TO THE EDITOR Acute myelogenous leukemia (AML) remains a challenging disease. Although responsible for only 15–20% of childhood leukemias, it amounts to up to a 30% of all leukemia-related deaths.1 Over the last three decades, there has been significant progress in the treatment of this disease, mostly due to the introduction of multiple drug regimens, treatment intensification, and improvements in postremission therapy. Early mortality has been addressed by a number of authors, since it is responsible for a significant proportion of all AML deaths.1, 2, 3, 4, 5, 6, 7 In general, it is said that approximately half of early deaths are due to treatment toxicity.1 In developing countries, such as Chile, treatment toxicity may be even higher, mostly from infectious complications. Children under 15 years comprise approximately 40% of the population of Chile, or 5 600 000 million people. Most of these patients (over 75%) receive their care through the National Health Services. Since 1987, children with the most common pediatric malignancies receive uniform standardized treatment in all of Chile. To that effect, the National Program for Antineoplastic Drugs for Children (PINDA) was founded, and is responsible for creating and supervising treatment programs for pediatric cancers and leukemia. Currently, there are 11 accredited treatment centers throughout the country, mostly in the capital, Santiago. PINDA has developed three treatment protocols for AML, on 1987, 1992, and 1998, based on Berlin–Francfort–Muenster (BFM) protocols. The objective of this publication is to analyze the results of the first two protocols, PINDA 87 and 92, both based on BFM-83, with emphasis on treatment toxicity and early mortality. All patients under 15 years of age with newly diagnosed AML in Chile, between March 1st, 1987, and July 31st, 1999, are included, with the exception of patients with Down's syndrome, secondary myeloblastic leukemia, and myelosarcoma. These patients were treated in 11 Chilean hospitals, according to the protocols described below. The diagnosis was made by morphology, and cytochemichal stains at each treating center, without a central review of pathology. Immunophenotype was studied at Hospital Salvador one center in Santiago. Cytogenetic studies were performed for a group of patients. All cases were classified according to the French–American–British classification (FAB). All patients were treated alike, with no differences for risk groups. The two treatment protocols were quite similar, both based on AML-BFM 83.10 The first protocol, PINDA 87 was used from March 1st, 1987 to November 30th, 1991, and the second, PINDA 92, from January 1st, 1992 to January 30th, 1998. The treatment protocols consisted of an induction phase (ADE chemotherapy), followed by an 8-week consolidation phase, followed by maintenance chemotherapy. The nutritional status of patients at presentation was assessed, and, in general, it was similar to the general population, with a low frequency of malnourished children. Dietary guidelines were followed during treatments. The great majority of patients presented with severe infections at diagnosis 0 during the first 3 weeks, which were resolved at each treatment center. The PINDA guidelines for febrile neutropenia during treatment were followed. Some supportive care measures employed by the different centers for each protocol were not homogeneous. For example, there were some problems in the early years for platelet transfusions, and leukapheresis was not available at all hospitals. The main end points of the analysis are disease- and event-free survival, and early mortality. Follow-up was carried out at each treating center, and is updated as of June, 2003. The definitions of complete response (CR), early death (ED), no response (NR), and CNS involvement, are the BFM definitions.8 Event-free survival was calculated from date of diagnosis to the last follow-up or first event (failure to achieve remission, resistant leukemia, relapse, second malignancy, or death from any cause). Disease-free survival was defined for patients achieving a complete remission, and it is calculated from the date of remission to the last follow-up, or date of relapse or second malignancy. The statistical analysis was carried out with the NCSS 2004 software (NCSS, Kaysville, UT, USA). Actuarial survival was calculated with the Kaplan–Meier method, and the log-rank test was used to assess differences in survival. Differences between populations were calculated with the 2 test. No statistical analysis of risk factors, or of differences in outcome for treatment centers was performed. Demographic and tumor characteristics of patients treated under PINDA 87 (n=106), and PINDA 92 (n=151) are detailed in Table 1. Median follow-up was 13 years for AML PINDA 87, and 8 years for AML PINDA 92. Comparative results for both protocols are displayed on Table 2. There were no significant differences in survival between protocols. Event-free survival for both protocols is shown in . The incidence of M3 leukemia is relatively high in Chile, accounting for 19.8% of these patients. Since this specific subtype has a better prognosis, thanks to the use of all-trans retinoic acid (ATRA),11 we analyzed the overall 5-year survival of M3 leukemia compared to non-M3 cases. Although the 5-year survival of M3 cases was higher than that observed for other types, this difference was not significant (43 and 33% for M3 and non-M3 leukemias, respectively (Figure 2). The data on how many patients received ATRA as part of their treatment is not available to perform a separate analysis. We studied two groups separately: patients who died before receiving treatment, and patients dying within 42 days of the beginning of treatment. In all, 14 patients in both protocols died before treatment. The characteristics of these patients are analyzed in Table 3. The FAB distribution of this group is similar to the global population. The proportion of patients dying before treatment with an initial WBC over 100 000, was significantly higher than for patients with lower counts (P=0.00003) lower counts. Six patients presented with significant intracranial or thoracic bleeding). In summary, the early death before treatment group includes a high proportion of children with WBCs over 100 000, and associated bleeding phenomena. A total of 48 patients died within the first 42 days of treatment. A high proportion of these are children under 2 years (36%), and with initial WBCs over 100 000 (26%). In all, 17 patients (40%) had hemorrhage: eight intracranial hemorrhages, three disseminated intravascular coagulation syndromes, three GI bleedings, three pulmonary hemorrhages, and one disseminated intravascular coagulation syndrome. All patients had infection: there were five bronchopneumonia, two typhlitis, one necrotizing enterocolitis, the others were diagnosed as febrile neutropenia, with or without septic shock. There was also heart failure in one case, and hydrothorax in another. An infectious microorganism was isolated in only 10 cases: Klebsiella enterobacter in four, Candida albicans in three, Staphylococcus in two (one was coagulase negative) Pseudomonas aeruginosa in one, and Enterococcus in one case. This study shows the reality in a developing country, where we have been able to establish a uniform treatment regimen for all the patients covered by the public health system, over a considerable period. Chile has been the only Latin American country where this has been accomplished. The long follow-up period with a minimum of 66 months is a warranty of the results. Our patient demographics are similar to other international publications on the subject, where the incidence of M3 leukemia is similar to that reported in Spain or Italy, and different from countries like the US or Germany. Overall survival with these protocols is inferior to that reported with the original BFM protocol,8 but the results are similar to other international results with contemporary treatment regimens. The difference with AML-BFM 83 may be due to our incidence of early death, since the rates of recurrences are comparable. Bone marrow transplant is a very useful treatment tool, which was used only in two patients, because it was unavailable in our country at the time. Early mortality is very high in this report. As a comparison, the early mortality reported on the BFM-83 protocol was 11%.8 In our experience, this is due to infection and bleeding as main causes. This high frequency was observed especially in the early years of the protocol, mostly in provincial hospitals, was associated with deficits in blood banking, and lack of platelets for transfusion – mostly on weekends or nights. This situation has improved in recent years. The same has happened for the treatment of infectious complications. The availability of better antibiotics and antifungals has helped reduce the mortality due to infection. Another important factor is the increasing implementation of isolation and Intensive Care Units for oncologic patients. A recent study from Malaysia, in which a treatment protocol based on BFM-83 was evaluated also shows a relatively high early mortality rate, which was improved with better supportive care, thus underscoring the importance of such measures in a health system with limited resources.8 These improvements are reflected in the AML PINDA 98 protocol, based on BFM 93, which will be the subject of another analysis. In protocol 98, for the first 79 patients, early mortality dropped to 15.2%, compared to 28 and 21% for protocols 87 and 92, respectively. The DFS is similar to BFM protocol, in other words, the incidence of relapse is similar to BFM studies, which emphasizes the value of improving early mortality from infection and hemorrhage. Since 1998, a single public hospital in Santiago has developed a bone marrow transplant program, which is intended to cover the needs for the whole pediatric population in the country, and will be an additional improvement in the care of our patients.

