Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Hyperglycemia During Childhood Cancer Therapy: Incidence, Implications, and Impact on Outcomes

Hyperglycemia During Childhood Cancer Therapy: Incidence, Implications, and Impact on Outcomes Hyperglycemia is a known complication of therapies used in the treatment of childhood cancer, particularly glucocorticoids and asparaginase. It has been linked to increased infection and reduced survival. With more limited data on hyperglycemia during childhood cancer treatment compared with adult cancer, impact on outcomes is less clear in this population. As additional glycemic-altering cancer agents including immune checkpoint inhibitors and targeted therapies make their way into pediatric cancer treatment, there is a more pressing need to better understand the mechanisms, risk factors, and adverse effects of hyperglycemia on the child with cancer. Thus, we utilized a systematic approach to review the current understand- ing of the incidence, implications, and outcomes of hyperglycemia during childhood cancer therapy. Published literature regarding glycemic patterns and Background evidence-based management of hyperglycemia during therapy Childhood cancer therapy is frequently complicated by hyper- for childhood cancer is lacking, even for ALL therapy, during glycemia, with prevalence ranging from 10% to 20%. This has which risk for hyperglycemia is well established. Very little is been most readily observed during treatment for acute lympho- known about hyperglycemia during treatment of other types of blastic leukemia (ALL) (1–3) and is most often associated with cancers, and data are lacking to guide identification of at-risk glycemic-altering chemotherapeutic and supportive agents, populations. Published findings are limited in their application particularly glucocorticoids (GCs) and asparaginase. Early recog- due to variability in blood glucose (BG) measurements (fasting nition and safe management of hyperglycemia are important in status, postprandial timing, single vs repeated episodes) and children to prevent acute complications including ketoacidosis definitions of hyperglycemia. Standard guidelines for screening, and hyperosmolar nonketotic coma (4). In addition, hyperglyce- diagnosing, and managing hyperglycemia during childhood mia during childhood cancer therapy has been associated with cancer therapy are needed. increased infections and lower survival (5), a finding replicated among adult cancer patients (6). Among patients undergoing treatment for ALL, increased Methods rates of hyperglycemia have been seen among older patients (10 years) and those with Down syndrome and CNS involve- Three electronic databases (Ovid, PubMed, and Medline) were ment (2), with mixed data on the association with obesity (2,7–9). searched using the following search terms: hyperglycemia, can- Treatment-related hyperglycemia is also described in chil- cer, childhood, management, and treatment. The terms were dren with solid tumors, occurring in 13% of children receiving used both as keywords and as MeSH terms to optimize the antiemetic dexamethasone during premetrexed therapy in a number of articles included. Reference lists were also searched Children’s Oncology Group Phase 2 trial (10). This is similar to for applicable publications. Inclusion criteria were manuscripts hyperglycemia rates reported in adult cancer patients receiving that reported findings in English, that were not single case antiemetic dexamethasone (up to 20%) (11). reports or isolated abstracts, and that were published between Received: February 28, 2019; Revised: June 11, 2019; Accepted: July 1, 2019 Published by Oxford University Press 2019. This work is written by US Government employees and is in the public domain in the US. 132 Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 A. Grimes et al. |133 level of greater than 126 mg/dL (13,14). Observed stress hyper- Records idenfied through Addional records glycemia frequently occurs in children with severe illness and/ database searching idenfied through other (n = 231) sources or steroid use (15). There is controversy regarding data and (n = 76) observations that guide pragmatic institutional approaches to hyperglycemia management to safely avoid adverse outcomes of suspected stress hyperglycemia in children. The lack of con- Records aer duplicates removed sensus to consistently define hyperglycemia across studies is (n = 222) not limited to childhood cancer literature. Of the 44 studies in 85 Duplicates this review, 29 included a definition of hyperglycemia with wide variability in this definition across studies, as outlined in Table 1. The most frequent definition used required a recorded Records screened Records excluded for non- BG value of at least 200 mg/dL on at least 2 occasions without re- (n = 222) relevance gard to fasting status (seen in 8 of 29 articles) followed by the oc- (n = 114) currence of any random single BG value of at least 200 mg/dL (5 of 29 articles). Some studies relied on standard toxicity grading Full-text arcles excluded, criteria and others accepted written documentation of a diagno- Full-text arcles assessed with reasons for inclusion sis of hyperglycemia or diagnostic code. Although most defini- (n = 64) (n = 108) tions relied on retrospective BG measurements, some evaluated 7 Case Reports response to oral glucose tolerance testing or measurements of 49 Non-pediatric insulin resistance. 2 Open study/no data published Studies included in 6 Late effects only qualitave review/synthesis Hyperglycemia Complications and Impact on (n = 44) Outcomes Findings regarding the impact of hyperglycemia on outcomes, Studies included in including complications, event free survival (EFS), and overall qualitave review/analysis survival (OS), are mixed. The largest retrospective study of hy- (n = 44) perglycemia during childhood cancer therapy (n¼ 1176) was performed on a cohort of Japanese children with ALL followed Figure 1. Flow diagram of search strategy. for a median duration of 62 months who received protracted L- asparaginase therapy with prednisolone throughout mainte- 1946 and 2019. The authors followed the Preferred Reporting nance therapy (16). In this cohort, hyperglycemic events were Items for Systematic Reviews and Meta-Analyses recommen- more frequent in maintenance than induction phases of ther- dations for this review. The search strategy is outlined in apy and reduced survival was linked to older age (10 years), in- Figure 1, and included publications are listed in Table 1. A total dependent of hyperglycemia. The largest cohort including of 222 articles were identified for potential inclusion after ac- infection complications (n¼ 871) showed no statistically signifi- counting for numerous duplicates (85), and an additional 114 cant difference regarding infection rate, EFS, and OS among were excluded for nonrelevance, leaving 108 articles that were children with ALL who experienced induction phase hypergly- screened for inclusion criteria with 44 identified for inclusion. cemia vs those who did not (8). A similar finding was reported In this review, we sought to determine the following among by Tsai et al (n¼ 133 patients) in which there was no statistically children receiving cancer therapy: the incidence of hyperglyce- significant difference between infectious episodes in the hyper- mia, risk factors for hyperglycemia, associated complications, glycemic vs euglycemic group (9). EFS was not reported. impact on outcomes, and methods for monitoring and manage- This is in contrast to the other, yet smaller (n< 200 in each) ment. Article synthesis was conducted using the integrative published studies that highlight a statistically significant in- methodology described by Whittemore and colleagues in which crease in infectious complications and lower EFS and OS among at least two researchers conduct an independent review of the ALL children who developed hyperglycemia during induction articles (12). If discrepancies were encountered, joint review (Zhang et al., n¼ 159; 5-year EFS: 62.9% vs 80.2% and OS 83.8% vs was employed until agreement was found. 94.9% in hyperglycemic vs euglycemic groups, respectively) (17). Sonabed et al. published similar results (n¼ 167 ALL patients; EFS: 68% vs 85% [P¼ .025] and OS: 74% vs 96% [P¼ .001] (5). Additionally, the risk of death was 6.2 times greater in those Incidence of Hyperglycemia in Childhood patients with BG levels greater than 200 mg/d, and overt hyper- Cancer Therapy glycemia was seen as an independent predictor of mortality in Among the 44 articles reviewed, reported incidence of hypergly- ALL even after controlling for risk group and type of steroid cemia in childhood cancer ranged considerably overall, from 3% used. Koltin et al. (n¼ 363) reported a similar association with to 63% (Table 1). A primary challenge in evaluating this rate is greater odds of prior hyperglycemia among ALL patients who the lack of consensus in how hyperglycemia is defined across died (2). studies. Variable definitions of pediatric hyperglycemia as well Another retrospective study limited to ALL induction phase as glucose variability have been cited to investigate the associa- (based on UKALL 2003, n¼ 144) showed that proven bacterial or tion of abnormal glucose homeostasis in critically ill children. fungal infection was more common in patients with overt The 2019 American Diabetes Association Guidelines and the hyperglycemia 200 mg/dL (odds ratio [OR] ¼ 4.1, P¼ .03) (18). 