Indian Journal of Pathology and Microbiology

: 2010  |  Volume : 53  |  Issue : 4  |  Page : 704--710

Retrospective review of pediatric patients with acute lymphoblastic leukemia: A single center experience

Safoorah Khalid1, Bushra Moiz2, Salman Naseem Adil2, Mohammad Khurshid2,  
1 Royal Darwin Hospital, Casuarina, Australia
2 Aga Khan University Hospital, Karachi, Pakistan

Correspondence Address:
Safoorah Khalid
Hematology Registrar, Royal Darwin Hospital, PO Box 41326, Casuarina, NT 0812


Objective: We reviewed the clinical details and treatment outcome of children with newly diagnosed acute lymphoblastic leukemia (ALL) to determine the significance of already established prognostic factors in our patients. Setting: A tertiary care hospital in Karachi, Pakistan. Study Design: This is a retrospective study. Materials and Methods: Children diagnosed with ALL were evaluated over a period of 17 years (January 1, 1989 to December 31, 2006). Data was collected by reviewing the medical records of the patients and the prognostic factors analyzed by us include age, gender, white blood cell count, central nervous system and mediastinal involvement at presentation, morphology and immunophenotype of the blast cells, and response to induction therapy. Results: There were 46 patients diagnosed during the study period and on regular follow-up. Forty five (97.8%) of these were in complete remission after 28 days of induction therapy. Thirty patients (65.2%) were alive and doing well at the time of study. Of these 30 patients, 26 (86.6%) remained relapse free while only four (13.3%) had relapsed. The remaining 16 patients (34.7%) did not survive including 11 (68.7%) who had a relapse. Only significant variables in terms of prognosis were age and ALL phenotype with a P value 0.04 and 0.03 respectively. Conclusion: We found that ALL is a frequent childhood hematological malignancy in our setting and is more prevalent in males and children less than ten years of age. Age and leukemia phenotype emerged as the important prognostic factors in pediatric ALL in our patients.

How to cite this article:
Khalid S, Moiz B, Adil SN, Khurshid M. Retrospective review of pediatric patients with acute lymphoblastic leukemia: A single center experience.Indian J Pathol Microbiol 2010;53:704-710

How to cite this URL:
Khalid S, Moiz B, Adil SN, Khurshid M. Retrospective review of pediatric patients with acute lymphoblastic leukemia: A single center experience. Indian J Pathol Microbiol [serial online] 2010 [cited 2020 Jul 11 ];53:704-710
Available from:

Full Text


Acute lymphoblastic leukemia (ALL) is a heterogeneous disease that results from the clonal proliferation of lymphoid progenitor cells with arrested maturation. This leads to the accumulation of lymphoblasts in the bone marrow which subsequently replace the normal hemopoietic precursors resulting in bone marrow failure. It is the most common malignancy in childhood accounting for nearly 30% of all pediatric cancers in the United States. [1] In Pakistan, lymphoid and hemopoietic malignancies are represented as the most common diagnostic group among all childhood malignancies [2] and acute leukemia has been found to be the second most common malignancy. Similarly, high frequency of ALL has been described in other neighboring countries like India, [3] Iran, [4] Thailand [5] and Oman. [6]

Newly diagnosed patients are stratified into prognostic groups that predict the risk of relapse. This is important because patients with favorable features are expected to respond satisfactorily to standard therapy thus minimizing the toxicity of treatment; whereas those with unfavorable features require more aggressive therapeutic strategies. Host related features include age, gender and race; disease related include white blood cell (WBC) count, morphology of the blast cells, immunophenotyping, cytogenetic and molecular genetic factors. Determination of these prognostic features at the time of diagnosis can assist an oncologist to modify treatment accordingly. [Table 1] summarizes the impact of different features on the outcome of pediatric ALL. These prognostic factors and their relation with outcome have been described extensively in various studies [7],[8],[9],[10],[11] ; however, the results obtained are not comparable and are widely debated. Apart from the features enlisted in [Table 1], a few studies have related the size of peripheral lymph nodes, liver and spleen with the outcome in ALL. [12],[13] An association between hemoglobin and platelet count with remission duration and relapse rate has also been noted by some. [12],[14],[15]{Table 1}

Before the advent of effective chemotherapy, ALL was considered a fatal disease. However, with improvements in the treatment, nearly 80% of the children have been cured. [7] This accomplishment is due to combination of multiagent chemotherapy, risk based intensification of therapy and central nervous system prophylaxis. Treatment is known to be the single most important prognostic factor, the relative prognostic power of disease characteristics varies from study to study. [1] Consequently, different sets of prognostic variables have been found to be useful in various clinical trials in pediatric ALL. These include age alone [16] ; age, central nervous system (CNS) involvement, immunophenotype of blast cells and deoxyribonucleic acid (DNA) index; age, sex, WBC count and hemoglobin level; and WBC count, hemoglobin and lymphadenopathy. [17],[18],[19] However, none of the treatment protocols have been found to be superior to others.

