Diet and Exercise Interventions in Pediatric Cancer Survivors and Effects on Cardiometabolic Disease Risk and Inflammaging Biomarkers: A Systematic Review.

Introduction
The incidence of childhood cancer has been steadily increasing over the past decade, with roughly 15,000 new cases diagnosed in children aged <15 y as estimated in 2025 [1]. Increased efficacy of antineoplastic treatments has also increased rates of survival. Currently, 5-y survival is >85%, increasing the estimated number of pediatric cancer survivors (PCSs) to be 420,000 individuals in the United States alone [2]. PCSs face a significant chronic disease burden compared with their nonsurvivor peers. By middle age, 99.9% of PCSs will experience chronic health conditions, with 96% experiencing severe chronic health conditions that outweigh the prevalence of community controls [3]. Early onset of diseases, such as cardiovascular disease [4] and type 2 diabetes in PCSs [5], may result in poor quality of life and increased risk for early mortality.
The onset of chronic disease at an earlier age suggests that PCSs may be experiencing accelerated cellular aging [6,7]. Atypical biological aging in PCSs may be due to chronic systemic inflammation, described as “inflammaging,” stemming from antineoplastic treatment [8] (Figure 1). Cancer treatment can induce epigenetic changes in immune cells, particularly in T-cell subsets that impact inflammation [9]. Daniel et al. [10] reported that PCSs have a higher frequency of type 1 cytokine-producing T cells and overactivation of p38 and mechanistic Target of Rapamycin Complex 1 in these cells posttreatment, resulting in elevated levels of circulating proinflammatory cytokines. Inflammaging in PCSs drives cell turnover, leading to rapid telomere attrition and increased pools of senescent cells [8], causing indiscriminate tissue damage, which ultimately predisposes this population to cardiometabolic dysfunction and chronic disease.
Lifestyle factors, including poor diet and lack of physical activity (PA), may also contribute to cardiometabolic risk factors that indirectly drive inflammaging in PCS (Figure 1). PCSs have higher rates of obesity [5,11,12], poor diet quality [[13], [14], [15], [16], [17], [18]], and physical inactivity [19] compared with community or sibling controls. Malnutrition during cancer therapy can decrease the efficiency of treatment and increase the risk of mortality in these patients [20,21]. Consequently, patients with cancer are encouraged to eat highly palatable, energy-dense foods, often high in sugar and fat that can promote obesity. Moreover, fatigue, decreased physical function, vision loss, mood disturbance, and taste changes can contribute to poor diet quality and low engagement in PA during and after treatment [[22], [23], [24], [25], [26]]. Poor diet quality and sedentary behavior may have an indirect impact on inflammaging in PCSs by fostering increased adiposity, insulin resistance, and cardiovascular disease risk markers, which drive chronic inflammation [27].
The gut microbiome may contribute to the inflammaging process [27], and antineoplastic treatment and lifestyle behaviors related to diet and exercise may adversely impact the gut microbiome of PCSs to promote chronic inflammation. Observational studies among patients with cancer have revealed a significant effect of cancer treatment, particularly chemotherapy and antibiotics, on the gut microbiota [28,29] and that these changes can persist into survivorship. Indeed, studies indicate that PCSs have lower alpha diversity and decreased relative abundance of microbial taxa such as Faecalbacterium [[30], [31], [32]]. Alpha diversity is a broad measure of gut microbial composition and has been negatively correlated with inflammation and cardiometabolic disease risk markers in population-based studies [[33], [34], [35], [36], [37]]. Low relative abundance of Faecalbacterium is associated with proinflammatory and metabolic disease, which may be due to this microorganism’s ability to successfully ferment complex carbohydrates to short-chain fatty acids [38]. Short-chain fatty acids can work locally on receptors in the lower bowel or systemically in circulation to regulate inflammatory response and enhance cardiometabolic health [[39], [40], [41], [42]].
Inflammaging in PCSs can potentially be mitigated by lifestyle changes either directly by decreasing chronic inflammation or indirectly by modifying cardiometabolic risk markers associated with chronic inflammation, such as increased adiposity, cardiovascular risk markers, insulin resistance, and gut dysbiosis (Figure 1). The goal of this systematic review was to identify diet and exercise interventions in PCSs and qualitatively summarize the impact of the interventions on outcomes related to inflammaging. Currently, there are no “gold standard” markers for “inflammaging”; therefore, relevant outcomes directly impacting inflammaging, such as immune activation, inflammatory markers, markers of accelerated biological aging, and indirect outcomes related to cardiometabolic risk markers and the gut microbiome will be explored (Supplementary Table 1).

