1 INTRODUCTION
Advancements in biology and cancer therapy for pediatric malignancies have resulted in dramatic improvement in survival of children with cancer. Current 5-year survival rates now exceed 80%, and the number of long-term survivors residing in the United States is estimated to reach 500,000 by 2020.1,2 Unfortunately, some childhood cancer survivors are at an increased risk of early onset chronic disease, and mortality compared to peers.3
Frailty, a phenotype characterized by reduced physiologic reserve, is not typically seen in young populations. However, frailty is prevalent among 7.9% of young adult childhood cancer survivors (mean age: 33.6 ± 8.1 years).4 This percentage is comparable to that of older adults (70 years of age) in the general population, where it is reported as 9.9%.5 Frailty increases the relative risk for chronic disease by 2.2 (95% confidence interval [CI]: 1.2, 4.2) among survivors of childhood cancer, who by 30 years of age, experience an average of 7.72 (95% CI: 7.26—8.18) chronic health conditions, 2.13 (95% CI: 2.03–2.22) of which are serious or life-threatening. This burden is not seen among community controls with no childhood cancer history, until they are at least 45 years of age, where the average number of any, and serious and/or life threatening chronic conditions per individual are estimated at 7.27 (95% CI: 6.39–8.15) and 1.96 (95% CI: 1.68–2.24), respectively.4,6-8 Frailty also increases risk for mortality among childhood cancer survivors (hazard ratio [HR]:) 2.6; 95% CI: 1.2–6.2),4 whose excess risk of death, primarily due to new primary cancers, cardiovascular disease (CVD), and pulmonary conditions, remains substantially elevated 30 years from primary cancer diagnosis (standardized mortality ratio: 6.9; 95% CI: 4.7–9.8).9, 10
Although recent advances in care for aging adults and other frail populations include promising pharmaceutical and/or nutraceutical interventions with biological or cellular targets,11–14 robust health is best preserved and often restored through engagement in health optimizing behaviors, including physical activity (PA).15, 16 Regular PA improves physical fitness, a cardinal sign of physiologic reserve.17, 18 While PA and physical fitness are related, they are separate constructs with two distinct definitions. PA is defined as repetitive bodily movements produced by muscle contractions and is usually viewed has a health behavior.19 Physical fitness is defined as the ability to carry out daily tasks without undue fatigue, and is mostly viewed as a physiological adaptation.20 Engaging in regular PA and optimizing physical fitness prevents and mitigates frailty, disease, and mortality in healthy members of the general population and among adults with chronic disease, including CVD, metabolic syndrome, and asthma.21–26 Adult survivors of childhood cancer who are more active are also less likely to be frail.27
Due in part to disease and treatment exposures, children with cancer and young cancer survivors are less likely than siblings or peers to engage in PA.28, 29 In fact, young survivors of childhood cancer are less physically fit than siblings, even after accounting for PA levels.30 This suggests that there may be other factors, such as subtle organ dysfunction, as a result of treatment, that impact the way childhood cancer survivors respond to PA.30, 31 Lack of engagement in PA over time among childhood cancer survivors likely exacerbates initial compromises in physical fitness and physiologic reserve, eventually leading to increased risk of chronic disease (Figure 1).32–36
Hypothesized model of the association between physical activity, physical fitness, and chronic disease
In the general population, participation in PA increases physical fitness, including metabolic health, muscular strength, and cardiopulmonary function.37–39 Increasing PA, and consequently physical fitness, also reduces risk for chronic disease, and may be a tool that childhood cancer survivors can use to optimize health.40, 41 Therefore, research examining the effects of PA and physical fitness on chronic disease in childhood cancer survivors has recently become a topic of interest. While this topic is not novel in the general population, it is relatively new in childhood cancer survivors. Many reviews on this topic in childhood cancer survivors focus on PA or physical fitness levels as outcomes,42, 43 rather than on associations between PA and physical fitness and chronic disease. Thus, this scoping review seeks to describe the impact of PA, and physical fitness on chronic disease, and mortality in childhood cancer survivors.
2 METHODS
2.1 Search strategy
A search was conducted for studies published in PubMed, Web of Science, CINAHL, and Cochrane database of systematic reviews via OVID. The search was performed on March 28, 2020 and included the following medical subject headings and text words: “survivor” and “childhood cancer” or “survivor” and “neoplasms” and “child,” in combination with each of the following terms: “physical activity” or “exercise” or “exercise therapy” or “exercise tolerance” or “physical exertion” or “physical fitness” and “chronic disease” or “organ dysfunction” or “late effect” or “mortality.” Additionally, each identified manuscript’s reference section was searched for relevant papers.
2.2 Inclusion and exclusion criteria
Manuscripts were retained if the study included childhood cancer survivors, and if the study design was cross-sectional, case-control, retrospective or prospective cohort. The exposure must have included either PA, physical fitness, or both and the outcomes either chronic disease or markers of chronic disease. We excluded randomized control trials, studies that were not peer reviewed, and studies that were not written in English.
