Introduction: Methotrexate (MTX) is an important and effective chemotherapeutic drug in the treatment of pediatric acute lymphoblastic leukemia (ALL). However MTX can induce toxicity, which can lead to amendments of treatment and subsequent impaired survival. The aim of this study was to identify metabolic and genetic determinants of MTX toxicity. Patients and Methods: In this prospective study, 134 Dutch pediatric ALL patients were treated with four high dosages MTX (HD-MTX: 5 g/m2) every other week according to the DCOG-ALL10 protocol. Toxicity was prospectively scored and a National Cancer Institute (NCI) grade ≥3 was considered clinically relevant toxicity. Plasma MTX levels were measured at 24 and 48 hours after each HD-MTX infusion. Erythrocyte folate, plasma folate and plasma homocysteine levels were determined at the start of protocol M. Seventeen single nucleotide polymorphisms (SNPs) in 7 candidate genes in the MTX pathway were analyzed. Results: Grade ≥3 mucositis occurred in 20% of the patients, skin toxicity in 7%, diarrhea in 3%, and neurotoxicity in 3%. Mucositis occurred especially after the first course compared to the other courses (p=0.006). Mucositis was not associated with plasma MTX, plasma folate or plasma homocysteine levels. Patients with mucositis had higher baseline levels of erythrocyte folate (median 1.2 μmol/L vs. 1.4 μmol/L, p<0.008). Wildtype rs7317112 in the ABCC4 gene was the only SNP associated with a higher frequency of mucositis (AA (39%) vs. AG/GG (15%), p=0.016). Conclusion: Mucositis is the most frequent occurring toxicity during the HD-MTX phase in pediatric ALL treatment, and occurs especially after the first MTX course. Only a higher baseline erythrocyte folate, which reflects the accumulation of MTX polyglutamates in mucosal cells, and the wild-type variant of rs7317112 SNP in ABCC4 were determinants of mucositis in pediatric ALL during MTX-HD treatment. Disclosures No relevant conflicts of interest to declare.

BACKGROUND Cure rates of pediatric acute lymphoblastic leukemia (ALL) have reached 90% in the developed countries. However, toxicity due to chemotherapeutic regimens occurs frequently but with great heterogeneity. This suggests that genetic variation is involved. In order to identify determinants of adverse effects, recent studies have investigated pharmacogenetic features in relation to toxicity. Most of these studies examined coding regions of the genome. Recently, it has been described that epi genetic regulators, such as micro-RNA's (miRNA), might also have an important regulatory function in genes involved in drug related toxicity. In a recent study 25 miRNA SNPs were found to be related to toxicity in pediatric ALL treatment (Lopez-Lopez, PLoS ONE, 2014). In pediatric ALL mucositis is one of the most frequent side effects during high dose methotrexate (MTX) treatment. AIM The aim of this study was to detect novel, epigenetic biomarkers that predict MTX related oral mucositis in pediatric acute lymphoblastic leukemia (B- and T cell) by studying single nucleotide polymorphisms (SNP) involved in miRNA levels and function. METHODS DNA was isolated from whole blood of 118 pediatric ALL patients that were treated with high dose MTX (5 gr/m2) according to the Dutch Childhood Oncology Group ALL-10 protocol. The recently published 25 SNPs, involved in miRNA function and located in the DROSHA, CNOT1, CNOT4, EIF2C1, GEMIN3, GEMIN4, MIR604, MIR453, MIR2110, MIR2053, MIR1294, MIR1206, DICER, XPO5 and TNRC6B genes, were selected for genotyping. Toxicity data during the consolidation phase were prospectively collected and documented according to the National Cancer Institute (NCI) v.3.0 score system. Mucositis NCI grade ≥ 3 (grade 3: confluent ulcerations, bleeding with minor trauma), was considered as clinical significant toxicity and was used as endpoint. RESULTS Mucositis was the only recurring toxicity in this prospectively well-documented cohort and therefore used as endpoint of this study. A selection of 20 of the previously identified 25 candidate SNPs was studied based on technical feasibility. In addition, 1 SNP in the XPO 5 gene was not considered for analysis because it was not in Hardy Weinberg equilibrium. Mucositis occurred in 19% of the patients in at least one of the MTX courses. Only the TT genotype of rs11866002 in the CNOT1 (CCR4-NOT complex, subunit 1) gene was associated with a higher risk of developing mucositis (NCI ≥ 3) compared to carries of CC/CT. The other 18 candidate SNPs analyzed did not show statistically significant associations. CONCLUSION The inter-patient variability of mucosal toxicity was not associated with most of our investigated SNPs which are involved in miRNA transcription and function. CNOT1 rs11866002 C>T was the only single nucleotide polymorphism associated with the occurrence of oral mucositis during pediatric acute lymphoblastic leukemia treatment. We acknowledge the Foundation Children Cancerfree (KiKa), Amstelveen, The Netherlands, for funding this research. Disclosures No relevant conflicts of interest to declare.