Methods
This systematic review was conducted in accordance with the PRISMA guidelines [43]. The study protocol was registered in PROSPERO (CRD42024511586). The review team consisted of 2 primary reviewers (KC and AB) and 1 secondary reviewer (KG).
To appropriately identify diet and exercise in the PCS population, a systematic search was conducted in PubMed and CINAHL using the search and Medical Subject Heading (MESH) terms that included but were not limited to “child,” “adolescent,” “pediatric,” “neoplasm,” “cancer,” oncology,” “tumor,” “survivor,” “diet,” “nutrition therapy, “exercise,” “dietary supplements,” “risk reduction behavior,” “gastrointestinal microbiome,” “cardiometabolic,” metabolic syndrome,” “chronic disease,” “obesity,” “immune system,” “inflammation,” and “cellular senescence.” The full search strategy is provided in Supplementary File 1. Additionally, clinicaltrials.gov was searched using the same search criteria for any ongoing studies or unpublished data. The final search for all databases was conducted in January 2025, with all relevant published articles included up until that date. Inclusion criteria for the articles were as follows: 1) focused on PCSs; 2) human participants; 3) clinical trials; 4) English language; and 5) outcomes related to cellular aging, cardiometabolic disease risk markers, inflammation, or gut microbiome. The exclusion criteria were as follows: 1) review articles, abstracts, design papers, and conference proceedings; 2) animal studies; 3) participants who were still undergoing cancer-related treatment; and 4) participants with a cancer diagnosis at an age >18 y.
Deduplication and data extraction were completed using the Covidence software platform [44]. The data extraction team consisted of the 2 primary reviewers (KC and AB) with conflicts discussed and resolved collaboratively. Conflicts that could not be resolved were reviewed and resolved by the third reviewer (KG). Titles and abstracts were independently reviewed for inclusion, and full-text analysis was completed for relevant outcome data.
Due to the heterogeneity of the included data, data extraction included empirical data quantifying significant differences between groups or within groups over time. Data were reported descriptively or, with continuous data, as a change in overall measurement (increased, decreased, or stayed the same), indicating significance as it relates to the test conducted within each study. Outcomes related to inflammaging, cardiometabolic risk, and the gut microbiome were determined before data extraction by all members of the review team based on the literature pertaining to these outcomes as they relate to chronic disease risk in PCSs. A full list of the outcomes searched for and included is provided in Supplementary Table 1. Data were extracted from a study if it included any one of the outcomes listed. Additional data were extracted as they relate to date of publication, country of origin, length of intervention, number of participants, age of participants, cancer type, survivorship status, and a description of the intervention. Missing data were noted as “not applicable.”
The quality assessment was completed independently by 2 reviewers (KC and AB) using the Cochrane Risk of Bias Tool [Risk of Bias 2.0 (RoB 2.0) or Risk of Bias In Nonrandomized Studies of Interventions (ROBINS-I)] [45,46]. The following domains were used for the RoB 2.0 tool: 1) randomization, 2) deviations from the intended intervention, 3) missing data, 4) measurement of the outcome, and 5) selection of the reported result. For the ROBINS-I tool, the following domains were assessed: 1) confounding, 2) selection of participants in the study, 3) classification of interventions, 4) deviations from intended interventions, 5) missing data, 6) measurement of outcomes, and 7) selection of the reported result.

