Unexpected Mortality from the Use of E. Coli L-Asparaginase During

Home , Asparaginase

Unexpected mortality from the use of E. coli L-asparaginase during remission induction therapy for childhood acute lymphoblastic leukemia: a report from the Taiwan Pediatric Oncology Group D-C Liang1, I-J Hung2, C-P Yang2, K-H Lin3, J-S Chen4, T-C Hsiao5, T-T Chang6, C-H Pui7, C-H Lee3, K-S Lin3 and the Taiwan Pediatric Oncology Group

Departments of Pediatrics, 1Mackay Memorial Hospital; 2Chang Gung Children’s Hospital at Linkou; 3National Taiwan University Hospital; 4National Cheng Kung University Hospital; 5Chang Gung Children’s Hospital at Kaohsiung; 6Kaohsiung Medical College Hospital, Taiwan; 7St Jude Children’s Research Hospital, and the University of Tennessee, College of Medicine, Memphis, Tennessee, USA

The relative efficacy and toxicity of E. coli L-asparaginase and Hacco Kogyo, Japan) was used as the third agent and adminis- epidoxorubicin used in remission induction therapy for child- tered at 5000 IU/m2/day on days 3–10 for eight doses; 115 of hood acute lymphoblastic leukemia (ALL) were assessed in a 118 children (97%) with standard-risk ALL achieved a com- randomized trial conducted in Taiwan. All patients had stan- 3 dard-risk ALL, defined as a leukocyte count Ͻ10 × 109/l and plete remission. In the Taiwan Pediatric Oncology Group were aged between 1 and 2 or 7 and 10 years, or a leukocyte (TPOG) 882A protocol (1988–1993), L-asparaginase (Leunase) count Ͻ50 × 109/l and were aged between 2 and 7 years, without was again used as the third agent at a dosage of 5000 IU/m2 evidence of a T cell or mature B cell immunophenotype, central thrice weekly for nine doses; 208 of 216 children (96.3%) nervous system leukemia or expression of two or more with standard-risk ALL attained a complete remission.4 myeloid-associated antigens. Ninety-three patients were ran- Because the relative efficacy as well as the acute and long- domized to receive E. coli L-asparaginase at 10 000 IU/m2 thrice weekly for nine doses and 108 to receive epidoxorubicin at term sequelae of an anthracycline and L-asparaginase used for 20 mg/m2 weekly for two doses during remission induction with remission induction were uncertain, we initiated a ran- daily prednisolone, weekly vincristine and, on day 22, a dose of domized trial – TPOG-ALL-93-SR (standard-risk) protocol etoposide plus cytarabine. Patients treated with L-asparaginase (1993–1997) – to address this issue. One group of patients had a significantly higher rate of fatal infection with or without were randomized to receive epidoxorubicin (epirubicin), and hemorrhage than did those who received epidoxorubicin dur- 5 ing remission induction (six of 93 vs none of 108, P = 0.009), the other group L-asparaginase. Although the dosage of E. coli resulting in a lower rate of complete remission in the former L-asparaginase varied widely between protocols (ie 5000 to group (93.6 vs 99.1%, P = 0.05). In addition, patients treated with 25 000 IU/m2),4,6–8 we chose the dosage of 10 000 IU/m2 L-asparaginase had a higher frequency of hyperglycemia and thrice weekly for nine doses to avoid undertreatment. The hypoalbuminemia. The overall rate of event-free survival was unexpectedly high mortality in our patients treated with lower in patients treated with L-asparaginase than in other L-asparaginase prompted us to report our early results. patients (P = 0.06); estimated 3-year rates were 72% (95% confidence interval, 55–89%) and 87.2% (78–96%), respectively. We conclude that L-asparaginase (Leunase) given at 10 000 IU/m2 for nine doses was poorly tolerated and resulted in excessive toxicity, both through its effects as a single agent and possibly Patients and methods through potentiation of etoposide. Keywords: acute lymphoblastic leukemia; L-asparaginase; Patients remission induction From September 1993 to July 1997, 201 children with newly diagnosed standard-risk ALL in Taiwan were enrolled in the nationwide TPOG-ALL-93-SR protocol. These cases rep- Introduction resented 46% of patients in all risk groups treated during the same time period. The protocol was approved by the Human Contemporary treatment programs featuring three or more Investigations Committee of each participating hospital and drugs during remission induction can induce complete informed consent was obtained from the parents of each remissions in more than 95% of children with acute lym- patient. The 110 boys and 91 girls ranged in age from 1 to phoblastic leukemia (ALL). For patients with low- or standard- 15 years (median, 3 years and 11 months). risk ALL, remission induction regimens often contain three drugs and can induce complete remissions in over 98% of patients. Two agents – a glucocorticoid and vincristine – are used in every induction regimen, and most protocols rely on Diagnosis an anthracycline or L-asparaginase as the third agent.1 In our Taipei Children’s Leukemia (TCL) 821 study (1982–1984), in The diagnosis of ALL was based on the morphologic, cyto- which prednisolone, vincristine and doxorubicin were used chemical and immunophenotypic characterization of leu- for remission induction, all 54 children with standard-risk ALL kemic blast cells in bone marrow and/or peripheral blood. 2 attained complete remission. In our subsequent TCL 842 Bone marrow and peripheral blood smears were stained with study (1984–1988), E. coli L-asparaginase (Leunase; Kyowa a modified version of the Romanowsky method9 and other standard techniques including periodic acid-Schiff, myelop- eroxidase, ASD-chloroacetate esterase and ␣-naphthyl butyr- Correspondence: D-C Liang, Division of Pediatric Hematology- Oncology, Department of Pediatrics, Mackay Memorial Hospital, 92, ate esterase. Immunophenotypic studies were performed by 10 Sec. 2, Chung-San North Road, Taipei 10449, Taiwan previously described methods. Standard-risk ALL was Received 3 August 1998; accepted 7 October 1998 defined by the following criteria: (1) leukocyte count E. coli L-asparaginase for childhood ALL D-C Liang et al 156 Table 1 Treatment schedules used in the TPOG-ALL-93-SRL and TPOG-ALL-93-SRE studies