2.3 Screening data extraction
In total, the search yielded 846 records and 15 additional records identified through other sources. After duplicates were removed, a total of 595 records remained for screening. One reviewer (MDW) initially screened the study titles and abstracts and reviewed the full texts to determine if the studies were still eligible. During the full-text screening, reference lists of the articles were screened for additional papers. If the reviewer was unsure about a paper during any point of the screening process, it was brought to a second reviewer (KKN) to decide. Finally, a narrative synthesis was conducted to summarize the articles that were retained, relevant to the topics outlined in this scoping review (Figure 2). The results from 11 primary research studies are discussed in this review (Table 1).
Manuscript selection flowchart
The effects of physical activity or physical fitness on chronic disease in childhood cancer survivor
| Author Country | Year published | Population | Study design | N | Age | Measure of physical activity/fitness | Measure of chronic disease | Results |
|---|---|---|---|---|---|---|---|---|
| Physical activity and risk factors for, or intermediate markers of, cardiovascular disease among childhood cancer survivors | ||||||||
|
Meacham56 United States |
2010 |
Adult survivors of childhood cancer 51.5% male 85.3% white |
Cross-sectional | 8599 survivors | <19 to >50 y |
Single question from the Youth Risk Behavior Surveillance Survey Classified as sedentary if they answered “no” to the following question: “During the past month, did you participate in any physical activities or exercises, such as running, calisthenics, gold, bicycling, swimming, wheelchair basketball, or walking for exercise.” |
Cardiovascular risk factors Obesity Hypertension Dyslipidemia |
In sedentary survivors ↑Risk of obesity ↑Risk of hypertension ↑Risk of dyslipidemia ↑Risk of any three risk factors |
|
Slater55 United States |
2015 |
Young adult survivors of childhood cancer 46.4% male 85.9% white Sibling controls 53.9% male 93.3% white |
Cross-sectional |
319 survivors; 208 controls |
Survivors: mean 14.6 y Controls: mean 13.6 y |
Modifiable Activity Questionnaire for Adolescents, past year leisure time physical activity Low physical activity: self-reported ≤60 min of moderate to vigorous activity per day. High physical activity: self-reported >60 min of moderate to vigorous activity per day. |
Cardiovascular risk factors Waist circumference Percent fat mass Abdominal subcutaneous fat Visceral fat Lean body mass Triglycerides HDL-C LDL-C Blood pressure Insulin sensitivity HOMA-IR cCSC cIMT |
In low PA survivors ↑Percent body mass ↑Subcutaneous fat High PA survivors had sharper reductions in waist circumference, percent fat mass, abdominal subcutaneous fat, and abdominal visceral fat than high PA controls. |
|
Howell57 United States |
2017 |
Adult survivors of childhood acute lymphoblastic leukemia 87.6% white 52.1% male Community controls 86.7% white 50.1% male |
Prospective |
330 survivors; 331 controls |
Survivors: mean 28.9 y Controls: mean 29.2 y |
Sedentary behavior measured by triaxial accelerometry Sedentary: more than or equal to the population mean of 60% time spent sedentary Active: less than the population mean of 60% time spent sedentary |
Hypertension High cholesterol Hypertriglyceridemia Obesity Abnormal glucose Graded with the Criteria for Adverse Events v. 4.03 |
Survivors spend more time in sedentary activity compared to controls In sedentary survivors ↑Risk of onset high cholesterol ↑Risk of any cardiovascular risk factor |
| Physical activity and physical fitness reduce chronic disease risk among childhood cancer survivors | ||||||||
|
Jones59 United States |
2014 |
Adult survivors of childhood Hodgkin lymphoma 53.2% male 91.0% white |
Prospective | 1187 survivors | Median: 31.2 y |
Single question from the Youth Risk Behavior Surveillance Survey Categories of fitness: 0 MET h/week (referent) 3 MET h/week 6 MET h/week 9 MET h/week 12 MET h/week 15–21 MET h/week |
Serious cardiovascular disease graded by the Common Terminology Criteria for Adverse Events v. 4.03 | Significant inverse relationship between time spent in vigorous activity and incident cardiovascular disease |
|
Wolfe60 United States |
2012 |
Young survivors of posterior fossa tumors 85.7% male 85.7% white |
Cross-sectional | 14 survivors | Mean: 14.4 y |
Maximal cardiopulmonary exercise stress test performed on a leg ergometer Peak VO2 in ml/kg/min VO2 was multiplied by 1.11 to reflect treadmill testing |
Late effects severity score for neurological, endocrinological, and visual/auditory outcomes | Significant positive correlation between physical fitness and improved neurological outcomes |
|