1998 report of a WHO consultation on diagnosis and classifica- Sonabed et al. demonstrated that patients with mild tion of diabetes mellitus define pediatric hyperglycemia as a BG hyperglycemia (BG 140–199 mg/dL) and overt hyperglycemia Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 134 | J Natl Cancer Inst Monogr, 2019, Vol. 2019, No. 54 Table 1. Studies included for review Cancer type Article Article type Age, y Hyperglycemia (%) Hyperglycemia definition Brain tumors Filho (2016) Retrospective cohort 1–12 45 of 105 (43%) >150 mg/dL, any Mekitarian Filho(2011) Retrospective cohort NR 124 of 198 (63%) 150 mg/dL, any Many cancers Busaidy (2015) Clinical trial 14–74 36 of 57 (63%) 100 mg/dL, fasting Goldman (2016) Review article n/a n/a n/a Hwangbo (2017) Review article n/a n/a n/a Shah (2017) Review article n/a n/a n/a Sopfe (2019) Retrospective cohort 0–30 151 of 344 (44%) Mean 126 mg/dL Walker (1988) Review article n/a n/a n/a Warwick (2013) Clinical Trial 3–23 9 of 72 (13%) *CTCAE v3.0 any grade Xiu (2014) Review article n/a n/a n/a Acute leukemia Baillargeon (2005) Retrospective cohort 2–18 17 of 155 (11%) 200 mg/dL 2 times Bani-Hashem (2009) Case control 3–11 NR/50 NR Banihashem (2014) Prospective cohort 1–14 3 of 29 (10%) 140 mg/dL 2 times Belgaumi (2003) Retrospective cohort <14 23 of 156 (15%) NR Bostrom (2013) Retrospective cohort <18 17 of 17 (100%) Diagnosis recorded Cetin (1994) Retrospective cohort NR 6 of 136 (4%) NR Dacou-Voutetakis (1983) Prospective case-control 3–15 9 of 15 (60%) IGT Impaired 2-h GTT 6 of 15 (40%) DM Dare (2013) Retrospective cohort 1–16 36 of 144 (25%) 140 mg/dL, any El-Fayoumi (2018) Case-control NR 6 of 70 (9%) >200 mg/dL, fasting Esbenshade (2013) Prospective cohort 3–11 1 of 34 (3%) Insulin resistance by HOMA (>3.15) Gatzioura (2016) Retrospective cohort 2–9 16 of 102 (16%) 200 mg/dL 2 times Gramatges (2013) Review article n/a n/a n/a Hijiya (2016) Review article n/a n/a n/a Howard (2002) Review article n/a n/a n/a Koltin (2012) Retrospective cohort 1–18 57 of 363 (16%) 200 mg/dL 2 times Laila (2016) Prospective cohort 1–15 4 of 50 (8%) 200 mg/dL 2 times Lebovic (2016) Retrospective cohort 1–21 16 of 183 (9%) Diagnosis recorded Lowas (2009) Retrospective cohort 2–18 33 of 162 (20%) 200 mg/dL 2 times Moschovi (2018) Review article n/a n/a n/a Moschovi (2010) Prospective cohort n/a n/a n/a Panigrahi (2016) Retrospective cohort 1–40 13 of 146 (9%) >100 mg/dL, fasting Pui (1981). Retrospective cohort 0–20 41 of 421 (10%) >140 mg/dL, fasting or >200 mg/dL, any Roberson (2008) Retrospective cohort 3-18 215 of 797 (27%) >200 mg/dL, any Roberson (2009) Retrospective cohort NR 141 of 871 (16%) 200 mg/dL, any Schmiegelow (2017) Review article n/a n/a n/a Sonabend (2008) Retrospective cohort 2–15 28 of 135 (21%) mild Mild: 140–200 mg/dL 47 of 135 (35%) overt Overt: >200 mg/dL Sonabend (2009) Retrospective cohort 1–18 41 of 167 (25%) mild Mild: 140–200 mg/dL 56 of 167 (34%) overt Overt: >200 mg/dL Spinola-Castro (2009) Retrospective cohort 1–18 12 of 311 (4%) 200 mg/dL 2 times Thu Huynh (2017) Review article n/a n/a n/a Tsai (2015) Retrospective cohort 1–18 22 of 133 (17%) >126 mg/dL, fasting or 200 mg/dL 2 times Warris (2016) Prospective cohort 3–16 2 of 50 (4%) 100 to <126 mg/dL, fasting Yeshayahu (2015) Case-control 13–21 4 of 30 (13%) 200 mg/dL 2 times Yoshida (2015) Retrospective cohort 1–18 69 of 1176 (6%) CTCAE v3.0 Grade 3–4 Zhang (2014) Retrospective cohort 1–16 38 of 159 (24%) >126 mg/dL, fasting or 200 mg/dL 2 times *CTCAE Grade 1 ¼>ULN–160 mg/dL, Grade 2 ¼>160–250 mg/dL, Grade 3 ¼>250–500 mg/dL, Grade 4 ¼>500 mg/dL, Grade 5 ¼ death. CTCAE ¼ Common Toxicity Criteria for Adverse Events; DM ¼ diabetes mellitus; GTT ¼ glucose tolerance test; HOMA ¼ homeostatic metabolic assessment; IGT ¼ impaired glucose tolerance; NR ¼ not recorded. (BG > 200 mg/dL) during induction were twice as likely to have in defining hyperglycemia (200 mg/dL), there were inconsis- documented infection than euglycemic patients (19). Patients tencies in whether two episodes or single episodes met inclu- with overt hyperglycemia were 4.2 times more likely to have sion criteria and varied widely in their duration of evaluation. bacteremia or fungemia, 3.8 times more likely to have cellulitis, In the hematopoietic stem cell transplant (HSCT) setting, and 4 times more likely to be admitted for febrile neutropenia Sopfe et al. demonstrated that increased mean glucose levels than the euglycemic group.) Although studies were consistent were associated with significantly worse treatment-related Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 A. Grimes et al. |135 mortality, OS, risk of severe graft-versus-host disease, and, for in univariate analysis (9,16,19). Other studies did not find an as- allogeneic HSCT patients, more infections and ICU days (20). sociation between BMI and hyperglycemia (2,4,27,31). Insulin re- Similar findings are reported in adult HSCT studies, which dem- sistance during therapy has been associated with increased BMI onstrated associations between hyperglycemia and infection, (32). In most analyses, sex is not associated with hyperglycemia length of hospital stay, organ dysfunction, graft-versus-host in ALL patients (2,8,9,19,27,29,31), although two did demonstrate disease, delayed hematopoietic recovery, and mortality (21,22). female sex as a risk factor in univariate but not multivariable Filho et al. observed among children with brain tumors ad- analysis (16,28). Three studies found that race or ethnicity mitted for elective tumor resection (n¼ 105) a statistically (4,8,19) and family history of diabetes (9,19,28) were not associ- higher complication rate in the postoperative period in patients ated with hyperglycemia in ALL patients. Trisomy 21 has been who developed hyperglycemia (23). Multivariable analysis dem- found in univariate analyses to be associated with hyperglyce- onstrated that peak BG level on postoperative day 1 was inde- mia (2,16), but in multivariable analysis, this did not hold true pendently associated with greater risk for complications, (2). including infections. Longer hospitalization and longer ICU stay Risk category of ALL is associated with hyperglycemia in were also recorded among individuals who were hyperglycemic some studies in univariate but not multivariable analysis, likely (BG 150 mg/dL) at the time of ICU admission. Mekitarian Filho because of its collinearity with age (2,9,16,19); other studies did et al. demonstrated a similar association with univariate not find any association (4,8,27). ALL disease- and treatment- analysis (n¼ 198), but this did not hold up with multivariable related risk factors such as lineage (4,9,27), protocol (4,8,9), and analysis (24). white blood cell count at diagnosis (2,9) have been found to not be associated with hyperglycemia or ketoacidosis. Type of ste- roid (prednisone vs dexamethasone) has generally not been as- Mechanisms of Hyperglycemia During sociated with risk of hyperglycemia (1,2,29), although in Childhood Cancer Therapy univariate analysis, one study did make this association (19). Different preparations and doses of asparaginase have been Patients undergoing pediatric cancer therapy have multiple po- evaluated as risk factors for hyperglycemia because of their tential risk factors for hyperglycemia, including GC exposure, known associations with hyperglycemia and insulin resistance hyperphagia, dietary choices, mucositis, pancreatitis, physio- (33). Lowas et al. demonstrated an increased risk with L-aspara- logic stress-mediated cortisol release, and deconditioning in ad- ginase compared with PEG-asparaginase (OR ¼ 3) in multivari- dition to drug-mediated effects on insulin production or able analysis (29). Yoshida and colleagues demonstrated that secretion and sensitivity. patients receiving protracted PEG-asparaginase had a higher The most well-established mechanism of these risk factors rate of hyperglycemia during maintenance chemotherapy, al- is the impact of GCs on insulin signaling pathways to cause though there was no direct comparison with other dosing regi- both insulin resistance and insulin insufficiency (25). High-dose mens to conclude a true difference (16). Further, Lebovic et al. GC therapy alters metabolism by increasing skeletal muscle ca- demonstrated increased risk with higher dose, the dose was tabolism and lipolysis. The resultant circulating amino acids also correlated with increased age and obesity, and no multivar- and fatty acids attenuate insulin signaling and reduce glucose iable analysis was performed (34). uptake by tissues. The amino acids provide substrate for hepatic No studies evaluating hyperglycemia risk factors in non-ALL gluconeogenesis and are stimulated by GC-enhanced expres- pediatric oncology populations were identified. Sopfe et al. eval- sion of hepatic phospho-enolpyruvate carboxykinase and glu- uated hyperglycemia in the pediatric HSCT population, and cose-6-phosphatase. Glucocorticoids also trigger pancreatic although this study did include patients with both malignant islet cells to inappropriately release glucagon, which is wors- and nonmalignant diseases, the following were identified as ened by escalating insulin resistance. Glucocorticoids directly risk factors for per-HSCT hyperglycemia: older age, malignancy downregulate glucose transporter 4 to reduce glucose uptake (vs nonmalignant diagnosis), post-HSCT steroids and/or mTOR and directly reduce biosynthesis and secretion. inhibitors, and allogeneic transplant (vs autologous). Other chemotherapy-related mechanisms for hyperglycemia Similar to studies in pediatric ALL and HSCT patients, adult include asparaginase interference with insulin production and studies have demonstrated that age is a risk factor for hypergly- secretion, mTOR (mammalian target of rapamycin) inhibitors cemia (21). Other risk factors include use of parenteral nutrition and tyrosine kinase inhibitor-initiated reduction in insulin sen- (specifically in adult HSCT oncology patients) (35–37), steroid sitivity, and immune checkpoint inhibitor-triggered dose and duration (38,39), and treatment with certain agents autoimmune-mediated pancreatic beta cell failure (26). including mTOR inhibitors (40–44), tyrosine kinase inhibitors (45), PD-1 inhibitors (42), and docetaxel, everolimus, temsiroli- mus, and androgen deprivation therapy (46). Risk Factors for Hyperglycemia During Cancer Lastly, although no studies directly evaluated stress hy- Therapy perglycemia in oncology patients, stress and critical illness In the pediatric oncology population, evaluation of risk factors have also been presumed to be risk factors in this population for hyperglycemia has focused on the ALL population; among (3,47–49). these studies, nearly all evaluations are limited to induction phase. The most consistently identified risk factor for hypergly- cemia in this population is increased age, with the most com- Monitoring for Hyperglycemia mon cutoff, 10 years or older, resulting in a range of increased odds from 3.6 to 37.2 in both univariate and multivariable analy- Currently, there are no established guidelines for monitoring ses (2,4,8,9,16,27–30). Increased body mass index (BMI) or over- hyperglycemia during childhood cancer therapy, in general or weight or obese status has been associated with increased odds among specific disease groups (15). Thus, no standard exists for (ORs ranging from 3.3 to 17.4) of hyperglycemia in several multi- how often or with which measurements such monitoring variable analyses (28,29), but in others, this was only significant should occur or when or how to intervene. This management Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 136 | J Natl Cancer Inst Monogr, 2019, Vol. 2019, No. 54 generally occurs through consultation with endocrinologists gluconeogenesis. Metformin has a very common side effect of and varies widely across institutions. There is a need for pub- gastrointestinal upset and a potential risk for hypoglycemia, he- lished data on glycemic patterns among children with cancer in patic toxicity, and rare lactic acidosis. Onset of action can re- order to determine appropriate monitoring and drive recom- quire weeks for full effect and it must be discontinued for mendations for management. imaging with contrast. However, Adeberg et al. determined that Even in ALL, for which the most data are available, there is among diabetic glioblastoma patients, those taking metformin conflicting data on the incidence and severity hyperglycemia had improved progression-free survival (10.13 vs 6.7 months, during ALL induction and subsequent therapy. The variable P¼ .018) (50). rates of glycemic abnormalities reported are difficult to com- The recent, expanding development of type 2 diabetes medi- pare due to differences in steroids, definitions of hyperglyce- cations that focus on many of the same pathways that are dis- mia, duration and timing of evaluation, as well as differences in rupted by GCs offer insight into both therapies and mechanisms regional ethnicities, obesity rates, baseline insulin resistance, of insulin actions and signaling for children receiving these and underlying genetic risks for hyperglycemia. drugs. New type 2 diabetes therapies include inhibitors of the Pharmacokinetics differ between GC preparations, with pred- enzyme that regulates the dipeptidyl peptidase 4 peroxisome nisolone typically peaking at 4–8 hours and having a duration of proliferator-activated receptor signaling pathway to enhance 12–16 hours. The GC peak action is likely associated with fluctu- insulin action (DPP-4 inhibitors), agonists of incretin hormones ations in mealtime insulin resistance and exaggerated post- that stimulate the pancreas to release insulin from b cells and prandial hyperglycemia. Thus, fasting chemistry profiles may reduce competing glucagon secretion, and SGLT2 competitive be deceivingly normal and are not ideal when screening for agents to allow for renal excretion of glucose when rising into peak hyperglycemia, regardless of etiology (GC-induced, other hyperglycemia range. These each offer unique potential for GC- chemotherapeutic agents, underlying pancreatitis). There is of- induced hyperglycemia, yet none are approved for pediatric ten a delay in recognizing hyperglycemia, partly due to differing use. Each also has associated side effects, which may be espe- surveillance approaches. Comparing pre- and postprandial cially concerning in an immunocompromised patient. blood sugars at the peak of GC action is the most logical strategy The ability to appropriately study and compare the effects of to appropriately recognize GC-related hyperglycemia; however, the various hyperglycemic therapies in children with cancer is medical teams and families may resist this approach as too in- further complicated by the stress of many clinical unknowns vasive. Overall, dysglycemia assessment using the HgbA1c test when approaching the families of these very ill children. Steroid is not helpful in the setting of acute hyperglycemia and therapy itself typically results in a gradual cumulative onset of anemia. Insulin therapy, if necessary, requires monitoring for underlying skeletal muscle insulin resistance, lipotoxicity, insu- fasting and pre- and postprandial effects once initiated. The lin synthesis or signaling dysregulation, and hepatic gluconeo- lack of evidence-based guidelines for glucose monitoring, insu- genesis, which leads to increasing risk for hyperglycemia. These lin initiation, and subsequent insulin adjustment in childhood metabolic changes in the setting of decreased activity and often cancer makes interpretation of the current literature coinciding infection and stress make management challenging. challenging. Utilization of monitoring tools that have been suc- The initial management then requires adjustment because cessful in juvenile diabetes such as continuous glucose moni- therapy often ends abruptly (eg, 28-day GC therapy); addition- toring may offer insight into glycemic patterns during cancer ally, patients may require intermittent stress dosing when ill. therapy. Children with ALL also receive chemotherapy with pancreatitis risk (asparaginase, GCs), varied dosages or types of GC therapy, appetite changes, and scheduled invasive procedures and radio- logic testing that require customized adjustments in steroid- Hyperglycemia Management During Cancer induced hyperglycemia management. The need for immediate Therapy management flexibility and the limited approved options for di- Steroid-induced hyperglycemia is a multifactorial complication abetes medications in children account for the preference for of steroid-enhanced gluconeogenesis and diminished glucose rapid onset insulin analogues (onset of action within 15 minutes utilization. Insulin remains the mainstay of available therapy; and duration of <4 hours) to be used for acute hyperglycemia however, evidence for insulin requirements for safe dosing regi- correction dosing and coverage of postprandial hyperglycemia mens and the role for continuous glucose monitoring applica- and long-acting basal insulin (duration of 22–40 hours without tions and potential pump management are needed. Replacing peaks in action) as the accepted standard of care for initial pedi- insulin requires expertise and caution to safely enhance com- atric ALL management. Further study is needed to determine promised insulin secretion and overcome insulin resistance predictors of response across treatment protocols and cancer during ALL induction when the cumulative effect of steroid types with consistent, randomized approaches to monitoring therapy becomes evident. Likewise, rapid reversal of insulin re- and glycemic therapy to determine best approaches for this placement is required to avoid outpatient postinduction hypo- unique population of youth. glycemia. Insulin therapies have advanced during the past decade to allow more physiologic basal-bolus therapies. The choice to use the newer, rapid-acting analogues of insulin affect physiologic glucose monitoring needs with regard to meal ad- Limitations vice, insulin duration, and basal insulin requirements. The com- plex glycemic metabolic pathways of GC-induced There are extensive limitations in reporting hyperglycemia- hyperglycemia can also be manipulated with insulin secreta- related complications and survival during childhood cancer gogues; however, these carry a known risk for hypoglycemia therapy: 1) all published clinical studies are retrospective; 2) the and the potential for exacerbation of pancreatic beta cell ex- majority of publications are limited to ALL and are limited to haustion. Metformin therapy has been proposed as having po- hyperglycemia occurring during induction therapy; 3) small co- tential to enhance insulin sensitivity and reduce hort sizes; 4) variable definitions of hyperglycemia; and 5) Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 A. Grimes et al. |137 11. Jeong Y, Han HS, Lee HD, et al. A pilot study evaluating steroid-induced dia- different treatment protocols and evaluation periods limit betes after antiemetic dexamethasone therapy in chemotherapy-treated comparisons. cancer patients. Cancer Res Treat. 2016;48(4):1429–1437. A consensus statement for the evaluation and management 12. Whittemore R, Knafl K. The integrative review: updated methodology. J Adv of hyperglycemia in childhood cancer is lacking, and there are Nurs. 2005;52(5):546–553. 13. Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes insufficient data to describe glycemic patterns during childhood mellitus and its complications. Part 1: diagnosis and classification of diabetes cancer therapies, including therapies with proven association mellitus provisional report of a WHO consultation. Diabet Med. 1998;15(7): with hyperglycemia such as GCs and asparaginase. Data are 539–553. 14. American Diabetes Association. 2. Classification and diagnosis of diabetes: lacking in children regarding newer agents associated with hy- standards of medical care in diabetes-2019. Diabetes Care. 2019;42(Suppl 1): perglycemia in adults, including targeted therapies and im- S13–S28. mune checkpoint inhibitors. A future study that includes 15. Patki VK, Chougule SB. Hyperglycemia in critically ill children. Indian J Crit Care Med. 2014;18(1):8–13. prospective evaluation of glycemic patterns in children with 16. Yoshida H, Imamura T, Saito AM, et al. Protracted administration of L-aspar- cancer receiving glycemic-altering therapies is needed. In addi- aginase in maintenance phase is the risk factor for hyperglycemia in older tion, a more comprehensive review of national hospital data patients with pediatric acute lymphoblastic leukemia. PLoS One. 2015;10(8): e0136428. would help to capture incident rates of hyperglycemia in a 17. Zhang BH, Wang J, Xue HM, Chen C. Impact of chemotherapy-related hyper- larger number of children with cancer across various institution glycemia on prognosis of child acute lymphocytic leukemia. Asian Pac J types, sizes, and geographic locations as well as a more varied Cancer Prev. 2014;15(20):8855–8859. 