monitoring, 2 h after administration of the first dose of received a full course of urate oxidase and the 129 controls urate oxidase. Methemoglobinemia (11.4%) was found at that who received allopurinol. The distribution of leukemic cell time and peaked at 15.6% at 14 h. The child recovered com- Urate oxidase for hyperuricemia C-H Pui et al 1815 Table 1 Comparison of maximal blood chemistry values during treatment with urate oxidase or allopurinol

Test Urate oxidase group (n = 119) Allopurinol group (n = 129) P value

Uric acid (mg/dl) 2.3 (0–11.4)a 3.9 (1.3–11.5) Ͻ0.001 Latic dehydrogenase (U/l) 1412 (375–14 619) 1032 (213–28 245) 0.16 Phosphorous (mg/dl) 5.6 (3.5–9.0) 5.5 (2.5–14.2) 0.29 BUN (mg/dl) 11 (3.0–46.0) 24 (7.0–77.0) Ͻ0.001 Creatinine (mg/dl) 0.6 (0.3–2.6) 0.7 (0.2–2.6) 0.01 Calcium (mg/dl) 9.5 (7.5–12.0) 9.4 (7.7–12.0) 0.26 aMedian (range).

pletely without treatment 4 days later and was subsequently diagnosed to have glucose-6-phosphate dehydrogenase deﬁciency. Urate oxidase was not repeated in this child or the other six patients with allergic reactions.

