The survival of pediatric Acute Lymphoblastic Leukemia (ALL) has improved dramatically in the last few decades, whicht leads to an increased number of survivors entering into adolescent and adult age groups.[1] There are several reviews and meta-analyses demonstrating higher rates of Insulin resistance, type 2 diabetes and the components of the metabolic syndrome in pediatric cancer survivors compared to the general population.[2],[3][4] A study from Childhood Cancer Survivor Study revealed that survivors demonstrated an increased risk for diabetes mellitus regardless of treatment exposures compared to siblings.[5] There is limited data in literature regarding the prevalence of impaired fasting glucose and insulin resistance in survivors of childhood leukemia from Low middle income countries like India. This study aimed to estimate the prevalence of IR, Impaired fasting glucose (IFG) (prediabetes, and diabetes mellitus) and to investigate the associated risk factors among childhood ALL survivors in India.

Abbreviations

ALL           Acute Lymphoblastic Leukemia

BP             Blood Pressure

BMI         Body Mass Index

CRT       Cranial Radiotherapy

DM         Diabetes Mellitus

FBS        Fasting Blood sugar

HOMA IR Homeostasis Model Assessment Insulin Resistance

IR               Insulin Resistance

This was a prospective cross sectional study conducted in the Department of Pediatric Oncology, Regional Cancer Centre, Trivandrum, Kerala, South India after getting approval from the Institutional Ethical Committee (HEC No. 09/2018).  We treated children with ALL aged ≤14 years according to the modified BFM 95 protocol. (Supp-1)

After completion of treatment, children were routinely followed up every month for the first year, 2 monthly in 2nd year, 4 monthly in 3rd year 6 monthly till 5 years, and annually thereafter. Children diagnosed with ALL at age ≤ 14 years and have completed at least two years of follow- up after treatment completion were enrolled. Survivors with preexisting DM were excluded. At the time of enrollment, weight, height and body mass index (BMI), were recorded. Puberty status was assessed by Tanner stage. Stage I tanner was categorized as prepubertal , stage II, III and IV was categorized as pubertal and stage V was categorized as post pubertal. Weight was measured using an electronic machine and height was measured on the stadiometer. Waist circumference was measured at the midpoint between the lower margin of the least palpable rib and the top of the iliac crest, using a stretch?resistant tape. A family history of DM was defined by the presence of DM (documented history of diabetes or current therapy with oral hypoglycemic agents/Insulin) in parents, grandparents, or siblings in the family.

After an overnight fast of 12 hours, fasting blood samples were collected at time of enrollment. Fasting insulin was estimated using quantitative enhance chemiluminescence immunoassay using Vitros XT 7600 and fasting glucose levels were analyzed by glucose oxidase /peroxidase methods using  dry chemistry method (reagent coated slides)

HOMA-IR was calculated by the formula:

Fasting insulin (µU/mL) × FBS (mg/dL)

405

ALL survivors were considered as insulin resistant if [1-4]:

• Prepubertal - HOMA-IR ≥ 2.5 / Fasting Insulin> 15 µU/mL
• Pubertal - HOMA-IR ≥ 3.16 / Fasting Insulin> 30 µU/ml
• Postpubertal - HOMA-IR ≥ 2.5 / Fasting Insulin> 20 µU/ml
• Prediabetes/ Impaired fasting glucose was defined as fasting blood glucose concentration (FBS) of 100-125 mg/dl while diabetes was defined by FBS of ≥126 mg/dl. Oral glucose tolerance test (OGTT) was not done due to logistic reasons.

Obesity was defined based on Body Mass Index (BMI) (kg/m²) standard deviation scores or z- scores. To classify overweight/ obesity, a BMI WHO z score of ≥1.03 was used in childhood survivors ≤ 19 years old while adult cut off of BMI value ≥ 25 Kg/m² was used for survivors > 19 years old.

