Incidence and Outcome of Osteonecrosis in Children and Young Adults with Acute Lymphoblastic Leukaemia Treated on a Dexamethasone Containing Protocol: Results of the Medical Research Council UK Trial ALL 2003.

There is concern that dexamethasone is associated with a higher risk of ostenecrosis than prednisolone when used in older children and young adults with Acute Lymphoblastic Leukaemia. UKALL 2003 is an ongoing randomised controlled trial in which patients receive dexamethasone 6 mg/m² for 29 days during induction and 5 days each month during maintenance in combination with Vincristine. Additionally, they receive 10 mg/m² for 14 days in an alternate week schedule during each of two delayed intensifications. The trial is investigating whether treatment can be adjusted according to Minimal Residual Disease (MRD) levels in children and young adults with ALL. The trial opened in October 2003 to patients aged 1 – 21 years (increased to 25 years recently) and at 30/06/08 had accrued 1864 patients.

Intensity of treatment is determined by risk stratification based on NCI risk, blast karyotype, and morphological marrow response at day 8/15 of induction to allocate patients to one of 3 treatment regimens (A, B and C – escalating intensity from A to C). NCI standard risk (SR) patients receive a 3 drug induction whilst high risk (HR) patients receive 4 drugs including daunorubicin. Pegylated E. Coli Asparaginase (Medac, Oncospar) is used throughout treatment. Only patients with overt CNS disease (CNS3) at presentation receive cranial radiotherapy, and high dose Methotrexate is not part of the protocol, although Regimen C includes escalating intermediate dose IV Methotrexate without folinic acid rescue given during two courses of Capizzi interim maintenance . CNS directed therapy comprises 19 – 26 doses of intrathecal methotrexate given over 2 – 3years depending on regimen and gender.

The actuarial incidence of symptomatic AVN (confirmed on MRI) is 4% overall and 1%, 13% and 16% in patients < 10 years, 10–15 years and > 16 years respectively (p <0.00005 ≤ 10 vs > 10). There is no difference in incidence by sex (M vs F 3.5% vs 4%, P =0.7) or treatment intensity (A= 1%, B =8%, C = 6%, p (stratified by age) = 0.7). Onset of symptoms was during maintenance in the vast majority of affected patients (40/41). The sites involved were hips = 19, knees = 14, ankles/feet = 7, shoulders = 4, elbows = 4, spine = 3 and multi-focal = 13. Management varied from centre to centre and consisted of switching to prednisolone or stopping steroids followed by no further treatment in 19, surgery in 8 (core decompression = 4, hip replacement = 4) and administration of bisphosphonates in 2 cases. Eighteen patients have resulting moderate to severe disability, 5 have recovered and the remaining have ongoing need for analgesia after a median follow-up of 33 weeks.

AVN is a problem predominantly of older children and young adults. The incidence associated with a dexamethasone containing protocol in that age group is similar to that reported with prednisolone (CCG 1882, J Clin Oncol 18:3262–3272).