Treatment Dosage Day of treatment

Remission induction (5 week) Vincristine 1.5 mg/m2 i.v. 1,8,15,22 Prednisolone 60 mg/m2 p.o. 1–29 L-asparaginase (SRL group) 10 000 IU/m2 i.m. 2,4,6,8,10,12,15,17,19 or Epidoxorubicin (SRE group) 20 mg/m2 i.v. 1,8 Etoposide 300 mg/m2 i.v. 22 Cytarabine 300 mg/m2 i.v. 22 TIT (triple intrathecal therapy) Age-dependenta 1,22 Consolidation treatment (11 week) Methotrexate 1000 mg/m2 i.v. 1,8,22,36,50,64 Mercaptopurine 50–75 mg/m2 p.o. 1–78 TIT therapy Age-dependenta 1,8,15,22 Reinduction treatment (2 week) Vincristine 1.5 mg/m2 i.v. 1,8 Dexamethasone 10 mg/m2 p.o. 1–15 L-asparaginase 10 000 IU/l i.m. 1,3,5,8,10,12 Epidoxorubicin 20 mg/m2 1,8 TIT therapy Age-dependenta 1 Continuation treatment (110 week) Mercaptopurine 50–75 mg/m2 p.o. daily Methotrexate 20–30 mg/m2 p.o. weekly 8-week cycle Vincristine 1.5 mg/m2 i.v. 1,8 Dexamethasone 10 mg/m2 p.o. 1–15 TIT therapy Age-dependenta 1

aAge-adjusted doses of TIT treatment Drug/age (year) 1–2 2–3 Ͼ3 Body surface area у1m2

Methotrexate (mg/m2) 8 10 12 15 Hydrocortisone (mg/m2) 8 10 12 15 Cytarabine (mg/m2) 16202430

