Obesity in Childhood Cancer Survivors: Call for Early Weight Management^,,

Abstract

A high prevalence of obesity and cardiometabolic conditions has been increasingly recognized in childhood cancer survivors. In particular, survivors of pediatric acute lymphoblastic leukemia have been found to be at risk of becoming overweight or obese early in treatment, with increases in weight maintained throughout treatment and beyond. Nutrition plays an important role in the etiology of obesity and cardiometabolic conditions and is among the few modifiable factors that can prevent or delay the early onset of these chronic conditions. However, nutritional intake in childhood cancer survivors has not been adequately examined and the evidence is built on data from small cohorts of survivors. In addition, the long-term impact of cancer diagnosis and treatment on survivors’ nutritional intake as well as how survivors’ nutritional intake is associated with chronic health conditions have not been well quantified in large-scale studies. Promoting family-based healthy lifestyles, preferably at a sensitive window of unhealthy weight gain, is a priority for preventing the early onset of obesity and cardiometabolic conditions in childhood cancer survivors.

Introduction

Dramatic improvements in the diagnosis and treatment of cancer in childhood have led to a rapidly growing cohort of survivors, now estimated to exceed 420,000 in the United States (1, 2). However, this success has brought the recognition that childhood cancer survivors have an elevated risk of premature mortality and serious morbidity (3–6). Data from the St. Jude’s Lifetime Cohort of childhood cancer survivors revealed that, at 45 y of age, the estimated prevalence was 95% for having 1 chronic condition and 80% for having a severe, disabling, or life-threatening condition (3). Notably, childhood cancer survivors are 7 times more likely to die from cardiac causes than the general population (4, 6). They experience a significantly high risk of developing cardiovascular disease (CVD)¹⁰ (7–9) and related factors such as hypertension (10, 11), dyslipidemia (10, 11), insulin resistance or diabetes (10–13), and obesity (14, 15) at a young age. Chronic conditions diagnosed in young adulthood are strong predictors for subsequent mortality (2, 4, 6). Obesity adds additional risks to the elevated chronic health conditions already experienced by childhood cancer survivors. Identifying risk factors for obesity and obesity-related chronic conditions is a major priority for improving the long-term health of childhood cancer survivors. We provide a critical review of the current state of research and knowledge of obesity and associated risk factors in survivors of childhood cancer.

Current Status of Knowledge

Prevalence of obesity in childhood cancer survivors

Although malnutrition due to cancer-related anorexia and cachexia still represents an important concern in cancer care for survivors of childhood cancer (16), overconsumption of food calories and reduced energy expenditure due to physical inactivity are also apparent. The Childhood Cancer Survivor Study (CCSS), a consortium of 26 pediatric oncology centers in the United States and Canada, assessed the BMI of 1451 adult survivors of acute lymphoblastic leukemia (ALL), the most commonly diagnosed cancer in children and adolescents, and of 2167 same-sex siblings between 1995–1996 and 2002–2003 (17). Findings of this longitudinal study revealed that the prevalence of obesity [defined as BMI (in kg/m²) ≥30] increased in both the survivors and their siblings, but the survivors experienced a greater increase in obesity rate than their siblings (13.5% vs. 7.5%). It should be noted that childhood cancer survivors enrolled in this original cohort of the CCSS were mostly young adults (mean age: 32.3 y) and were long-term survivors (mean interval from diagnosis: 25.1 y) (17).