18. Dare JM, Moppett JP, Shield JP, Hunt LP, Stevens MC. The impact of hypergly- population of children in terms of diagnosis, age, ethnicity, and cemia on risk of infection and early death during induction therapy for acute BMI. Children receiving cancer therapy should be evaluated for lymphoblastic leukemia (ALL). Pediatr Blood Cancer. 2013;60(12):E157–E159. potential risks for hyperglycemia (age 10 years, Down syn- 19. Sonabend RY, McKay SV, Okcu MF, Yan J, Haymond MW, Margolin JF. drome, CNS involvement, obesity, planned therapy that Hyperglycemia during induction therapy is associated with increased infec- tious complications in childhood acute lymphocytic leukemia. Pediatr Blood includes glycemic-altering agents) and should undergo moni- Cancer. 2008;51(3):387–392. toring with serial BG assessments with a recording of fasting 20. Sopfe J, Pyle L, Keating AK, et al. Malglycemia is associated with poor out- status. Initial intervention should include insulin therapy based comes in pediatric and adolescent hematopoietic stem cell transplant patients. Blood Adv. 2019;3(3):350–359. on lack of data around use of metformin in children receiving 21. Storey S, Von Ah D. Impact of malglycemia on clinical outcomes in hospital- cancer therapy and risk of hepatotoxicity. Potential future use ized patients with cancer: a review of the literature. Oncol Nurs Forum. 2012; of closed-loop systems may provide insight into timing, dura- 39(5):458–465. 22. Hammer MJ, Casper C, Gooley TA, O’Donnell PV, Boeckh M, Hirsch IB. The tion, and incidence of dysglycemia in children with cancer and contribution of malglycemia to mortality among allogeneic hematopoietic optimize risk identification in addition to improved safety for cell transplant recipients. Biol Blood Marrow Transplant. 2009;15(3):344–351. monitoring and management. 23. Filho NO, Alves RL, Fernandes AT, Castro FSP, Melo JRT, Modolo NSP. Association of increased morbidity with the occurrence of hyperglycemia in the immediate postoperative period after elective pediatric neurosurgery. J Neurosurg Pediatr. 2016;17(5):625–629. 24. Mekitarian Filho E, Carvalho WB, Cavalheiro S, Horigoshi NK, Freddi NA, Note Vieira GK. Hyperglycemia and postoperative outcomes in pediatric neurosur- gery. Clinics (Sao Paulo). 2011;66(9):1637–1640. Affiliations of authors: UT Health Science Center San 25. van Raalte DH, Ouwens DM, Diamant M. Novel insights into glucocorticoid- Antonio, Department of Pediatrics, San Antonio, TX (AG, JL); mediated diabetogenic effects: towards expansion of therapeutic options? Cook Children’s Medical Center, Fort Worth, TX (AM, RH); Eur J Clin Invest. 2009;39(2):81–93. Children’s Hospital Colorado, Aurora, CO (JS). 26. Gallo M, Muscogiuri G, Felicetti F, et al. Adverse glycaemic effects of cancer therapy: indications for a rational approach to cancer patients with diabetes. Metab Clin Exp. 2018;78:141–154. 27. Gatzioura I, Papakonstantinou E, Dimitriadou M, et al. Glucose levels be- References fore the onset of asparaginase predicts transient hyperglycemia in chil- 1. Belgaumi AF, Al-Bakrah M, Al-Mahr M. Dexamethasone-associated toxicity dren with acute lymphoblastic leukemia. Pediatr Blood Cancer. 2016;63(7): during induction chemotherapy for childhood acute lymphoblastic leukemia 1181–1184. is augmented by concurrent use of daunomycin. Cancer. 2003;97(11): 28. Baillargeon J, Langevin AM, Mullins J, et al. Transient hyperglycemia in 2898–2903. Hispanic children with acute lymphoblastic leukemia. Pediatr Blood Cancer. 2. Koltin D, Sung L, Naqvi A, Urbach SL. Medication induced diabetes during in- 2005;45(7):960–963. duction in pediatric acute lymphoblastic leukemia: prevalence, risk factors 29. Lowas SR, Marks D, Malempati S. Prevalence of transient hyperglycemia dur- and characteristics. Support Care Cancer. 2012;20(9):2009–2015. ing induction chemotherapy for pediatric acute lymphoblastic leukemia. 3. Schmiegelow K, Muller K, Mogensen SS. Non-infectious chemotherapy- Pediatr Blood Cancer. 2009;52(7):814–818. associated acute toxicities during childhood acute lymphoblastic leukemia 30. Banihashem A, Ghasemi A, Ghaemi N, Moazzen N, Amirabadi A. Prevalence therapy. F1000Res. 2017;6:444. of transient hyperglycemia and diabetes mellitus in pediatric patients with 4. Roberson JR, Raju S, Shelso J, Pui CH, Howard SC. Diabetic ketoacidosis during acute leukemia. Iran J Ped Hematol Oncol. 2014;4(1):5–10. therapy for pediatric acute lymphoblastic leukemia. Pediatr Blood Cancer. 31. Spinola-Castro AM, Siviero-Miachon AA, Andreoni S, Tosta-Hernandez PD, 2008;50(6):1207–1212. Macedo CR, Lee ML. Transient hyperglycemia during childhood acute lym- 5. Sonabend RY, McKay SV, Okcu MF, Yan J, Haymond MW, Margolin JF. phocytic leukemia chemotherapy: an old event revisited. Clin Adv Hematol Hyperglycemia during induction therapy is associated with poorer survival Oncol. 2009;7(7):465–472. in children with acute lymphocytic leukemia. J Pediatr. 2009;155(1):73–78. 32. Esbenshade AJ,Simmons JH,KoyamaT,Lindell RB,FriedmanDL. Obesity 6. Tieu MT, Lovblom LE, McNamara MG, et al. Impact of glycemia on survival of and insulin resistance in pediatric acute lymphoblastic leukemia wor- glioblastoma patients treated with radiation and temozolomide. J Neurooncol. sens during maintenance therapy. Pediatr Blood Cancer. 2013;60(8): 2015;124(1):119–126. 1287–1291. 7. Pui CH, Burghen GA, Bowman WP, Aur RJ. Risk factors for hyperglycemia in 33. Cetin M, Yetgin S, Kara A, et al. Hyperglycemia, ketoacidosis and other com- children with leukemia receiving L-asparaginase and prednisone. J Pediatr. plications of L-asparaginase in children with acute lymphoblastic leukemia. J 1981;99(1):46–50. Med. 1994;25(3–4):219–229. 8. Roberson JR, Spraker HL, Shelso J, et al. Clinical consequences of hyperglyce- 34. Lebovic R, Pearce N, Lacey L, Xenakis J, Faircloth CB, Thompson P. Adverse mia during remission induction therapy for pediatric acute lymphoblastic effects of pegaspargase in pediatric patients receiving doses greater than 3, leukemia. Leukemia. 2009;23(2):245–250. 750 IU. Pediatr Blood Cancer. 2017;64(10):e26555. 9. Tsai MC, Huang HH, Chou YY, Cheng CN, Chen JS, Lin SJ. Risk factors for hy- 35. Sheean P, Braunschweig C. The incidence and impact of dextrose dose on hy- perglycemia during chemotherapy for acute lymphoblastic leukemia among perglycemia from parenteral nutrition (PN) exposure in hematopoietic stem Taiwanese children. Pediatr Neonatol. 2015;56(5):339–345. cell transplant (HSCT) recipients. JPEN J Parenter Enteral Nutr. 2006;30(4): 10. Warwick AB, Malempati S, Krailo M, et al. Phase 2 trial of pemetrexed in chil- 345–350. dren and adolescents with refractory solid tumors: a Children’s Oncology 36. Sheean PM, Freels SA, Helton WS, Braunschweig CA. Adverse clinical conse- Group study. Pediatr Blood Cancer. 2013;60(2):237–241. quences of hyperglycemia from total parenteral nutrition exposure during Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 138 | J Natl Cancer Inst Monogr, 2019, Vol. 2019, No. 54 hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2006; 43. Busaidy NL, Farooki A, Dowlati A, et al. Management of metabolic effects as- 12(6):656–664. sociated with anticancer agents targeting the PI3K-Akt-mTOR pathway. J Clin 37. Karnchanasorn R, Malamug LR, Jin R, Karanes C, Chiu KC. Association of hy- Oncol. 2012;30(23):2919–2928. perglycemia with prolonged hospital stay but no effect on engraftment after 44. Montero N, Pascual J. Immunosuppression and post-transplant hyperglyce- autologous hematopoietic stem cell transplantation. Endocr Pract. 2012;18(4): mia. Curr Diabetes Rev. 2015;11(3):144–154. 508–518. 45. Shah RR. Hyperglycaemia induced by novel anticancer agents: an undesir- 38. Healy SJ, Nagaraja HN, Alwan D, Dungan KM. Prevalence, predictors, and out- able complication or a potential therapeutic opportunity? Drug Saf. 2017; comes of steroid-induced hyperglycemia in hospitalized patients with hema- 40(3):211–228. tologic malignancies. Endocrine. 2017;56(1):90–97. 46. Hershey DS, Bryant AL, Olausson J, Davis ED, Brady VJ, Hammer M. 39. Pidala J, Kim J, Kharfan-Dabaja MA, et al. Dysglycemia following glucocorti- Hyperglycemic-inducing neoadjuvant agents used in treatment of solid coid therapy for acute graft-versus-host disease adversely affects transplan- tumors: a review of the literature. Oncol Nurs Forum. 2014;41(6):E343–E354. tation outcomes. Biol Blood Marrow Transplant. 2011;17(2):239–248. 47. Barua R, Templeton AJ, Seruga B, Ocana A, Amir E, Ethier JL. Hyperglycaemia 40. Khan KH, Wong M, Rihawi K, et al. Hyperglycemia and phosphatidylinositol and survival in solid tumours: a systematic review and meta-analysis. Clin 3-kinase/protein kinase B/mammalian target of rapamycin (PI3K/AKT/ Oncol (R Coll Radiol). 2018;30(4):215–224. mTOR) inhibitors in phase I trials: incidence, predictive factors, and manage- 48. Storey S, Ah D, Hammer M. Measurement of hyperglycemia and impact on ment. Oncologist. 2016;21(7):855–860. health outcomes in people with cancer: challenges and opportunities. Oncol 41. Goldman JW, Mendenhall MA, Rettinger SR. Hyperglycemia associated with Nurs Forum. 2017;44(4):E141–E151. targeted oncologic treatment: mechanisms and management. Oncologist. 49. Dungan KM, Braithwaite SS, Preiser JC. Stress hyperglycaemia. Lancet. 2009; 2016;21(11):1326–1336. 373(9677):1798–1807. 42. Hwangbo Y, Lee EK. Acute hyperglycemia associated with anti-cancer medi- 50. Adeberg S, Bernhardt D, Harrabi SB, et al. Metformin influences progression cation. Endocrinol Metab. 2017;32(1):23–29. in diabetic glioblastoma patients. Strahlenther Onkol. 2015;191(12):928–935. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JNCI Monographs Oxford University Press

Hyperglycemia During Childhood Cancer Therapy: Incidence, Implications, and Impact on Outcomes

Loading next page...