BACKGROUND:The National Chilean Pediatric Oncology Group, PINDA, reports the first prospective, nonrandomized trial for acute lymphoblastic leukemia (ALL), using a modified version of the Berlin-Frankfurt-Munster protocol (ALL BFM 86). The aim of this study was to classify immunophenotypes, to decrease cranial irradiation, and to assess whether this protocol would improve the survival rate.PROCEDURE:From June, 1987, to June, 1992, 444 unselected children were diagnosed with ALL. Of them, 425 were evaluable. Therapy was stratified by risk. Standard-risk (SR) and high-risk (HR) patients received protocols I, M, II, and maintenance therapy. Very-high-risk (VHR) patients received protocol E instead of protocol M. All patients received a prephase treatment consisting of prednisone and intrathecal methotrexate (MTX). HR and VHR patients received cranial irradiation (12-18 Gy). The following changes were made to the ALL BFM 86 protocol: in protocol M, MTX 1 g/m2 instead of 5 g/m2; in protocol E, citarabine 1 g/m2 instead of 2 g/m2; mithoxantrone and ifosfamide were substituted by teniposide and cyclophosphamide.RESULTS:Immunophenotypes: pro-B-ALL, 14%; common ALL, 67.4%; pre-B-ALL, 4.3%; T-ALL, 10%; undifferentiated leukemia (AUL), 4.3%. The overall 5-year event-free survival (EFS) rate was 60% +/- 2% (SE). The 5-year EFS rate for each risk group was: SR 75%, HR 62%, VHR 28%, with a median follow-up of 6.5 years (range 4.5-9.5 years). The cumulative incidence of central nervous system (CNS) relapse was 5.4%.CONCLUSIONS:We have been able successfully to perform a nationwide study. Our strategy to adapt the BFM protocol to our population of patients trial was effective in improving the EFS. The immunophenotype distribution is similar to that in other reported series.