Discussion

In this study, non-recombinant urate oxidase was more effective than allopurinol in the prevention and treatment of hyperuricemia, as in the small series reported by Masera et al.17 Uric acid levels decreased to 4.1 mg/dl or lower (median, 0.6 mg/dl) in all of our patients with ALL or NHL during treatment with urate oxidase, despite the concomitant administration of cytoreductive chemotherapy. The results we report probably underestimate the clinical potential of the urate oxidase, since patients with non-B cell ALL were studied during treatment that is much less intensive than standard remission induction regimens for this disease.24 Importantly, the more rapid uricolytic effects of urate oxidase compared to allopuri- Figure 1 Comparison of plasma uric acid levels at diagnosis and nol (Figure 1) would be expected to shorten or eliminate during the first 2 days of up-front window therapy in patients receiving delays in chemotherapy due to hyperuricemia, a major con- non-recombinant urate oxidase (n = 119) or allopurinol (n = 129). sideration in the treatment of patients with B cell ALL or Median values (mg/dl) were significantly lower in the urate oxidase advanced-stage NHL.5 group on both the first (median, 1.0 vs 3.4 mg/dl, P Ͻ 0.001) and second (1.1 vs 2.8 mg/dl, P Ͻ 0.001) days of study. The bars represent Hyperphosphatemia with consequent hyperphosphaturia is quartile ranges. another important cause of acute renal failure due to tumor lysis.5,25 In theory, the use of urate oxidase should facilitate the excretion of phosphorous, as it would obviate the need to alkalinize the urine, and phosphorus is more soluble in acidic urine. This prediction could not be tested in the present study because of the small number of patients at high risk of massive tumor cell lysis (eg those with B cell ALL or advanced-stage B cell NHL). Nonetheless, we would emphasize that none of our B cell ALL or stage III or IV Burkitt NHL cases treated with urate oxidase had renal failure requiring dialysis. By contrast, acute renal failure prompted hemodialysis in 10 of 40 patients with B cell ALL or advanced-stage NHL treated previously at this center26 and 28 of 133 treated recently in the Pediatric Oncology Group (six of whom died of complications).27 We attributed this difference to our recent use of urate oxidase and the adoption of ‘reduction phase’ therapy as pioneered by French investigators, which is associated with a lesser degree of tumor cell lysis than conventional therapy.23 The frequency of allergic reactions complicating the use of non-recombinant urate oxidase has been uncertain. Masera et al17 did not observe any adverse effects in their study of 30 patients, in contrast to the 4.5% incidence of allergic reactions, primarily bronchospasm and hypoxemia, noted here. Figure 2 Plasma uric acid levels in patients with B cell ALL or NHL, determined before and after treatment with chemotherapy and Whether this observation reflects inherent differences in the non-recombinant urate oxidase. Each data point represents an individ- allergic status of our patients and those studied by Masera et ual patient. al is not clear. Conceivably, the frequency of antecedent Urate oxidase for hyperuricemia C-H Pui et al 1816 exposure to either urate oxidase itself or to contaminants from 12 Ablin A, Stephens BG, Hirata T, Wilson K, Williams HE. Nephro- Aspergillus could differ between these study populations. pathy, xanthinuria, and orotic aciduria complicating Burkitt’s lym- Because of the abrupt onset of the reaction, diphenhydramine, phoma treated with chemotherapy and allopurinol. Metabolism 1972; 21: 771–778. epinephrine and oxygen were prepared and available at the 13 Shohet I, Aladjem M, Lotan D, Katznelson D, Sperling D, Kende bedside for each of our patients before administration of each G. Acute renal failure complicating Burkitt’s lymphoma treated dose of urate oxidase. Thus, all reactions were managed with chemotherapy and allopurinol. Intl J Pediatr Nephrol 1980; promptly leading to their rapid resolution. It remains to be 1: 240–241. determined if the more purified recombinant product now 14 Brogard JM, Coumarso D, Franckhauser J, Stahl A, Stahl J. Enzy- under investigation will be less allergenic than Uricozyme. matic uricolysis: a study of the effect of a fungal urate-oxydase. Rev Eur Etudes Clin Biol 1972; 17: 890–895. We recommend the use of urate oxidase in patients with 15 Gardais J. L’urate-oxydase dans le traitement des hemopathies malignant diseases who have frank hyperuricemia or are at malignes. Quest Med 1976; 29: 1199–1203. high risk of developing this complication. 16 Robert A, Corberand J, Regnier CL. Utilisation de l’urate-oxydase dans la prevention et le traitement des hyperuricemies dan 20 cas de leucose aigue de l’enfant. Rev Med Toulouse 1976; 12 Acknowledgements (Suppl. 