Treatment of children with acute lymphoblastic leukaemia (ALL) aims to cure all patients with as little toxicity as possible and, if possible, to restrict further intensification of chemotherapy to patients with an increased risk of relapse. However in Medical Research Council (MRC) trial UKALL X two short myeloablative blocks of intensification therapy given at weeks 5 and 20 were of benefit to children in all risk groups. The successor trials, MRC UKALL XI and MRC ALL97, tested whether further intensification would continue to benefit all patients by randomising them to receive, or not, an extended third intensification block at week 35. After a median follow-up of 4 years (range 5 months to 8 years), 5 year projected event-free survival was superior at 68% for the 894 patients allocated a third intensification compared with 60% for the 887 patients who did not receive one (odds ratio 0.75, 95% CI 0.63–0.90, 2P = 0.002). This difference was almost entirely due to a reduced incidence of bone marrow relapses in the third intensification arm (140 of 891 in the third intensification arm vs 171 of 883 in the no third intensification, 2P = 0.02). Subgroup analysis suggests benefit of the third intensification for all risk categories. Overall survival to date is no different in the two arms, indicating that a greater proportion of those not receiving a third intensification arm and subsequently relapsing can be salvaged. These results indicate that there is benefit of additional intensification for all risk subgroups of childhood ALL.
Risk categorisation in childhood acute lymphoblastic leukaemia (ALL) has become increasingly complex and now combines clinical factors such as WBC, age, gender and speed of response with identification of cytogenetic and molecular genetic abnormalities,12 and most recently the monitoring of minimal residual disease by molecular methods.345 The aim of such strategies is to identify patients at high risk of treatment failure who may benefit from more intensive treatment and those in whom treatment may, perhaps, be safely reduced to avoid possible long-term morbidity.6 Although there are some sub-groups of patients with ALL who have a much better prospect of cure than others,789 there is also good evidence that it is these very patients who have benefited most from intensified treatment protocols.1011
In keeping with results of other studies conducted in the last 15 years,1213 introduction of intensification therapy following the induction of morphological remission improved outcome for all children with ALL in the Medical Research Council United Kingdom Acute Lymphoblastic Leukaemia X(MRC UKALL X).14 This was obtained with two short 5-day intensification blocks more closely resembling acute myeloid leukaemia-type treatment regimens, rather than the extended 8 week intensification introduced by the German Berlin–Frankfurt–Munster (BFM) group,13 the latter approach having been investigated in randomised trials by the American CCG.12 The successor UK trials, MRC UKALL XI and ALL97, tested in a randomised fashion whether a third block of extended intensification at 35 weeks would benefit patients when added to the UKALL X regimen including the two short blocks.
Patients and methods
1781 consecutive children with ALL were randomised to a third intensification block or not in two MRC studies, UKALL XI and ALL97, between March 1992 and July 1998. UKALL XI accrued 2090 patients from August 1990 to March 1997 but the third intensification randomisation was only introduced in March 1992. ALL97 continued accruing patients to this randomisation until June 1998, when results were revealed by an independent data monitoring committee after the fifth interim analysis and randomisation halted in order for all subsequent patients to receive the third block. This report contains follow-up of all randomised patients to October 1998.
Eligibility criteria for entry to the two studies were different. UKALL XI recruited all ALL patients aged 1–15 years inclusive, except mature B cell; whilst ALL97 randomised children aged 1–18 years considered to have standard risk ALL by the Oxford hazard score,7 based on initial WBC, age and gender but excluded patients in whom cytogenetic or molecular genetic studies showed Ph′-positive ALL, balanced translocations involving chromosome 11q23 or near-haploidy.
894 (50%) of the 1781 randomised patients were allocated three intensification blocks (third block arm) and 887 received only two intensification blocks (non-third block arm). Characteristics of third block and non-third block recipients are described in Table 1 which shows that the groups were matched for age, gender, white cell count (WCC) at presentation and other presenting features.
Details of the randomisation scheme and treatment schedules are shown in Figure 1 and Table 2. All patients received a 4 week induction course followed by two 5-day intensification blocks at weeks 5 and 20 identical to the schedule associated with the best outcome in UKALL X.14 They were then randomised to receive, or not, an 8 week third intensification block at week 35. All patients received a total of 100 (UKALL XI) or 105 (MRC ALL97) weeks of treatment.