Statistical analysis of the data was done by using SPSS Version 11.0. Continuous variables were expressed as Mean ± SD and categorical variables using frequencies and proportions. Association between two categorical variables was assessed using Chi-square/ Fishers exact test. The normality was assessed using the Kolmogorov Smirnov Z test. A comparison between the two groups has been done using Student T-test. ANOVA test was used to compare the HOMA index at different stages of puberty and further Bonferroni multiple comparison tests were done to find the significantly differing groups. The risk factors for Insulin Resistance in univariate analysis were estimated using binary logistic regression and further multivariate logistic regression was done using the forward likelihood ratio method. P-value <0.05 was considered to be statistically significant.

A total of 202 survivors of childhood ALL were enrolled in the study (Figure-1). The survivor cohort included 58.4% males and 41.6% females. The mean age at ALL diagnosis and survivor enrollment was 5.2± 3.2 years and 13.2± 3.8 years respectively. One hundred and fifteen survivors (57%) received cranial radiotherapy. Forty-one (20%) of the survivor cohort was overweight/obese at enrollment. Family history of diabetes was present in 42% (n=84) survivors

(Table-1). The proportions of Tanner staging were not different between groups of girls and boys (p-value= 0.292).

Table 1: Baseline characteristics of childhood ALL survivors in survivors of pediatric ALL (n=202).

All- Acute Lymphoblastic Leukemia, Crt- Cranial Radiotherapy, Mpal- Mixed Phenotypic Acute Leukemia Overweight/Obesity

Forty-one (20%) of the survivor cohort was overweight/obese at enrollment. Mean BMI z score at diagnosis was -1.4± 1.59, which showed a statistically significant increase to -0.26± 1.61 at survivor study enrolment (p-value: 0.001). A significant difference was observed between the BMI Z- scores at the time of diagnosis and at study enrolment in both female (p-value =0.001) and male survivors (p-value = 0.001).

Insulin Resistance, Prediabetes, and Diabetes in Survivors of Pediatric ALL

Seventeen (8.4%) childhood ALL survivors met the criteria for prediabetes. (Table-2) None of the 202 subjects had diabetes based on the fasting glucose value. The prevalence of IR in survivors according to HOMA-IR was 31.2% (n=63). Fasting hyperinsulinemia was detected in 8.4% (n=17) survivors. Female survivors were more likely to have hyperinsulinemia (13.1% vs 5.1%; p-value: 0.001) while male survivors were more likely to have prediabetes(11.9% vs 3.6%; p-value:0.04). (Table-2)Mean fasting insulin and mean HOMA-IR of the survivor population were 11.2± 6.8µU/ml and 2.5±1.58 respectively. A statistically significant difference was observed in the mean insulin levels between males (9.9 ±5.6) and female (12.9± 8.07mIU/L) survivors (p-value: 0.002).Similarly, the mean HOMA-IR was elevated (2.85±1.85) in females as compared to males (2.25±1.35)(p-value: 0.009).

Table 2: Impaired fasting glucose, Fasting Insulin and Insulin resistance among pediatric.

Impact of Puberty on IR

There was no significant difference in mean HOMA-IR values between different pubertal stages in boys and girls (Table-2). But, among pubertal stage, there was a significant difference between the different Tanner stages with respect to HOMA IR for girls. Hence we did Bonferroni multiple comparison tests, which showed the mean difference between Tanner stages II (Mean (SD)=2.194(1.426)) and III (Mean (SD)=4.125(3.091) )was statistically significant (p- value:0.047) while that between Tanner Stage I (Mean (SD) =2.263(1.851)) and stage III (Mean (SD) =4.125(3.091) )was only marginally significant(p-value =0.096)

The mean age of the Insulin resistant survivor was 14.2 ± 3.4 years and 12.7 ±4 years for insulin sensitive survivors (p-value:0.007). Among survivors with IR, the mean value of fasting blood sugars and BMI z score at survivor enrollment was significantly higher (p-value: 0.034 and 0.001 respectively) as compared to Insulin sensitive group (Table 3). Insulin resistant survivors received higher cumulative dose of steroids and higher cumulative dose of L Asparaginase as compared to Insulin sensitive survivors.

Table 3: Comparison of the clinical characteristics and laboratory parameters of survivors of pediatric ALL with Insulin Resistance and survivors with Insulin Sensitive.