Ͻ10 × 109/l for patients aged 1 to 2 years or 7 to 10 years; vincristine, L-asparaginase and epidoxorubicin. Continuation Ͻ50 × 109/l for those aged 2 to 7 years; and (2) non-T, non- treatment consisted of weekly methotrexate and daily mercap- B immunophenotype, no central nervous system (CNS) leuke- topurine with pulses of dexamethasone and vincristine every mia, as defined by the presence of any leukemic lymphoblasts 8 weeks. CNS-directed therapy was based on triple intrathecal identified on cytocentrifuged smears of cerebrospinal fluid,11 treatment with methotrexate, hydrocortisone and cytarabine; or the presence of cranial nerve palsies and the lack of cranial irradiation was not given to any patient. There were expression of two or more myeloid-associated antigens no significant differences in age, gender or presenting leuko- (CD11, CD13, CD14, CD15 or CD33). cyte count between the two groups (Table 2).

Randomization and treatment

The patients were randomized within each participating insti- Table 2 Presenting features of the study groups tution into two treatment groups (SRL and SRE), which differed = = only in remission induction therapy. During remission induc- Category SRL (n 93) SRE (n 108) tion, all patients received prednisolone and vincristine; Age (year) patients in the SRL group were given additional E. coli L-aspar- 2 Median 3.8 3.9 aginase (Leunase) at 10 000 IU/m intramuscularly thrice Range 1.5–8.6 2–9.9 weekly for nine doses, while those in the SRE group were Sex treated with epidoxorubicin 20 mg/m2 intravenously on days Male 49 61 1 and 8 for two doses (Table 1). Upon attaining complete Female 44 47 remission, all patients received consolidation treatment con- Leukocyte count × 109/l sisting of six doses of high-dose methotrexate (1 g/m2 intra- Median 8.1 6.4 Range 0.4–43.4 0.7–45 venously over 24 h) and daily mercaptopurine over 11 weeks, followed by reinduction treatment with dexamethasone, E. coli L-asparaginase for childhood ALL D-C Liang et al 157 Table 3 Comparisons of complications during remission induc- Results tion therapy During remission induction, 19 of 93 patients in the SRL No. of P value groups, compared with eight of 108 in the SRE group, episodes developed a severe infection (P = 0.01, Table 3). Five patients in the SRL group died as a result of this complication; one SRL SRE additional patient in this group died of severe infection and upper gastrointestinal tract bleeding during induction therapy. Severe infection 19 8 0.01 Sepsis 15 7 Details of cases with fatal complications are summarized in Pneumonia 2 1 Table 4. Hence, altogether, six of 93 patients in the SRL group, Typhlitis 1 0 as compared with none of the 108 in the SRE group, died Perianal abscess 1 0 during remission induction (P = 0.009). As expected, patients Disseminated fungal infection 0 0 in the SRL group were also more likely than those in the SRE Hyperglycemia 6 1 0.05 group to have hyperglycemia and hypoalbuminemia (Table Hypoalbuminemia 6 1 0.05 Cerebral thrombosis 2 0 NS 3). There were no significant differences in other compli- Pancreatitis 0 0 NS cations. Consciousness change 1 1 NS Only one patient in the SRE group failed to attain complete Seizure 0 1 NS remission, while in the SRL group the complete remission rate Febrile neutropenia 14 24 NS was significantly lower (93.6 vs 99.1%, P = 0.05) due to the Hemorrhage 2 2 NS excess number of deaths during this phase of therapy (Table Leukemia lysis syndrome 2 0 NS No complications 40 54 NS 5). The overall EFS probability was likewise worse in the SRL group (Figure 1; P = 0.06). Estimated rates at 3 years post-diag- NS, not significant. nosis were 72% (55–89% CI), and 87% (78–96%), respectively. The two groups did not differ significantly with respect to DFS (Figure 2; P = 0.43). By the Cox regression analysis, treatment group was the only variable with a significant Statistical analysis impact on prognosis. Patients treated by the SRL protocol had a 1.56-fold (0.52–4.6 CI) higher risk of failure as compared to Comparisons of the frequencies of mortality, toxicity and com- those in the SRE group. plications during remission induction, as well as the complete remission rates between the two study groups, were assessed with Fisher’s exact test. Sex distributions were compared with the ␹2 test, and distributions of age and presenting leukocyte Discussion counts with the two-sample t-test. Event-free survival and disease-free survival were estimated by the Kaplan–Meier We attribute the increased frequencies of severe infection and method and compared with the two-sided log-rank test. mortality in this study to treatment with L-asparaginase. By Ninety-five percent confidence intervals (CIs) were calculated depleting asparagine, this agent inhibits protein synthesis, by Greenwood’s formula. Event-free survival (EFS) was ana- resulting in a variety of metabolic dearrangements and organ lyzed from the start of remission induction, and disease-free dysfunction, including immunosuppression.12 Since the rela- survival (DFS) from the time of attaining complete remission tive frequencies of febrile neutropenia in this study were not to the time of failure. Failures included death or relapse of significantly different (Table 3), we suggest that the primary any kind. In both analyses, cases were censored at the last adverse affect of L-asparaginase was inhibition of normal follow-up date, if no failure was observed. Follow-up data immune function, including synthesis of immunoglobulin. were accrued through August 1997. The impact of the treat- Additionally, L-asparaginase treatment increases unbound ment groups, presenting leukocyte count and age on clinical (active) portion of etoposide and its catechol metabolites due outcome was analyzed by both univariate and multivariate to hypoalbuminemia13–15 and decreased systemic clearance.14 Cox regression models. The increased concentration of unbound etoposide and its metabolites could be expected to enhance myelosuppression.