 
/lp/oxford-university-press/hyperglycemia-during-childhood-cancer-therapy-incidence-implications-o700ZV1liy
Publisher
Oxford University Press
Copyright
Copyright © 2022 Oxford University Press
ISSN
1052-6773
eISSN
1745-6614
DOI
10.1093/jncimonographs/lgz022
Publisher site
See Article on Publisher Site

Abstract

Hyperglycemia is a known complication of therapies used in the treatment of childhood cancer, particularly glucocorticoids and asparaginase. It has been linked to increased infection and reduced survival. With more limited data on hyperglycemia during childhood cancer treatment compared with adult cancer, impact on outcomes is less clear in this population. As additional glycemic-altering cancer agents including immune checkpoint inhibitors and targeted therapies make their way into pediatric cancer treatment, there is a more pressing need to better understand the mechanisms, risk factors, and adverse effects of hyperglycemia on the child with cancer. Thus, we utilized a systematic approach to review the current understand- ing of the incidence, implications, and outcomes of hyperglycemia during childhood cancer therapy. Published literature regarding glycemic patterns and Background evidence-based management of hyperglycemia during therapy Childhood cancer therapy is frequently complicated by hyper- for childhood cancer is lacking, even for ALL therapy, during glycemia, with prevalence ranging from 10% to 20%. This has which risk for hyperglycemia is well established. Very little is been most readily observed during treatment for acute lympho- known about hyperglycemia during treatment of other types of blastic leukemia (ALL) (1–3) and is most often associated with cancers, and data are lacking to guide identification of at-risk glycemic-altering chemotherapeutic and supportive agents, populations. Published findings are limited in their application particularly glucocorticoids (GCs) and asparaginase. Early recog- due to variability in blood glucose (BG) measurements (fasting nition and safe management of hyperglycemia are important in status, postprandial timing, single vs repeated episodes) and children to prevent acute complications including ketoacidosis definitions of hyperglycemia. Standard guidelines for screening, and hyperosmolar nonketotic coma (4). In addition, hyperglyce- diagnosing, and managing hyperglycemia during childhood mia during childhood cancer therapy has been associated with cancer therapy are needed. increased infections and lower survival (5), a finding replicated among adult cancer patients (6). Among patients undergoing treatment for ALL, increased Methods rates of hyperglycemia have been seen among older patients (10 years) and those with Down syndrome and CNS involve- Three electronic databases (Ovid, PubMed, and Medline) were ment (2), with mixed data on the association with obesity (2,7–9). searched using the following search terms: hyperglycemia, can- Treatment-related hyperglycemia is also described in chil- cer, childhood, management, and treatment. The terms were dren with solid tumors, occurring in 13% of children receiving used both as keywords and as MeSH terms to optimize the antiemetic dexamethasone during premetrexed therapy in a number of articles included. Reference lists were also searched Children’s Oncology Group Phase 2 trial (10). This is similar to for applicable publications. Inclusion criteria were manuscripts hyperglycemia rates reported in adult cancer patients receiving that reported findings in English, that were not single case antiemetic dexamethasone (up to 20%) (11). reports or isolated abstracts, and that were published between Received: February 28, 2019; Revised: June 11, 2019; Accepted: July 1, 2019 Published by Oxford University Press 2019. This work is written by US Government employees and is in the public domain in the US. 132 Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 A. Grimes et al. |133 level of greater than 126 mg/dL (13,14). Observed stress hyper- Records idenfied through Addional records glycemia frequently occurs in children with severe illness and/ database searching idenfied through other (n = 231) sources or steroid use (15). There is controversy regarding data and (n = 76) observations that guide pragmatic institutional approaches to hyperglycemia management to safely avoid adverse outcomes of suspected stress hyperglycemia in children. The lack of con- Records aer duplicates removed sensus to consistently define hyperglycemia across studies is (n = 222) not limited to childhood cancer literature. Of the 44 studies in 85 Duplicates this review, 29 included a definition of hyperglycemia with wide variability in this definition across studies, as outlined in Table 1. The most frequent definition used required a recorded Records screened Records excluded for non- BG value of at least 200 mg/dL on at least 2 occasions without re- (n = 222) relevance gard to fasting status (seen in 8 of 29 articles) followed by the oc- (n = 114) currence of any random single BG value of at least 200 mg/dL (5 of 29 articles). Some studies relied on standard toxicity grading Full-text arcles excluded, criteria and others accepted written documentation of a diagno- Full-text arcles assessed with reasons for inclusion sis of hyperglycemia or diagnostic code. Although most defini- (n = 64) (n = 108) tions relied on retrospective BG measurements, some evaluated 7 Case Reports response to oral glucose tolerance testing or measurements of 49 Non-pediatric insulin resistance. 2 Open study/no data published Studies included in 6 Late effects only qualitave review/synthesis Hyperglycemia Complications and Impact on (n = 44) Outcomes Findings regarding the impact of hyperglycemia on outcomes, Studies included in including complications, event free survival (EFS), and overall qualitave review/analysis survival (OS), are mixed. The largest retrospective study of hy- (n = 44) perglycemia during childhood cancer therapy (n¼ 1176) was performed on a cohort of Japanese children with ALL followed Figure 1. Flow diagram of search strategy. for a median duration of 62 months who received protracted L- asparaginase therapy with prednisolone throughout mainte- 1946 and 2019. The authors followed the Preferred Reporting nance therapy (16). In this cohort, hyperglycemic events were Items for Systematic Reviews and Meta-Analyses recommen- more frequent in maintenance than induction phases of ther- dations for this review. The search strategy is outlined in apy and reduced survival was linked to older age (10 years), in- Figure 1, and included publications are listed in Table 1. A total dependent of hyperglycemia. The largest cohort including of 222 articles were identified for potential inclusion after ac- infection complications (n¼ 871) showed no statistically signifi- counting for numerous duplicates (85), and an additional 114 cant difference regarding infection rate, EFS, and OS among were excluded for nonrelevance, leaving 108 articles that were children with ALL who experienced induction phase hypergly- screened for inclusion criteria with 44 identified for inclusion. cemia vs those who did not (8). A similar finding was reported In this review, we sought to determine the following among by Tsai et al (n¼ 133 patients) in which there was no statistically children receiving cancer therapy: the incidence of hyperglyce- significant difference between infectious episodes in the hyper- mia, risk factors for hyperglycemia, associated complications, glycemic vs euglycemic group (9). EFS was not reported. impact on outcomes, and methods for monitoring and manage- This is in contrast to the other, yet smaller (n< 200 in each) ment. Article synthesis was conducted using the integrative published studies that highlight a statistically significant in- methodology described by Whittemore and colleagues in which crease in infectious complications and lower EFS and OS among at least two researchers conduct an independent review of the ALL children who developed hyperglycemia during induction articles (12). If discrepancies were encountered, joint review (Zhang et al., n¼ 159; 5-year EFS: 62.9% vs 80.2% and OS 83.8% vs was employed until agreement was found. 94.9% in hyperglycemic vs euglycemic groups, respectively) (17). Sonabed et al. published similar results (n¼ 167 ALL patients; EFS: 68% vs 85% [P¼ .025] and OS: 74% vs 96% [P¼ .001] (5). Additionally, the risk of death was 6.2 times greater in those Incidence of Hyperglycemia in Childhood patients with BG levels greater than 200 mg/d, and overt hyper- Cancer Therapy glycemia was seen as an independent predictor of mortality in Among the 44 articles reviewed, reported incidence of hypergly- ALL even after controlling for risk group and type of steroid cemia in childhood cancer ranged considerably overall, from 3% used. Koltin et al. (n¼ 363) reported a similar association with to 63% (Table 1). A primary challenge in evaluating this rate is greater odds of prior hyperglycemia among ALL patients who the lack of consensus in how hyperglycemia is defined across died (2). studies. Variable definitions of pediatric hyperglycemia as well Another retrospective study limited to ALL induction phase as glucose variability have been cited to investigate the associa- (based on UKALL 2003, n¼ 144) showed that proven bacterial or tion of abnormal glucose homeostasis in critically ill children. fungal infection was more common in patients with overt The 2019 American Diabetes Association Guidelines and the hyperglycemia 200 mg/dL (odds ratio [OR] ¼ 4.1, P¼ .03) (18). 