BMI- Body Mass Index, FBS -Fasting Blood sugar, SD- Standard deviation.

Risk Factors for Developing Insulin Resistance on Univariate and Multivariate Analysis

The impact of demographic, anthropomorphic, and other ALL related variables on the presence of IFG was explored On the univariate analysis, variables significantly associated with a higher risk of IR were overweight/obese survivors (OR=3.865, 95%CI=1.895- 7.885), older age at survivor enrollment (OR= 1.009, 95%CI=1.002 -1.015) and longer follow up duration ( ≥ 5 years) (OR=1.918, 95% CI=1.050-3.504).

On multivariate analysis, only older age at enrollment (OR=1.012, 95% CI: 1.004- 1.019; p- value=0.002) and overweight or obesity at survivor enrollment (OR=4.985, 95% CI: 2.319- 10.718; p-value=0.001) were significant predictors of IR in pediatric ALL survivors. We did not detect any significant influence of gender, ALL risk category, cranial irradiation and family history of DM diagnosis on IR (Table-4). Similarly, prediabetes in survivors was not found to significantly influence IR (p-value: 0.703).

Table 4: Univariate and multivariate analysis for risk factors associated with Insulin Resistance in survivors of paediatric Acute Lymphoblastic Leukemia.

ALL- Acute Lymphoblastic Leukemia, BMI -Body mass index, OR- Odds Ratio. CI- Confidence Interval

Improvement in the cure rate of pediatric ALL is a blessing to patients and families and gratifying to physicians, but this advancement has come at a significant price in terms of short term and long term morbidities. Survivors have increased risk of long term morbidities including obesity, insulin resistance and diabetes. Twenty six survivors (12.8%) were obese and 15 survivors (7.5%) were overweight in our study. A metanalysis done by Midha et al. which includes 92,862 school going Indian children (age= 2 to 18 years) showed the prevalence of obesity to be 3.39% . As our study lacks a control arm, this meta-analysis was used to compare the prevalence of obesity among our survivor cohort and among the normal Indian children population. The results of our study also showed an increased prevalence of obesity in childhood leukemia survivors as compared to the normal population.

In the present study, we observed that 8.4 % of childhood ALL survivors had impaired fasting glucose. Ranjani et al conducted a community-based cross-sectional survey in children (6-11 years old) and adolescents (12-19 years old) in 1519 urban south Indian children (777 (51%) boys and 742(48.8%) girls) and revealed that 3.7% children had glucose intolerance (Four (0.3%) with diabetes and 52 (3.4%) with prediabetes (impaired fasting glucose and/or impaired glucose tolerance). [10] The study population was comparable to our survivor cohort in terms of ages of the subjects and gender proportion. Our finding of 8.4 % IFG was higher than the prevalence observed in this population-based study. This population based study used Oral glucose tolerance test (OGTT) along with fasting glucose, for detection of prediabetes and diabetes. But, we used only fasting glucose level in our study. In another study from northern India, by Mohapatra et al (n= 75 ALL survivors (Median age±SD = 11.9 (± 2.8)) similar prevalence of 13% was reported using fasting glucose [11].

Frank diabetes was not seen in any of our childhood ALL survivors which is similar to observation by Karakurt et al.[12]Another recent report on 1,044 adult survivors of childhood cancer (median attained age: 33.97 years) enrolled in the St. Jude Lifetime Cohort Study reported that 7.47 % met criteria for Diabetes. Head on comparison is not meaningful because these survivors were much older than our survivor cohort (33.97 ± .14 vs. 13.2 ± 3.8 years).

Insulin resistance was present in 31.2% of patients, which is greater than the reported prevalence of 17 % among survivors in the Northern India and in the earlier population-based study from South India, in which the prevalence was 6 %. [11],[10]A study from St Jude by Nottage et al and from Turkey by Karakurt et al (n=44) reported a prevalence of 49.9% and 27.2% respectively, nearly similar to the observation in our study.[13][12]The better socio economic standards & living conditions in the state of Kerala compared to other Indian states could have contributed to this difference [14].

