¹Clinica di Oncoematologia Pediatrica, Dipartimento di Pediatria, Università di Padova, Padova, Italy

²Cattedra di Ematologia, Università 'La Sapienza', Roma, Italy

³IV Divisione di Pediatria Ematologia-Oncologia, 'Istituto G. Gaslini', Genova, Italy

⁴Divisione di Ematologia Pediatrica, Ospedale 'Bambin Gesù', Roma, Italy

⁵Clinica Pediatrica II, Università di Bari, Bari, Italy

⁶Dipartimento di Scienze Pediatriche, Ospedale Sant'Orsola-Malpighi, Università di Bologna, Italy

⁷Oncoematologia Pediatrica, Policlinico San Matteo, Università di Pavia, Italy

Correspondence to: Dr S Cesaro, Pediatric Oncology-Hematology Clinic, University of Padova, Via Giustiniani 3, 35128-Padova, Italy

^aPresent address: Divisione di Pediatria, Azienda Ospedale Mellino Mellini, Chiari (Brescia), Italy

This retrospective study from the Italian Association of Pediatric Hematology Oncology-Bone Marrow Transplant Group (AIEOP-TMO) reports the results of consolidation with high-dose melphalan and autologous hematopoietic stem cell transplantation (auto-HSCT) in patients with acute myeloid leukemia (AML) in first complete remission (CR1). From October 1994 to July 1999, 20 patients (median age 9.9 years, range 0.11-16.2) were treated in six centers. Eighteen had de novo AML and two had secondary AML. According to BFM criteria, 10 were classified as standard- and 10 as high-risk patients, respectively. The median time from diagnosis to CR1 and from diagnosis to Auto-HSCT were 1.1 months (range 0.8-1.6) and 4.3 months (range 3.1-6.2), respectively. Purging with either mafosfamide (three) or in vivo interleukin-2 (four) was performed in seven of 20 patients. Melphalan was administered at a dosage of 150-220 mg/m² (median 180). Median total number of nucleated cells infused was 2.5 ´ 10⁸/kg (range 1.1-8.9). The myeloablative regimen was well tolerated with no toxic death, veno-occlusive disease or life-threatening complications. All patients had hematopoietic recovery in a median time of 27 days for neutrophils and 44 days for platelets. Eight of 20 patients relapsed after a median time of 7.2 months from transplant (range 5.7-15.9). Six of them died (five of progression of disease and one of sepsis) while the remaining two patients are alive in CR2. The 3-year cumulative probability of survival and event-free-survival (EFS) is 62% and 56%, respectively. This study showed that in pediatric patients with AML consolidation of CR1 with high-dose melphalan allows survival and EFS to be obtained comparable to other auto-HSCT or chemotherapy published series with a potential sparing effect both on duration of treatment (with respect to chemotherapy) and on long-term side-effects (with respect to auto-HSCT with TBI or busulfan containing regimens). Bone Marrow Transplantation (2001) 28, 131-136.

acute myeloid leukemia; pediatric autologous bone marrow transplantation; pediatric autologous hematopoietic stem cell transplantation; melphalan

Prognosis of acute myeloid leukemia (AML) in childhood has considerably improved over the last decades. The strategy accounting for this success has been the adoption of intensive induction chemotherapy followed by intensive post-remission treatment with either high-dose ara-C or myeloablative regimens followed by stem cell transplantation.¹ Allogeneic bone marrow transplantation is the best option for AML patients in first remission provided that an HLA compatible family donor is available.^2,3 Unfortunately, only 25-30% of potential candidates have an HLA-identical family donor. In the remaining patients, autologous hematopoietic stem cell transplantation (auto-HSCT) can be an alternative approach and its results in some series compare favorably with those of allogeneic BMT.² Although based on limited data, results so far reported on auto-HSCT in childhood AML in first complete remission (CR1) showed a leukemia-free survival ranging from 41% to 87%.^4,5,6,7

The best conditioning regimen is still a matter of discussion because of the many different myeloablative schemes used, some of which included total body irradiation (TBI).² Actually, the role of TBI in favoring leukemia eradication in AML patients has been not established so far by means of randomized trials. Moreover, the recent higher cure rates give much emphasis to prevention of long-term effects possibly associated with TBI or cranial irradiation.^8,9,10

We report a retrospective analysis of the Italian Association of Pediatric Hematology Oncology-Bone Marrow Transplant Group (AIEOP-TMO) regarding AML patients in CR1 who underwent auto-HSCT and received high-dose melphalan (L-PAM) as conditioning regimen.