Table 4 Characteristics of SRL patients with fatal complications during induction therapy

No. Age at Sex Presenting At diagnosis of complications diagnosis leukocyte × 9 (years) count 10 /l Days from initial Neutrophil Type of complications treatment count/mm3

1 3 Male 16 28 40 Sepsis 2 4 Female 15.6 31 0 Typhlitis 3 4 Male 8.9 28 12 Pneumonia 4 2 Female 36 23 0 Perianal abscess; necrotizing enterocolitis; sepsis 5 6 Male 5.6 20 7700 Septic shock; upper Gl bleeding 6 3 Female 42 3 112 Sepsis E. coli L-asparaginase for childhood ALL D-C Liang et al 158 Table 5 Treatment results by study groups ducts are based on their half-lives. PEG-L-asparaginase, which has the longest half-life, is usually administered at 2500 IU/m2 SRL SRE P value every other week for two doses in cases of newly diagnosed (n = 93) (n = 108) ALL. By contrast, Erwinia L-asparaginase, with the shortest half-life, is generally given at 10 000 IU/m2 three times per Death during induction 6 0 0.009 week for six to 12 doses. In the USA, the dosages of E. coli Complete remission rate 93.6% 99.1% 0.05 Relapse 5 4 L-asparaginase (Elspar, Merck Sharp & Dohme, USA) range Bone marrow 4 2 from 6000 to 10 000 IU/m2 given two to three times per week CNS 1 1 for six to 12 doses. In this study, the dosage of L-asparaginase Bone marrow + testis 0 1 (Leunase) was higher than those used in our two previous Death in remission 2 5 studies, because we did not wish to underdose our patients, % 3-year EFS with 95% CI 72 (55–89) 87 (78–96) 0.06 % 3-year DFS with 95% CI 80 (65–99) 89 (81–99) 0.43 thus compromising the comparison with epidoxorubicin, and because the dosage was the one recommended by Nesbit and associates,16 on the basis of results with an E. coli preparation (Crasnitin) produced by Bayer (UK, Germany, Italy, Indeed, the onset of fatal complications in four patients Denmark, Norway). occurred 1 to 9 days after the etoposide therapy on day 22. Why then was the Leunase preparation so poorly tolerated In conventional dosages, L-asparaginase treatment is gener- in this study? Until recently, L-asparaginase compounds ally well tolerated and not associated with severe infection. derived from different strains of E. coli were assumed to have Three forms of the enzyme are available commercially – one comparable potency. However, recent studies demonstrated derived from Erwinia chrysanthemi, the other prepared from that the enzyme activities and half-lives of these products dif- E. coli and the third representing a polyethylene glycol (PEG) fer substantially.17–20 For example, the Crasnitin preparation form of the E. coli product. The dosages of these three pro- was found to have a lower enzyme activity and a shorter half-