Increased obesity rates in comparison to the general population have also been reported in many other studies, and these studies tend to vary appreciably by length of survivorship and treatment protocols. For survivors of childhood cancer aged <20 y, the BMI z score or percentile is often used to evaluate weight status, rather than the absolute BMI, because an increased BMI is part of normative/adolescent development and also varies by sex (18). The BMI z score or percentile can be calculated on the basis of age- and sex-specific mean BMI of a reference population, such as the 2000 CDC growth charts for children aged 2–20 y (19). The BMI z score indicates the number of SDs the measurement is away from the age- and sex-specific mean in the reference population, whereas the BMI percentile indicates the position of a child’s BMI relative to children of the same sex and age in the reference population. A BMI z score >0 indicates a higher-than-average BMI as would a BMI >50th percentile. In our recent meta-analysis, a synthesis of 20 studies over the past 35 y, we identified a mean BMI z score of 0.83 in 1743 survivors of childhood ALL. This summary BMI z score suggests that young survivors, mostly preteenagers and adolescents who were off-treatment <10 y, had a substantially higher BMI than the reference population. Importantly, obesity is prevalent in young pediatric ALL survivors regardless of receipt of cranial radiation therapy (CRT), sex, and age at diagnosis (15).

Early onset of obesity in childhood cancer survivors

Accumulating evidence suggests that obesity occurs early during survivorship. Recent data from the CCSS suggest that 24-y-old childhood cancer survivors have the same cumulative incidence of severe or disabling, life-threatening, or fatal health conditions as 50-y-old siblings (5). Several longitudinal studies have examined weight patterns in patients with childhood ALL at key time points during and after treatment, such as at diagnosis, at the end of induction, at the end of consolidation, during maintenance, and years after treatment completion. These studies consistently identified a significant increase in the rate of overweight (BMI: 85–94.9th percentile) or obesity (BMI ≥95th percentile) during treatment. For example, in a retrospective cohort of 83 patients with childhood ALL diagnosed between 1985 and 2010, our research team found the percentage of being overweight or obese increased from 20% at diagnosis to ∼40% at the end of treatment (Figure 1) (20). In 183 patients with childhood ALL, Esbenshade et al. (21) similarly reported an increase in the overweight/obesity rate from 36% at diagnosis to 49% at the end of treatment; and the study by Withycombe et al. (22) in 1017 patients with high-risk ALL from the Children’s Oncology Group (COG) found an increase in the overweight/obesity rate from 27% at diagnosis to 42% at the end of treatment. The most recent data from 269 standard-risk patients with ALL treated without CRT in COG also detected an increase in overweight/obesity rate from 14% at diagnosis to 39% at the end of therapy (23). The prevalence of overweight/obesity in pediatric patients with ALL at the end of treatment (39–40%) appears to be higher than the prevalence in the general population, which ranged between 33% and 34% in 6395 children and adolescents aged 6–19 y in the 2003–2006 NHANES (24). And these studies all observed a rapid weight gain during the induction and early maintenance cycles of the treatment (20, 21, 23, 25).

FIGURE 1.

Changes in prevalence of overweight and obesity during and after treatment in childhood ALL survivors. Values are prevalences (%), n = 83. Reproduced from reference 20 with permission. © 2014 Wiley Periodicals, Inc. ALL, acute lymphoblastic leukemia; Consol, consolidation; Induct, induction.

To further describe the trajectory of weight changes during treatment, we recently completed a meta-analysis from 21 studies that assessed the longitudinal trend of BMI z score (26). The results supported the notion that pediatric patients with ALL gained substantial weight during early treatment (i.e., from diagnosis to start of maintenance) and again during maintenance. Overall, the mean increase in BMI z score during treatment was 0.8 in 1514 pediatric patients with ALL (Figure 2). To illustrate this, consider a 5-y-old boy with an average BMI (i.e., BMI z score = 0 or BMI = 50th percentile) at diagnosis; by the end of treatment at age 7–8 y, this increase corresponds to 4 pounds (1.8 kg) of extra weight gain, compared with his peers, assuming normal growth in height. These findings provide strong support that the treatment phase is a sensitive window for unhealthy weight gain and potentially obesity-related cardiometabolic abnormalities in pediatric patients with ALL.

FIGURE 2.

Changes in BMI z score from diagnosis in childhood ALL survivors: a meta-analysis. Values are means (95% CIs), n = 1514. Reproduced from reference 26 with permission. ALL, acute lymphoblastic leukemia; tx, treatment. © 2015 Wiley Periodicals, Inc.