Phillips61 United States |
2019 |
Survivors of childhood acute lymphoblastic leukemia 51.0% male 86.5% white Community controls 48.2% male 89.2% white |
Cross-sectional |
341 survivors 288 controls |
Survivors: median 28.5 y Controls: median 32.2 y |
Submaximal cardiopulmonary exercise test Peak VO2 converted to METs. 1 MET = VO2 of 3.5 ml/kg/min |
Neurocognitive testing | A one MET increase resulted in significant increases in verbal ability, focused attention, verbal fluency, working memory, dominate motor speed, nondominate motor speed, memory, and academics |
| Physical activity and physical fitness reduce mortality risk in childhood cancer survivors | ||||||||
|
Cox63 United States |
2013 |
Adult survivors of childhood cancer deceased 49% male 86% white Adult survivors of childhood cancer alive 51% male 89% white |
Matched case-control |
445 cases 7162 controls |
Cases: mean 30.5 Controls: mean 28.8 |
Single question from the Youth Risk Behavior Surveillance Survey Categories of activity 0 days per week 1–2 days per week 3+ days per week (referent) |
All-cause mortality Cause-specific mortality |
Increased odds of all-cause mortality in those performing less days of PA. Increased odds of cardiovascular or pulmonary morality in survivors who perform less days of PA |
|
Scott64 United States |
2018 |
Adult survivors of childhood cancer 81.0% white 52.8% male |
Prospective | 15,450 survivors | Median 25.9 y | Single question from the Youth Risk Behavior Surveillance Survey |
All-cause mortality Cause-specific mortality |
Vigorous exercise >0 MET/H significantly reduced cumulative incidence of all-cause and cause specific mortality Change in exercise habits decreased risk of all-cause mortality |
|
Ness65 United States |
2019 |
Survivors of childhood cancer 51.1% male 84.1% white Community Controls 48.8% male 90.2% white |
Prospective |
1260 survivors 285 Controls |
Mean 35.7 y in survivors exposed to anthracycline/chest radiation Mean 35.3 y in survivors not exposed Controls: 34.1 y |
Peak oxygen consumption (VO2 peak) Exercise intolerance: ≤85% sex- and age-predicted VO2 |
Organ system impairment measured with cardiac imaging; autonomic response; pulmonary function; muscle strength; neurosensory integrity Mortality measured with national death index search |
↓Fitness in all survivors compared to control group ↑All-cause mortality in those who have exercise intolerance. ↑Odds of exercise intolerance in those who had impaired global longitudinal strain; chronotropic intolerance; impaired strength; and impaired pulmonary function |
| Comparing physical fitness and physical activity as separate constructs, and examining their ability to reduce cardiovascular disease risk factors | ||||||||
|
Slater69 United States |
2015 |
Adult survivors of childhood cancers who underwent hematopoietic cell transplantation 56.3% male 91.6% white Sibling controls 54.6% male 92.4% white |
Cross-sectional |
119 survivors 66 controls |
Survivors: mean 27.4 y Controls: mean 25.1 y |
Physical activity Modifiable Activity Questionnaire Low: <2.5 h/week of moderate to vigorous physical activity High: ≥2.5 h/week of moderate to vigorous physical activity Physical fitness 6-min walk test Low ≤588.9 m High >588.9 m |
Cardiovascular risk factors Waist circumference Percent fat mass Abdominal subcutaneous fat Visceral fat Lean body mass Triglycerides HDL-C LDL-C Blood pressure Insulin Sensitivity HOMA-IR cCSC cIMT |
In low PA survivors ↑Waist circumference In low physically fit survivors ↑Waist circumference ↑Percent fat mass ↓Insulin sensitivity |
|
Lemay70 Canada |
2019 |
Adult survivors of childhood ALL 49.6% male |
Cross-sectional | 246 survivors | 22.2 y |
Physical activity Minnesota Leisure Time Physical Activity Questionnaire Low activity: <150 min of moderate to vigorous leisure-time physical activity High activity: ≥150 min of moderate to vigorous leisure-time physical activity Physical fitness Submaximal cardiopulmonary exercise test Examined in increments of 10% |
BMI Total body fat % Waist circumference Dyslipidemia Insulin resistance Metabolic syndrome Reduced ejection fraction Hypertension Cognitive health Depression Bone mineral density |
Preventative fraction of PA ↑Percentage body fat ↑Depression ↑Bone mineral density Preventative fraction of physical fitness ↑BMI ↑Percent body fat ↑Waist Circumference ↑Dyslipidemia ↑Depression |
- % = percent, y = years, ↑ = increased, ↓ = decreased HDL-C = high density lipoprotein cholesterol, LDL-C = low density lipoprotein cholesterol, PA = physical activity, HOMA-IR = homeostasis model assessment of insulin resistance, cCSC = carotid cross-sectional compliance, cIMT = carotid intima-media thickness, MET = metabolic equivalent, > = more than, MET/H = Metabolic equivalent hours.
