Germline Gene Aberrations Are Common in High-Risk Adult and Pediatric Acute Lymphoblastic Leukemia Patients

Abstract

Personalized medicine involves a comprehensive analysis of factors affecting a disease. Family history is an important but not a definitive indicator of inherited predisposition to malignancy and thus studying the germline gene aberrations alongside somatic variants is warranted. The significance of germline predisposition has been increasingly recognized in acute myeloid leukemia and is noted in the latest WHO classification.1,2,3Despite the recent progress in acute lymphoblastic leukemia (ALL) therapies, many adult patients with ALL still do poorly and there is a need for new biomarkers and therapy targets. The aim of our study was to identify and determine the frequency of germline mutations in known ALL genes, to discover new genes associated with ALL predisposition, and to compare the germline genetic background and respective consequences of pediatric and adult high-risk ALL. We examined exome sequencing data from biobanked samples of adult (50) and pediatric (68) patients with high-risk ALL (Finnish Hematological Registry and Biobank - FHRB, and clinical repositories). First, a candidate-gene analysis consisted of 92 genes previously associated with germline predisposition to ALL or syndromes predisposing to ALL. Variants with minor allele frequency of <0.01 in the Genome Aggregation Database were considered. Missense variants were considered significant if ≥2/3 algorithms (CADD, DANN, Revel) classified it as pathogenic. We also reviewed literature, public databases and the American College of Medical Genetics classification (ACMG) in filtering the variants. Clinical characteristics of the patients were retrieved from hospital records and the Finnish Hematological Registry. Second, an unbiased approach was applied to find novel genes predisposing to ALL by checking pathogenic variants in the same gene in at least two (adult/pediatric) patients and filtering by gene ontologies DNA repair, cell cycle, and lymphocyte differentiation; and by COSMIC cancer census genes. In both analyses, only statistically significantly more common variants in our series compared to normal population were included. We also conducted a mutational signature analysis on the samples. Our analysis (Table 1) demonstrates that 8% of adult and 10% pediatric study patients carried a pathogenic or likely pathogenic mutation in their germline in known ALL predisposing genes. All these mutations were at least 30-fold more frequent in our study series compared with allele frequencies in the normal population (p<0.05). Four pediatric patients were identified to suffer from undiagnosed syndromes, which predispose to ALL (Li-Fraumeni and Noonan syndromes). We also found recurring aberrations in new genes with biological relevance to ALL, such as MUTYH and IL21R, potentially associating with ALL predisposition. Final results of the mutational signature analyses are pending. In conclusion, our results emphasize that germline predisposition is not rare among high-risk ALL patients. In addition to pediatric ALL patients, we show contributing germline variants also in adult patients. At least 20% of the adult ALL patients are transplanted and a potential germline basis of the disease should be considered when choosing the donor. Our analysis also reveals new information on the biology of high-risk ALL and may contribute to the future studies seeking for therapy options in this challenging patient category. Despite the anxiety that acknowledging inheritable factors may cause in patients, families, and caretakers, we encourage clinicians to integrate carefully interpreted germline data into patient care.