Background:As most children with acute lymphoblastic leukaemia (ALL) achieve long-term survival, minimising late effects of treatment is a priority. Acute lymphoblastic leukaemia survivors treated historically with protocols including cranial irradiation demonstrate increased weight gain.Methods:We retrospectively studied all 134 patients treated on the MRC/UKALL97 protocol (without cranial irradiation as standard therapy) at a single centre, with 77 inclusions. Height-, weight- and body mass index (BMI) standard-deviation scores (SDS) were recorded at diagnosis and annually until 3 years out (YO) from end of treatment (EoT); changes across time were explored using a univariate model (significance P⩽0.001 to account for multiple comparisons).Results:Whole-group height SDS was lower from 1 year into treatment until 2 YO, whereas weight- and BMI-SDS remained higher until 3 YO. In females, height-SDS was lower until EoT, but higher weight- and BMI-SDS persisted until 3 YO. In males, height-SDS was lower at EoT and at 2 YO; differences in BMI-SDS had resolved by 2 YO. By WHO criteria, more patients were overweight or obese at 3 YO than at diagnosis (P=0.01).Conclusion:Survivors of childhood ALL, particularly females, exhibit adverse changes in height-, weight- and BMI-SDS, which arise during treatment and persist into follow-up. Patients should be supported with appropriate dietary and lifestyle advice during ALL treatment and follow-up, which may minimise these changes and reduce associated long-term morbidity.

Aims Over 80% of children with acute lymphoblastic leukaemia (ALL) achieve long-term survival, thus surveillance for treatment late-effects is important. Chemotherapy schedules without cranial-irradiation are associated with body-mass-index (BMI) increases in ALL survivors which persist at 3 years from end-of-treatment (3YO), particularly in females,1 although no such UK cohort has been followed to final-height. Here, we examined annual changes in height/weight/BMI from 3YO-6YO and at final-height in our published cohort of 134 patients treated on MRC/UKALL97 protocol.1 Methods Data updated August 2011, 40 months since last analysis. Of 77 original inclusions, 27 (10M, 17F) had attained final-height; defined as age>16 years or completion of puberty. Height-, weight- and BMI- standard-deviation-scores (SDS) generated from age- and population-referenced normative data were recorded at ALL diagnosis, end-of-treatment, annually from 3YO-6YO and final-height. Changes with time from diagnosis in whole-group and gender sub-groups and between-group comparison for gender were explored using a univariate model (post-hoc analysis with Dunnett's two-sided test for multiple comparisons; p-value≤0.001 significant). Results Height-, weight- and BMI-SDS are shown in figure 1. Males had no change in weight-SDS, but reductions in height-SDS which persisted to 5YO, reflected in increased BMI-SDS to 5YO. Females had no change in height-SDS, but increased weight- and BMI-SDS at all time-points from 3YO-6YO. 27 patients at final-height were considered separately [median (range) follow-up 7 years (1-11)]. There were no differences in height in either gender subgroup. In females, but not males, persisting increases in weight-SDS (1.4 v 0.7, p<0.00001) and BMI-SDS (1.5 v 0.6, p<0.0001) were evident at final-height (figure 1). Conclusions Female survivors of childhood ALL exhibit adverse changes in height-, weight- and BMI-SDS which arise during treatment, persist into follow-up and remain elevated at final-height, predominantly due to excessive weight gain. In contrast, although males display transient increases in BMI during treatment and early follow-up, mainly due to disease/therapy on linear growth, these have resolved by final-height. Potential mechanisms for this sexually dimorphic phenomenon include increased dietary intake, altered metabolism and/or reduced exercise, and warrant further investigation. Assessment for co-morbidities, e.g. metabolic syndrome, and effectiveness of dietary/lifestyle intervention strategies should be undertaken in this group to minimise associated long-term morbidity.