Materials and methods

The patients analyzed were retrieved from the AIEOP registry that collects data from all pediatric bone marrow transplant centers in Italy. After a formal adhesion to the study, a specially designed case report form was sent to the principal investigator of each center to obtain more details regarding both the disease and the transplant procedure.

Eligibility criteria

The study included pediatric patients with a diagnosis of AML who underwent auto-HSCT in CR1 after two or more courses of induction treatment and were conditioned only with high doses of L-PAM. Patients with promyelocytic leukaemia were excluded from the analysis. From October 1994 to July 1999, 20 patients from six transplant centers were eligible. There were 10 males and 10 females with a median age at diagnosis of 9.9 years (range 0.11-16.2). The main demographic and clinical characteristics are listed in Table 1.

Diagnosis

AML was classified according to the French-American-British (FAB) co-operative group.¹¹ Trisomy 21 and secondary AML were reported in one and two cases, respectively. Secondary AML was diagnosed after an interval of 2 and 4 years in patients treated for Hodgkin's lymphoma and yolk salk tumor, respectively. The patients were divided into two subgroups (standard- and high-risk group) according to BFM definitions: children with FAB morphology M1, M2 and Auer rods, M4eo, t(8;21), inversion of chromosome 16 and blast cell reduction on day +15 to 5% were included in the standard-risk group, whereas all other patients were considered to be at high risk.¹² Ten patients belonged to the standard-risk and nine patients belonged to the high-risk group, whereas data were incomplete for one patient. Banded cytogenetic study, FISH or RT-PCR analysis was performed at diagnosis on the marrow aspirate in 17 out of 20 patients. The results are summarized in Table 1.

Definitions

A patient was considered to be in CR when clinical signs of leukemia had disappeared and bone marrow aspirate showed a normocellular marrow, fewer than 5% blasts, and no Auer rods. More than 5% of blasts in bone marrow aspirate and/or the presence of circulating blasts and/or extramedullary disease were defined as relapse.

The first of 3 consecutive days with a neutrophil count 0.5 ´ 10⁹/l or with an unsupported platelet count 50 ´ 10⁹/l were defined as time of myeloid and platelet engraftment, respectively. The criteria reported by Bearman et al¹³ were used to score regimen-related toxicity.

The long-term effects of treatment were investigated with specific questions in the data collection form regarding the occurrence of growth impairment, dysfunction of endocrine glands, cataract formation, central nervous system disorders, and secondary malignancy in patients who survived in remission more than 12 months from transplantation.

Induction treatment protocol and intensified consolidation in the post-remission phase

The AIEOP first-line protocol for AML (LAM 92) is shown in Figure 1. All but one of the patients received two or more courses of idarubicin, cytarabine and etoposide (ICE). One patient received a less intensive induction based on the association of cytarabine and daunomycin.¹⁴ Patients with an HLA-identical sibling donor were given an allogeneic BMT, whereas the remaining patients underwent auto-HSCT with or without purging. For patients who underwent auto-HSCT, the reasons for omitting TBI from conditioning were as follow: age less than 18 months at transplant (three patients), previous neurological toxicity (three patients), secondary AML (two patients) and policy of the center (12 patients).

Response to chemotherapy, harvest processing and infusion of hemopoietic stem cells

The median interval between diagnosis and CR1 was 1.1 months (range 0.8-1.7). Twelve patients underwent a second course, seven patients a third course and one patient a fourth course of chemotherapy before harvest.

The source of stem cells was bone marrow in 19 patients and peripheral blood in one patient. Three patients required a second harvest to achieve an adequate total number of total nucleated cells. The source of additional stem cells was bone marrow in one case and peripheral blood in two cases, respectively.

The median interval between diagnosis and auto-HSCT was 4.3 months (range 3.1-6.2). Purging was performed in seven out of 20 patients: three cases were purged with mafosfamide at a final concentration of 100 g/ml; four patients underwent an in vivo stimulation of NK activity with a 5-day course of interleukin-2 (IL-2) before harvest, followed by 12 weekly post-engraftment IL-2 courses.¹⁵ The marrow or peripheral stem cells were cryopreserved in 10% dimethylsulfoxide using a programmed rate freezer and stored in the vapor phase of liquid nitrogen. On day 0, the hemopoietic cells were thawed in a water bath at 37°C and infused.