Figure 1 Event-free survival (EFS) of 201 children with standard-risk ALL. The 3-year EFS in the SRE group was 87 vs 72% in the SRL group.

Figure 2 Disease-free survival (DFS) of 201 children with standard-risk ALL. The 3-year DFS in the SRE group was 89 vs 80% in the SRL group. E. coli L-asparaginase for childhood ALL D-C Liang et al 159 life than other E. coli derivatives. In a German study, the pro- ised Trials in Childhood Acute Lymphoblastic Leukemia. Clinical duct sold as Medac (Medac, Hamburg, Germany, produced Trial Service Unit: Oxford, UK, 1996, p 53. by Kyowa Hacco Kogyo, Japan, and having identical proper- 6 Rivera GK, Raimondi SC, Hancock ML, Behm FG, Pui C-H, Abro- mowitch M, Mirro J Jr, Ochs JS, Look AT, Williams DL, Murphy ties to the Leunase product) caused excessive toxicity when SB, Dahl GV, Kalwinsky DK, Evans WE, Kun LE, Simone JV, Crist 2 given at a dose of 10 000 IU/m at 3 day intervals for eight WM. Improved outcome in childhood acute lymphoblastic leuke- doses,21 prompting a 50% reduction in the dosage used in the mia with reinforced early treatment and rotational combination German frontline ALL and non-Hodgkin lymphoma trials.18 chemotherapy. Lancet 1991; 337: 61–66. Thus, in view of the toxicity results encountered in the present 7 Reiter A, Schrappe M, Ludwig W-D, Hiddemann W, Sauter S, Henze G, Zimmermann M, Lampert F, Havers W, Niethammer D, study, we elected to reduce the L-asparaginase (Leunase) dos- Odenwald E, Ritter J, Mann G, Welte K, Gadner H, Riehm H. age for the subsequent patients treated in the TPOG-ALL-93- Chemotherapy in 998 unselected childhood acute lymphoblastic SR by 50%. Since this amendment, L-asparaginase has been leukemia patients. Results and conclusions of the multicenter trial administered in reduced dosage to 11 patients, one of whom ALL-BFM 86. Blood 1994; 84: 3122–3133. developed sepsis, but there was no mortality. 8 Clavell LA, Gelber RD, Cohen HJ, Hitchcock-Bryan S, Cassady R, This study illustrates the profound adverse impact that can Tarbell NJ, Blattner SR, Tantravahi R, Leavitt P, Sallan SE. Four- result from a seemingly modest dosage adjustment of a single agent induction and intensive asparaginase therapy for treatment of childhood acute lymphoblastic leukemia. New Engl J Med agent in a multidrug regimen of cancer chemotherapy. Recent 1986; 315: 657–663. studies indicate that substitution of the Erwinia product for E. 9 Liang D-C, Ma S-W. A simple and rapid in situ staining method coli L-asparaginase, without a dosage adjustment for its shorter for granulocyte–macrophage colonies in agar culture. Int J Cell half-life, can also adversely affect treatment outcome.22 Cloning 1988; 6: 25–29. Together, these findings emphasize that in the design of mod- 10 Chuu W-M, Tien H-F, Lin D-T, Lin K-H, Su I-J, Chen B-W, Lin K- ern clinical trials of childhood ALL, dosage, administration S, Liang D-C. Infant leukemia: an analysis of nine Chinese patients. Am J Hematol 1990; 34: 246–251. modalities and schedules of any L-asparaginase products 11 Mahmoud HH, Rivera GK, Hancock ML, Krance RA, Kun LE, should be very carefully evaluated since products derived Behm FG, Ribeiro RC, Sandlund JT, Crist WM, Pui C-H. Low leu- from different strains of the same bacteria may display very kocyte counts with blast cells in cerebrospinal fluid of children different biological and clinical effects. with newly diagnosed acute lymphoblastic leukemia. New Engl J Med 1993; 329: 314–319. 