Both studies had additional randomisations. UKALL XI included a central nervous system (CNS)-directed treatment randomisation of high-dose intravenous methotrexate (HD-MTX) plus IT MTX vs intrathecal methotrexate only (IT MTX) for patients with presenting WCC <50 × 109/l which has been reported separately,15 and HD-MTX vs fractionated (24 Gy in 15 fractions of 1.6 Gy each) cranial radiotherapy for those with presenting WCC ⩾50 × 109/l. The randomisation for high risk patients with respect to irradiation continues to accrue patients in MRC ALL97. In addition, MRC ALL97 continues to accrue patients to a randomised comparison of dexamethasone vs prednisolone during induction and continuing treatment, and mercaptopurine vs thioguanine during continuing treatment. Patients with presenting WCC <50 × 109/l receive IT MTX alone as CNS-directed treatment in MRC ALL97.15
Individual centres in the UK obtained approval from their local research ethics committees, and informed consent from parents and patients, before entering patients into these protocols.
Randomisation method and statistics
Randomisation was done centrally on computer for both trials, and minimisation16 was used to balance within age, gender, WCC and CNS therapy allocated in UKALLXI, plus steroid and thiopurine allocated in ALL97. The planned sample size was 2500, for detection of a 5% difference with 80% power. No formal stopping rule was used, but the data monitoring committee was required to report the results to the steering committee if, in the light of results from an interim analysis and from any relevant randomised trials elsewhere, there was evidence beyond reasonable doubt that one treatment arm was clearly indicated or clearly contraindicated.
The main analyses used the logrank method17 with death or any event (relapse or death) as the primary endpoints. Secondary endpoints were bone marrow relapse, relapse without bone marrow involvement and death in first remission. Analyses of these excluded the patients who died without achieving remission. Estimates of event rate ratios within subgroups were calculated using the one-step approximation to the odds ratio.18 Cox regression analysis19 was used to confirm the independence of variables by use of the SAS statistical package.20
Of the 1781 randomised children, one in the no third block arm was given non-protocol treatment prior to achievement of remission and died of GVHD following an allogeneic bone marrow transplant in first remission. This patient has been counted as a death in remission in the survival analyses. Additionally, two patients were found to have high risk features (one with MLL gene rearrangement and one with near haploidy) after they had been randomised and were subsequently treated on the concurrent Medical Research Council high-risk protocol, ALL-HR1. These patients are all included in the intention-to-treat analyses to avoid post hoc bias.
Five patients died during induction treatment (two third block and three non-third block). Fifty-four patients failed to achieve a complete remission (31 third block and 23 non-third block) after 28 days of induction treatment. Of these, 52 remitted after receiving the first intensification block alone or with additional courses as further induction therapy, while two died with resistant disease (see Table 3). These non-remitters at day 28 had a poor outcome with 5-year EFS of 35% (95% CI = 11–59%).
At a median follow-up of 4 years (range 5 months to 8 years 1 month), 5-year disease-free survival is significantly better for patients in the third block arm (68%, 95% CI = 64–72%) compared with the no third block arm (60%, 95% CI = 56–64%) (Figure 2, OR 0.75, logrank 2P = 0.002). This difference was primarily due to fewer bone marrow relapses amongst patients who received three intensifications (third block 24% vs non-third block 30% at 5 years, 2P = 0.02, Figure 3), although the incidence of extramedullary relapses was also non-significantly lower (Figure 4, third block 9% vs non-third block 11% at 5 years, 2P = 0.3). The effect on event-free survival remained significant (2P = 0.003) if those with events before week 35, the time at which the third block was given, were excluded.
Overall survival was not significantly affected by the third block (Figure 5, third block 87% vs no third block 85% at 5 years, 2P = 0.2), since there was somewhat better salvage of no third block arm relapses with 58% survival 3 years post-relapse against 52% for those relapsing in the third block arm (not significant; 2P = 0.2).
There was no statistically significant difference in the effect of third intensification between subgroups by age, white blood count, immunophenotype or international risk groups (Figure 6). However, the number of patients with T cell disease was small so there is some uncertainty about the benefit or not for them, and there were only 48 patients in the risk group defined by age 10 years or more and white cell count 50 × 109/l or above. Analyses within groups according to CNS-directed therapy allocation also gave very similar results for the effect of the third block in each group, but with only 16 events among the ALL97 patients due to the length of follow-up, it is not possible to split the analysis by steroid or thiopurine allocation.