Background: Osteonecrosis (ON) and decline of bone mineral density (BMD) are serious side effects during and after treatment of childhood acute lymphoblastic leukemia (ALL). It is unknown whether ON and low BMD co-occur in the same patients, and whether these two osteogenic side effects can influence each other’s development during pediatric ALL treatment. Methods: BMD and the incidence of symptomatic ON were prospectively assessed in 466 patients with ALL (4-18 years of age) treated according to the dexamethasone-based Dutch Child Oncology Group-ALL9 protocol. Symptomatic ON was defined as persistent pain in arms or legs not caused by vincristine administration, and confirmed by magnetic resonance imaging. Bone mineral density of the lumbar spine (BMDLS) (n=466) and of the total body (BMDTB) (n=106) were measured by dual X-ray absorptiometry at ALL diagnosis, after 32 weeks of treatment, at cessation of treatment (109 weeks) and 1 year after cessation of treatment. BMD was expressed as age-matched and gender-matched standard deviation scores (SDS; Z-score). Multivariate linear mixed models were adjusted for age at diagnosis. Results: Thirty patients (6.4%) suffered from ON. At cessation of treatment, mean BMDLS was -1.28 SDS (SD: 1.27, n=332; p<0.01) and BMDTB was -0.74 SDS (SD: 1.29, n=65; p<0.01) lower in ALL patients compared to their healthy peers. At baseline, BMDLS and BMDTB did not differ between patients who developed or who did not develop ON (mean BMDLS ON+: -0.90 vs. ON-: -1.14, p=0.36; mean BMDTB ON+: 0.07 vs. ON-: 0.25 p=0.65). At cessation of treatment, patients with ON seem to have a trend for a lower mean BMDLS (ON+: -1.68 vs. ON-: -1.31, p=0.18) and they have a lower mean BMDTB (ON+: -1.91 vs. ON-: -0.59, p=0.01) than patients without ON. Multivariate analyses showed that BMDTB change during follow-up was significantly different for patients with ON than without ON (interaction group time, p=0.04). Between BMD measurements before and after the diagnosis, patients with ON seemed to have a more rapid decline of BMDTB than in patients of the same age without ON (mean BMDTB difference -1.13 vs. -0.62, p=0.10). Conclusion: We conclude that symptomatic ON and low BMD during antileukemic treatment co-occur in pediatric ALL patients. BMD status at ALL diagnosis does not seem to precede ON. However, the development of ON seems to aggravate BMD decline during antileukemic treatment, most likely due to bone destruction and the advised physical immobilization. Disclosures No relevant conflicts of interest to declare.

