BACKGROUND/AIM: Secondary forms of diabetes are often understudied and underdiagnosed in children and adolescents with cancer. The objectives of this cohort study were to study the incidence and risk factors for hyperglycaemia in leukaemia and lymphoma patients. METHODS: The investigators retrospectively collected 15 years of data from paediatric patients treated for acute lymphoblastic leukaemia (ALL), Hodgkin's lymphoma (HL), and non-Hodgkin's lymphoma (NHL) immediately at cancer diagnosis. They studied risk factors for hyperglycaemia in univariate and multivariate analyses.
BACKGROUND: Children and adolescents diagnosed with acute lymphoblastic leukaemia (ALL), Hodgkin's lymphoma (HL), and non-Hodgkin's lymphoma (NHL) are treated with specific and individual chemotherapy protocols sometimes combined with radiotherapy and/or hematopoietic stem cell transplant (HSCT). Thanks to research initiatives allowing constant re-evaluation of these protocols, survival rate of childhood cancer exceeds 83%. However, the effectiveness of these treatments is not without consequences: 50% of childhood cancer survivors (CCS) develop endocrine sequelae including metabolic syndrome and glucose metabolism disorders such as diabetes, insulin resistance and impaired glucose tolerance (IGT). In the general population, diabetes confers a 2 to 3 times increased risk of cardiovascular disease and corresponds to 12-55% of cases of end-stage renal disease worldwide, being as such the 7th expected leading cause of death by 2030. In CCS, the incidence of hyperglycaemia is still ill-defined and might range between 11 and 35% of cases. Moreover, despite the whole body of evidence that asparaginase, steroids and total body irradiation increase the risk of developing hyperglycaemia and diabetes, risk factors are missing and - asides from treatments - understudied (e.g., pre-existing obesity, sex, age, ethnicity, family history of diabetes, etc.). AIM: DIAB-ONCO. The purpose of this study was to assess the incidence and associated risk factors of developing hyperglycaemia in children and adolescents diagnosed with ALL, HL and NHL. Deciphering the factors associated with the onset of hyperglycaemia in paediatric patients treated for cancer will provide leverage for lifestyle or therapeutic intervention from a prevention perspective in newly diagnosed patients. INTERVENTION: The DIABONCO retrospective study is being carried out in collaboration with the Paediatric Haematology and Oncology (Institut Roi Albert II) of Cliniques universitaires Saint-Luc in Belgium (Brussels). The local ethical committee (Saint-Luc and UCL Hospital-Faculty Ethics Committee) approved this study protocol (approval number 2018/20MAR/122) and the study was conducted in accordance with the Declaration of Helsinki. This investigation included patients receiving treatment protocols conferring a diabetogenic risk. This included total body, cranial and abdominal irradiation (respectively TBI, CI and AI), steroids and L-asparaginase. Our cohort was therefore composed of patients treated for acute lymphoblastic leukaemia (ALL), Hodgkin's lymphoma (HL) and non-Hodgkin's lymphoma (NHL). The patients were stratified according to the presence or absence of hyperglycaemia during the treatment protocol and during clinical follow-up, which ended in August 2020. The groups were called the " hyperglycaemia-positive ALL, NHL or HL " and the "hyperglycaemia-free ALL, NHL or HL". INCLUSION and EXCLUSION CRITERIA: All children and adolescents aged 0 to 18 years treated with the aforementioned diabetogenic treatment protocols and diagnosed at Cliniques universitaires Saint-Luc with ALL, NHL or HL between January 2004 and December 2019 were included. Patients with an incomplete file or a history of the following conditions were excluded: previous diabetes (i.e., type 1, type 2, neonatal or monogenic diabetes), pancreatitis, steatosis, Down syndrome, pancreas and liver surgery, kidney disease and previous cancer other than leukaemia and lymphoma.
Cliniques Universitaires Saint-Luc - UCLouvain
Brussels, Belgium
Variables of interests - personal patient data and patient's family history
The investigators collected personal patient data such as sex, date of birth, country of origin, weight (kilograms), height (centimeters) and gestation at birth, complications during pregnancy (pre- or post-term, events, foetal macrosomia), dysmaturity, hypoglycaemia and hyperglycaemia in the neonatal period, the presence or absence of previous overweight (weight and height used to obtain BMI \> 85th centile)/obesity (BMI \> 95th centile), endocrine disease, autoimmune disease, acanthosis nigricans, sickle cell anaemia, any chronic treatment, date of death if patient died. Regarding the patient's family history, the investigators registered the presence or absence of previous gestational diabetes, polycystic ovarian syndrome, infertility, dystocia, consanguinity, diabetes, metabolic syndrome, sickle cell anaemia, pancreatic or liver surgery.
Time frame: Through study completion, an average of two years
Variables of interests - Cancer and treatment data
The investigators also gathered information about the primary diagnosis and its treatment: type of cancer, diagnosis date, stage and localization of the tumour, anthropometric data on diagnosis (weight in kilograms, height in meters), tanner stage, blood pressure (systolic and diastolic), treatments protocols (presence or absence of steroids, asparaginase, radiotherapy and HSCT) and the presence of treatment side effects such as steroid-resistant disease, allergy to asparaginase, pancreatitis and steatosis induced by treatment protocol.
Time frame: Through study completion, an average of two years
Variables of interests - Blood glucose and hyperglycemia data
When the patient developed hyperglycaemia more than twice, the date, the anthropometric data of onset (weight in kilograms, height in meters), the blood pressure, the treatment for the hyperglycaemia (e.g., insulin therapy, metformin), its doses per day, and its duration were reported. To obtain the number of blood glucose levels recorded, all blood glucose measurements from the start of treatment protocol until the end of our study were counted. The duration of blood glucose monitoring was evaluated by counting blood glucose measurements performed without an interruption of more than six months and deceased patients were excluded. To evaluate the percentage of patients having been tested for blood glucose after the maintenance phase, the investigators included only ALL patients treated before 2015 and HL patients treated before 2017 to have a sufficient delay between the end of maintenance phase and the end of our study for the metabolic outcome monitoring.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.
Study Type
OBSERVATIONAL
Enrollment
303
Time frame: Through study completion, an average of two years