Body mass index (BMI) and quality of life among long-term survivors of childhood acute lymphoblastic leukemia

Introduction

Survival rates for acute lymphoblastic leukemia (ALL) have improved in the last few decades with five-year survival of standard risk pediatric ALL now exceeding 90%¹. Adult survivors of childhood cancer are at increased risk of obesity (11–56%), metabolic syndrome, poor cardiac outcomes, and type-2 diabetes^2–10. With improved survival, greater attention has been dedicated to treatment-related late effects and quality of life (QOL).

Excess weight is known to contribute to decreased QOL in individuals with chronic diseases, including adult cancer survivors^11,12. Fewer studies have described the relationship between body mass index (BMI) and QOL in long-term pediatric cancer survivors, and in the studies that did, findings were mixed^13,14. We examined the relationship between BMI and QOL among childhood ALL survivors who returned for long-term follow-up (LTFU) after completion of therapy. We hypothesized that being overweight/obese compared to a healthy weight at the time of survey completion would be associated with poorer QOL using validated measures.

Methods

Results

A total of 90 individuals completed the survey; 83 provided QOL data and 58 were eligible for this analysis. Most were non-Hispanic white (97%) and 47% were male (Table 1). The median age was 3.9 years at ALL diagnosis (range: 1.3–19.5) and 21 years at the time of survey (range: 9–43), with a median time of 11 years since diagnosis. Participants with overweight/obesity were further from their diagnosis of ALL than healthy weight/underweight participants (median 15 (range: 5–35) vs. 9 (range: 5–32) years since diagnosis, p=0.15). All survivors were treated with standard of care ALL chemotherapy protocols, and 41% received cranial radiation. Half (51%) of the participants were overweight/obese at time of survey completion vs. 15% at time of diagnosis, with a median BMI at time of survey of 26.2 (adults; range: 19.2–47.4), and a median BMI percentile for children of 78.3 (range: 17.6–99.7). The association of overweight/obesity at time of survey with overweight/obesity at diagnosis (p = 0.05) and with receipt of cranial radiation (p = 0.06) were close to statistically significant.

Mean physical (51.5±9.0) and psychosocial (50.4±10.8) QOL scores were consistent with population normative data, suggesting that survivors are functioning well overall (Table 2). Compared to underweight/healthy weight status, obesity/overweight status was associated with significantly poorer psychosocial QOL (47.1±13.2 vs. 54.0±6.0, P=0.01, effect size Cohen’s d=0.67) and this remained statistically significant after adjusting for age and sex (P=0.03). Overweight/obesity at time of survey was associated with poorer physical QOL (compared to underweight/healthy weight status; 49.9±8.6 vs. 53.2±9.3, P=0.17, effect size Cohen’s d=0.37), however this difference was not statistically significant.

Discussion

While obesity has been recognized as a late effect of pediatric ALL therapy, the extent to which this affects long-term survivors QOL remains unknown. In our population, half of our study participants were overweight/obese at time of survey, compared with 15% at time of diagnosis. QOL scores were similar to general population means across all weight groups, as was the prevalence of overweight/obesity in this population²⁰. Overweight/obesity were associated with poorer physical QOL (unadjusted), and, when adjusted for age and sex, with poorer psychosocial QOL. Estimated effect sizes were small for psychosocial QOL and medium for psychological QOL²¹.

The large proportion of long-term ALL survivors who are overweight/obese, and the long-term increase in overweight/obesity after finishing treatment for ALL highlight a need for weight management, nutrition, and exercise interventions in survivors of childhood ALL²². Our findings suggest that such interventions may have co-benefits for survivors’ long-term QOL. Furthermore, participants with overweight/obesity were on average longer out from diagnosis, which suggests that time might be a factor in weight gain after ALL, but we could not test this given the cross-sectional nature of our study. Previous studies have examined the effectiveness of patient weight management interventions in primarily adult cancer populations; such interventions and their long-term effectiveness have been less studied for survivors of childhood cancers^23–25. Further research is necessary to evaluate the long-term feasibility, sustainability, and timing of such interventions among childhood cancer survivors, as these may have significant implications for survivors’ long-term QOL.

This study is consistent with other studies that found associations of obesity with poorer QOL but expands the data to include overweight survivors as well. However, this study has some limitations. This is a cross-sectional analysis and we could not determine the directionality of the associations we measured. Our sample size was small due to the single-site nature of the study, and participants were almost exclusively white, which may not be generalizable to the broader pediatric ALL survivor population. Underweight and healthy weight status were aggregated into one group due to small sample size, but given the understudied nature of underweight status and QOL in survivors of pediatric cancer, future studies may benefit for examining this category separately from healthy weight status.

These data demonstrate that pediatric ALL treatment carries with it a substantial burden for weight management issues well after therapy is completed, with possible implications for QOL. Weight management interventions are needed to increase awareness and to reduce weight gain in the years following treatment for childhood ALL. In addition to such interventions, future research should address the interplay between treatment for pediatric ALL and other factors, for example diet, physical activity, and sleep, and their combined contribution to unhealthy weight gain during and after ALL.

Data availability