Thirty-seven patients with Ewing sarcoma were treated in the First National Chilean Trial for Ewing's Sarcoma (1986-1991), which comprised the St. Jude Ewing's 78 Study. All patients received cyclophosphamide, doxorubicin, vincristine, and Dactinomycin for a total treatment period of about 10 months, and all prescribed therapy was administered. Local therapy consisted of irradiation (RT) to the primary tumor, complete surgical resection, or a combination of both surgery and RT. Twenty-nine of these patients had localized tumors, 24% had pelvic primary tumors, 21 were males, and 20 were greater than 10 years of age at diagnosis. Twenty-one patients had tumors that were greater than 8 cm in largest diameter. Fourteen of the 29 patients with localized disease remain disease free at 23 to 91 months from diagnosis. Fourteen patients have died of-tumor-related complications and 1 of a secondary malignancy. Relapse was local only in 4, metastatic in 9, and local plus metastatic in 1. Only 1 of the 8 patients with metastatic disease at presentation remains disease free. Toxicity consisted primarily of myelosuppression and mucositis. We conclude that this form of relative intense multimodal therapy for children/adolescents with localized Ewing sarcoma is curative in about half of affected children as in the original St. Jude study, and that it can be safely given in a developing country, provided that careful attention to supportive care and treatment planning is given. Although these results represent improvement in outcome for our patients, more effective therapy is needed for children with Ewing sarcoma, especially those with metastatic disease at presentation.

We have analyzed the sequence of 40 VDJ rearrangements of the immunoglobulin heavy chain gene locus on 32 unselected children from Chile with precursor B cell ALL at diagnosis. Rearrangements were derived by PCR with VH gene family-specific primers and sequenced directly. The number of VDJ rearrangements, and the pattern of VH, DH and JH gene usage was identical to the one reported by groups from developed countries. CDR3 regions represented an unbiased repertoire; VH to JH joinings were in frame in 36% of cases. Absent N nucleotides in the DJ border, suggestive of fetal origin of ALL, were seen in 9/40 rearrangements but they did not correlate with younger age. More than one rearrangement was sequenced in six patients, representing independent events with no signs of clonal evolution. One patient was analyzed at first bone marrow relapse showing persistence of one rearrangement and evolution of a second one which conserved the DJ border. The subset of B cell precursors which suffer malignant transformation to ALL appear to be common in different parts of the world.

Acute lymphoblastic leukemia (ALL) is the most frequent childhood cancer. The leukemic cells of ALL patients show several well defined numeric and structural chromosomal abnormalities which are universally known for its prognostic implications. We studied a group of 44 children with ALL, to investigate the incidence of chromosome aberrations in ALL, its lymphocyte lineage and some clinical feature associations, and the finding of non previously described aberrations. A high proportion of patients (79.5%) showed chromosomal abnormalities. Most of them had a pseudodiploid karyotype (46 chromosomes), characterized mainly by a translocation. In relation to chromosome number, 27% of them were hyperdiploid with more than 50; 9% hyperdiploid between 47-50 and 7% hypodiploid (less than 46). Among structural aberrations found, were the following recurrent translocations: t (1; 19), t (4; 11), t (9; 22) in 6.8%, 9.1% and 2.3% of cases respectively, all related to an early B immunophenotype. Other translocations found, compromised regions 7q22, 9p21 - 24. Two new translocations in ALL were found: t (1; 5)(q23; q33), apparently balanced, and t (13; 21)(q14; q22), unbalanced. Other recurrent structural changes found were: deletion (6q), (7q), (9p), (11q), (12p), inversion (3q), isochromosome (7q), maker chromosomes and double minutes. The distribution of chromosome abnormalities in this group of patients was in agreement with previous reports from other investigators.

• Prognosis for ALL remains poor. While about 80-90% of adult patients will have complete remissions at some point during treatment, about half will relapse, so overall cure rate is around 40%.1 • The main treatment for ALL currently used is long-term chemotherapy and various combination chemotherapy regimens.1 • Because leukemia cells spread widely throughout the bone marrow and to many other organs, it is not possible to cure this type of cancer by surgery.1 • For patients who do not respond to chemotherapy, stem cell transplant is often the best chance for a cure in the advanced stage of the disease.6 – Additional therapeutic options are needed to help patients achieve hematologic remission so they may be eligible for transplant.6 FACTS AND FIGURES

Journal of Pediatric Hematology/Oncology (JPHO) reports on major advances in the diagnosis and treatment of cancer and blood diseases in children. The journal publishes original research, commentaries, historical insights, and clinical and laboratory observations.