12 Calabresi P, Chabner BA. Antineoplastic agents. In: Gilman GA, Rall TW, Nies AS, Taylor P (eds). Goodman & Gilman’s The Phar- macological Basis of Therapeutics, edn 8. Pergamon: New York, 1991, pp 1209–1263. Acknowledgements 13 Stewart CF, Fleming RA, Arbuck SG, Evans WE. Prospective evalu- ation of a model for predicting etoposide plasma protein binding in cancer patients. Cancer Res 1990; 50: 6854–6856. We thank Wan-Hui Chang (Institute of Epidemiology, College 14 Evans WE, Rodman JH, Relling MV, Petros WP, Stewart CF, Pui of Public Health, National Taiwan University, Taipei, Taiwan) C-H, Rivera GK. Differences in teniposide disposition and pharm- for the assistance in statistical analysis. This work was sup- acodynamics in patients with newly diagnosed and relapsed acute ported by grant No. TPOG 86–01 from the Childhood Cancer lymphocytic leukemia. J. Pharmacol Exp Ther 1992; 260: 71–77. 15 Relling MV, Nemec J, Schuetz EG, Schuetz JD, Gonzalez FJ, Kor- Foundation of ROC, Taipei, Taiwan, by grants (CA 20180 and zekwa KR. O-demethylation of epipodophyllotoxins is catalyzed CA 21765) from the US National Cancer Institute, and by by human cytochrome P450 3A4. Mol Pharmacol 1994; 45: American Lebanese Syrian Associated Charities (ALSAC). 352–358. 16 Nesbit ME, Ertel I, Hammond GD. L-asparaginase as a single agent in acute lymphoblastic leukemia: survey of Children Cancer Study Group. Cancer Treat Rep 1981; 65 (Suppl. 4): 101–107. 17 Boos J, Werber G, Ahlke E, Schulze-Westhoff P, Nowak-Gottl U, Wurthwein G, Verspohl EJ, Ritter J, Jurgens H. Monitoring of aspar- References aginase activity and asparagine levels in children on different asparaginase preparations. Eur J Cancer 1996; 32: 1544–1550. 1 Pui C-H. Childhood leukemias. New Engl J Med 1995; 332: 18 Ahlke E, Nowak-Gottl U, Schulze-Westhoff P, Werber G, Borste 1618–1630. H, Wurthwein G, Jurgens H, Boos J. Dose reduction of asparagin- 2 Lin S-T, Liang D-C, Hung I-J, Yang C-Y, Yang Y-M, Chen H-N, ase: under pharmacokinetic and pharmacodynamic control during Hsieh Y-L, Lin C-Y, Lin M-T, Twu B-H, Law K-L, Tsai Y-M, Lin K- induction therapy in children with acute lymphoblastic leukemia. H, Peng C-T, Chen C-H, Chen P-M. Treatment of childhood acute Br J Haematol 1997; 96: 675–681. lymphoblastic leukemia with protocol TCL 821: a report of Taiwan 19 Boos J. Pharmacokinetic and drug monitoring of L-asparaginase Children’s Cancer Study Group. J Formos Med Assoc 1989; 88: treatment. Int J Clin Pharmacol Therapeut 1997; 35: 946–965. 48–56. 20 Nowak-Gottl U, Ahlke E, Klosel K, Jurgens H, Boos J. Changes 3 Yang C-P, Lin S-T, Liang D-C, Hung I-J, Yang Y-M, Chen H-N, in coagulation and fibrinolysis in childhood acute lymphoblastic Hsieh Y-L, Law K-L, Lin M-T, Twu B-H, Lin K-H, Tsai Y-M, Peng leukemia re-induction therapy using three different asparaginase C-T, Chen P-M. Treatment of childhood acute lymphoblastic leu- preparations. Eur J Paediatr 1997; 35: 96–98. kemia with protocol TCL-842 in Taiwan: The Taiwan Children’s 21 Sutor AH, Niemeyer C, Sauter S, Witt I, Kaufmehl K, Romback A, Cancer Study Group. J Formos Med Assoc 1993; 92: 431–439. Brandis M, Riehm H. Gerinnungs veranderungen bei Behandlung 4 Lin K-H, Chen J-S, Hung I-J, Yang C-P, Jaing T-H, Lin D-T, Chen mit den Protokollen ALL-BFM-90 und NHL-BFM-90. Klin Padiatrie R-L, Jou S-T, Liang D-C, Chen S-H, Chang T-T, Chiou S-S, Hsieh 1992; 204: 264–273. Y-L, Chang T-K, Lin M-T, Hsiao T-C, Peng C-T, Wang C-C, Cheng 22 Otten J, Suciu S, Lutz P, Benoit Y, Robert A, Thyss A, Plouvier E, S-N, Ong C-W, Chen B-W, Lin S-T, Chen H-N, Chen W-D, Lin Ferster A, Mechiwaud F, Mazingue F, Brock P, Vilmer E, Solbu K-S. Seven year experience of childhood acute lymphoblastic leu- G, Waterkeyn C, Philippe N. The importance of L-asparaginase kemia – a report from Taiwan Pediatric Oncology Group (TPOG). (A’ASE) in the treatment of acute lymphoblastic leukemia in chil- Med Pediatr Oncol 1995; 25: 294 (Abstr. P-63). dren: results of the EORTC 58881 randomized phase trial showing 5 Liang D-C. TPOG-ALL-93-SR (standard risk). In: Greaves L, Elphin- greater efficiency of Escherichia coli as compared to Erwinia stone T, Sinclair D (eds). International Register of Current Random- A’ASE. Blood 1996; 88 (Suppl. 1): 669 (Abstr. 2663). E. coli L-asparaginase for childhood ALL D-C Liang et al 160