Conditioning regimen and post-transplant supportive therapy

The patients were conditioned with an L-PAM dose ranging from 150 to 220 mg/m² (median 180 mg/m²), administered by infusion over 15 min after dilution in normal saline. The total number of nucleated cells infused were 2.5 ´ 10⁸/kg (range 1.1-8.9).

All patients had an indwelling central venous catheter and were nursed in single room or laminar air-flow beds or HEPA filtered rooms. Eight of the 20 patients received antifungal prophylaxis with fluconazole at a dose of 6 mg/kg/day until engraftment. During the aplastic phase, irradiated and filtered platelet and erythrocyte transfusions were administered in order to maintain platelets >10 ´ 10⁹/l and hemoglobin >80 g/l, respectively. Empirical broad-spectrum antibiotic therapy was started in the case of fever. Antifungal treatment was added in the case of suspected or proven fungal infection or in the case of fever persisting more than 3 consecutive days despite broad-spectrum antibiotic treatment.

Filgrastim at a dose of 5 g/kg/day was used in six of 20 patient for a median time of 21 days (range 2-27) according to the practice of centers. Trimethoprim-sulphamethoxazole prophylaxis for Pneumocystis carinii pneumonia was given once neutrophil engraftment had occurred.

Statistical analysis

Data were collected on Microsoft Access database and analyzed as of 31 March 2000. For overall and event-free survival (OS and EFS) analysis treatment failures included death from any cause and both relapse and death from any cause, respectively. The OS and EFS curves after transplantation (starting point) were calculated by the Kaplan-Meier method. Results are expressed as probability (%) with 95% confidence limits. The impact of age at diagnosis, sex, FAB morphology, leukocyte count at diagnosis, age at transplant, time interval between diagnosis and CR1 and between diagnosis and auto-HSCT, number of courses of chemotherapy before auto-HSCT (< or > of 2), purging, total number of nucleated cells infused, use of growth factors, on clinical outcome was tested in univariate analysis. The SAS package (SAS Institute, Cary, NY, USA) was used for statistical analysis of the data. The Mann-Whitney test was used for continuous variables and either ² test or Fisher's exact test for the categorical variables. A P value of <0.05 was considered significant.

Results

All patients engrafted with a median time of 27 days for neutrophils (range 18-84) and 44 days for platelet recovery (range 16-257), respectively. During the pre-engraftment period, the median transfusion requirement was 4 units of leukocyte-depleted red cells (range 1-18) and 6 units of single donor apheresis platelets (range 2-19). Two patients were never transfused. Purging delayed significantly engraftment for platelets (P = 0.002) but not for neutrophils (P = 0.2). Major complications were infection (fever of unknown origin in seven cases, four cases of microbiologically determined infections due to Candida guillermondii, Staphylococcus epidermidis, Pseudomonas aeruginosa and cytomegalovirus, respectively) occurring during the first 100 days post transplant with a median duration of 9 days (range 4-23). All proven or suspected infections disappeared with a median of 9 days of treatment (range 4-23). Transplant-related toxicity was mild or moderate and limited to oral mucosa (grade 1: one patient; grade 2: six patients), gastrointestinal system (grade 1: seven patients), kidney (grade 1: one patient; grade 3: one patient) and liver (grade 1: one patient; grade 2: one patient). One patient experienced severe kidney damage leading to renal insufficiency. No case of veno-occlusive disease or toxic death related to transplant was reported.

Outcome

Overall, 14 of the 20 patients are alive with a median follow-up time of 2.1 years (range 0.6-5.4): 12 are in CR1 and two in CR2 4.5 and 15.5 months after relapse, respectively.

Eight patients have relapsed with a median time to relapse of 7.2 months (range 5.7-15.9) months after transplant. Six of eight children died, five of progression of disease, and one of early sepsis while the remaining two patients are alive in CR2 after chemotherapy and a second auto-HSCT, respectively.