Background: Body mass index (BMI: kg/m2) and change in BMI during treatment might influence treatment outcome of pediatric patients with acute lymphoblastic leukemia (ALL). However, previous studies in pediatric acute lymphoblastic leukemia reported contradictory results. Therefore, we studiedthe influence of (change in) BMI on treatment outcome in pediatric ALL patients who were treated according to a dexamethasone-based protocol (Dutch Childhood Oncology Group [DCOG] ALL-9). Patients and Methods: Data on body composition were prospectively collected from a cohort of newly diagnosed Dutch pediatric ALL patients (N=762, age 2-17 years), treated from 1997-2004. BMI at diagnosis was expressed as standard deviation scores (SDS) and categorized into underweight (≤–1.8SDS), or normal weight and overweight (>–1.8SDS). BMI loss was defined after 32 weeks of treatment. Dual X-ray absorptiometry scans were performed in a nested single center cohort (n=106) to assess the contribution of %fat and lean body mass to BMI. Multivariate analyses were corrected for age at diagnosis and risk treatment grpup. Results: Multivariate analyses showed that patients with underweight (8%) had an increased risk of relapse (Hazard Ratio (HR) 1.79, 95% CI: [1.04-3.10], and a similar overall survival (HR 1.10 [0.57-2.10]). BMI loss during the first 32 weeks of treatment was associated with a decreased overall survival (HR 2.10 [1.14-3.87]), and a similar risk of relapse (HR 1.27 [0.70-2.30]) compared to patients without BMI loss. Dual X-ray absorptiometry revealed that BMI loss mainly consisted of a loss of lean body mass and gain in %fat. Conclusion: Underweight at diagnosis is associated with an increased risk of relapse and a BMI loss early during treatment is associated with an increased mortality. This suggests that these patients might benefit from exercise interventions and a high-quality nutrient diet during therapy. Disclosures No relevant conflicts of interest to declare.

Purpose: Osteopenia is a potential complication of childhood cancer treatment, but the magnitude of this problem in survivors is unknown. We examined (determinants of) bone mineral density (BMD) status in long-term survivors of adult childhood cancer (CCS). Methods: This retrospective single-center cohort study included 346 subjects with the most common types of childhood cancer. Acute leukemia or lymphoma was diagnosed in 273 (81%) of the patients. All subjects had a median age at diagnosis of 7.0 years (range: 0.1-16.8 years) and a median follow-up time of 16.7 years (range 5.6-39.9 years). Total body BMD (BMDTB) and BMD of the lumbar spine (BMDLS) were measured by dual-X-ray absorptiometry. Osteopenia was defined as BMD standardized deviation score (SDS) below -1. Twelve candidate single nucleotide polymorphisms (SNPs) in 11 genes (COL1A1, TNFSF11, TNFRSF11, TNRFSA11B, VDR, ESR1,WLS, LRP5, MTHFR, MTRR, IL6) were investigated. Results: Survivors had a lower BMDTB and BMDLS (mean SDS: -0.55, p<0.001; and -0.30, p<0.001, respectively) as compared to healthy peers. This was similar in cancer survivors of leukemia and lymphoma (mean SDS: -0.49, p<0.001; and -0.32, p<0.001, respectively). Osteopenia (BMDTB and/or BMDLS) was present in 45% of the survivors (46% of the leukemia and lymphoma survivors). Multivariate logistic regression analyses identified age at diagnosis <12 years, age >30 years at follow-up, male gender, underweight at follow-up, carriers of the minor allele of rs2504063 (LRP5), treatment with cranial-spinal radiotherapy and prednisone as independent prognostic factors for osteopenia. In survivors of leukemia and lymphoma, we identified low BMI at follow-up, carriers the minor allele of rs2504063 (LRP5), treatment with cranial-spinal radiotherapy and cyclophosphamide use as prognostic factors for osteopenia. Conclusions: This large cohort of childhood cancer survivors with the long follow-up identified osteopenia in 45% of CCS. This indicates that greater awareness for low BMD is warranted, especially in survivors who are older than 30 years, male, have underweight, have a genetic predisposition, and who were treated with cranial-spinal radiotherapy and/or steroids. Disclosures No relevant conflicts of interest to declare.