Twenty-two deaths occurred during first complete remission (CR). Of these patients, six were recipients of the third block and 16 were not. Inspection of the causes of non-relapse deaths (Table 3) reveals that the excess deaths in non-third block recipients were primarily due to infection and secondary leukaemia: only one death occurred during or shortly after the third block.
The third block was associated with moderate excess morbidity. Antibiotics were used in 53% of third block recipients but only four patients (0.6%) received amphotericin, and 58% of patients requiring antibodies received only 1 week or less. Although 75% of patients experienced some degree of neutropenia, it was severe (<0.1 × 109/l) in only 15% and did not persist beyond 1 week in 80% of these patients. Sixty-eight percent of patients had neutropenia of less than 0.5 × 109/l. Intravenous feeding was required in only 1.8% of patients, suggesting that mucositis was an insignificant problem in the vast majority. Gut toxicity occurred in only 20% of patients and was severe in 0.3%. Two-thirds of patients required blood product support but moderate to severe bleeding was reported in only 3% of patients. These treatment-related complications resulted in 25% of patients being unable to complete the full third block. Additionally, 51% had their treatment interrupted due to complications during the third block. Of these, 72%, 25% and 3% suffered one, two and three such interruptions respectively, resulting in a 1 to 7 week delay in completion of their treatment.
Cox regression analysis showed that age, gender, WCC, late remission and third block allocation were all of independent prognostic importance. The Oxford hazard score14 (0.22 + log [WCC + 1] + 0.0043 × age2 − 0.39 × sex; boy = 1, girl = 2) remained valid for defining a small group of very high risk patients (score > 0.8) in both the third block arm (5 year EFS = 41%, 95% CI = 28–55%) and the non-third block arm (5 year EFS = 39%, 95% CI = 25–53%), consisting of about 6% of those randomised.
The results of MRC UKALL XI and ALL97 presented here re-inforce and extend the findings of UKALL X that intensification of treatment is of benefit to all children with ALL. This randomised comparison in a large group of children of all risk categories indicates that the addition of a third block of intensification therapy at 35 weeks improved 5 year event-free survival by about 8%. Patients of all risk categories had a superior outcome with three blocks of intensification therapy, with no suggestion of less benefit in lower risk patients. The extra intensification offered protection against marrow relapse, although extra-medullary disease was also marginally reduced in the third intensification group.
There have been a number of recent trials demonstrating the benefit of various intensified regimens in selected higher risk patients with ALL2122 but two recent randomised trials have involved patients with standard risk ALL. The German BFM 86 included a randomisation to a further 4 week intensification of treatment at 12 months for standard risk patients with no improvement in event-free survival but only one quarter of eligible patients were randomised.13 The CCG randomised over 1200 patients to receive double delayed intensification, intensified vincristine and prednisolone, or standard treatment at 9–10 months. Patients receiving double delayed intensification had a 7.5% improvement in EFS at 5 years.23
It is uncertain whether it was the mere application of a further block of intensification which led to the improved outcome in UKALL XI and ALL97, or whether the design of the schedule has made an important contribution. The 8-week intensification regimen tested here was similar to the prolonged schedules originally devised by the German BFM group24 and later modified by the American CCG12 group of investigators. It is a departure from the short, but extremely myelosuppressive, 5 day blocks shown to be effective in UKALL X, not just in duration but also in containing a re-induction component and different drug combinations. In the absence of any randomised comparisons of the two types of schedule it is impossible to know if the more prolonged treatment is better. It is however possible that the more prolonged schedule provides increased dose intensity over a prolonged duration at a time when minimal residual disease levels may be starting to rise in a proportion of patients.25
Although the toxicity of the third block was only moderate it precluded completion of the full 8 weeks in around 25% of patients. Remarkably, only one of 900 patients given the third block in multiple centres died from complications which could in any way be related to it directly. It is possible that the improvement in outcome with a third block would be even greater if the full block was administered to a higher proportion of patients which the moderate morbidity and low mortality should permit.