Results

Overview of search and studies meeting inclusion criteria
Our search retrieved 255 studies; 16 studies met the inclusion criteria (Figure 2). The 16 studies are described in Table 1 [[47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62]]. Two studies implemented a dietary intervention [47,48] 6 studies implemented an exercise intervention [[49], [50], [51], [52], [53], [54]], and 8 studies implemented a combined exercise and diet intervention [[55], [56], [57], [58], [59], [60], [61], [62]]. Almost half of the studies (n = 7) were single-arm studies [48,50,52,53,56,57,62] 8 were randomized controlled trials (RCTs) [47,51,54,[58], [59], [60], [61],55], and 1 study was a nonrandomized trial [49]. Fourteen studies measured weight status or body composition [47,48,[50], [51], [52], [53], [54],[55], [57], [58], [59], [60], [61], [62]] 6 studies measured cardiovascular-related risk markers [47,48,52,54,58,59] 6 studies measured markers of insulin sensitivity [47,48,50,52,54,59] 4 studies measured markers of inflammation or immune system function [47,49,48,56], and 1 study included measurements related to the gut microbiome [56]. Nine studies were conducted in the United States [47,51,53,56,[58], [59], [60], [61],[55], [63]] 5 in western Europe [48,50,52,54,62] 1 in Canada [49], and 1 in Slovakia [56].
The length of the interventions varied greatly, with 1 study employing a single time point [49]; 10 studies reported interventions of 1–9 mo in length [47,48,[50], [51], [52], [53],56,59,60,62]; 3 studies were a year [54,57,55]; 1 study was 2 y [58]; and 1 study did not report the duration and instead used the number of sessions as the goal for the intervention [61]. Participant age was also variable; 6 studies enrolled child participants 2–14 y of age [49,56,57,59,61,55]; 5 studies enrolled adolescents, 15–21 y of age, and young adults, 22–39 y of age [48,52,54,58,60]; and 5 studies included children as well as adolescents and young adults (AYA) [47,50,51,53,62]. Eight of the studies allowed survivors of any type of childhood cancer [47,48,51,54,57,[60], [61], [62]] 6 studies enrolled only survivors of acute lymphoblastic leukemia (ALL) [49,52,56,58,59,55], and 2 studies enrolled participants who had previously undergone hematopoietic stem cell transplantation (HSCT) [50,53]. The length of time in survivorship varied greatly across the studies, with some enrolling PCSs who had recently ended active treatment or were in maintenance therapy in studies that required survivorship of ≥10 y.

Diet-only interventions
Two studies were diet intervention studies [47,48]; 1 study tested daily grape juice in conjunction with a typical diet, and the other study provided personalized nutrition counseling. Blair et al. [47] conducted a crossover trial of daily grape juice supplementation (6 fluid ounces) for 4 wk among 24 AYA PCSs of any type of childhood cancer with a survivorship status of ≥3 y. The control arm consumed 6 fluid ounces of apple juice daily, which had a similar nutrient profile but lacked the flavonoids that were the nutrients of interest in the intervention arm [47]. Compliance to the intervention was not reported. Daily grape juice consumption was not associated with improvement in inflammatory markers (C-reactive protein [CRP]) or other cardiometabolic disease risk markers (BMI, blood pressure, fasting glucose and insulin, HDLs, LDLs, triglycerides, and total cholesterol) compared with the apple juice control arm [47].
Quidde et al. [48] implemented a 12-wk, single-arm intervention advocating for the Deutsche Gesellschaft fur Ernahrung [63] recommendations, which focus on a diet rich in fiber, fruits, and vegetables. Twenty-three AYA PCSs of any type of childhood cancer at any stage of survivorship were enrolled and received 4 nutrition counseling sessions (2 in person and 2 via phone) from a registered dietitian [48]. Compliance with the intervention was reported to be 70%, which was based off the retention. There was no pre–post change in cardiometabolic disease risk markers [waist-to-hip ratio, A1c (glycosylated hemoglobin), total cholesterol, LDL, HDL, or CRP] [48].