2.4 Synthesis
We grouped the studies by the outcomes the study examined. The groups included risk factors of CVD, clinically evident CVD and neurological outcomes, mortality, and a section examining the differential effects of PA and fitness. The studies in each group were then summarized. Additional information such as study design, demographics of the population, measurements of the exposure and outcomes can be found in Table 1.44
3 RESULTS
3.1 Summary of the characteristics of the studies
We included 11 studies in our scoping review. Of the 11 studies, 8 studies focused on the associations between PA, physical fitness and either clinically evident CVD, or CVD risk factors. Thus, most of this review will be focused on CVD as the outcome. The other three studies reported on the associations between PA, physical fitness, neurological outcomes, and mortality. Two studies examined the differential impact of PA on CVD risk factors and physical fitness on CVD risk factors.
Of the 11 studies, 6 were cross-sectional, 4 studies used a prospective cohort design, and 1 study used matched case-control design. Nearly all studies were conducted in the United States, with one study being conducted in Canada. All studies were published in between 2010 and 2019, indicating a recent interest in this topic. Results from the review are described in the sections below with additional information provided in Table 1.
3.2 Physical activity, and risk factors for, or intermediate markers of, cardiovascular disease among childhood cancer survivors
Clinically evident CVD is often defined as a distant end point. Therefore, investigators examining the cardiovascular health of the general population have identified a set of markers that either indicate early damage or predict future disease (Table 2).45–47 These indices are used as surrogates or intermediate markers of disease in both cross sectional studies and intervention studies to identify persons most at risk for adverse outcomes and/or to select those who might be responsive to intervention.48–54 These indices have also been used to examine early cardiovascular health in childhood cancer survivors, discussed below.55–57
Markers of early cardiovascular disease and how they are measured
| Markers | Measured outcomes |
|---|---|
| Obesity |
Body mass index Waist circumference Percent body fat Abdominal subcutaneous fat Abdominal visceral fat Lean body mass |
| Hypertension |
Systolic blood pressure Diastolic blood pressure |
| Hyperlipidemia |
Triglycerides High density lipoprotein cholesterol Low density lipoprotein cholesterol |
| Insulin homeostasis dysfunction |
Insulin sensitivity Homeostasis model assessment of insulin resistance |
| Structural or functional heart defects |
Carotid cross-sectional distensibility Carotid cross-sectional compliance Carotid intima-media thickness |
Associations between PA, measured by questionnaires of habitual activity or with activity monitors, and CVD risk factors are reported in three key papers among childhood cancer survivors with mean ages ranging from 14 to 50 years of age.55–57 Slater et al. reported that young childhood cancer survivors who were physically active ≥60 min per day (min/day) had lower percent fat mass (24.4 ± 1.3 vs. 29.8 ± 0.9%, p < 0.01) and higher lean body mass (41.3 ± 0.7 vs. 39.5 ± 0.5 kg, p < 0.01) compared to less active (<60 min/day) survivors.55 The effects of high PA on body composition were more pronounced in cancer survivors than in controls, with greater differences in waist circumference (2.1 cm vs. 1.0 cm, pinteraction = 0.04), percent body fat (5.4 vs. 2.7%, pinteraction = 0.04), abdominal subcutaneous fat (29.8 vs. 9.2 cm3, pinteraction = 0.02), and abdominal visceral fat (4.9 vs. 0.2 cm3, pinteraction < 0.01). PA was not associated with lipid values, glucose homeostasis, blood pressure, or carotid artery measures. Fortunately, as survivors age, PA can positively impact not only body composition, like the younger survivors, but also blood pressure, lipid values, and glucose homeostasis.56 In an older cohort of childhood cancer survivors, Meacham et al. reported sedentary survivors had increased odds of obesity (odds ratio: 1.3; 95% CI: 1.1–1.5), hypertension (OR: 1.5; 95% CI: 1.2–1.8), dyslipidemia (OR: 1.3; 95% CI: 1.0–1.6), diabetes (OR: 1.7; 95% CI: 1.2–2.3), and having a combination of any three cardiovascular risk factors (OR: 1.7; 95% CI: 1.1–2.6).56 Measuring PA with accelerometers, which is more objective than self-report, confirm the previous results.57 Howell et al. reported longitudinal (mean follow-up 5.2 ± 1.5 years) associations between PA (assessed via accelerometry) and CVD risk factors among acute lymphoblastic leukemia survivors and community controls.57 Survivors completed, on average, fewer min/day of moderate to vigorous PA compared to controls (17.3, 95% CI: 15.6–19.1 vs. 21.2, 95% CI: 19.5–22.9 min/day, p < 0.01). Importantly, survivors who spent >60% per day of their time being sedentary had increased risk of developing high total cholesterol (HR: 2.52, 95% CI: 1.12–5.64), or any cardiovascular risk factor (HR: 1.96, 95% CI: 1.16–3.30) when compared to those not sedentary. These data, while observational, suggest that participating in PA can help reduce CVD risk factors in survivors, even after they transition from early to long-term survivorship.