Importantly, in our meta-analysis, pediatric patients with ALL did not return to their pretreatment weight and were consistently more overweight or obese than their noncancer peers beyond treatment completion. From diagnosis to several years after treatment completion, the mean increase in BMI z score remained high, ranging from 0.5 to 0.9 (Figure 2) (26). Although studies with a longer length of follow-up are required to further evaluate the obesity rate in long-term survivors, the existing literature suggests that the unhealthy weight gain that occurred early in treatment is unlikely to be reversed after children complete cancer treatment (26).

Predictors for obesity in childhood cancer survivors

Weight gain is generally considered as a consequence of positive energy balance (i.e., amounts of energy intake exceed amounts of energy expenditure). There is a growing body of evidence suggesting that childhood cancer survivors have high energy intakes due to overconsumption of foods with high energy density and corresponding low expenditure due to sedentary lifestyles.

Nutritional intake.

The nutritional intake in childhood cancer survivors has not been adequately examined, and the existing evidence has mostly featured small cohorts of survivors. Some studies examined the adherence of survivors’ diet to the current dietary guidelines (27–31). Although different diet quality indices and assessment methods were used to capture dietary intake, existing studies all reported poor diet quality in childhood cancer survivors (27–29, 32, 33). The mean diet quality score was either less than half (28) or only slightly above half (27, 29, 32, 33) of the maximum score, ranging from 33% to 56%. For example, we found that the mean Healthy Eating Index (HEI)–2010 score was 52.7 in 22 childhood leukemia or lymphoma survivors (33), a value that is similar to the mean HEI-2005 of 55.5, which was previously reported in 91 childhood cancer survivors (27). The HEI measures the adherence to the Dietary Guidelines for Americans, with a maximum score of 100 (34). A higher HEI indicates a better diet quality. Overall, childhood cancer survivors showed a low percentage of consuming ≥5 servings of vegetables and fruits each day (27, 29, 32, 35). In particular, survivors had a low adherence to consumption of green vegetables (27), and the actual intake relative to the recommended intake was <50% (28, 29). Survivors also were found to consume fewer whole grains (30) and dietary fiber (28, 29, 33), but they derive a higher percentage of calories from fat (28, 30, 35), saturated fat (27, 28), and refined carbohydrates (35) compared with the recommended intake. One study reported that childhood cancer survivors consumed 10% more total calories than estimated energy requirements (36). On the other hand, childhood cancer survivors had a good adherence to the recommended intakes of dairy or milk (27, 28) and total-protein foods or meat and beans (27, 28), although the adherence to seafood and plant protein recommendations remained poor (33).

Few studies have assessed the dietary intake of a broad category of micronutrients and minerals in childhood cancer survivors. One obstacle for studying nutritional intake is the lack of a valid method to quantify habitual intake. Two common methods used to assess diet are FFQs and repeated 24-h dietary recalls (37). FFQs are designed to measure a person’s typical diet over a defined period of time, and 24-h dietary recalls involve a recall of food intake in the preceding 24 h (37). The use of FFQs has been associated with substantial underreporting of dietary intake and therefore is not appropriate to estimate absolute intakes of nutrients and minerals (38, 39). Repeated 24-h dietary recalls, on the other hand, provide more valid estimates and are often used as the reference method to validate FFQs (40). Using repeated 24-h dietary recalls over 12 mo, we observed a low intake of vitamin D and calcium in 22 childhood cancer survivors (33). Only 4% of the survivors met the recommended intake for vitamin D, whereas 29% met the recommended intake for calcium (33). Cohen et al. (36), using 3-d food records to assess dietary intake in 50 childhood cancer survivors, found that 32% of the survivors did not meet the recommended intake for calcium. A high percentage of childhood cancer survivors were found not to meet the recommended intake for folate and iron (50% and 44%, respectively) (36). We also found a very low adherence to recommended potassium intake, which may be explained by low consumptions of vegetables and fruits (33). In contrast, existing studies consistently reported a high sodium intake in childhood cancer (27, 28, 32, 33), with a mean sodium intake >3000 mg/d, although the upper limit is 2300 mg/d in the 2010 Dietary Guidelines (34).