Males had 3.8 times higher risk of developing prediabetes than females in our population. However, other studies did not find any association of gender and Prediabetes.[2],[3] Cranial Radiotherapy was considered as a risk factor for IR and IFG in different studies, in historic times where higher dose of CRT was used.[15] But in our study, there was no significant difference in the fasting insulin and HOMA-IR of survivors who received cranial radiotherapy and those who did not. The absence of effect of CRT on fasting insulin and HOMA-IR could be attributed to the lower doses of CRT used (only one patient received CRT of 24 Gy) and also partly to the younger age of the survivor cohort in our study.

We observed that the survivors with a longer follow up of ≥ 5 years had increased insulin resistance as compared to survivors with shorter follow up of <5 years( p= value= 0.03) in univariate analysis. It could be explained by the fact that with longer follow-up, most of the children would have reached puberty and pubertal state increases the IR; leading on to insulin- resistant state. Hence these survivors need to be closely followed up further to assess the precise pattern of IR when they get older.

Current study demonstrated that survivors of childhood ALL, who were overweight/ obese were more likely to be insulin resistant than their counterparts, which was in concordance with the study done by Moran et al from USA [16]. In keeping with the previous report, we also found the increased risk for IR among survivors who were older at enrolment [17]. We did not find the impact of puberty on insulin resistance in our survivors. But, Karakurt et al from Turkey(n= 44)showed that the incidence of IR significantly increased with the progression of the pubertal stage, although, the gender-wise impact of puberty on HOMA IR was not separately analyzed in their study [12]. We observed a gender wise difference in the form of an increase in mean HOMA IR in Tanner stage 3 compared to Tanner stage 2 and 1 in girls.

Studies have shown that higher doses of Cranial radiotherapy may induce growth hormone deficiency, which may cause insulin resistance and diabetes mellitus [18]. We have not measured GH levels of our survivor cohort but only a few (one patient received CRT of 24 Gy) of our survivors have received higher doses of CRT (> 18 Gy)

 The strength of our study is that we could include a large number of ALL survivors (n=202) for a reasonable duration of follow up (mean follow up duration= 5 year 5 months).Our study has some limitations. First, oral glucose tolerance test (OGTT) was not performed. Our study did not have a control group, however we compared the results of our cohort with a large population based study (n= 1519) from South India. Although, the gold standard test for evaluating IR is the euglycemic hyperinsulinemic clamp, its use is limited to clinical practice owing to the time and cost involved. Hence, we used HOMA IR, which is a simpler, less invasive and reliable technique to estimate IR.

Prevalence of Insulin Resistance and prediabetes were higher in Indian survivors of childhood ALL. Males have increased risk of developing prediabetes. Although none of our survivor was diabetic, 8.4 % of survivors were prediabetic, suggesting that this outcome will become more prevalent as survivors age. There was an increased prevalence of IR among overweight or obese survivors. A longer follow up is required to know the precise pattern of IR in peripubertal adolescents. Our study stresses the need to periodically screen pediatric ALL survivors especially the older/ overweight/obese survivors for IR.

Patients Consent

 Informed consent/assent was obtained for the study.

Ethical Approval

The study was approved by Institutional Ethical Committee (HEC No. 09/2018)

Conflict of Interest

Authors disclose no conflict of interest.

Financial Support

This work was funded and supported by the Institutional fund of the Regional Cancer Centre.

We acknowledge the entire nursing team for the excellent state-of-the-art clinical care given to the patients.