3.3 Physical activity and physical fitness reduce chronic disease risk among childhood cancer survivors
While previous studies indicate that PA is an important tool in the management of CVD risk factors, eventual development of CVD may be unavoidable in some childhood cancer survivors because of previous exposure to cardiotoxic agents as part of their cancer therapy.58 Nevertheless, data from a prospective observational study shows that, even among individuals with these exposures, childhood cancer survivors who report more PA have lower risk for clinically evident CVD.59 In a study that included adult survivors of Hodgkin Lymphoma, Jones et al., ascertained PA by self-report, and reported participation in vigorous PA ≥9 metabolic equivalent (MET) h/week reduced risk for coronary artery disease (relative risk [RR]: 0.53; 95% CI: 0.32–0.89), valve replacement (RR: 0.36; 95% CI: 0.14–0.95), or any grade 3–5 cardiovascular event (RR: 0.49; 95% CI: 0.31–0.76) when compared to reported participation in PA < 9 MET h/week.59 In this study, duration of self-reported vigorous exercise also had a graded, inverse relationship with any serious, life threatening, disabling or fatal cardiovascular event, with risk ratios of 0.87 (95% CI: 0.56–1.34) in those who performed 3–6 MET h/week, 0.45 (95% CI: 0.26–0.80) for 9–12 MET h/week, and 0.47 (95% CI: 0.23–0.95) for 15 to 21 MET h/week, compared to those who performed 0 MET h/week (ptrend = 0.002).59
While most physical fitness and/or PA studies among childhood cancer survivors have focused on associations between PA and CVD, increasing physical fitness has potential benefit for other organ systems. While this research is well established in the general population, it is still novel in childhood cancer survivors.21 Despite the novelty, observational research indicates increasing physical fitness among childhood cancer survivors may also provide neurological benefit.60, 61 Two relevant studies report associations between cardiorespiratory fitness and neurological function in both adolescent (mean age 14.4 years) and young adult (mean age 28.5 years) childhood cancer survivors.60, 61 Among adolescent survivors of medulloblastoma, cardiorespiratory fitness was measured using a maximal cardiopulmonary exercise test (CPET) on a stationary cycle to ascertain aerobic capacity (VO2peak), and late effects (including neurologic outcomes) with a system that assigns a score from 0 (no impairment) to 2 (impairment requiring medical intervention).60, 62 Aerobic capacity was associated with lower neurological scores (partial r: –0.69, p < 0.05). In a cohort of young adult survivors of childhood cancer, a one MET higher performance on CPET was associated with better verbal ability (beta [β]: 0.10; 95% CI: 0.024–0.18; p < 0.01), focus (β: 0.096; 95% CI: 0.0045–0.19; p = 0.04), verbal fluency (β: 0.13; 95% CI: 0.053–0.20; p < 0.01), working memory (β : 0.07; 95% CI: 0.0062–0.13; p = 0.03), dominant hand motor speed (β: 0.15; 95% CI: 0.0062–0.24; p < 0.01), nondominant hand motor speed (0.15; 95% CI: 0.058–0.25; p < 0.01), visual-motor speed (β: 0.010; 95% CI: 0.031–0.16; p < 0.01), memory (β: 0.11; 95% CI: 0.035–0.18; p < 0.01), reading (β: 0.050; 95% CI: 0.0050–0.095; p = 0.03), and math (β: 0.098; 95% CI: 0.039–0.16; p < 0.01).61
3.4 Physical activity and physical fitness reduce mortality risk in childhood cancer survivors
There is observational evidence that childhood cancer survivors who are more active or more physically fit have lower risk for mortality when compared to those less active or physically fit, even among survivors with clinically evident organ dysfunction and exposure to cardiotoxic therapies.63–65 Cox et al. used a matched case-control design to examine associations between PA, all-cause and cause-specific mortality (determined via National Death Index).63 Cases (survivors who died) and controls (survivors who did not die) self-reported PA at baseline, and were matched by primary diagnosis, age at baseline questionnaire, and survival time. Length of follow-up time from baseline to death was categorized into tertiles, <5.7 years (n = 148), 5.7–8.9 years (n = 148), and > 8.9 years (n = 149). Among survivors who were <10.7 years from diagnosis, those who reported 0 days per week of PA had increased odds of all-cause mortality compared to those who reported 3+ days per week of PA (OR: 2.06, 95% CI: 1.18–3.61).63 In survivors who were 10.7–18.9 years from diagnosis, those who reported only 1–2 days per week of PA had increased odds of dying compared to survivors who reported 3+ days per week of PA (OR: 2.61, 95% CI: 1.50–4.53).63 In survivors who were >18.9 years from diagnosis who reported 0 days per week of PA had increased odds of dying compared to those who reported 3+ days per week of PA (OR: 2.13, 95% CI: 1.27–3.56).63 Finally, survivors who reported 1–2 days per week of PA had increased odds of dying from cardiac or pulmonary reasons compared to survivors who performed 3+ days per week of PA (OR: 2.07, 95% CI: 1.00–4.30).63 More recently, Scott et al. examined associations between self-reported vigorous PA and both all-cause mortality and nonaccidental causes of death among adult childhood cancer survivors.64 Survivors who reported 15–18 MET h/week of vigorous PA (compared to those who reported 0 MET h/week) had reduced risk for all-cause (RR: 0.58; 95% CI: 0.42–0.80) and nonaccidental death (RR: 0.63; 95% CI: 0.43–0.92).64 Additionally, in survivors who completed a follow-up questionnaire (n = 2279), a reported increase in PA (6 MET h/week) over 7.9 (6.4–9.4) years was associated with 13% reduction of all-cause mortality.64 While both studies indicated strong associations between PA and mortality, independent of baseline chronic disease, they relied on self-report to ascertain both PA and baseline chronic disease.