Prognostic features that have been defined so far as the predictors of outcome in pediatric ALL varies in different populations and so their importance is expected to vary in different centers. It can thus be assumed that features which are important in one population may not be appropriate for the other population as observed by Magrath et al. [20] in 2005.

The present work aims to evaluate demographic, clinical and laboratory features of pediatric patients suffering from acute lymphoblastic leukemia (FAB L1 and L2) and analyze their prognostic factors and disease outcome.

 Materials and Methods

This was a retrospective analysis of medical records of children up to 16 years of age who were diagnosed as ALL over a period of 17 years (January 1, 1989 to December 31, 2006) at a tertiary care hospital in Pakistan. All patients were diagnosed on the basis of morphological and cytochemical characterization of leukemic cells. French American British (FAB) classification was used for morphological classification of ALL. Results of immunophenotyping and cytogenetics were taken into consideration when available.

No information about the patient's identity was included in the data and individual records were accessed using medical record numbers. The prognostic factors which were evaluated include age, gender, WBC count, CNS and mediastinal involvement at presentation, morphology and immunophenotype of the blast cells, and response to induction therapy. Patients with L3 morphology were excluded as it constitutes a different treatment and prognostic group. The facility for cytogenetic analysis was not available at our institute during the early study period; therefore the results were not evaluated in determining the outcome. Presence or absence of mediastinal mass was assessed through chest X-ray and/or computed tomography scan in each patient.

Due to the physicians' preferences and experiences, six treatment protocols were used in these patients at our institute. The protocols that were used include Berlin-Frankfurt-Munich (BFM), [21] ALL 1997, [22] Medical Research Council childhood ALL randomized trials, UKALL X and UKALL XI, Medical Research Council Working Party on childhood leukemia (UKALL X [23] and UKALL XI [24] ) and MRC UKALL XII, Medical Research Council trial for adult patients with ALL (UKALL XII [25] ). None of the patient underwent bone marrow transplantation. Remission status was assessed at day 28 of induction therapy by performing the blood counts and bone

Complete remission was defined by less than 5% blast cells in the bone marrow and normalization of peripheral blood counts at four weeks after starting induction therapy. [26] Relapse was defined as the reappearance of more than 20% blast cells in the marrow or the presence of localized leukemic infiltrates at any site after completion of induction chemotherapy. [26] Outcome was defined as a lack of response to induction therapy, a relapse after achieving complete remission or death due to any cause. [26]

Statistical Analysis

All statistical analyses were performed using SPSS Version 15.0 (SPSS Inc. Chicago, IL, USA). Descriptive statistics were calculated for each of the variables. These included mean± standard deviation (± 2 SD) for all continuous data. The relationship of each prognostic factor to the treatment outcome was analyzed by Pearson Chi square test; however, Fisher's exact test was used when the number of cases in any category was less than 20%. P-values less than 0.05 were considered statistically significant.


Medical records of 96 patients were reviewed. However, 50 (52%) patients were excluded either because of missing data (47 patients), referral to other hospitals for treatment (2 patients) or failure to follow up (1 patient). Forty six patients with ALL were on regular follow-up and thus evaluated. These patients had an age ranging from 0.9 to 16 years (median age ± SD 6.0 ± 4.7 years). There were 35 (76.1%) males and 11 (23.9%) females. Twenty seven patients (58.7%) were in the age group of one to nine years. Two patients (4.3%) were less than one year old and 17 (37%) were more than nine years old at the time of initial diagnosis. The clinical characteristics that were noted at diagnosis include: lymphadenopathy in 25 (54.3%), hepatomegaly in 25 (54.3%), splenomegaly in 20 (43.4%), CNS involvement in two (4.3%) and enlarged mediastinum in one (2.2%) patients. Testicular involvement was not seen in any male patient (n=35) at the time of initial diagnosis.