Appendix: Participating principal TPOG investigators

Institution Investigators

Chang Gung Children’s Hospital at Linkou Iou-Jih Hung, Chao-Ping Yang, Tang-Her Jaing National Taiwan University Hospital, Taipei Kai-Hsin Lin, Dong-Tsamn Lin, Rong-Long Chen, Shian-Tarng Jou Mackay Memorial Hospital, Taipei Der-Cherng Liang, Shu-Huey Chen, Hsi-Che Liu National Cheng Kung University Hospital, Taiwan Jiann-Shiuh Chen Chang Gung Children’s Hospital at Kaohsiung Tyh-Chen Hsiao Kaohsiung Medical College Hospital, Kaohsiung Tai-Tsung Chang, Shyh-Shin Chiou Changhua Christian Hospital, Changhua Mahng-Tung Lee, Ming-Tsan Lin China Medical College Hospital, Taichung Ching-Tien Peng Tri-Service General Hospital, Taipei Shin-Nan Cheng Veterans General Hospital Taichung San-Ging Shu, Chain-Lan Chein Veterans General Hospital Taipei Yuh-Lin Hsieh Taipei Medical College Hospital, Taipei Shou-Tien Lin Tzu-Chi General Hospital, Hualien Kong-Ling Kam Kuang Tien General Hospital, Taichung Chen-Woei Wang Chung Shan Medical College Hospital, Taichung Wan-Der Chen Sun Yat-Sen Cancer Center, Taipei Bow-Wen Chen Mennonite Christian Hospital, Hualien Hsin-Nan Chen