The 3-year Kaplan-Meier estimate of OS and EFS was 62% (confidence interval: 37-87) and 56% (confidence interval: 33-79), respectively (Figure 2). Among the factors tested in univariate analysis, only a white blood cell count at diagnosis >10 ´ 10⁹/l resulted significantly associated with relapse (P < 0.012). Among 10 patients evaluable for long-term effects, one case of subclinical hypothyroidism, in the patient affected by Down's syndrome, and one case of mild hypogonadism, in the patient previously treated with alkylating agents and thoraco-abdominal irradiation for metastatic Hodgkin's lymphoma at diagnosis, was reported.

Discussion

The optimal post-remission treatment for AML is still a matter of controversy. Allogeneic BMT from an HLA matched sibling is currently considered the most effective anti-leukemic treatment available, but only 25-30% of patients have a suitable donor.¹⁶ Moreover, recently, several authors have shown, by means of studies based on 'biological randomization', that more intensive chemotherapy regimens can result in an overall survival comparable to that of sibling BMT.^17,18,19,20 The use of an unrelated volunteer as bone marrow donor is not indicated for patients in CR1 after initial induction treatment due to the still high transplant-related mortality and high incidence of acute and chronic graft-versus-host disease. However, this option may be acceptable for children with particularly adverse prognostic features such as complex cytogenetic abnormalities, late CR1 or megakaryocytic AML.²¹ Although fewer data are available for children than for adults, the scenario is the same for pediatric patients. Moreover, some questions have not been addressed by controlled trials such as the need for using TBI in conditioning regimens for auto-HSCT, the efficacy of in vitro purging for auto-HSCT and the optimal administration of high-dose cytarabine (dose, number of courses) before auto-HSCT.

Our results compare favorably with other previously published pediatric auto-HSCT studies for AML in CR1.^4,6,7 Tiedemann et al⁶ reported a 5-year EFS of 87%, using a conditioning regimen based on chemotherapy without TBI (and, in this case, without purging). A possible explanation of the good results of that study is the optimal quality of CR1 obtained with an intensive induction treatment including high-dose cytarabine. The total dose of cytarabine, ranging from 12 to 30 g, was modulated on the results of bone marrow aspirate performed at day 14. Indeed, there is evidence of a superiority of high-dose vs intermediate-dose cytarabine in patients with AML²² and the introduction of at least one course of high-dose cytarabine is the common feature of all studies showing results of chemotherapy treatment that approach or equal those obtained with auto-HSCT or allogeneic bone marrow transplantation.^1,2,12 Also, Bonetti et al⁵ recently reported a 5-year disease-free survival of 68% for patients given auto-HSCT after a conditioning regimen comprising TBI and high-dose L-PAM. In this latter study, major differences that could in part explain the better results obtained were not in the induction treatment (that was given according to AIEOP protocols for most patients (39 out of 53) but in the auto-HSCT procedure that included, besides TBI, in vivo or in vitro bone marrow purging in all patients but one. By contrast, the results of our study combined a mix of purged and unpurged autografts. Only three patients received a bone marrow purged in vitro with mafosfamide, a drug able to eliminate residual leukemic cells.²³ Potentially, the extension of the in vitro purging to all patients could improve the EFS but, to date, such data have not been tested in a randomized study.² Four patients received a 5-day course of IL-2 before marrow collection followed by 12 IL-2 courses after engraftment in order to stimulate immune functions and NK cell activity.¹⁵ This strategy has been suggested by results of small pilot studies on relapsed patients or patients in CR2^24,25 but its efficacy deserves further confirmation.