Little is known about whether childhood cancer survivors’ dietary intake differs from that of the general population. The only study that compared dietary intake of childhood cancer survivors with that of siblings did not find that the HEI-2005 and its 12 components differed significantly between the 91 survivors and 30 siblings (27). Poor nutritional intake can significantly exacerbate chronic health conditions experienced by the survivors, who are at higher risk of developing these conditions at a much younger age than the general population (5).

Physical activity.

A substantial number of studies assessed physical activity level in childhood cancer survivors, including several publications from the CCSS in which self-reported physical activity was assessed by using adapted questions from the Behavior Risk Factor Surveillance System (41–49). Two CCSS studies reported that survivors were 60–70% more likely to be physically inactive than their siblings (47) or healthy controls (44). Overall, various reports from the CCSS found that ∼50–70% of survivors of childhood cancer did not meet the CDC guidelines for physical activity, although definitions of meeting the physical activity guidelines varied across studies. Other studies involving other patient cohorts have shown similar patterns of reduced physical activity in childhood cancer survivors (32, 50, 51). When accelerometers were used to objectively measure levels of physical activity in 1 study, <5% of childhood cancer survivors met the CDC guidelines for physical activity (51). Studies that assessed levels of physical activity in adolescent (11–18 y) and younger adult (19–25 y) survivors of childhood cancer on the basis of self-report (35, 52–64) also observed that a substantial proportion of the adolescent and younger adult survivors, ranging between 50% and 80%, did not adhere to CDC guidelines for physical activity (35, 52, 58, 61, 63, 64). Adolescents experience the steepest decline in physical activity across all age groups (65). However, because methods used to assess physical activity vary substantially in existing studies, it cannot be directly determined whether the poor adherence to physical activity guidelines in adolescent survivors is similar to that reported for the general adolescent population, such as in NHANES (66, 67).

Studies that compared levels of physical activity in childhood cancer survivors with those without cancer reported varying results. Although some reports suggested that survivors are less active (56, 62, 68), others reported the opposite or no difference (60, 69). However, when the longitudinal trend of physical activity levels was evaluated, a significant decline in physical activity during treatment was reported in 97 adolescent survivors (age range: 15–20 y) who were asked to recall levels of physical activity during and after treatment. The proportion who were physically inactive increased from 26.4% at diagnosis to 84.5% at the end of treatment and persisted beyond treatment completion (54, 55).

A few studies provided empirical evidence on the levels of total energy expenditure in childhood cancer survivors. Our study (70) and the study by Reilly et al. (71) assessed total energy expenditure in childhood cancer survivors using the doubly labeled water method. The doubly labeled water method provides an accurate and objective assessment on energy expenditure in free-living individuals and is considered as the reference method to measure total energy expenditure (72, 73). In 17 childhood cancer survivors (median age: 11.5 y), we found a mean total energy expenditure of 2173 kcal/d (70), similar to the mean energy expenditure of 2150 kcal/d in 20 childhood cancer survivors (mean age: 10.9 y) (71). Importantly, childhood cancer survivors have an energy deficit of nearly 500 kcal/d compared with the estimated energy requirement for the recommended level of physical activity (70). This energy gap could be partially addressed by promoting an average of 60 min/d of moderate-to-vigorous physical activity such as brisk walking and swimming (∼250–400 kcal/d). However, physical activity alone may be insufficient to reverse an energy gap of 500 kcal/d. Dietary modification to facilitate a reduction in energy intake may also be needed to achieve energy balance in childhood cancer survivors.