Exercise-only interventions
There were 6 interventions that focused on exercise [[50], [51], [52], [53], [54],49]. Four employed a mix of resistance and aerobic exercise [[50], [51], [52],54] 1 focused on resistance training alone [53], and 1 intervention focused on a single bout of aerobic exercise [49]. Davis et al. [50] enrolled 24 children and AYA survivors who underwent HSCT and had been in survivorship for at least a year in a single-arm 6-mo supervised resistance and aerobic exercise intervention consisting of 45-to-60-min sessions 2–3 times per week. Compliance with the intervention was 85% based on participants engaging with the exercise regimen at least 2 times per week [50]. No change in body composition or body fat percentage as measured by dual-energy X-ray absorptiometry (DEXA) was reported; however, fasting insulin and HOMA-IR were significantly decreased [50]. Jarvela et al. [52] enrolled 17 AYA ALL survivors, with a survivorship status between 11 and 22 y, in a single-arm, 16-wk, remote resistance training and aerobic exercise intervention. The resistance training consisted of muscle training sessions 3–4 times per week with 8 exercises to strengthen gluteal and lower limb muscles, shoulders and upper limb muscles, abdominal muscles, and back muscles [52]. Participants were instructed to do as many repetitions as possible for each exercise and repeat the cycle 3 times per session [52]. Additionally, there was a recommendation for aerobic exercise that included an activity of the participant’s choice, such as walking, jogging, or aerobics at least 3 times per week for 30 min per session [52]. Compliance to the intervention was not measured due to lack of self-reported data [52]. This exercise intervention also resulted in a reduction in fasting insulin, HOMA-IR, diastolic blood pressure, waist circumference, waist-to-hip ratio, and body fat percentage [52]. No changes in fasting glucose, triglycerides, LDL cholesterol, HDL cholesterol, body weight, or BMI were reported at postintervention [52].
In a larger, 12-mo RCT, Rueegg et al. [54] enrolled 151 children who survived any type of childhood cancer, with a survivorship status of at least 5 y, in a remote exercise plan of 30 min of strength training and 2 h of intensive aerobic training per week or a control condition in which participants were asked to keep their PA “constant” for 1 y. Compliance with the intervention was <50%, based on participants meeting at least two-thirds of the individual goal [54]. There was no statistically significant difference in waist circumference, HDL cholesterol, triglycerides, diastolic blood pressure, systolic blood pressure, or fasting glucose between the intervention and control arm following the intervention [54].
One nonrandomized study conducted by Ladha et al. [49] was designed to determine the impact of an acute bout of exercise on the immune system of children in the maintenance period phase for ALL compared with healthy controls. The exercise consisted of the participant running on a treadmill for 10 min, followed by 10 min of walking, repeating the sequence for a total of 3 cycles [49]. Compliance was not measured due to the design of the study [49]. The intervention resulted in a significant change in absolute neutrophil count as measured in whole blood at 4 time points (pre-exercise, postexercise, 1 h postexercise, and 2 h postexercise) in the PCS group, but no significant changes were observed between groups [49]. Ketterl et al. [53] employed a single-arm resistance training intervention in 20 AYA survivors who underwent HSCT with a survivorship status of at least 80 d [53]. Participants were given 3 sessions of in-person instruction to follow a home-based exercise routine for 12 wk [53]. The exercise regimen was tailored to each individual participant and included completing 1–2 sets of 8–10 exercises, 8–12 repetitions of each exercise, 2–3 days per week [53]. Compliance with the intervention was 89% based on participants completing the coaching calls and exercising at least twice per week [53]. No changes in fat or lean body mass as measured by DXA were reported postintervention [53].