Initial blood counts (mean ± SD) at the time of presentation were as follows: hemoglobin: 7.8 ± 2.4g/dl (range 3.7- 12.6g/dl); WBC count: 32.7 ± 67.0 x 10 9 /L (range 1.1- 340 x10 9 /L) and platelets: 64 ± 94 x 10 9 /L (range 7- 521 x10 9 /L). Forty patients (87%) had WBC count less than 50 x 10 9 /L while six (13%) had a white cell count greater than 50 x 10 9 /L.

Serum lactate dehydrogenase (LDH) levels were available for 35 patients (73.9%) and were normal (< 400IU/L) in two (5.8%) and increased in 33 (94.1%) patients. Mean LDH levels ± SD were 2294.2 ± 2615.2 IU/L. According to FAB classification, L1 and L2 morphology was observed in 18 patients (39.1%) and 17 patients (37%) respectively. In 11 patients (23.9%), morphological classification was not available.

Immunophenotyping by flowcytometry was studied in 31 patients (67.3%). Twenty eight patients (90.3%) had B-cell ALL with CD10 positivity in 20 patients (64.5%) Three patients (9.7%) were evaluated to have T-cell phenotype.

Karyotyping results were available for 12 patients (26.2%) only which showed normal cytogenetics in nine patients and; hyperdiploidy, 47XX and 48XY each in one patient. Because of the paucity of data on karyotyping, it was not evaluated as a prognostic factor by us.

At day 28 of induction therapy, 45 patients (97.8 %) were found to be in complete remission and the percentage of bone marrow blasts was less than 5%. One patient (2.2%) received compromised treatment due to financial constraints and he achieved remission after another two weeks of chemotherapy. Of the 46 evaluable patients, 30 (65.2%) were alive and doing well at the time of study including 26 (86.6%) who remained relapse free throughout their course and four patients (13.3%) who had a relapse. Median time to follow-up of these patients was 46 months. The other 16 patients (34.7%) did not survive including 11 (68.7%) who had a relapse. Median time to relapse was twenty five months. Among the non salvageable patients, 12 (75%) were males and four (25%) were females. Nine patients (59.9%) had CNS relapse while systemic relapse was observed in four patients (26.6%). Testicular relapse was observed in only two male patients (13.3%). The mortality among the relapsed patients was 73.3% (11 patients).

[Table 2] shows the distribution of various prognostic factors in relation to the outcome in 46 patients with pediatric ALL. This shows that CNS and mediastinal involvement, WBC count and morphological type of the blast cells at presentation, and response to induction therapy had no prognostic importance; however CNS and mediastinal involvement at presentation and delayed response to induction therapy were associated with poor outcome. Treatment related toxicities were observed in six patients including disseminated fungal infection in two, combined pseudomonas and herpes infection, febrile neutropenia, tumor lysis syndrome and vincristine related neuropathy in one each. All six patients succumbed to their complications following therapy.

We found age at the time of presentation (P= 0.04) and ALL phenotype (P= 0.03) as the only prognostic factors with significant influence on the outcome in childhood ALL. Patients in age group 1-9 years (n=27) fare well and had statistically significant better survival compared to children less than one year of age (n=2) who had 100% mortality. Similarly, patients with CD10 positive B-cell ALL (n=20) did well with a better outcome as compared to both CD10 negative B-cell ALL (n=8) and T-cell phenotype (n=3).{Table 2}

Different treatment protocols used in the study were not compared for the difference in prognosis. However, it was observed that patients who were treated with BFM protocol (n=5) responded very well (P= 0.05). Clinical and laboratory features of patients who relapsed and/ or expired are summarized in [Table 3]. It was observed that all expired patients without relapse (n=5) had massive hepatosplenomegaly and this was a significant finding. Similarly significant lymphadenopathy and splenomegaly were present in all the patients who relapsed and expired (n=11).{Table 3}


Various clinical and laboratory features present at the time of initial diagnosis can predict the likelihood that a patient will remain in remission or not. Most of the classical prognostic factors described in ALL were analyzed in this study. The patients who had an outcome were found to have one or more of the poor prognostic features. However, only age and ALL phenotype were found to have significant influence on the outcome when analyzed. In many studies, age was found to be a strong prognostic factor in childhood ALL; and this feature retained its significance in our study also (p value 0.04). [18],[27] Patients in the age group 1 to 9-years age group were found to have the best prognosis among all age groups. It was also observed in our study that patients less than one year old showed a poor outcome and this is in concordance with other studies which have shown extraordinarily high failure rates among infants. [28],[29] This may be due to the presence of other unfavorable presenting features in this age group such as hyperleucocytosis, CNS disease, lack of CD10 expression and cytogenetic abnormalities involving 11q23 gene rearrangements.