The results of the prior studies were further validated in childhood cancer survivors whose physical fitness and baseline chronic disease were measured objectively.65 The authors examined the associations between physical fitness (using CPET to ascertain VO2peak), and mortality in childhood cancer survivors. Survivors were categorized into two groups: those who were exposed to cardiotoxic therapies (chest radiation/anthracyclines), and those not exposed to cardiotoxic therapies. Cancer-free community controls were also examined. Organ system impairment was accounted for, and defined as impaired ejection fraction (ejection fraction < 53%), impaired global longitudinal strain (global longitudinal strain > 1.5 standard deviations above sex-, age-, and vendor-specific means), chronotropic incompetence (<80% of age- and sex-predicted HR reserve on the stress test), impaired pulmonary function (forced expiratory volume < 80%), impaired muscular strength (peak torque knee extension z-score < 1.5 standard deviations of control group), and impaired neurosensory integrity (modified total neuropathy score > 5).65 Both exposed (VO2peak = 25.7 ± 8.6 ml/kg/min) and nonexposed (VO2peak = 26.8 ± 8.4 ml/kg/min) survivors had significantly lower aerobic capacity compared to community controls (VO2peak = 32.7 ± 7.8, p < 0.01). After accounting for organ system dysfunction, with a median of 4 years of follow-up, the HR for National Death Index determined all-cause mortality in both exposed and nonexposed survivors was 3.93 (95% CI: 1.09–14.14) in those with exercise intolerance compared to those who were normal.65
3.5 Comparing physical fitness and physical activity as separate constructs, and examining their ability to reduce cardiovascular disease risk factors
Most studies evaluating associations between PA and chronic disease or mortality among childhood cancer survivors use PA as a surrogate for physical fitness. This is because there is observational evidence that suggest both participating in more PA and better physical fitness are associated with better health outcomes in this population.57, 59, 65 Thus, PA and physical fitness are highly correlated. However, PA and physical fitness are not equivalent. This is an important distinction, because there is some indication that childhood cancer survivors may not respond to regular PA when compared to peers. Even when survivors and peers report similar levels of PA, survivors’ performance on physical fitness measures is less robust.66 The differences between PA and physical fitness may be methodological due known biases with self-reported PA.67, 68 However, several observational studies have evaluated both PA and physical fitness in childhood cancer survivors and demonstrate that physical fitness is negatively associated with more adverse health outcomes than is PA.69, 70
The differential association between physical fitness (measured by submaximal or maximal exercise tests), PA (measured by questionnaires of habitual activity), and CVD risk factors among young adult childhood cancer survivors a mean age of 22–25 years is reported in two studies.69, 70 Slater et al. found only one outcome associated with PA: waist circumference.67 Survivors who reported ≥2.5 h/week of PA had lower values (81.9 ± 2.5 cm) compared to survivors who reported <2.5 h/week of PA (88.6 ± 3.1 cm, p = 0.009).69 However, in a subset of the sample (82 survivors, 33 controls) who completed a 6-min walk test to measure physical fitness, and who were categorized with low or high physical fitness using the median walk distance (588.9 m) as the cut point, physical fitness was associated with a greater number of outcomes. Survivors with high physical fitness had, on average, lower waist circumference (77.8 ± 2.6 vs. 87.8 ± 2.5 cm, p < 0.01), lower percent fat mass (33.6 ± 1.8% vs. 39.4 ± 1.7%, p < 0.01), and higher insulin sensitivity (10.9 ± 1.0 mg/kg/min vs. 7.4 ± 1.1 mg/kg/min, p < 0.01) compared to survivors with low physical fitness.69 Similar results are reported in a study by Lemay et al., who found that ≥150 min/week of moderate or vigorous PA was associated with three significant preventive fractions (PF): body fat >25% males/ > 35% females (PF: 0.55, 95% CI: 0.10–0.78), depression (Brief Symptom Inventory T-score ≥63, PF: 0.81, 95% CI: 0.39–0.94), and low bone mineral density (≤ –1 z-score on dual-energy x-ray absorptiometry [PF: 0.60, 95% CI: 0.20–0.80]).70 However, every 10% increase in physical fitness (measured by a maximal CPET using a leg ergometer) was associated with five fractions: BMI ≥30 kg/m2 (PF: 0.24, 95% CI: 0.11–0.36), body fat (PF: 0.22, 95% CI: 0.03–0.46), waist circumference (males: ≥102 cm/females: ≥88 cm [PF: 0.25, 95% CI: 0.08–0.38]), HDL-C (males: <1.03/females: <1.3 mmol/L [PF: 0.21, 95% CI: 0.03–0.38]), and depression (Brief Symptom Inventory T-score ≥63 [PF: 0.26, 95% CI: 0.02–0.43]).70