Combined diet and exercise interventions
Eight of the studies combined diet and exercise. Friedman et al. [58] randomized 274 AYA PCSs of ALL with a survivorship status of at least 5 y to an intervention arm, including 2 y of biweekly remote coaching sessions and daily website interaction, or a control arm that received mailed brochures. The intervention was designed to promote a 5% calorie reduction with dietary pattern recommendations based on the United States Dietary Guidelines for Americans and 180 min of exercise per week [58]. Compliance with the intervention, based on the completion of coaching calls, was reported to be 15% [58]. No statistically significant changes in body weight, diastolic blood pressure, systolic blood pressure, LDL cholesterol, HDL cholesterol, or triglycerides were reported between the intervention and control arm postintervention [58]. Similarly, Huang et al. [59] randomized 38 PCSs of ALL with a survivorship status of at least 2 y to either a remote intervention focused on individual nutrition goals, including calorie reduction and dietary choices, and an exercise goal of 1 h of PA and 15,000 steps daily or control for 4 mo. Compliance with this intervention was 80% based on the amount of curriculum received by the participants [59]. No statistically significant differences were observed in body weight, BMI, A1c, fasting glucose, triglycerides, total cholesterol, diastolic blood pressure, or systolic blood pressure between the intervention arm and the control arm at postintervention [59]. Moyer-Miller et al. [55] conducted a 12-mo RCT with 14 children who survived ALL and were in maintenance therapy. The intervention consisted of nutrition education materials to promote a healthy diet pattern, daily multivitamin, and 15–20 min of PA per week vs. the control group that was asked to maintain habitual PA patterns. Compliance with the intervention was not reported [55]. There was no statistically significant change in body weight, BMI, or lean body mass as measured by peripheral quantitative computed tomography between the 2 study arms at postintervention [55].
Stern et al. [61] implemented a dyad-based RCT, where child survivors were enrolled with a caregiver. Fifty-three dyads of caregivers and PCSs of any cancer type with a survivorship status of 6 mo to 4 y were randomly assigned to a diet and exercise group or control group [61]. The intervention group received 6 sessions with the research team, either in-person group or individual sessions or telephone sessions, to promote dietary and PA change [61]. The control group received enhanced usual care consisting of a 1-h wellness session addressing the role of diet and exercise in pediatric overweight using material from the publicly available WeCan! Manual [61,64]. Compliance with the intervention was not reported [61]. Although there were no statistically significant differences between groups at the end of the intervention, a significant decrease in BMI was reported in the intervention arm from baseline to postintervention in PCSs [61]. Vanderlooven et al. [62] conducted a single-arm trial that enrolled 23 AYA PCSs with a survivorship status of at least 6 mo in an intervention that included 2 supervised exercise sessions, 1 at-home training session, and monthly exercise and nutrition coaching sessions for 4 mo [62]. Compliance with the intervention was not reported [62]. No statistically significant differences were observed following the intervention, but lean body mass as measured by bioelectrical impedance analysis significantly increased from the period of postintervention to 1-y postintervention [62].
Two studies assessed the use of dietary supplements combined with exercise [56,60]. Bielik et al. [56] enrolled 16 child PCSs of ALL with a survivorship status of 1–3 y in a single-arm study that evaluated the use of a once-daily probiotic consisting of 20 billion CFUs of Lacticaseibacillus paracasei subspecies Paracasei CNCMI-1518 with 25–45 min of remote online exercise twice weekly for 8 wk. The exercise component was designed to improve endurance and muscle strength [56]. Exercise consisted of 2–3 series of 10–15 repetitions each [56]. Compliance with the intervention was not reported [56]. There was a significant increase in gut microbiome alpha diversity (richness), whereas alpha diversity evenness decreased significantly at postintervention [56]. A statistically significant increase in the relative abundance of Lactobacillus casei as well as a nonsignificant increase in lactate-fermenting bacteria Veillonella ratti and Veillonella rogosae was observed postintervention [56]. However, there was no change in CRP, and this inflammatory marker was not correlated with the relative abundance of any microbial taxa [56]. Krull et al. [60] enrolled 67 AYA PCSs of any type of cancer with a survivorship status of >10 y to in an RCT for 24 wk. The study examined the impact of supervised resistance training 3 times weekly with 21 g of whey protein supplement or a placebo of isocaloric sucrose [60]. The exercise regimen consisted of 6 machine exercises, which included leg press, leg extension, chest press, horizontal row, and alternating vertical lateral pull and abdominals with biceps and triceps every other session [60]. Compliance with the exercise intervention ranged between 67% and 75%, and compliance with intake of the supplement was 84% [60]. There was no difference between groups for lean body mass postintervention. However, a within-arm significant increase in lean body mass as measured by DXA was observed in both study arms from baseline to postintervention [60].
Blair et al. [57] conducted a single-arm gardening study to examine the feasibility of a nonconventional diet and PA intervention among cancer survivors for 1 y. Both survivors of adult-based cancers and PCSs were enrolled, with 4 participants qualifying as PCS of any type of cancer; the mean age of these participants was 9.8 y with an unreported survivorship status [57]. Compliance with the intervention was not reported [57]. The team observed an increase in BMI in the PCS following the yearlong gardening intervention; however, these results were not statistically evaluated [57]. Among the PCSs enrolled, only 50% were considered overweight at baseline, but all 4 PCS participants (100%) were overweight by the end of the study [57].