1998 report of a WHO consultation on diagnosis and classifica- Sonabed et al. demonstrated that patients with mild tion of diabetes mellitus define pediatric hyperglycemia as a BG hyperglycemia (BG 140–199 mg/dL) and overt hyperglycemia Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 134 | J Natl Cancer Inst Monogr, 2019, Vol. 2019, No. 54 Table 1. Studies included for review Cancer type Article Article type Age, y Hyperglycemia (%) Hyperglycemia definition Brain tumors Filho (2016) Retrospective cohort 1–12 45 of 105 (43%) >150 mg/dL, any Mekitarian Filho(2011) Retrospective cohort NR 124 of 198 (63%) 150 mg/dL, any Many cancers Busaidy (2015) Clinical trial 14–74 36 of 57 (63%) 100 mg/dL, fasting Goldman (2016) Review article n/a n/a n/a Hwangbo (2017) Review article n/a n/a n/a Shah (2017) Review article n/a n/a n/a Sopfe (2019) Retrospective cohort 0–30 151 of 344 (44%) Mean 126 mg/dL Walker (1988) Review article n/a n/a n/a Warwick (2013) Clinical Trial 3–23 9 of 72 (13%) *CTCAE v3.0 any grade Xiu (2014) Review article n/a n/a n/a Acute leukemia Baillargeon (2005) Retrospective cohort 2–18 17 of 155 (11%) 200 mg/dL 2 times Bani-Hashem (2009) Case control 3–11 NR/50 NR Banihashem (2014) Prospective cohort 1–14 3 of 29 (10%) 140 mg/dL 2 times Belgaumi (2003) Retrospective cohort <14 23 of 156 (15%) NR Bostrom (2013) Retrospective cohort <18 17 of 17 (100%) Diagnosis recorded Cetin (1994) Retrospective cohort NR 6 of 136 (4%) NR Dacou-Voutetakis (1983) Prospective case-control 3–15 9 of 15 (60%) IGT Impaired 2-h GTT 6 of 15 (40%) DM Dare (2013) Retrospective cohort 1–16 36 of 144 (25%) 140 mg/dL, any El-Fayoumi (2018) Case-control NR 6 of 70 (9%) >200 mg/dL, fasting Esbenshade (2013) Prospective cohort 3–11 1 of 34 (3%) Insulin resistance by HOMA (>3.15) Gatzioura (2016) Retrospective cohort 2–9 16 of 102 (16%) 200 mg/dL 2 times Gramatges (2013) Review article n/a n/a n/a Hijiya (2016) Review article n/a n/a n/a Howard (2002) Review article n/a n/a n/a Koltin (2012) Retrospective cohort 1–18 57 of 363 (16%) 200 mg/dL 2 times Laila (2016) Prospective cohort 1–15 4 of 50 (8%) 200 mg/dL 2 times Lebovic (2016) Retrospective cohort 1–21 16 of 183 (9%) Diagnosis recorded Lowas (2009) Retrospective cohort 2–18 33 of 162 (20%) 200 mg/dL 2 times Moschovi (2018) Review article n/a n/a n/a Moschovi (2010) Prospective cohort n/a n/a n/a Panigrahi (2016) Retrospective cohort 1–40 13 of 146 (9%) >100 mg/dL, fasting Pui (1981). Retrospective cohort 0–20 41 of 421 (10%) >140 mg/dL, fasting or >200 mg/dL, any Roberson (2008) Retrospective cohort 3-18 215 of 797 (27%) >200 mg/dL, any Roberson (2009) Retrospective cohort NR 141 of 871 (16%) 200 mg/dL, any Schmiegelow (2017) Review article n/a n/a n/a Sonabend (2008) Retrospective cohort 2–15 28 of 135 (21%) mild Mild: 140–200 mg/dL 47 of 135 (35%) overt Overt: >200 mg/dL Sonabend (2009) Retrospective cohort 1–18 41 of 167 (25%) mild Mild: 140–200 mg/dL 56 of 167 (34%) overt Overt: >200 mg/dL Spinola-Castro (2009) Retrospective cohort 1–18 12 of 311 (4%) 200 mg/dL 2 times Thu Huynh (2017) Review article n/a n/a n/a Tsai (2015) Retrospective cohort 1–18 22 of 133 (17%) >126 mg/dL, fasting or 200 mg/dL 2 times Warris (2016) Prospective cohort 3–16 2 of 50 (4%) 100 to <126 mg/dL, fasting Yeshayahu (2015) Case-control 13–21 4 of 30 (13%) 200 mg/dL 2 times Yoshida (2015) Retrospective cohort 1–18 69 of 1176 (6%) CTCAE v3.0 Grade 3–4 Zhang (2014) Retrospective cohort 1–16 38 of 159 (24%) >126 mg/dL, fasting or 200 mg/dL 2 times *CTCAE Grade 1 ¼>ULN–160 mg/dL, Grade 2 ¼>160–250 mg/dL, Grade 3 ¼>250–500 mg/dL, Grade 4 ¼>500 mg/dL, Grade 5 ¼ death. CTCAE ¼ Common Toxicity Criteria for Adverse Events; DM ¼ diabetes mellitus; GTT ¼ glucose tolerance test; HOMA ¼ homeostatic metabolic assessment; IGT ¼ impaired glucose tolerance; NR ¼ not recorded. (BG > 200 mg/dL) during induction were twice as likely to have in defining hyperglycemia (200 mg/dL), there were inconsis- documented infection than euglycemic patients (19). Patients tencies in whether two episodes or single episodes met inclu- with overt hyperglycemia were 4.2 times more likely to have sion criteria and varied widely in their duration of evaluation. bacteremia or fungemia, 3.8 times more likely to have cellulitis, In the hematopoietic stem cell transplant (HSCT) setting, and 4 times more likely to be admitted for febrile neutropenia Sopfe et al. demonstrated that increased mean glucose levels than the euglycemic group.) Although studies were consistent were associated with significantly worse treatment-related Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 A. Grimes et al. |135 mortality, OS, risk of severe graft-versus-host disease, and, for in univariate analysis (9,16,19). Other studies did not find an as- allogeneic HSCT patients, more infections and ICU days (20). sociation between BMI and hyperglycemia (2,4,27,31). Insulin re- Similar findings are reported in adult HSCT studies, which dem- sistance during therapy has been associated with increased BMI onstrated associations between hyperglycemia and infection, (32). In most analyses, sex is not associated with hyperglycemia length of hospital stay, organ dysfunction, graft-versus-host in ALL patients (2,8,9,19,27,29,31), although two did demonstrate disease, delayed hematopoietic recovery, and mortality (21,22). female sex as a risk factor in univariate but not multivariable Filho et al. observed among children with brain tumors ad- analysis (16,28). Three studies found that race or ethnicity mitted for elective tumor resection (n¼ 105) a statistically (4,8,19) and family history of diabetes (9,19,28) were not associ- higher complication rate in the postoperative period in patients ated with hyperglycemia in ALL patients. Trisomy 21 has been who developed hyperglycemia (23). Multivariable analysis dem- found in univariate analyses to be associated with hyperglyce- onstrated that peak BG level on postoperative day 1 was inde- mia (2,16), but in multivariable analysis, this did not hold true pendently associated with greater risk for complications, (2). including infections. Longer hospitalization and longer ICU stay Risk category of ALL is associated with hyperglycemia in were also recorded among individuals who were hyperglycemic some studies in univariate but not multivariable analysis, likely (BG 150 mg/dL) at the time of ICU admission. Mekitarian Filho because of its collinearity with age (2,9,16,19); other studies did et al. demonstrated a similar association with univariate not find any association (4,8,27). ALL disease- and treatment- analysis (n¼ 198), but this did not hold up with multivariable related risk factors such as lineage (4,9,27), protocol (4,8,9), and analysis (24). white blood cell count at diagnosis (2,9) have been found to not be associated with hyperglycemia or ketoacidosis. Type of ste- roid (prednisone vs dexamethasone) has generally not been as- Mechanisms of Hyperglycemia During sociated with risk of hyperglycemia (1,2,29), although in Childhood Cancer Therapy univariate analysis, one study did make this association (19). Different preparations and doses of asparaginase have been Patients undergoing pediatric cancer therapy have multiple po- evaluated as risk factors for hyperglycemia because of their tential risk factors for hyperglycemia, including GC exposure, known associations with hyperglycemia and insulin resistance hyperphagia, dietary choices, mucositis, pancreatitis, physio- (33). Lowas et al. demonstrated an increased risk with L-aspara- logic stress-mediated cortisol release, and deconditioning in ad- ginase compared with PEG-asparaginase (OR ¼ 3) in multivari- dition to drug-mediated effects on insulin production or able analysis (29). Yoshida and colleagues demonstrated that secretion and sensitivity. patients receiving protracted PEG-asparaginase had a higher The most well-established mechanism of these risk factors rate of hyperglycemia during maintenance chemotherapy, al- is the impact of GCs on insulin signaling pathways to cause though there was no direct comparison with other dosing regi- both insulin resistance and insulin insufficiency (25). High-dose mens to conclude a true difference (16). Further, Lebovic et al. GC therapy alters metabolism by increasing skeletal muscle ca- demonstrated increased risk with higher dose, the dose was tabolism and lipolysis. The resultant circulating amino acids also correlated with increased age and obesity, and no multivar- and fatty acids attenuate insulin signaling and reduce glucose iable analysis was performed (34). uptake by tissues. The amino acids provide substrate for hepatic No studies evaluating hyperglycemia risk factors in non-ALL gluconeogenesis and are stimulated by GC-enhanced expres- pediatric oncology populations were identified. Sopfe et al. eval- sion of hepatic phospho-enolpyruvate carboxykinase and glu- uated hyperglycemia in the pediatric HSCT population, and cose-6-phosphatase. Glucocorticoids also trigger pancreatic although this study did include patients with both malignant islet cells to inappropriately release glucagon, which is wors- and nonmalignant diseases, the following were identified as ened by escalating insulin resistance. Glucocorticoids directly risk factors for per-HSCT hyperglycemia: older age, malignancy downregulate glucose transporter 4 to reduce glucose uptake (vs nonmalignant diagnosis), post-HSCT steroids and/or mTOR and directly reduce biosynthesis and secretion. inhibitors, and allogeneic transplant (vs autologous). Other chemotherapy-related mechanisms for hyperglycemia Similar to studies in pediatric ALL and HSCT patients, adult include asparaginase interference with insulin production and studies have demonstrated that age is a risk factor for hypergly- secretion, mTOR (mammalian target of rapamycin) inhibitors cemia (21). Other risk factors include use of parenteral nutrition and tyrosine kinase inhibitor-initiated reduction in insulin sen- (specifically in adult HSCT oncology patients) (35–37), steroid sitivity, and immune checkpoint inhibitor-triggered dose and duration (38,39), and treatment with certain agents autoimmune-mediated pancreatic beta cell failure (26). including mTOR inhibitors (40–44), tyrosine kinase inhibitors (45), PD-1 inhibitors (42), and docetaxel, everolimus, temsiroli- mus, and androgen deprivation therapy (46). Risk Factors for Hyperglycemia During Cancer Lastly, although no studies directly evaluated stress hy- Therapy perglycemia in oncology patients, stress and critical illness In the pediatric oncology population, evaluation of risk factors have also been presumed to be risk factors in this population for