LAUSR

Non-Hodgkin lymphoma (NHL) can occur at any age, and is common in adolescents and young adults (AYA, age 15–39 years). The clinical outcome of AYA NHL can differ compared to younger and older populations. Treatment challenges include choosing between the standard paediatric versus the adult treatment strategies and a low rate of patient enrolment in clinical trials. In addition, there is very limited information on age-related genomic profiles of NHLs to facilitate the clinical investigation of new treatment therapies specific for the AYA population. Next generation sequencing was performed to identify genomic alterations in 195 NHL patients including 75 AYA and 120 older adults (OA) across 11 types of NHL. The details of disease types and clinical characteristics are shown in Table I. Sequence data were assessed for base substitutions, short insertions/deletions, copy number alterations and gene fusions/rearrangements. The genomic profiling of AYA and OA groups revealed both similarities and differences. There were 4 5 genomic alterations (GA) identified per case in AYA compared to 5 1 GA in OA. The top two most commonly altered genes for both groups were the same, including IGH/IGK/IGL rearrangements (35% AYA vs. 33% OA) and TP53 (27% in both). TNFAIP3 was altered in 17% of AYA and 14% of OA. In diffuse large B cell lymphoma (DLBCL) IGH/IGK/IGL rearrangements were the most frequent alterations in both age groups (42% AYA vs. 36% OA). CDKN2A mutations were more common in OA DLBCL (36% OA vs. 7% AYA). Genes that were more frequently altered in AYA DLBCLs compared with OA DLBCLs were B2M (39% AYA vs. 14% OA), TNFAIP3 (36% AYA vs. 14% OA), CD58 (29% AYA vs. 9% OA), CIITA (19% AYA vs. 2% OA) and PDCD1LG2 (19% AYA vs. 0% OA). Genes that were more frequently altered in OA DLBCLs included KMT2D (also termed MLL2; 31% OA vs. 19% AYA), CDKN2B (26% OA vs. 3% AYA), MYD88 (21% OA vs. 3% AYA), CREBBP (19% OA vs. 7% AYA) and CD79B (16% OA vs. 0% AYA). Clinically relevant genomic alterations, defined as GA associated with approved drugs or novel therapies under evaluation in registered clinical trials, were identified in 59% of AYAs and 70% of OAs. ALK fusion and histone modification (HM) genomic alterations displayed different patterns overall and were identified in various NHL subtypes (Table I). Genomic alterations of many other genes and frequencies are shown in Figures S1–S4, and detailed data are shown in Tables S1–S3. ALK fusions were more frequently seen in AYA (7/75; 9%) compared to OA (1/120; <1%) with a significant Pvalue of 0 006. Further division of the age groups showed even more difference in the rate of ALK fusions. Anaplastic large cell lymphoma (ALCL) is more common in AYA and 4/5 of AYA ALCL patients harboured NPM1-ALK fusions (80%). ALK fusions were not identified in the 2 OA ALCLs (0%) (P = 0 14). Three DLBCL AYA patients (10%) had CLTC-ALK or SEC31A-ALK fusions, and only one single NPM1-ALK was found in 58 OA DLBCLs (2%). All ALK fusions have been previously observed in large B-cell lymphomas (Gascoyne et al, 2003; Van Roosbroeck et al, 2010). In a study of 11 ALK-positive NHL patients, 9 of who were AYA, all demonstrated response or stable disease upon treatment with crizotinib (Gambacorti Passerini et al, 2014). Alterations in HM genes have been commonly reported in many cancer types. In the present study, we observed that these alterations were predominantly seen in OA and much less commonly in AYA. The difference in the alteration frequency of HM genes between AYA (13/75; 17%) and OA (47/ 120; 39%) was highly significant (P = 0 001), and included alterations in ASXL1, CREBBP, EP300, EZH2, KDM5A, KMT2D, KMT2C (also termed MLL3), SETD2, and WHSC1 (also termed MMSET) (Fig 1). The rate of HM gene mutation varied with a further breakdown of age groups. HM gene