The other factor which emerged as an important prognostic factor in our study was ALL phenotype with a p value 0.03 and this is in agreement with other studies. [17],[30] CD10 positive phenotype was found to be a favorable feature in terms of prognosis followed by CD10 negativity. Although the number of patients is small, T-cell phenotype of blast cells continued to have a poor outcome.

Regarding gender, it has been reported in various studies that females have superior survival than males of same age group having ALL. [18],[31] However, we did not observe such gender based outcome. This may be because of small sample size of females (n= 11/ 46) in our study.

The size of peripheral lymph nodes, liver and spleen provide an indirect measurement of leukemic cell burden. [1] Therefore, several studies have demonstrated that massive lymphadenopathy, [32] hepatomegaly and/ or splenomegaly [33] impact adversely on remission duration and survival. With improvements in therapy, the importance of these factors has disappeared. However, we observed that among the patients who had an outcome, those who had extensive lymphadenopathy and hepatosplenomegaly at the time of presentation were found to have inferior survival.

CNS status and mediastinal involvement at presentation are also considered to be important features that can predict the outcome in patients with ALL. [4],[20] It has been reported previously that early response to induction therapy is also important for determining the prognosis in patients with ALL. [17],[24] Although not statistically significant, we found that patients with CNS (n=2) and mediastinum (n=1) involvement at presentation and those showing slow response to induction therapy (n=1) had a poor outcome.

The WBC count-prognosis relationship has been established in many studies on pediatric ALL [8],[18] and it has been observed that a count more than 50x10 9 /L at presentation is associated with adverse outcome. However, we did not find WBC count to be a significant prognostic factor and even patients with a very high white cell count responded very well to treatment and were still doing well at the time of study. This may be because of the presence of other favorable prognostic factors like age group and morphology of the blast cells in these patients (all three patients with high WBC count at presentation were in the 1 to 9 years age group and morphology of the blast cells revealed ALL L1).

Studies on the prognostic influence of morphologic type of blast cells have yielded controversial results. Few have denied any association between FAB type and prognosis [34] while others have reported L2 morphology to be associated with poor outcome. [35] However, our study negates any such significance of blast cell morphology. Despite advancements in childhood ALL management, relapsed disease status contributes significantly to the morbidity and mortality in these patients. Overall, less than one in four patients with relapse achieve long term survival. [1] A mortality of 73.3% was observed among the relapsed patients (n=15). This is in concordance with previously published reports. [36]

In our experience, many classical prognostic factors in pediatric ALL showed little or no significance. Since the prevalence of individual risk factors is variable in different ethnic groups due to variable genetic make up, prognostic factors that are important in one population may not be applicable to others.


Our study was limited by a number of factors. Being a retrospective study, the data of a large number of patients could not be retrieved. Secondly, karyotyping as a prognostic factor was not studied which could have yielded noteworthy results. Finally, because of various treatment protocols used by physicians, the outcome, based on therapy, could not be evaluated.


ALL was found to be a frequent childhood hematological neoplasm in our setting. The disorder was more prevalent in males and children under ten years of age. Age and leukemia phenotype emerged as the important prognostic factors in pediatric ALL in our patients.

Most relapses continue to occur in patients who are considered to have a good prognosis and vice versa, as observed in our study. Better predictors are thus needed to further refine therapy and at the same time avoiding dose related toxicity. Well designed prospective studies are required to focus on the different prognostic factors and see their impact on the outcome.