Bias assessment
The RoB 2 was used for the 8 studies that were RCTs [47,51,54,55,[58], [59], [60], [61]] (Figure 3) [[47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62]]. Two of the 6 trials had low or moderate areas of RoB [47,59], whereas 6 studies had at least 1 area of bias deemed high risk [51,54,58,60,61,55]. The domains with at least half of the studies at high risk included “deviation from intended intervention” [54,58,60,55] and “missing data” [51,54,58,60,61]. In the “selection of the reported result” domain, all studies were deemed low risk [47,51,54,55,[58], [59], [60], [61]].
Although no gold standard exists for assessment of bias in single-arm studies, the ROBINS-I tool developed for quasi-experimental studies has been widely used in the literature to evaluate single-arm trials [65,66]. We used ROBINS-I for 8 studies; 1 study was a quasi-experimental design [49], whereas the remaining 7 studies were single-arm trials [48,50,52,53,56,57,62] (Figure 3). Due to the pilot nature of most studies that examined both diet and exercise in PCS, most had at least 1 category that was deemed high risk for bias. Most studies were deemed high risk in the “measurement of outcomes” domain [48,52,53,56,57,49] and the “bias due to confounding” domain [50,52,53,56,57,49]. The remaining domains had studies as moderate risk or low risk.

Discussion
This is the first systematic review, to our knowledge, which explores the impact of diet and/or exercise interventions on cardiometabolic disease risk markers and surrogate markers of inflammaging in survivors of a pediatric cancer. We identified and narratively analyzed the empirical evidence from 16 intervention trials. None of the trials had outcomes directly related to biological aging, and studies had varying outcomes related to systemic inflammation or cardiometabolic disease risk markers, including body fat distribution, insulin resistance, and the gut microbiome. Of the studies included, more than half resulted in null [47,48,51,[53], [54], [55],58,59] or negative outcomes [57] related to inflammaging, 6 studies positively impacted indirect outcomes related to inflammaging via cardiometabolic risk markers or the gut microbiome [50,52,56,[60], [61], [62]], whereas only 1 study resulted in positive outcomes directly related to immune function [49]. Diet interventions attempted in this population, with and without exercise, appear to be the least successful in producing outcomes related to inflammaging in PCS.
Interventions that included an exercise component were successful in modifying some cardiometabolic disease risk markers, including body weight, body fat distribution, lean muscle mass, and insulin sensitivity. Increasing PA works to increase energy expenditure and build skeletal muscle mass. Changes in body composition can lead to improvements in endocrine function [67]. Davis et al. [50], however, observed improvement in endocrine function in PCSs absent of a change in weight or body composition. Exercise, both resistance based and aerobic, can improve insulin signaling independent of body composition changes by increasing skeletal muscle contraction, which leads to greater production and activation of glucose transporter type 4, a key mechanism in glucose homeostasis [68]. Studies that utilized a supervised component for the exercise or support in the form of caregivers or trained professionals, such as exercise physiologists, physical therapists, or registered dietitians, were the most consistent in reporting favorable outcomes [50,[60], [61], [62],49]. This type of design also reported high compliance [50,60], which may have aided in the self-efficacy of the participants for long-term adaptation, which was reflected in the change of outcomes observed at follow-up vs. immediately following the intervention [62].
The majority of trials reviewed here were not designed to look specifically at inflammaging or cardiometabolic disease risk markers, but instead, these results were explored as secondary outcomes. Only 2 studies included in this analysis with null outcomes had a primary endpoint related to bodyweight that was appropriately powered [58,59]. Consequently, it can be noted that there is modest evidence to support a diet and exercise intervention that employs recommendations and counseling for calorie reduction and increased moderate to vigorous PA that lacks efficacy for weight loss. All other null outcomes reported in this article may be more appropriately interpreted as a lack of evidence due to underpowering. Additionally, because of the secondary nature of the outcomes, many studies enrolled participants with cardiometabolic or inflammatory markers that are clinically in a “healthy” range. Thus, improvements may be difficult to discern if there is no risk at baseline.
Retention and compliance rates were low among participants, which may have contributed to a lack of change in many of the primary and secondary measures. Some interventions reported compliance with the intervention as low as 15% and 48% [54,58] by the end of the trial, with 1 study indicating that at least a quarter of the participants did not engage with the intervention at all [58]. Attrition rates consistently fell below the 80% threshold for bias [69,70] undermining the results and generalizability of the results.
Cancers considered that “childhood cancers” are heterogenous in terms of age of onset, and given their heterogeneity, patients are exposed to different treatment modalities. Participants in the studies reviewed here were heterogenous in terms of age of diagnosis, length of survivorship, cancer type, and treatment modality. Inflammaging may be more common among survivors undergoing specific treatments, such as cranial or full-body radiation, or those who frequently use corticosteroids, as these treatments are known to have long-term impact on metabolic health and the immune system [71,72]. Age at diagnosis may also have varying long-term effects on inflammaging and cardiometabolic disease risk markers, such as the gut microbiome, immune system, body weight, and resting metabolic rate, due to the dynamic changes and maturation throughout development. The gut microbiome and the immune system go through a significant period of maturation starting in infancy, through puberty, and continuing into young adulthood [73,74]. Environmental assaults, such as anticancer treatment, during this period could have a more lasting and significant impact on inflammaging and cardiometabolic disease risk. Therefore, the heterogeneity of characteristics among participants may have contributed to the null results observed in the existing studies. Larger studies that can analyze participants in subsets based on cancer type and age at diagnosis could help to better understand the impact of diet and exercise interventions on the accelerated aging and health risks in PCS; however, this would require multisite participation to capture a large sample size.
This systematic review is not without limitations. First, we found only 16 studies that implemented diet and exercise interventions in PCS with outcomes related to inflammaging, cardiometabolic risk factors, or the gut microbiome. Inflammaging and chronic disease risk are significant in this population, and poor diet quality and lack of exercise may be key players, as reported in several observational studies conducted in this population [15,[75], [76], [77]]. However, few interventional studies that met our defined criteria have been attempted in PCSs. As previously stated, the interventions were heterogeneous, and few studies implemented a control group or placebo group; therefore, cautious interpretation of the results is warranted. Inflammaging and cardiometabolic risk encompass a spectrum of outcomes, such as obesity, endocrine function, cardiovascular risk, systemic inflammation, and gut microbiome diversity, composition, and function. No diet or exercise intervention has been attempted in PCSs that encompass measures related to each of these outcomes, which limits understanding of how changes in lifestyle might impact inflammaging and cardiometabolic risk to mitigate long-term chronic disease.