The conditioning regimen with only high-dose L-PAM was well tolerated with no toxic death, veno-occlusive disease or life-threatening complications. Extrahematological regimen-related toxicity was mild or moderate and limited mainly to oral and gastrointestinal symptoms. These findings confirm the good tolerability of high-dose L-PAM in pediatric auto-HSCT as previously reported by Tiedemann et al⁶ and compares favorably with the treatment-related mortality of different regimens. In fact, a randomized study of autologous transplantation vs intensive consolidation chemotherapy for AML in childhood reported a treatment-related mortality as high as 15% in the ABMT arm with a conditioning regimen based on busulfan and cyclophosphamide²⁶ while adding TBI to high-dose L-PAM is associated with a TRM incidence as high as 4%, with hemorrhage and infection being the causes of death.⁵ Another aspect deserving particular emphasis, concomitant with the improved cure rate, is the prevention of long-term effects. Hypothyroidism, growth impairment and cataract formation have been associated with TBI containing regimens whilst sterility has been described after both TBI⁸ and busulfan containing regimens.² Also, the strategy to cure pediatric AML with chemotherapy alone without auto-HSCT intensification such as BFM trials,¹² although apparently less toxic, has a potential risk of the development of neuro-cognitive abnormalities and endocrine dysfunction because of the administration of cranial irradiation.²⁷ In this study, the short follow-up did not allow any definitive conclusions about long-term side-effects of high-dose L-PAM but the only side-effects observed so far are mild dysfunction of thyroid and gonads that were considered more likely associated with constitutional underlying disease (Down's syndrome) in one patient and to previous first-line treatment with alkylating agents and thoraco-abdominal irradiation in the other.

In conclusion, this retrospective study showed that in pediatric patients with AML, consolidation of CR1 with high-dose L-PAM achieved survival and EFS rates comparable to other published series of auto-HSCT and chemotherapy with a potential sparing effect on the duration of treatment (with respect to chemotherapy) and on the long-term side-effects (with respect to auto-HSCT with TBI or busulfan containing regimens). Whether the use of TBI in addition to L-PAM increases the chance of cure of children with AML in CR1 should be studied in future randomized trials.

Acknowledgements

We would like to thank Raffaele Micciolo, MD, for help in collecting data, Chiara Tommasi, PhD, for assistance with the statistical analysis and Judith Kingston, MD, for reviewing the English style of the manuscript.

1 Lowenberg B, Downing JR, Burnett A. Acute myeloid leukemia. New Engl J Med 1999; 341: 1051-1062, MEDLINE

2 Gorin NC. Autologous stem cell transplantation in acute myelocytic leukemia. Blood 1998; 92: 1073-1090, MEDLINE

3 Zander AR, Berger C, Kroger N et al. High dose chemotherapy with busulfan, cyclophosphamide, and etoposide as conditioning regimen for allogeneic bone marrow transplantation for patients with acute myeloid leukemia in first complete remission. Clin Cancer Res 1997; 3: 2671-2675, MEDLINE

4 Klingebiel T, Pession A, Paolucci P et al. Autologous vs allogeneic BMT in AML: the European experience. Report of the EBMT-Pediatric Diseases Working Party. Bone Marrow Transplant 1996; 18: (Suppl. 2) 49, MEDLINE

5 Bonetti F, Zecca M, Pession A et al. Total-body irradiation and melphalan is a safe and effective conditioning regimen for autologous bone marrow transplantation in children with acute myeloid leukemia in first remission. J Clin Oncol 1999; 17: 3729-3735, MEDLINE

6 Tiedemann K, Waters KD, Tauro GP et al. Results of intensive therapy in childhood acute myeloid leukemia, incorporating high-dose melphalan and autologous bone marrow transplantation in first complete remission. Blood 1993; 82: 3730-3738, MEDLINE

7 Woods WG, Kobrinsky N, Buckley J et al. Intensively timed induction therapy followed by autologous or allogeneic transplantion for children with acute myeloid leukemia or myelodysplastic syndrome: a Childrens Cancer Group pilot study. J Clin Oncol 1993; 11: 1448-1457, MEDLINE

8 Liesner RJ, Leiper AD, Hann IM, et al. Late effects of intensive treatment for acute myeloid leukemia and myelodysplasia in childhood. J Clin Oncol 1994; 12: 916-924, MEDLINE

9 Leahey AM, Teunissen H, Friedman DL et al. Late effects of chemotherapy compared to bone marrow transplantation in the treatment of pediatric acute myeloid leukemia and myelodysplasia. Med Pediatr Oncol 1999; 32: 163-169, MEDLINE

10 Cohen A, Rovelli A, Bakker B et al. Final height of patients who underwent bone marrow transplantation for hematologiacal disorders during childhood: a study by the Working Party for Late Effects-EBMT. Blood 1999; 93: 4109-4115, MEDLINE

11 Bennett JM, Catovsky D, Daniel MT et al. Proposed revised criteria for the classification of acute myeloid leukemia. A report of the French-American-British cooperative group. Ann Intern Med 1985; 103: 620-625, MEDLINE