1Whitlock JA, Gaynon PS. Acute Lymphoblastic leukemia in children. In: Greer JP, Foerster J, Lukens JN, Rodgers GM, Paraskevas F, Glader B, editors. Wintrobe's Clinical Hematology. USA: Lippincott Williams and Wilkins; 2004. p. 2143-68.
2Bhurgri Y. Childhood lymphoma and leukemia. J Pak Med Assoc 2006;56:147-8.
3Swaminathan R, Rama R, Shanta V. Childhood cancers in Chennai, India, 1990-2001: incidence and survival. Int J Cancer 2008;122:2607-11.
4Karimi M, Yarmohammadi H, Sabri MR. An analysis of prognostic factors and the five year survival rate in childhood acute lymphoblastic leukemia. Med Sci Monit 2002;8:792-6.
5Kamsa-ard S, Wiangnon S, Suwanrungruang K, Jetsrisuparb A, Horsith S. Trends in incidence of childhood leukemia, Khon Kaen, Thailand 1985-2002. Asian Pac J Cancer Prev 2006;7:75-8.
6Knox-Macaulay HH, Brown LC. Descriptive epidemiology of de novo acute leukemia in the Sultanate of Oman. Leuk Res 2000;24:589-94.
7Pui CH, Evans WE. Acute lymphoblastic leukemia. N Engl J Med 1998;339:605-15.
8Dopfer R, Henze G, Bender-Gφtze C, Ebell W, Ehninger G, Friedrich W, et al. Six year experience with a comprehensive approach to the treatment of recurrent childhood acute lymphoblastic leukemia (ALL- REZ BFM 85): A relapse study of the BFM group. Blood 1991;78:1166-72.
9Griffin TC, Shuster JJ, Buchanan GR, Murphy SB, Camitta BM, Amylon MD. Slow disappearance of peripheral blood blasts is an adverse prognostic factor in childhood T-cell acute lymphoblastic leukemia: a Pediatric Oncology study group. Leukemia 2000;14:792-5.
10Gaynon PS, Desai AA, Bostrom BC, Hutchinson RJ, Lange BJ, Nachman JB, et al. Early response to therapy and outcome in childhood acute lymphoblastic leukemia: a review. Cancer 1997;80:1717-26.
11Silverman LB, Gelber RD, Young ML, Dalton VK, Barr RD, Sallan SE. Induction failure in acute lymphoblastic leukemia of childhood. Cancer 1999;85:1395-404.
12Miller DR, Leikin S, Albo V, Sather H, Karon M, Hammond D. Prognostic factors and therapy in acute lymphoblastic leukemia of childhood. Cancer 1983;51:1041-9.
13Schwenn MR, Blattner SR, Lynch E, Weinstein HJ. Hic-COM: a 2 month intensive chemotherapy regimen for children with stage III and IV Burkitt's lymphoma and B- cell acute lymphoblastic leukemia. J Clin Oncol 1991;9:133-8.
14Robison LL, Sather HN, Coccia P. Assessment of the interrelationship of prognostic factors in childhood acute lymphoblastic leukemia. Am J Pediatr Hematol Oncol 1980;2:5-13.
15Zippin C, Cutler SJ, Reeves WJ Jr, Lum D. Variation in survival among patients with acute lymphocytic leukemia. Blood 1971;37:59-72
16Mφricke A, Zimmermann M, Reiter A, Gadner H, Odenwald E, Harbott J, et al. Klin Pediatr 2005;217:310-20.
17Hussein H, Sidhom I, Naga SA, Hussein H, Sidhom I, Naga SA, Amin M, et al. Outcome and prognostic factors of acute lymphoblastic leukemia in children at the National Cancer Institute, Egypt. J Pediatr Hematol Oncol 2004;26:507-14.
18Ng SM, Lin HP, Ariffin WA, Zainab AK, Lam SK, Chan LL. Age, sex, hemoglobin level and white cell count at diagnosis are important prognostic factors in children with acute lymphoblastic leukemia treated with BFM type protocol. J Trop Pediatr 2000;46:338-43.
19Advani S, Pai S, Venzon D, Adde M, Kurkure PK, Nair CN, et al. Acute lymphoblastic leukemia in India: an analysis of prognostic factors using a single treatment regimen. Ann Oncol 1999;10:167-76.
20Magrath I, Shanta V, Advani S, Adde M, Arya LS, Banavali S, et al. Treatment of acute lymphoblastic leukemia in countries with limited resources; lessons from use of a single protocol in India over a twenty year period. Eur J Cancer 2005;41:1570-83.
21Riehm H, Reiter A, Schrappe M, Berthold F, Dopfer R, Gerein V, et al. Corticosteroid dependent reduction of leukocyte count in blood as a prognostic factor in acute lymphoblastic leukemia in childhood (therapy study ALL-BFM 83). Klin Pediatr 1987;199:151-60.