Mechanisms for obesity and CVD risk in childhood cancer survivors

Although poor dietary intake and sedentary behaviors can be fueled by the “obesogenic” environment to which the entire society is exposed, childhood cancer survivors have a unique factor that may put them at a higher risk of obesity than their peers—that is, the cancer treatment received at a very young age. Cancer and its treatment can affect obesity and CVD risk through complex pathways (Figure 3). Several regimens for treating hematologic malignancies include corticosteroids. Corticosteroids are known to be critically involved in regulating energy intake, storage, and mobilization. The prolonged use of corticosteroids has shown effects on body composition associated with increases in percentage of body fat in pediatric ALL survivors (74). Two studies in patients with ALL receiving maintenance therapy reported a significant increase in caloric intake while patients were receiving corticosteroids (75, 76). Results from animal studies also provide compelling evidence that passive elevation in glucocorticoids provokes a dose-response increase in energy intake—in particular, a proportional increase in the intake of calories from high-energy-dense foods (77). Although patterns of dietary intake were originally thought to be acute responses to cancer treatment, recent evidence indicates that children had difficulty reversing unhealthy eating habits after treatment completion (78), and a high proportion of childhood cancer survivors report taste preferences that interfere with adherence to a low-fat diet (79).

FIGURE 3.

A conceptual model of obesity and CVD risk in childhood cancer survivors. CVD, cardiovascular disease.

CRT is a common treatment for brain tumors and is used in selected leukemia regimens to prevent central nervous system relapse. CRT can directly damage the hypothalamic-pituitary region, impairing signaling reception from hormones such as ghrelin and leptin that regulate hunger and appetite (14, 80–82). These findings suggest that cancer treatment may have a long-lasting impact on survivors’ diet. Nevertheless, whether survivors’ nutritional intake patterns are associated with patient and treatment characteristics has not been well quantified. We previously found that that long-term survivors (time from diagnosis: ≥10 y) had a lower diet quality than more recent survivors (time from diagnosis: <5 y) (33), and another study (27) found that childhood cancer survivors treated with CRT had a lower diet quality than those treated without CRT. However, both studies were based on small cohorts of survivors. Large-scale studies with systematic assessments on patient and treatment characteristics are required to further quantify the impact of cancer diagnosis and treatment on nutritional intake in childhood cancer survivors.

Cancer treatment can also directly affect the levels of physical activity in childhood cancer survivors. Receipt of CRT may decrease muscle mass and strength and impair balance and postural control (44). Anthracyclines and thoracic radiation, used in treating selected youth with cancer, have been known to cause left ventricular dysfunction and subsequent impairment in cardiovascular fitness (8). Vincristine, used in all phases of ALL treatment, is an established risk factor for neuropathy, which often results in reduced muscle strength, impaired balance, and other mobility limitations. These physical impairments can lead to reduced physical activity in childhood cancer survivors (58). In addition to contributing to obesity risk through causing positive energy balance, some chemotherapeutic drugs such as l-asparaginase (83–85) and prednisone (86) can directly impair β cell function and lipid synthesis, leading to hyperinsulinemia and dyslipidemia. Obesity and cardiometabolic conditions can further contribute to serious morbidity and premature mortality in childhood cancer survivors.

Family environment also plays an important role in shaping children’s dietary and activity behaviors (87). Family factors and parental control can be particularly important for children diagnosed with cancer at a young age. In response to the impact of the disease and treatment complexity, parents may be permissive on issues such as diet and physical activity while the child is undergoing cancer treatment. After the completion of treatment, parents find difficulties in reversing the unhealthy eating habits and sedentary lifestyle that have been established during cancer treatment (78). Taken together, treatment-induced positive energy balance and metabolic abnormalities, coupled with negative behavioral adaptions that occur early in treatment, tend to persist beyond treatment completion and may cause early onset of obesity and CVD risk in childhood cancer survivors (Figure 3).

Lifestyles and chronic health conditions in childhood cancer survivors

Many established CVD risk factors in the general population are preventable or modifiable (88). Nutrition and physical activity play important roles in the etiology of chronic health conditions and are among the few modifiable behaviors that can prevent or delay the early onset of chronic diseases. Although intervention studies have provided strong evidence that behavioral modifications reduce CVD risk factors and events in the general population, few studies have examined nutrition, physical activity, and chronic health conditions in childhood cancer survivors. Studies that evaluated nutritional intake in association with obesity and cardiometabolic conditions reported inconsistent findings (27–30). Among the few studies that assessed physical activity with chronic health conditions, 1 large cross-sectional study from the CCSS reported a 10% increased prevalence of obesity in association with low levels of vigorous physical activity (<30 min/d for at least 3 d/wk) (45), and 2 studies reported that low levels of physical activity were associated with low bone mineral density (60, 62).