4 DISCUSSION
With more than 500,000 childhood cancer survivors living in the United States today,1,2 and 62.3% of them at risk for chronic disease,3 interventions to ameliorate chronic disease are of interest and relevance. Our review shows that modifiable lifestyle choices, including PA, are likely effective in improving physical fitness and reducing the risk for future organ dysfunction. Most research has been focused on improving physical fitness to CVD risk, due to its association with early mortality in childhood cancer survivors.71 However, because childhood cancer survivors are also at increased risk for early onset endocrine,72, 73 musculoskeletal,74 and severe neurological disorders,75, 76 and because PA and associated gains in physical fitness mitigates risk for these diseases in other populations,77–79 observational and interventional work is needed to determine if increasing PA and improving physical fitness decreases the risk of future chronic disease in childhood cancer survivors.
The robust literature describing biological mechanisms responsible for associations between PA, physical fitness, and CVD in noncancer populations provides some insight into why PA and/or better physical fitness may reduce, but not completely eliminate CVD risk among childhood cancer survivors.80–82 Acute and consistent bouts of moderate and/or vigorous PA influence cardiovascular homeostasis by stimulating specific cellular signaling pathways that initiate cellular adaptations of cardiac tissue, eventually resulting in beneficial or physiologic hypertrophy (Figure 3).
Key molecular and cellular cardiac outcome from habitual physical activity
During childhood, adolescence, and young adulthood, in order to respond to the changing needs of the body, cardiac tissue has robust growth and regenerative capacity.83 Cardiac hyperplasia (new cell formation) exceeds cell death, significantly influencing cardiac growth.84 Physiologic hypertrophy is normal, and typically characterized by a modest increase in ventricular volume with a coordinated growth in wall and septal thicknesses, and/or a reduction in left ventricular chamber dimension, with an increase in free wall and septal thicknesses. In childhood cancer survivors, physiologic cardiac hypertrophy is blunted.85 Survivors who were unexposed and exposed to cardiotoxic therapies (anthracyclines and chest radiation) have decreased left ventricular mass and wall thickness compared to sibling controls.85
An abnormal hypertrophic response to exercise among childhood cancer survivors may be due, in part, to abnormal molecular responses to exercise. In noncancer populations, molecular responses to exercise contribute to cardiomyocyte growth by influencing intracellular signaling pathways, gene expression, and ribonucleic acid transcription/translation, resulting in contractile protein accumulation within the cell and increased cell mass.86–91 There is also evidence that resident endogenous cardiac stem and progenitor cells (eCSC) are activated under increased cardiac workloads and contribute to cellular hypertrophy.92, 93 Unfortunately, anthracycline treatment, often included as part of therapy for childhood cancer, interferes with mechanisms of cardiovascular adaptation by increasing oxidative stress, causing premature myocyte apoptosis,94 and targeting and destroying myocyte eCSC.95 Young heart tissue in children with cancer is also sensitive to radiation induced deoxyribonucleic acid damage, resulting in activation of the p53 pathway and further inducing cellular apoptosis.96 It is possible that chemotherapy-induced dysregulation of eCSC turnover and radiation-induced apoptosis during childhood reduce the number of healthy adult myocytes, impairing adaptive growth when young survivors transition into adulthood.