22Hann I, Vora A, Richards S, Hill F, Gibson B, Lilleyman J, et al. Benefit of intensified treatment for all children with acute lymphoblastic leukaemia: results from MRC UKALL XI and MRC ALL97 randomized trials. UK Medical Research Council's Working Party on Childhood Leukaemia. Leukemia 2000;14:356-63.
23Chessells JM, Bailey C, Richards SM. Intensification of treatment and survival in children with lymphoblastic leukemia: Results of UK Medical Research Council Trial UKALL X. Lancet 1995;345:143-8.
24Hann I, Vora A, Harrison G, Harrison C, Eden O, Hill F, et al. UK Medical Research Council's Working Party on Childhood Leukaemia. Determinants of outcome after intensified therapy of childhood lymphoblastic leukemia: results from Medical Research Council United Kingdom acute lymphoblastic leukemia XI protocol. Br J Haematol 2001;113:103-14.
25Goldstone AH, Richards SM, Lazarus HM, Tallman MS, Buck G, Fielding AK, et al. In adults with standard risk acute lymphoblastic leukemia (ALL) the greatest benefit is achieved from a matched sibling allogeneic transplant in first complete remission and an autologous transplant is less effective than conventional/ maintenance chemotherapy in all patients: final results of the international ALL trial (MRC UKALL XII/ ECOG E2993). Blood 2008;111:1827-33.
26Reiter A, Schrappe M, Ludwig WD, Hiddemann W, Sauter S, Henze G, et al. Chemotherapy in 998 unselected childhood acute lymphoblastic leukemia patients: results and conclusion of the multicenter trial ALL-BFM 86. Blood 1994;84:3122-33.
27Reaman GH, Sposto R, Sensel MG, Lange BJ, Feusner JH, Heerema NA, et al. Treatment outcome and prognostic factors for infants with acute lymphoblastic leukemia treated on two consecutive trials of Children's Cancer Group. J Clin Oncol 1999;17:445-55.
28Frankel LS, Ochs J, Shuster JJ, Dubowy R, Bowman WP, Hockenberry-Eaton M, et al. Therapeutic trial for infant acute lymphoblastic leukemia: The Pediatric Oncology group experience (POG 8493). J Pediatr Hematol Oncol 1997;19:35-42.
29Ferster A, Bertrand Y, Benoit Y, Boilletot A, Behar C, Margueritte G, et al. Improved survival for acute lymphoblastic leukemia in infancy: The experience of EORTC Childhood leukemia cooperative group. Br J Haematol 1994;86:284-90.
30Donadieu J, Auclerc MF, Baruchel A, Leblanc T, Landman-Parker J, Perel Y, et al. Critical study of prognostic factors in childhood acute lymphoblastic leukemia: differences in outcome are poorly explained by the most significant prognostic variables. Br J Haematol 1998;102:729-39.
31Shuster JJ, Wacker P, Pullen J, Humbert J, Land VJ, Mahoney DH Jr, et al. Prognostic significance of sex in childhood B- precursor acute lymphoblastic leukemia: a Pediatric Oncology Study Group. J Clin Oncol 1998;16:2854-63.
32Chilcote RR, Coccia P, Sather HN, Robison LL, Baehner RL, Nesbit ME Jr, et al. Mediastinal mass in acute lymphoblastic leukemia. Med Pediatr Oncol 1984;12:9-16.
33Miller DR, Leikin S, Albo V, Sather H, Karon M, Hammond D. Prognostic factors and therapy in acute lymphoblastic leukemia of childhood. Cancer 1983;51:1041-9.
34Miller DR, Krailo M, Bleyer WA, Lukens JN, Siegel SE, Coccia PR, et al. Prognostic implications of blast cell morphology in childhood acute lymphoblastic leukemia: A report from the Children's Cancer study group. Cancer Treat Rep 1985;69:1211-5.
35Gaynon PS, Qu RP, Chappell RJ, Willoughby ML, Tubergen DG, Steinherz PG, et al. Survival after relapse in childhood acute lymphoblastic leukemia: impact of site and time to first relapse, the Children's Cancer group experience. Cancer 1998;82:1387-95.
36Chessells JM. Relapsed lymphoblastic leukemia in children: a continuing challenge. Br J Haematol 1998;102:423-38.