Future directions
Many considerations should be made when designing diet and exercise interventions to reduce inflammaging and chronic disease risk among survivors of pediatric cancer. These include more singular focus on diet-based studies, further exploration of interventions that directly impact the gut microbiome in PCSs, and considerations of the study design.

Dietary interventions
Although there is some indication that exercise may be beneficial for improving endocrine function and increasing lean body mass, as well as stimulating the immune system, adherence seems to diminish beyond 6 mo, and significant involvement of the health care team is needed for compliance [48,50]. The impact of dietary change on inflammaging in PCSs is more difficult to interpret based on the studies that have been conducted in this population. Designs that include a minor change in the form of a supplement (grape juice, probiotics, or protein powder) lacked sufficient evidence to make any discernable conclusions [47,56,60]. Most studies designed a diet-based intervention that included counseling or education for dietary pattern change or calorie reduction [48,55,58,59,61,62]; however, only 1 study [48] collected diet data in the form of food logs and quantified diet improvement via the “Healthy Eating Index—European Prospective Investigation into Cancer and Improving diet quality” [78]. Diet data may be critical to understand the impact of dietary nutrients, energy intake, or overall dietary patterns on inflammaging in PCS; thus, future studies should include diet data collection in the design of a nutrition-based intervention.
Although evidence is limited, there is potential for the efficacy of diet-based interventions with and without weight loss to be effective in terminating the cycle of inflammaging; however, the diet interventions utilizing dietary pattern change or daily calorie restriction that have been attempted thus far in PCS have been largely ineffective or without sufficient evidence to support improvement in markers of inflammaging and cardiometabolic risk and exhibit low compliance [58,59]. Additionally, calorie restriction or dietary change can be prohibitive for under-resourced participants, and cancer survivors have cited barriers to dietary pattern change due to fatigue and long-term taste changes [[22], [23], [24], [25], [26]]. Despite these challenges, targeting dietary changes in PCS, utilizing interventions that may be accessible and easily implemented, is imperative for addressing long-term chronic disease risk. One accessible intervention may be time-restricted eating (TRE). TRE is a type of intermittent fasting in which individuals limit caloric consumption to a specific time window in a 24-h period, fasting for the remainder of the circadian cycle, and has shown efficacy in improving cardiometabolic risk in the general population [79]. Unlike caloric restriction, TRE interventions often result in high compliance due to the relatively easy implementation of the behavior change [80].