hyperglycemia has focused on the ALL population; among (3,47–49). these studies, nearly all evaluations are limited to induction phase. The most consistently identified risk factor for hypergly- cemia in this population is increased age, with the most com- Monitoring for Hyperglycemia mon cutoff, 10 years or older, resulting in a range of increased odds from 3.6 to 37.2 in both univariate and multivariable analy- Currently, there are no established guidelines for monitoring ses (2,4,8,9,16,27–30). Increased body mass index (BMI) or over- hyperglycemia during childhood cancer therapy, in general or weight or obese status has been associated with increased odds among specific disease groups (15). Thus, no standard exists for (ORs ranging from 3.3 to 17.4) of hyperglycemia in several multi- how often or with which measurements such monitoring variable analyses (28,29), but in others, this was only significant should occur or when or how to intervene. This management Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 136 | J Natl Cancer Inst Monogr, 2019, Vol. 2019, No. 54 generally occurs through consultation with endocrinologists gluconeogenesis. Metformin has a very common side effect of and varies widely across institutions. There is a need for pub- gastrointestinal upset and a potential risk for hypoglycemia, he- lished data on glycemic patterns among children with cancer in patic toxicity, and rare lactic acidosis. Onset of action can re- order to determine appropriate monitoring and drive recom- quire weeks for full effect and it must be discontinued for mendations for management. imaging with contrast. However, Adeberg et al. determined that Even in ALL, for which the most data are available, there is among diabetic glioblastoma patients, those taking metformin conflicting data on the incidence and severity hyperglycemia had improved progression-free survival (10.13 vs 6.7 months, during ALL induction and subsequent therapy. The variable P¼ .018) (50). rates of glycemic abnormalities reported are difficult to com- The recent, expanding development of type 2 diabetes medi- pare due to differences in steroids, definitions of hyperglyce- cations that focus on many of the same pathways that are dis- mia, duration and timing of evaluation, as well as differences in rupted by GCs offer insight into both therapies and mechanisms regional ethnicities, obesity rates, baseline insulin resistance, of insulin actions and signaling for children receiving these and underlying genetic risks for hyperglycemia. drugs. New type 2 diabetes therapies include inhibitors of the Pharmacokinetics differ between GC preparations, with pred- enzyme that regulates the dipeptidyl peptidase 4 peroxisome nisolone typically peaking at 4–8 hours and having a duration of proliferator-activated receptor signaling pathway to enhance 12–16 hours. The GC peak action is likely associated with fluctu- insulin action (DPP-4 inhibitors), agonists of incretin hormones ations in mealtime insulin resistance and exaggerated post- that stimulate the pancreas to release insulin from b cells and prandial hyperglycemia. Thus, fasting chemistry profiles may reduce competing glucagon secretion, and SGLT2 competitive be deceivingly normal and are not ideal when screening for agents to allow for renal excretion of glucose when rising into peak hyperglycemia, regardless of etiology (GC-induced, other hyperglycemia range. These each offer unique potential for GC- chemotherapeutic agents, underlying pancreatitis). There is of- induced hyperglycemia, yet none are approved for pediatric ten a delay in recognizing hyperglycemia, partly due to differing use. Each also has associated side effects, which may be espe- surveillance approaches. Comparing pre- and postprandial cially concerning in an immunocompromised patient. blood sugars at the peak of GC action is the most logical strategy The ability to appropriately study and compare the effects of to appropriately recognize GC-related hyperglycemia; however, the various hyperglycemic therapies in children with cancer is medical teams and families may resist this approach as too in- further complicated by the stress of many clinical unknowns vasive. Overall, dysglycemia assessment using the HgbA1c test when approaching the families of these very ill children. Steroid is not helpful in the setting of acute hyperglycemia and therapy itself typically results in a gradual cumulative onset of anemia. Insulin therapy, if necessary, requires monitoring for underlying skeletal muscle insulin resistance, lipotoxicity, insu- fasting and pre- and postprandial effects once initiated. The lin synthesis or signaling dysregulation, and hepatic gluconeo- lack of evidence-based guidelines for glucose monitoring, insu- genesis, which leads to increasing risk for hyperglycemia. These lin initiation, and subsequent insulin adjustment in childhood metabolic changes in the setting of decreased activity and often cancer makes interpretation of the current literature coinciding infection and stress make management challenging. challenging. Utilization of monitoring tools that have been suc- The initial management then requires adjustment because cessful in juvenile diabetes such as continuous glucose moni- therapy often ends abruptly (eg, 28-day GC therapy); addition- toring may offer insight into glycemic patterns during cancer ally, patients may require intermittent stress dosing when ill. therapy. Children with ALL also receive chemotherapy with pancreatitis risk (asparaginase, GCs), varied dosages or types of GC therapy, appetite changes, and scheduled invasive procedures and radio- logic testing that require customized adjustments in steroid- Hyperglycemia Management During Cancer induced hyperglycemia management. The need for immediate Therapy management flexibility and the limited approved options for di- Steroid-induced hyperglycemia is a multifactorial complication abetes medications in children account for the preference for of steroid-enhanced gluconeogenesis and diminished glucose rapid onset insulin analogues (onset of action within 15 minutes utilization. Insulin remains the mainstay of available therapy; and duration of <4 hours) to be used for acute hyperglycemia however, evidence for insulin requirements for safe dosing regi- correction dosing and coverage of postprandial hyperglycemia mens and the role for continuous glucose monitoring applica- and long-acting basal insulin (duration of 22–40 hours without tions and potential pump management are needed. Replacing peaks in action) as the accepted standard of care for initial pedi- insulin requires expertise and caution to safely enhance com- atric ALL management. Further study is needed to determine promised insulin secretion and overcome insulin resistance predictors of response across treatment protocols and cancer during ALL induction when the cumulative effect of steroid types with consistent, randomized approaches to monitoring therapy becomes evident. Likewise, rapid reversal of insulin re- and glycemic therapy to determine best approaches for this placement is required to avoid outpatient postinduction hypo- unique population of youth. glycemia. Insulin therapies have advanced during the past decade to allow more physiologic basal-bolus therapies. The choice to use the newer, rapid-acting analogues of insulin affect physiologic glucose monitoring needs with regard to meal ad- Limitations vice, insulin duration, and basal insulin requirements. The com- plex glycemic metabolic pathways of GC-induced There are extensive limitations in reporting hyperglycemia- hyperglycemia can also be manipulated with insulin secreta- related complications and survival during childhood cancer gogues; however, these carry a known risk for hypoglycemia therapy: 1) all published clinical studies are retrospective; 2) the and the potential for exacerbation of pancreatic beta cell ex- majority of publications are limited to ALL and are limited to haustion. Metformin therapy has been proposed as having po- hyperglycemia occurring during induction therapy; 3) small co- tential to enhance insulin sensitivity and reduce hort sizes; 4) variable definitions of hyperglycemia; and 5) Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 A. Grimes et al. |137 11. Jeong Y, Han HS, Lee HD, et al. A pilot study evaluating steroid-induced dia- different treatment protocols and evaluation periods limit betes after antiemetic dexamethasone therapy in chemotherapy-treated comparisons. cancer patients. Cancer Res Treat. 2016;48(4):1429–1437. A consensus statement for the evaluation and management 12. Whittemore R, Knafl K. The integrative review: updated methodology. J Adv of hyperglycemia in childhood cancer is lacking, and there are Nurs. 2005;52(5):546–553. 13. Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes insufficient data to describe glycemic patterns during childhood mellitus and its complications. Part 1: diagnosis and classification of diabetes cancer therapies, including therapies with proven association mellitus provisional report of a WHO consultation. Diabet Med. 1998;15(7): with hyperglycemia such as GCs and asparaginase. Data are 539–553. 14. American Diabetes Association. 2. Classification and diagnosis of diabetes: lacking in children regarding newer agents associated with hy- standards of medical care in diabetes-2019. Diabetes Care. 2019;42(Suppl 1): perglycemia in adults, including targeted therapies and im- S13–S28. mune checkpoint inhibitors. A future study that includes 15. Patki VK, Chougule SB. Hyperglycemia in critically ill children. Indian J Crit Care Med. 2014;18(1):8–13. prospective evaluation of glycemic patterns in children with 16. Yoshida H, Imamura T, Saito AM, et al. Protracted administration of L-aspar- cancer receiving glycemic-altering therapies is needed. In addi- aginase in maintenance phase is the risk factor for hyperglycemia in older tion, a more comprehensive review of national hospital data patients with pediatric acute lymphoblastic leukemia. PLoS One. 2015;10(8): e0136428. would help to capture incident rates of hyperglycemia in a 17. Zhang BH, Wang J, Xue HM, Chen C. Impact of chemotherapy-related hyper- larger number of children with cancer across various institution glycemia on prognosis of child acute lymphocytic leukemia. Asian Pac J types, sizes, and geographic locations as well as a more varied Cancer Prev. 2014;15(20):8855–8859. 