Common Genetic Variants in Trypsin Regulating Genes Are Associated with AsparAginase-Associated Pancreatitis in Children with Acute Lymphoblastic Leukemia: A Ponte Di Legno Toxicity Working Group Study

Background: Asparaginase-associated pancreatitis (AAP) is a well-known toxicity of childhood acute lymphoblastic leukemia (ALL) therapy. Recent multi-trial group phenotyping of 465AAP caseshas documented severe complications to AAP, including 8% risk of needing assisted mechanical ventilation, 26% risk of developing pancreatic pseudocysts and 9% risk of developing persisting diabetes ( Wolthers et al. Lancet Oncology, 2017) . Investigation of host genome variation associated with AAP has been limited by varying phenotype definition, inclusion criteria and small study sizes. Objectives and Methods: To investigate genetic variants associated with risk of developing AAP, this genome-wide association study reports data on 1544 children (1.0−17.9 years) from 10 ALL trial groups treated with ALL from January 2000−January 2016. The Ponte di Legno toxicity working group consensus definition (Schmiegelow et al. Lancet Oncology, 2016) was used to diagnose AAP: At least two of i) amylase, pancreatic amylase, or pancreatic lipase >3x upper normal limit (UNL), ii) abdominal pain, iii) imaging compatible with AAP. Controls included children treated for ALL with verified completion of intended asparaginase therapy, 78% of whom (1024/1320) received at least 8 injections of PEG-asparaginase without developing AAP. Germline DNA obtained after clinical remission was genotyped on Illumina Infinium Omni2.5exome-8 BeadChip arrays. Association analyses were done in PLINK and annotation in Ensembl. Results: Of 1564 patients passing genotype quality control, 244 had AAP. 205 of 244 (84%) of cases and 1185/1320 (90%) of controls were of European ancestry. Median age was 8.1 years (IQR 4.3−13.1) and 5 (IQR: 3−9) for cases and controls, respectively. After filtering, 1401908 single nucleotide polymorphisms (SNPs) with a minor allele frequency above 1% were analyzed. In logistic regression analysis, adjusting for age and ancestry, the variant rs62228256 (reference allele=C, minor allele=T (C>T)) on 20q13.2 had the strongest association to AAP (OR=3.75; 95% CI 2.33−6.04; p=5.2∙10 -8 ). rs62228256 is located in a non-coding region without known regulatory effects. rs13228878 (A>G; OR=0.61; 95% CI 0.5−0.76; p=7.1∙10 -6 ) and rs10273639 (C>T; OR=0.62; 95% CI 0.5−0.77; p =1.1∙10 -5 ) were among the top 30 SNPs most significantly associated to AAP. They are in high linkage disequilibrium (R 2 =0.94) and located in the PRSS1-PRSS2 locus on chromosome 7. The rs13228878 A risk allele was not associated with level of amylase (p=0.1) or lipase (p=0,68) at diagnosis of AAP, age at diagnosis of AAP (p=0.63), or risk of pseudocysts (p=0.78). Using identical diagnostic criteria for pancreatitis, the major C allele in rs10273639 has been associated with pancreatitis risk in adults ( Whitcomb et al. Nature Genetics, 2012; Masson et al. Gut, 2017 ) with identical risk allele and similar odds ratios. PRSS1 and PRSS2 encode cationic and anionic trypsinogen, respectively. rs10273639 is an expressive quantitative locus for PRSS1 and the C risk-variant is associated with elevated expression of trypsinogen in pancreatic tissue. Gain of function mutations in PRSS1 , known from hereditary pancreatitis, lead to increased autoactivation, increased intra-acinar trypsin levels, and increased risk of auto-digestion leading to pancreatitis. Further investigation of previously validated SNPs known to regulate trypsin activation gave the following results for associations with AAP; rs17107315 in pancreatic secretory trypsin inhibitor ( SPINK1; OR=2.87; 95% CI 1.36−5.8; p=4∙10 -3 ), rs10436957 in chymotrypsin C ( CTRC ; OR=0.69; 95% CI 0.53−0.89; p=5∙10 -3 ) and rs4409525 in Claudin-2 ( CLDN2 ; OR=1.41; 95% CI 1.08−1.83; p=1∙10 -2 ). In total, 207 out of 244 cases were homozygous for the risk allele in rs13228878 (n=104), rs17107315 (n=1), rs10436957 (n=165) and/or rs4409525 (n=16). However, no significant additive effect of having more than one risk allele was found. Conclusion: Children who develop AAP possess the same pancreatitis risk variants as adults with non-asparaginase associated pancreatitis. This shared genetic disposition may facilitate research into pathogenesis and identification of effective interventions towards AAP. Disclosures No relevant conflicts of interest to declare.