12 Creutzig U, Zimmermann M, Ritter J et al. Definition of a standard-risk group in children with AML. Br J Haematol 1999; 104: 630-639, MEDLINE

13 Bearman SI, Appelbaum FR, Buckner CD et al. Regimen related toxicity in patients undergoing bone marrow transplantation. J Clin Oncol 1998; 6: 1562-1568,

14 Amadori S, Testi AM, Aricò M et al. Prospective comparative study of bone marrow transplantation and postremission chemotherapy for childhood acute myelogenous leukemia. J Clin Oncol 1993; 11: 1046-1054, MEDLINE

15 Messina C, Zambello R, Rossetti F et al. Interleukin-2 before and/or after autologous bone marrow transplantation for pediatric acute leukemia patients. Bone Marrow Transplant 1996; 17: 729-735, MEDLINE

16 Margolis DA, Casper JT. Allogeneic transplantation for acute myeloid leukemia in children. In: Thomas DE, Blume KG, Forman SJ (eds). Hematopoietic Stem Cell Transplantation. Blackwell Science: Malden, MA, 1999, pp 835-848.

17 Cassileth PA, Harrington DP, Appelbaum FR et al. Chemotherapy compared with autologous or allogeneic bone marrow transplantation in the management of acute myeloid leukemia in first remission. New Engl J Med 1998; 339: 1649-1656, MEDLINE

18 Keating S, de Witte T, Suciu S et al. The influence of HLA-matched sibling donor avaibility on treatment outcome for patients with AML: an analysis of AML 8A study of EORTC Leukemia Cooperative Group and GIMEMA. Br J Haematol 1998; 102: 1344-1353, MEDLINE

19 Stevens RF, Hann IM, Wheatley K, Gray RG. Marked improvements in outcome with chemotherapy alone in paediatric acute myeloid leukemia: results of the United Kingdom Medical Research Council's 10th AML trial. Br J Haematol 1998; 101: 30-40,

20 Burnett AK, Goldstone AH, Stevens RMF et al. Randomized comparison of addition of autologous bone marrow transplantation to intensive chemotherapy for acute myeloid leukemia in first remission: results of MRC AML 10 Trial. Lancet 1998; 351: 700-708, MEDLINE

21 Sierra J, Storer B, Hansen JA et al. Unrelated donor marrow transplantation for acute myeloid leukemia: an update of the Seattle experience. Bone Marrow Transplant 2000; 26: 397-404, MEDLINE

22 Mayern RJ, Davies RB, Schiffer CA et al. Intensive postremission chemotherapy in adults with myeloid leukemia. New Engl J Med 1994; 331: 896-903, MEDLINE

23 Sharkis SJ, Santos GW, Colvin OM. Elimination of acute myelogenous leukemia cells from marrow and tumor suspensions in the rat with 4-hydroperoxycyclofosfamide. Blood 1980; 55: 521-523, MEDLINE

24 Meloni G, Foa R, Vignetti M et al. Interleukin-2 may produce prolonged remissions in advanced acute myelogenous leukemia. Blood 1994; 84: 2158-2163, MEDLINE

25 Meloni G, Vignetti M, Andrizzi C et al. Interleukin-2 for the treatment of advanced acute myelogenous leukemia patients with limited disease: update experience with 20 cases. Leuk Lymphoma 1996; 21: 429-435, MEDLINE

26 Ravindranath Y, Yeager AM, Chang MN et al. Autologous bone marrow transplantation versus intensive consolidation chemotherapy for acute myeloid leukemia in childhood. New Engl J Med 1996; 334: 1428-1434, MEDLINE

27 Hertzberger H, Huk WJ, Ueberall M et al. CNS late effects after ALL therapy in chilhood. Part I: Neuroradiological findings in long-term survivors of chilhood ALL-An evaluation of the interferences between morphology and neuropsychological performance. Med Pediatr Oncol 1997; 28: 387-400, MEDLINE

Figure 1 Outline of AIEOP protocol (LAM 92 P) for childhood acute myeloid leukemia.

Figure 2 Overall survival (OS) and event-free survival (EFS) after high-dose melphalan and stem cell transplantation in AML patients in first complete remission. Confidence interval in parentheses. n: number of patients; E: events.

Table 1 Main demographic and clinical characteristics of the eligible patients at diagnosis