Conclusions

Childhood cancer survivors have a significantly elevated risk of obesity and cardiometabolic conditions. Poor dietary intake patterns and low levels of physical activity can both contribute to obesity and CVD risk in this population. Although treatment exposures alone or in combination also contribute to elevated cardiometabolic risk in childhood cancer survivors, it is important to note that the attributable fraction identified in 1713 adult survivors of childhood cancer was <50%, ranging from 9.3% for hypertension to 15.5% for dyslipidemia, 41.7% for diabetes, and 42.1% for obesity (3). This leaves ample room for improvement through behavioral interventions. Therefore, it is critical to further improve our understanding of the lifestyle behaviors in childhood cancer survivors, especially for dietary intake patterns for which few data from large-scale cohorts exist. Although dietary guidelines have been developed for cancer survivors, such as those developed by the American Cancer Society (89) and the World Cancer Research Fund/American Institute for Cancer Research (90), these guidelines focus on survivors with malignancies other than childhood cancer. The long-term follow-up guidelines for childhood cancer survivors, developed by the COG, do not have a specific focus on nutrition (91). The lack of specific dietary guidelines for childhood cancer survivors may be explained by the fact that the nutritional intake and its associations with chronic health conditions are still poorly understood in this population. Improvement of our knowledge of nutritional intake in childhood cancer survivors is critical to developing evidence-based dietary recommendations for this population. In addition, little knowledge exists for the association between diet, physical activity, and chronic health conditions in childhood cancer survivors and, in particular, to determine whether survivors who had specific treatment exposure experience differential associations for nutrition and chronic health outcomes compared with those who are not at risk. Existing studies have focused primarily on survivors of pediatric ALL; and we know very little about the impact of other cancer diagnoses, age at diagnosis, and time from diagnosis on lifestyle and chronic health conditions in childhood cancer survivors. Large-scale studies with systematic assessments on patient and treatment characteristics are required to further quantify the impact of nutritional intake by cancer- and treatment-related characteristics so that targeted recommendations can be made to improve diet quality in this population.

Cancer diagnosis may serve as a “teachable moment” and trigger positive lifestyle changes in the survivors. Few interventions are designed to promote lifestyle modifications in this population, in particular in young survivors who are at a sensitive window of unhealthy weight gain and development of modifiable CVD risk factors. Many programs for cancer treatment and survivorship care do not include weight management. This disparity has been recognized by the Institute of Medicine in its 2013 workshop summary outlining recommendations for improving the care and outcomes for adolescent children with cancer (92). In particular, an increasing body of evidence now suggests that unhealthy weight gain and the development of CVD risk factors occur early in treatment and persist beyond treatment completion. If the intervention is initiated late in survivorship (78), reversing the unhealthy eating habits and sedentary lifestyle established during treatment may become difficult.

Family environment plays a highly influential role in shaping children’s diet and activity behaviors (87), in particular for younger children (3–11 y old) given their limited autonomy and dependence on adult caregivers. For families with children who have survived cancer, family environment can be even more important because a close parent-child relationship is often expected (52). Previous research provides convincing evidence that interventions targeting parents exclusively or targeting parents and children together yield greater success in preventing childhood obesity than those targeting children alone (93–95). Therefore, weight-management programs in young childhood cancer survivors need to engage parents as the “agents of change,” emphasizing parents’ roles in transitioning the family toward healthy eating and exercise through modifying both physical and social family environment, and help parents resume a more authoritative role with parenting after a child’s diagnosis and treatment of cancer. Promoting family-based healthy lifestyles, preferably at a sensitive window of unhealthy weight gain, is becoming a priority for preventing the early onset of CVD risk factors in childhood cancer survivors.