The role of the gut microbiome in inflammaging and chronic disease risk in PCSs
Gut microbes have a direct impact on the immune system and may play an integral role in the inflammaging process. PCSs have lower relative abundance of bacteria that suppress inflammation in survivorship, which may be contributing to the inflammaging process [[30], [31], [32]]. Bielik et al. [56] demonstrated that exercise combined with a probiotic was successful in increasing the alpha diversity of a group of PCSs without changes in systemic inflammation. Probiotic efficacy in modifying immune system function may be impacted by habitual dietary intake [81]. Dietary components, such as macronutrients, micronutrients, and prebiotic fibers, may impact the functional capacity of the probiotic organisms [[81], [82], [83]]. Bielik et al. [56] did not measure dietary intake from the participants, which may have offered insight into the dynamic interaction of the dietary components with the probiotic.
The use of probiotics, prebiotics, and synbiotics has been shown to modify gut microbiota composition and metabolism and reduce inflammation in the healthy populations [84,85] and survivors of breast cancer [86]. The use of these types of supplements may interrupt the cycle of chronic inflammation by restoring microbial diversity within the gut and supplying metabolites that enhance gut barrier function and limit activation of the immune system, thus reducing inflammation. Additionally, microbial metabolites, such as short-chain fatty acids, activate receptors related to intestinal motility and hormone secretion [39,87], which would further promote cardiometabolic health. Future studies exploring the prevention of chronic disease risk in PCSs should consider implementing an intervention that targets the gut microbiome to curtail the inflammaging process.

Study design
Many diet and exercise interventions have been attempted in AYA survivors of PCS, but few found clinically significant changes in inflammaging and cardiometabolic disease risk markers, which may have been due to low compliance and retention. AYA PCS may be particularly vulnerable as they transition from the care of the pediatric oncology team to a health care team to support their medical needs as adults [88]. Because follow-up in general medicine may be less frequent than oncology follow-ups and differences in protocols related to managing chronic disease, the immediate needs of PCS mitigating inflammaging and chronic disease development may not be met; thus, designing a low-cost, easily accessible lifestyle intervention for PCS is warranted.
Consistent engagement by the health care team as well as in-person supervision for exercise did produce the most successful results in PCS; however, this may not be clinically feasible for the long term. Use of technology and engagement via mobile devices did result in body composition changes in adolescent participants in a diet and exercise intervention [59], and access to technology and use of online platforms for coaching and dietary counseling may reduce the burden of consistent engagement and improve compliance and efficacy. Additionally, it may be critical to involve skilled and licensed practitioners, such as registered dietitians [62] and exercise physiologists or physical therapists [53,62], in the design and implementation of the interventions.
Although inflammaging is a surrogate marker of accelerated aging, direct measures of biological aging may lend greater insight into the impact of diet and PA behavior change in chronic disease development in PCS. Measures related to telomere length, cellular senescence, or DNA methylation should be considered in future studies [89].
In conclusion, PCSs are burdened with increased risk for chronic disease at an early age due to cancer-related treatment and subsequent lifestyle behaviors. Changes in diet and exercise behaviors could improve cardiometabolic risk markers and inflammaging, and interventions that have an exercise component have suggested that aerobic and resistance training can improve insulin sensitivity, body weight, body composition, immune function, and gut microbiome in PCSs. Diet-based interventions have shown limited success in PCSs. Diet trials designed to be accessible and specifically target the gut microbiome in relation to cardiometabolic health and inflammation are warranted.

Author contributions
The authors’ responsibilities were as follows — KC: designed research; KC and AB: conducted research; KC, AB, BD, KCh and MLS: analyzed data; KC: wrote the paper; KG: had primary responsibility for final content; and all authors: read and approved the final manuscript.

Data availability
Data described in the manuscript, code book, and analytic code will not be made available because it is qualitative in nature and can be referenced in the original manuscripts analyzed.

Funding
This publication was made possible by Grant Number T32CA057699 from the National Cancer Institute (KC), Grant Number L30CA295383 from the National Cancer Institute, and the University of Illinois ChicagoApplied Health Sciences Interdisciplinary Grant (KG).

Conflict of interest
There were no competing interests reported for any of the authors listed.