Past cancer treatment also interferes with other cardiac and vascular adaptative mechanisms, including endothelial cell response. Endothelial cells line the walls of the heart chambers, coronary vessels, and peripheral vasculature and are mechanosensitive.97, 98 In healthy individuals, these cells respond to sheer forces and increase venous return by increasing production of nitric oxide, a potent vasodilator.97, 98 Vasodilation improves luminol blood flow, tissue elasticity, vascular diameter, and blood flow to all of the body’s organs.99–103 However, in childhood cancer survivors, treatment exposures reduce system wide nitric oxide reserves (causing endothelial dysfunction).95
The relationship between PA, physical fitness, and health is interdependent.79, 104, 105 As the heart and the rest of the body adapt, physical fitness increases, enabling increased frequency, duration, and intensity of PA. However, childhood cancer survivors do not appear to reap the same rewards from PA as their peers,30 and have lower physiological reserve than the cancer-free population.106–108 For survivors of childhood cancer, low physiological reserve may limit healthy, physiological adaptation from repeated bouts of PA, thus, increasing the risk of age-related chronic diseases. However, this does not mean that childhood cancer survivors should not engage in exercise. The current literature demonstrates that survivors who engage in PA, even if they have past exposure to cardiotoxic therapies, have reduced risk of adverse cardiac outcomes. More studies are needed to determine the most effective interventions (intervention timing, behavioral strategies, delivery mechanisms, exercise dose) to improve physical fitness in survivors to a level that there is an impact on chronic disease risk.
While this review gives insight into how PA and physical fitness impact chronic disease among childhood cancer survivors, there are limitations that should be discussed. First, this review only examined observational studies. This was an intentional choice given other systematic reviews have examined exercise interventions in childhood cancer survivors.109, 110 Furthermore, most of the interventions focus on improving physical fitness as an outcome, with some including early CVD risk factors. Synthesizing the observational research allowed us to examine more late effects, such as clinically evident CVD, and mortality, in an older cohort. Thus, this review does not address if changing PA or physical fitness through an intervention will influence cardiovascular outcomes or mortality among childhood cancer survivors Also, most of the studies examining the association between PA or physical fitness and chronic disease were cross-sectional, limiting an evaluation of the temporal association between PA/fitness and chronic disease. Second, we made no attempt to identify unpublished studies such as dissertations, raising the possibility that some studies were missed. Third, the majority of studies reviewed used self-reported PA.55, 59 Measurement error resulting from seasonal, response, and recall bias are common when persons are asked about their PA, impacting accuracy when used as a substitute for cardiopulmonary fitness.67, 68, 111 Fourth, when comparing the differences between PA and physical fitness and their impact on chronic disease, one of the two studies used the 6-min walk test as the measure for physical fitness. In childhood cancer survivors, the 6-min walk test has been associated with decreased muscular strength and neuropathy.112, 113 Therefore, the 6-min walk test may predict overall physical function rather than purely cardiopulmonary fitness. Fortunately, the 6-min walk test is valid when predicting childhood cancer survivors’ VO2.114 Finally, only 11 studies were discussed in this review. Thus, the association we describe between PA, physical fitness and chronic disease is preliminary. The conclusions may change as more research is done. Systematic reviews and meta-analysis will be needed as future studies are published on this topic.
Childhood cancer survivors who perform PA or who are physically fit have improved markers of CVD, decreased risk for clinically evident CVD, and mortality compared to less active/fit survivors. The literature indicates that physical fitness is more strongly associated with future health than self-reported or measured PA. Unfortunately, childhood cancer survivors do not appear to benefit as much as their peers from regular PA. Cytotoxic cancer therapies are known to impact organ function, reducing physiologic reserve and associated exercise responses. Thus, research is needed to determine if and which specific exercise interventions (with documented duration, intensity, frequency, and timing) best perturb cardiovascular adaptation and maximize health benefits in childhood cancer survivors. Mechanistic work is also needed to better understand why survivors respond differently than their peers to exercise interventions and to identify supplement or pharmaceutical mechanisms that might enhance survivors exercise response. Nevertheless, while the research community awaits new insights into how PA and physical fitness play a role in chronic disease risk, inactive childhood cancer survivors should be encouraged to be more physically active, and be referred to exercise specialists with expertise adapting programs to address the unique physiological characteristics of this population.
ACKNOWLEDGMENTS
Financial support for this work provided by a Cancer Center Core Grant, P30CA021765 Charles Roberts, from the National Cancer Institute and by the National Institute of Health. The authors would like to acknowledge Tracie Gatewood for her assistance with article preparation.
AUTHOR CONTRIBUTIONS
Conceptualization: M.D.W. and K.K.N. Methodology: M.D.W. and K.K.N. Investigation: M.D.W. Resources: M.D.W., K.K.N, and C.G.G. Drafting of the manuscript: M.D.W., K.K.N., and C.G.G. Critical revision of the manuscript for important intellectual content: M.D.W., K.K.N., C.G.G., E.R.F., and R.E.P. Visualization: M.D.W. and C.G.G. Supervision: K.K.N.
CONFLICT OF INTEREST
The authors have stated explicitly that there are no conflicts of interest in connection with this article.