1. Arora RS, Arora B. Acute leukemia in children: A review of the current Indian data. South Asian J Cancer. 2016; 05: 155-160.
2. Williams HE, Howell CR, Chemaitilly W, Wilson CL, Karol SE, Nolan VG, et al. Diabetes mellitus among adult survivors of childhood acute lymphoblastic leukemia: A report from the St. Jude Lifetime Cohort Study. Cancer. 2020; 126: 870-878.
3. Neville KA, Cohn RJ, Steinbeck KS, Johnston K, Walker JL. Hyperinsulinemia, impaired glucose tolerance, and diabetes mellitus in survivors of childhood cancer: Prevalence and risk factors. J Clin Endocrinol Metab. 2006; 91: 4401-4407.
4. Oeffinger KC, Adams-Huet B, Victor RG, Church TS, Snell PG, Dunn AL, et al. Insulin resistance and risk factors for cardiovascular disease in young adult survivors of childhood acute lymphoblastic leukemia. J Clin Oncol. 2009; 27: 3698-3704.
5. Meacham LR, Sklar CA, Li S. Diabetes mellitus in long-term survivors of childhood cancer - Increased risk associated with radiation therapy: A report for the childhood cancer survivor study. Arch Intern Med. 2009; 169:1381-1388.
6. Madeira IR, Miranda Carvalho CN, Gazolla FM, De Matos HJ, Borges MA, Bordallo MAN. Ponto de corte do indice Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) avaliado pela curva Receiver Operating Characteristic (ROC) na deteccao de sindrome metabólica em crianças pre-puberes com excesso de peso. Arq Bras Endocrinol Metabol. 2008; 52:1466-1473.
7. Keskin M, Kurtoglu S, Kendirci M, Atabek ME, Yazici C. Homeostasis model assessment is more reliable than the fasting glucose/insulin ratio and quantitative insulin sensitivity check index for assessing insulin resistance among obese children and adolescents. Pediatrics. 2005; 115.
8. Singh Y, Garg MK, Tandon N, Marwaha RK. A Study of insulin resistance by HOMA-IR and its cut-off value to identify metabolic syndrome in urban Indian adolescents. JCRPE J Clin Res Pediatr Endocrinol. 2013; 5: 245-251.
9. Reaven GM. Role of insulin resistance in human disease. Diabetes. 1988; 37:1595- 1607.
10. Ranjani H, Sonya J, Anjana RM, Mohan V. Prevalence of glucose intolerance among children and adolescents in urban South India (ORANGE-2). Diabetes Technol Ther. 2013; 15:13-19.
11. Mohapatra S, Bansal D, Bhalla AK, Verma Attri S, Sachdeva N, Trehan A, et al. Is there an increased risk of metabolic syndrome among childhood acute lymphoblastic leukemia survivors? A developing country experience. Pediatr Hematol Oncol. 2016; 33: 136- 149.
12. Karakurt H, Sarper N, Kiliç SÇ, Gelen SA, Zengin E. Screening survivors of childhood acute lymphoblastic leukemia for obesity, metabolic syndrome, and insulin resistance. Pediatr Hematol Oncol. 2012; 29: 551-561.
13. Nottage KA, Ness KK, Li C, Srivastava D, Robison LL, Hudson MM. Metabolic syndrome and cardiovascular risk among long-term survivors of acute lymphoblastic leukaemia - From the St. Jude Lifetime Cohort. Br J Haematol. 2014; 165: 364-374.
14. International Institute for Population Sciences (IIPS) and ICF. 2018. National Family Health Survey.
15. Surapolchai P, Hongeng S, Mahachoklertwattana P, Pakakasama S, Winaichatsak A, Wisanuyothin N, et al. Impaired glucose tolerance and insulin resistance in survivors of childhood acute lymphoblastic leukemia: Prevalence and risk factors. J Pediatr Hematol Oncol. 2010; 32: 383-389.
16. Gurney JG, Ness KK, Sibley SD, O'Leary M, Dengel DR, Lee JM, et al. Metabolic Syndrome and Growth Hormone Deficiency in Adult Survivors of Childhood Acute Lymphoblastic Leukemia.
17. Moran A, Jacobs DR, Steinberger J, Hong CP, Prineas R, Luepker R, et al. Insulin resistance during puberty: Results from clamp studies in 357 children. Diabetes. 1999; 48: 2039-2044.  
18. Sklar C, Mertens A, Walter A, Mitchell D, Nesbit M, O'Leary M, et al: Final height after treatment for childhood acute lymphoblastic leukemia: comparison of no cranial irradiation with 1800 and 2400 centigrays of cranial irradiation. J Pediatr. 1993; 123: 59- 64.