This report adds to the evidence that additional intensification has benefit for all currently defined risk group patients.2324 Whether a group of patients can be defined with sufficiently low risk of relapse to warrant exclusion from protocols involving further intensification of treatment is unproven at present. The use of newer technologies to detect minimal residual disease has identified patients in whom persistent disease after early treatment predicts a high risk of relapse.25 However, it is not known whether patients with no detectable minimal residual disease after some months can be deemed ‘cured’ and may safely curtail treatment. Pending such information it is perhaps more appropriate to explore improvements in treatment for standard risk patients by alterations in therapy which combine a possibility of improved survival with low risk of late morbidity. Such investigations of steroid treatment and thiopurines are in progress in ALL 97 and in randomised trials performed by other collaborative groups.5
Rubnitz JE, Look AT . Molecular genetics of childhood leukemias J Pediatr Hematol Oncol 1998 20: 1–11
Chessells JM, Swansbury GJ, Reeves B, Bailey CC, Richards SM . Cytogenetics and prognosis in childhood lymphoblastic leukaemia: results of MRC UKALL X. Medical Research Council Working Party in Childhood Leukaemia Br J Haematol 1997 99: 93–100
Brisco MJ, Condon J, Hughes E, Neoh SH, Sykes PJ, Seshadri R, Toogood I, Waters K, Tauro G, Ekert H . Outcome prediction in childhood acute lymphoblastic leukaemia by molecular quantification of residual disease at the end of induction Lancet 1994 343: 196–200
Cave H, van der Werff ten Bosch, Suciu S, Guidal C, Waterkeyn C, Otten J, Bakkus M, Thielemans K, Grandchamp B, Vilmer E . Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia. European Organization for Research and Treatment of Cancer–Childhood Leukemia Cooperative Group New Engl J Med 1998 339: 591–598
van Dongen JJ, Seriu T, Panzer-Grumayer ER, Biondi A, Pongers-Willemse MJ, Corral L, Stolz F, Schrappe M, Masera G, Kamps WA, Gadner H, van Wering ER, Ludwig WD, Basso G, de Bruijn MA, Cazzaniga G, Hettinger K, van der Does-van den Berg A, Hop WC, Riehm H, Bartram CR . Prognostic value of minimal residual disease in acute lymphoblastic leukaemia in childhood Lancet 1998 352: 1731–1738
Ochs J, Mulhern R . Long-term sequelae of therapy for childhood acute lymphoblastic leukaemia Baillières Clin Haematol 1994 7: 365–376
Chessells JM, Richards SM, Bailey CC, Lilleyman JS, Eden OB . Gender and treatment outcome in childhood lymphoblastic leukaemia: report from the MRC UKALL trials Br J Haematol 1995 89: 364–372
Chessells JM, Hall E, Prentice HG, Durrant J, Bailey CC, Richards SM . The impact of age on outcome in lymphoblastic leukaemia; MRC UKALL X and XA compared: a report from the MRC Paediatric and Adult Working Parties Leukemia 1998 12: 463–473
Pui CH . Recent advances in the biology and treatment of childhood acute lymphoblastic leukemia Curr Opin Hematol 1998 5: 292–301
Hutchinson RJ, Neerhout S, Bertolone S, Cooper HA, Roskos R, Tannous R, Wells L, Heerema N, Hummell D, Sailoer S, Sather H, Trigg M and the Children's Cancer Study Group . Should therapy be intensified for patients with good risk acute lymphoblastic leukaemia Am Soc Hematol 1996 A2662
Rivera GK, Pinkel D, Simone JV, Hancock ML, Crist WM . Treatment of acute lymphoblastic leukemia. 30 years’ experience at St Jude Children's Research Hospital New Engl J Med 1993 329: 1289–1295
Tubergen DG, Gilchrist GS, O'Brien RT, Coccia PF, Sather HN, Waskerwitz MJ, Hammond GD . Improved outcome with delayed intensification for children with acute lymphoblastic leukemia and intermediate presenting features: a Childrens Cancer Group phase III trial J Clin Oncol 1993 11: 520–526
Reiter A, Schrappe M, Ludwig WD, Hiddemann W, Sauter S, Henze G, Zimmermann M, Lampert F, Havers W, Niethammer D, Odenwald E, Ritter J, Mann G, Welte K, Gadner H, Riehm H . Chemotherapy in 998 unselected childhood acute lymphoblastic leukemia patients. Results and conclusions of the multicenter trial ALL-BFM 86 Blood 1994 84: 3122–3133