6): 1093–1100. 17 Masera G, Jankovic M, Zurlo MG, Locasciulli A, Rossi MR, Uderzo C, Recchia M. Urate-oxidase prophylaxis of uric acid- This work was supported in part by grants from the National induced renal damage in childhood leukemia. J Pediatr 1982; 100: Cancer Institute (PO1 CA 20180 and P30 CA-21765) and the 152–155. American Lebanese Syrian Associated Charities. We thank 18 Jankovic M, Zurlo MG, Rossi E, Edefonti A, Cossu MM, Giani M, M Heim, K Juneau, and V Jones for data management and J Masera G. Urate-oxidase as hypouricemic agent in a case of acute Gilbert for editorial review. tumor lysis syndrome. Am J Pediatr Hematol Oncol 1985; 7: 202–204. 19 Synold TW, Relling MV, Boyett JM, Rivera GK, Sandlund JT, Mah- References moud H, Crist WM, Pui C-H, Evans WE. Blast cell methotrexate- polyglutamate accumulation in vivo differs by lineage, ploidy, and 1 Rieselbach RE, Bentzel CJ, Cotlove E, Frei E III, Freireich EJ. Uric methotrexate dose in acute lymphoblastic leukemia. J Clin Invest acid excretion and renal function in the acute hyperuricemia of 1994; 94: 1996–2001. leukemia: pathogenesis and therapy or uric acid nephropathy. Am 20 Masson E, Relling MV, Synold TW, Liu Q, Schuetz JD, Sandlund J Med 1964; 37: 872–884. JT, Pui C-H, Evans WE. Accumulation of methotrexate polyglutam- 2 Holland P, Holland NH. Prevention and management of acute ates in lymphoblasts is a determinant of antileukemic effects in hyperuricemia in childhood leukemia. J Pediatr 1968; 72: 358– vivo: a rationale for high-dose methotrexate. J Clin Invest 1996; 366. 97: 73–80. 3 Krakoff IH, Murphy ML. Hyperuricemia in neoplastic disease in 21 Pui C-H, Rivera GK, Hancock ML, Sandlund JT, Ribeiro RC, Boyett children: prevention with allopurinol, a xanthine oxidase inhibi- JM, Relling MV, Evans WE, Crist WM. Risk-adapted findings for tor. Pediatrics 1968; 41: 52–56. ALL – from St Jude Children’s Research Hospital. In: Buchner T, 4 Jones DP, Mahmoud H, Chesney RW. Tumor lysis syndrome: Hiddemann W, Wo¨rmann B, Schellong G, Ritter J, Creutzig J (eds). pathogenesis and management. Pediatr Nephrol 1995; 9: 206– Acute Leukemias VI. Prognostic Factors and Treatment Strategies. 212. Haematology and Blood Transfusions. Springer-Verlag: Berlin, 5 Cohen LF, Balow JE, Magrath IT, Poplack DG, Ziegler JL. Acute Heidelberg, 1997, pp 629–637. tumor lysis syndrome: a review of 37 patients with Burkitt’s lym- 22 Masson E, Synold TW, Relling MV, Schuetz JD, Sandlund JT, Pui phoma. Am J Med 1980; 68: 486–491. C-H, Evans WE. Allopurinol inhibits de novo purine synthesis in 6 Jones DP, Stapleton FB, Kalwinsky D, McKay CP, Kellie SJ, Pui C- lymphoblasts of children with acute lymphoblastic leukemia. H. Renal dysfunction and hyperuricemia at presentation and Leukemia 1996; 10: 56–60. relapse of acute lymphoblastic leukemia. Med Pediatr Oncol 23 Patte C, Leverger G, Rubie H, Bertrand Y, Coze C, Mećhinaud F, 1990; 18: 283–286. Lutz P, Michon J, Baruchel A, Courbon B for the SFOP. High cure 7 DeConti RC, Calabresi P. Use of allopurinol for prevention and rate in B-cell (Burkitt’s) leukemia in the LMB 89 protocol of the control of hyperuricemia in patients with neoplastic disease. New SFOP (French Pediatric Oncology Society). Proc ASCO 1993; 12: Engl J Med 1966; 274: 481–486. 317 (Abstr). 8 Andreoli SP, Clark JH, McGuire WA, Bergstein JM. Purine 24 Pui C-H. Childhood leukemias. New Engl J Med 1995; 332: excretion during tumor lysis in children with actue lymphocytic 1618–1630. leukemia receiving allopurinol: relationship to acute renal failure. 25 Tsokos GC, Balow JE, Spiegel RJ, Magrath IT. Renal and metabolic J Pediatr 1986; 109: 292–298. complications of undifferentiated and lymphoblastic lymphomas. 9 Klinenbeg JR, Goldfincer SE, Seegmiller JE. The effectiveness of Medicine 1981; 60: 218–229. the xanthine oxidase inhibitor allopurinol in the treatment of gout. 26 Stapleton FB, Strother DR, Roy S III, Wyatt RJ, McKay CP, Murphy Ann Intern Med 1965; 62: 639–647. SB. Acute renal failure at onset of therapy for advanced stage Burk- 10 Greene ML, Fujimoto WY, Seegmiller JE. Urinary xanthine itt lymphoma and B cell acute lymphoblastic lymphoma. Pedi- stones – a rare complication of allopurinol therapy. New Engl J atrics 1988; 82: 863–869. Med 1969; 280: 426–427. 27 Bowman WP, Shuster JJ, Cook B, Griffin T, Behm F, Pullen J, Link 11 Band PR, Silverberg DS, Henderson JF, Ulan RA, Wensel RH, M, Head D, Carroll A, Berard C, Murphy S. Improved survival for Banerjee TK, Little AS. Xanthine nephropathy in a patient with children with B-cell acute lymphoblastic leukemia and stage IV lymphosarcoma treated with allopurinol. New Engl J Med 1970; small noncleaved-cell lymphoma: A Pediatric Oncology Group 283: 354–357. Study. J Clin Oncol 1996; 14: 1252–1261.