18. Dare JM, Moppett JP, Shield JP, Hunt LP, Stevens MC. The impact of hypergly- population of children in terms of diagnosis, age, ethnicity, and cemia on risk of infection and early death during induction therapy for acute BMI. Children receiving cancer therapy should be evaluated for lymphoblastic leukemia (ALL). Pediatr Blood Cancer. 2013;60(12):E157–E159. potential risks for hyperglycemia (age 10 years, Down syn- 19. Sonabend RY, McKay SV, Okcu MF, Yan J, Haymond MW, Margolin JF. drome, CNS involvement, obesity, planned therapy that Hyperglycemia during induction therapy is associated with increased infec- tious complications in childhood acute lymphocytic leukemia. Pediatr Blood includes glycemic-altering agents) and should undergo moni- Cancer. 2008;51(3):387–392. toring with serial BG assessments with a recording of fasting 20. Sopfe J, Pyle L, Keating AK, et al. Malglycemia is associated with poor out- status. Initial intervention should include insulin therapy based comes in pediatric and adolescent hematopoietic stem cell transplant patients. Blood Adv. 2019;3(3):350–359. on lack of data around use of metformin in children receiving 21. Storey S, Von Ah D. Impact of malglycemia on clinical outcomes in hospital- cancer therapy and risk of hepatotoxicity. Potential future use ized patients with cancer: a review of the literature. Oncol Nurs Forum. 2012; of closed-loop systems may provide insight into timing, dura- 39(5):458–465. 22. Hammer MJ, Casper C, Gooley TA, O’Donnell PV, Boeckh M, Hirsch IB. The tion, and incidence of dysglycemia in children with cancer and contribution of malglycemia to mortality among allogeneic hematopoietic optimize risk identification in addition to improved safety for cell transplant recipients. Biol Blood Marrow Transplant. 2009;15(3):344–351. monitoring and management. 23. Filho NO, Alves RL, Fernandes AT, Castro FSP, Melo JRT, Modolo NSP. Association of increased morbidity with the occurrence of hyperglycemia in the immediate postoperative period after elective pediatric neurosurgery. J Neurosurg Pediatr. 2016;17(5):625–629. 24. Mekitarian Filho E, Carvalho WB, Cavalheiro S, Horigoshi NK, Freddi NA, Note Vieira GK. Hyperglycemia and postoperative outcomes in pediatric neurosur- gery. Clinics (Sao Paulo). 2011;66(9):1637–1640. Affiliations of authors: UT Health Science Center San 25. van Raalte DH, Ouwens DM, Diamant M. Novel insights into glucocorticoid- Antonio, Department of Pediatrics, San Antonio, TX (AG, JL); mediated diabetogenic effects: towards expansion of therapeutic options? Cook Children’s Medical Center, Fort Worth, TX (AM, RH); Eur J Clin Invest. 2009;39(2):81–93. Children’s Hospital Colorado, Aurora, CO (JS). 26. Gallo M, Muscogiuri G, Felicetti F, et al. Adverse glycaemic effects of cancer therapy: indications for a rational approach to cancer patients with diabetes. Metab Clin Exp. 2018;78:141–154. 27. Gatzioura I, Papakonstantinou E, Dimitriadou M, et al. Glucose levels be- References fore the onset of asparaginase predicts transient hyperglycemia in chil- 1. Belgaumi AF, Al-Bakrah M, Al-Mahr M. Dexamethasone-associated toxicity dren with acute lymphoblastic leukemia. Pediatr Blood Cancer. 2016;63(7): during induction chemotherapy for childhood acute lymphoblastic leukemia 1181–1184. is augmented by concurrent use of daunomycin. Cancer. 2003;97(11): 28. Baillargeon J, Langevin AM, Mullins J, et al. Transient hyperglycemia in 2898–2903. Hispanic children with acute lymphoblastic leukemia. Pediatr Blood Cancer. 2. Koltin D, Sung L, Naqvi A, Urbach SL. Medication induced diabetes during in- 2005;45(7):960–963. duction in pediatric acute lymphoblastic leukemia: prevalence, risk factors 29. Lowas SR, Marks D, Malempati S. Prevalence of transient hyperglycemia dur- and characteristics. Support Care Cancer. 2012;20(9):2009–2015. ing induction chemotherapy for pediatric acute lymphoblastic leukemia. 3. Schmiegelow K, Muller K, Mogensen SS. Non-infectious chemotherapy- Pediatr Blood Cancer. 2009;52(7):814–818. associated acute toxicities during childhood acute lymphoblastic leukemia 30. Banihashem A, Ghasemi A, Ghaemi N, Moazzen N, Amirabadi A. Prevalence therapy. F1000Res. 2017;6:444. of transient hyperglycemia and diabetes mellitus in pediatric patients with 4. Roberson JR, Raju S, Shelso J, Pui CH, Howard SC. Diabetic ketoacidosis during acute leukemia. Iran J Ped Hematol Oncol. 2014;4(1):5–10. therapy for pediatric acute lymphoblastic leukemia. Pediatr Blood Cancer. 31. Spinola-Castro AM, Siviero-Miachon AA, Andreoni S, Tosta-Hernandez PD, 2008;50(6):1207–1212. Macedo CR, Lee ML. Transient hyperglycemia during childhood acute lym- 5. Sonabend RY, McKay SV, Okcu MF, Yan J, Haymond MW, Margolin JF. phocytic leukemia chemotherapy: an old event revisited. Clin Adv Hematol Hyperglycemia during induction therapy is associated with poorer survival Oncol. 2009;7(7):465–472. in children with acute lymphocytic leukemia. J Pediatr. 2009;155(1):73–78. 32. Esbenshade AJ,Simmons JH,KoyamaT,Lindell RB,FriedmanDL. Obesity 6. Tieu MT, Lovblom LE, McNamara MG, et al. Impact of glycemia on survival of and insulin resistance in pediatric acute lymphoblastic leukemia wor- glioblastoma patients treated with radiation and temozolomide. J Neurooncol. sens during maintenance therapy. Pediatr Blood Cancer. 2013;60(8): 2015;124(1):119–126. 1287–1291. 7. Pui CH, Burghen GA, Bowman WP, Aur RJ. Risk factors for hyperglycemia in 33. Cetin M, Yetgin S, Kara A, et al. Hyperglycemia, ketoacidosis and other com- children with leukemia receiving L-asparaginase and prednisone. J Pediatr. plications of L-asparaginase in children with acute lymphoblastic leukemia. J 1981;99(1):46–50. Med. 1994;25(3–4):219–229. 8. Roberson JR, Spraker HL, Shelso J, et al. Clinical consequences of hyperglyce- 34. Lebovic R, Pearce N, Lacey L, Xenakis J, Faircloth CB, Thompson P. Adverse mia during remission induction therapy for pediatric acute lymphoblastic effects of pegaspargase in pediatric patients receiving doses greater than 3, leukemia. Leukemia. 2009;23(2):245–250. 750 IU. Pediatr Blood Cancer. 2017;64(10):e26555. 9. Tsai MC, Huang HH, Chou YY, Cheng CN, Chen JS, Lin SJ. Risk factors for hy- 35. Sheean P, Braunschweig C. The incidence and impact of dextrose dose on hy- perglycemia during chemotherapy for acute lymphoblastic leukemia among perglycemia from parenteral nutrition (PN) exposure in hematopoietic stem Taiwanese children. Pediatr Neonatol. 2015;56(5):339–345. cell transplant (HSCT) recipients. JPEN J Parenter Enteral Nutr. 2006;30(4): 10. Warwick AB, Malempati S, Krailo M, et al. Phase 2 trial of pemetrexed in chil- 345–350. dren and adolescents with refractory solid tumors: a Children’s Oncology 36. Sheean PM, Freels SA, Helton WS, Braunschweig CA. Adverse clinical conse- Group study. Pediatr Blood Cancer. 2013;60(2):237–241. quences of hyperglycemia from total parenteral nutrition exposure during Downloaded from https://academic.oup.com/jncimono/article/2019/54/132/5567550 by DeepDyve user on 16 July 2022 138 | J Natl Cancer Inst Monogr, 2019, Vol. 2019, No. 54 hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2006; 43. Busaidy NL, Farooki A, Dowlati A, et al. Management of metabolic effects as- 12(6):656–664. sociated with anticancer agents targeting the PI3K-Akt-mTOR pathway. J Clin 37. Karnchanasorn R, Malamug LR, Jin R, Karanes C, Chiu KC. Association of hy- Oncol. 2012;30(23):2919–2928. perglycemia with prolonged hospital stay but no effect on engraftment after 44. Montero N, Pascual J. Immunosuppression and post-transplant hyperglyce- autologous hematopoietic stem cell transplantation. Endocr Pract. 2012;18(4): mia. Curr Diabetes Rev. 2015;11(3):144–154. 508–518. 45. Shah RR. Hyperglycaemia induced by novel anticancer agents: an undesir- 38. Healy SJ, Nagaraja HN, Alwan D, Dungan KM. Prevalence, predictors, and out- able complication or a potential therapeutic opportunity? Drug Saf. 2017; comes of steroid-induced hyperglycemia in hospitalized patients with hema- 40(3):211–228. tologic malignancies. Endocrine. 2017;56(1):90–97. 46. Hershey DS, Bryant AL, Olausson J, Davis ED, Brady VJ, Hammer M. 39. Pidala J, Kim J, Kharfan-Dabaja MA, et al. Dysglycemia following glucocorti- Hyperglycemic-inducing neoadjuvant agents used in treatment of solid coid therapy for acute graft-versus-host disease adversely affects transplan- tumors: a review of the literature. Oncol Nurs Forum. 2014;41(6):E343–E354. tation outcomes. Biol Blood Marrow Transplant. 2011;17(2):239–248. 47. Barua R, Templeton AJ, Seruga B, Ocana A, Amir E, Ethier JL. Hyperglycaemia 40. Khan KH, Wong M, Rihawi K, et al. Hyperglycemia and phosphatidylinositol and survival in solid tumours: a systematic review and meta-analysis. Clin 3-kinase/protein kinase B/mammalian target of rapamycin (PI3K/AKT/ Oncol (R Coll Radiol). 2018;30(4):215–224. mTOR) inhibitors in phase I trials: incidence, predictive factors, and manage- 48. Storey S, Ah D, Hammer M. Measurement of hyperglycemia and impact on ment. Oncologist. 2016;21(7):855–860. health outcomes in people with cancer: challenges and opportunities. Oncol 41. Goldman JW, Mendenhall MA, Rettinger SR. Hyperglycemia associated with Nurs Forum. 2017;44(4):E141–E151. targeted oncologic treatment: mechanisms and management. Oncologist. 49. Dungan KM, Braithwaite SS, Preiser JC. Stress hyperglycaemia. Lancet. 2009; 2016;21(11):1326–1336. 373(9677):1798–1807. 42. Hwangbo Y, Lee EK. Acute hyperglycemia associated with anti-cancer medi- 50. Adeberg S, Bernhardt D, Harrabi SB, et al. Metformin influences progression cation. Endocrinol Metab. 2017;32(1):23–29. in diabetic glioblastoma patients. Strahlenther Onkol. 2015;191(12):928–935.

Journal

JNCI MonographsOxford University Press

Published: Sep 1, 2019

Keywords: hyperglycemia; childhood cancer

References