Chessells JM, Bailey C, Richards SM . Intensification of treatment and survival in all children with lymphoblastic leukaemia: results of UK Medical Research Council trial UKALL X. Medical Research Council Working Party on Childhood Leukaemia Lancet 1995 345: 143–148
Hill F, Hann I, Gibson B, Eden OB, Richards S . Comparison of high dose methotrexate with continuing intrathecal methotrexate versus intrathecal methotrexate alone in low white blood count childhood acute lymphoblastic leukaemia: preliminary results from the UKALLXI randomised trial Blood 1998 92: (Suppl. 1) 398a
White SJ, Freedman LS . Allocation of patients to treatment groups in a controlled clinical study Br J Cancer 1978 37: 849–857
Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, Mantel N, McPherson K, Peto J, Smith PG . Design and analysis of randomized clinical trials requiring prolonged observation of each patient. II Analysis and examples Br J Cancer 1977 35: 1–39
Early Breast Cancer Trialists’ Collaborative Group . Systemic treatment of early breast cancer by hormonal cytotoxic, or immune therapy. 133 randomised trials involving 31 000 recurrences and 24000 deaths among 75000 women Lancet 1992 339: 1–15
Cox D . The contribution of statistical methods to cancer research Cancer 1991 67: 2428–2430
SAS/STAT User's Guide SAS Institute Inc: Cary, NC 1989
Steinherz PG, Gaynon PS, Breneman JC, Cherlow JM, Grossman NJ, Kersey JH, Johnstone HS, Sather HN Trigg ME, Uckun FM, Bleyer WA . Treatment of patients with acute lymphoblastic leukemia with bulky extra-medullary disease and T-cell phenotype or other poor prognostic features: randomized controlled trial from the Children's Cancer Group Cancer 1998 82: 600–612
Nachman JB, Sather HN, Sensel MG, Trigg ME, Cherlow JM, Lukens JN, Wolff L, Uckun FM, Gaynon PS . Augmented post-induction therapy for children with high-risk acute lymphoblastic leukaemia and a slow response to initial therapy New Engl J Med 1998 338: 1663–1671
Lange B, Sather H, Weetman R, Bostrom B, McGuire P, Rackoff W, Arthur D, Trigg M for the Children's Cancer Group . Double delayed intensification improves outcome in moderate risk paediatric acute lymphoblastic leukaemia – CCG 1891 study ASH 1997 90: Suppl. 1: 559a
Henze G, Fengler R, Reiter A, Ritter J, Riehm H . Impact of early intensive reinduction therapy on event-free survival in children with low-risk acute lymphoblastic leukemia Hamatol Bluttransfus 1990 33: 483–488
Morley A . Quantifying leukemia Lancet 1998 339: 627–629
We are grateful to all the physicians who entered patients to these studies. Members of the MRC childhood leukaemia working party during these studies were: Bailey CC, Barton C, Broadbent V, Caswell M, Chessells JM, Darbyshire PJ, Dempsey SI, Durrant J, Eden OB (Chairman), Gibson B, Goodman A, Gray R, Hann I, Haworth C, Forman K, Hill F, Jenney M, Kernahan J, King DJ, Kinsey SE, Madden M, Mann JR, Martin J, Meller ST, Mitchell C, Oakhill A, Radford M, Reid MM, Richards SM, Smyth O, Stevens RF, Thomas A, Vargha-Khadem F, Vora AJ, Walker D, Webb D, Wheatley K, Will A, Windebank K. We also thank J Burrett for data management and all those who worked in the randomisation office.
About this article
Emergent Radiation for Leukemic Optic Nerve Infiltration in a Child Receiving Intrathecal Methotrexate
Practical Radiation Oncology (2019)
Nanoliposomal L-Asparaginase and Its Antitumor Activities in Lewis Lung Carcinoma Tumor-Induced BALB/c Mice
Advances in Materials Science and Engineering (2019)
Genome-wide association study identifies susceptibility loci for B-cell childhood acute lymphoblastic leukemia
Nature Communications (2018)
A genome-wide association study identifies risk loci for childhood acute lymphoblastic leukemia at 10q26.13 and 12q23.1
Near-haploid and low-hypodiploid acute lymphoblastic leukemia: two distinct subtypes with consistently poor prognosis