Abstract
Introduction: "B-others" is an heterogeneous Ph negative adult ALL subtype that continue to have a poor prognosis and need to be better molecular defined by identifying genetic alterations that predict the risk of treatment failure and developing novel and targeted therapies.
Aims: Analyze adult B-other ALL or "triple negative" (BCR-ABL, E2A-PBX, MLL-AF4) patients (pts) by whole exome sequencing (WES) NGS-target sequencing and by SNP array to discover new altered genes involved in leukemogenesis.
Patients and Methods: 183 samples relative to 136 adult "B-other"-ALL pts, were sequenced by WES (44pts/93 samples), additional 92 B-Other pts were analyzed with NGS target sequencing to validate the 8 most mutated genes (PAX5, JAK2, TP53, PTPN11, CREBBP, PTPN11, PIEZO2, CUL3) based on lower frequency variant detection tool MuTect and 14/44 were also genotyped with SNP arrays (6 SNP 6.0, 8 CytoScan HD). Furthermore 103 B-ALL SNP arrays (79 Ph+; 24 Ph-) were used as a comparative ALL cohort.
MuTect and VARSCAN tools to call mutations (Single Nucleotide Variants=SNVs and/or INDELs) were used and we selected variants with a minor allele frequency (MAF) lower than 0.05 and filtered using dbSNP138. Pathway enrichment analysis (PEA; over-representation test, FDR=0.05) was performed for REACTOME, KEGG and databases.
Results: We identified 1762 SNVs (1515 non-synonymous, 86 frameshift indel,112 nonframeshift indel,46 stopgain and 3 stoploss) in 1683 genes. The average number of somatic SNVs was 38 per case (range 9-112). 80 genes were recurrently mutated in at least 3 pts.
We identified, for the first time, three innovative signaling deregulated in our B-Others cohort: 1) PAK2/FLT3 signaling (i.e.PAK2;CRIPAK;FLT3;MYC;CASP7;APAF1;DLC1;UBC;E2F1;TP53;TLR4;PLEC; PIP4K2C;JAK2;IFNAR2;BPI;CUL3;CNN2;NRAS;PTPN11;MTOR;NF1;KRAS;UBA1EZH2); 2) The "Death receptor signaling" (CYLD;SPPL2A;UBC;TNFRSF1A;BARD1;DDX6;ACTN2;CUL3;AFF4;HDAC6;MTOR;UBXN11;BAHD1; KIF26B;LNX1;EFHC1;CRIPAK;PLEKHN1); 3) Interferon Signaling: (PTPN11 ; JAK2;IFNAR2;TRIM29;STAT2;NEDD4;UBC;FLNB;NCAM1;UBA1; NUPL1;EIF4E).
We found some mutated genes (KRAS, NRAS, PAX5, JAK2, TP53, PTPN11, CREBBP) previously reported to be involved in B-Other ALL, confirming the pivotal roles of these gene in ALL. Regarding the frequency (Tab 1), TP53 was the most mutated gene and that between these gene is the only one associated to a worst OS (p=0.004). PTPN11, involved in interferon signaling , mutational rate is of 11%. By SNP arrays CNA analysis on the 80 recurrently mutated genes revealed that PTPN11 is significantly altered (mainly in LOH) compared Ph- and Ph+ ALL (p=0.03; p=0.01). For the first time we described: ATXN1 gene as mutated, involved on Notch signaling; ANKLE1, an endonuclease, and PKHD1L1 two genes highly mutated but without a clear involvement in canonical pathways. Surprisingly a little known gene PKHD1L1, a homolog of the autosomal recessive polycystic kidney disease gene, is mutated in 7.2% of the pts. CRIPAK is a novel negative regulator of the Pak1 (p21-activated protein kinase 1) was found to be mutated in 14.4% of pts. This gene is an interactor of significantly enriched "Death Receptor Signalling" and "Activation of HOX genes during differentiation" pathways and its role has to be better defined in leukemogenesis.
Conclusions: Point mutations are the prevalent mechanism identified in our pts cohort (88.8%). Analysis of SNVs confirmed mutations in important genes known to be involved in leukemogenesis, like TP53 mutations that significantly affect OS. The impact of PTPN11 high mutation and CNA rate need to be further investigated. The meaning of frequent mutations not previously described has to be evaluate. The found deregulated signaling could be have a prognostic and therapeutic effect, in particular our analyses identified p21-activated protein kinase PAK2 as a novel key nodal point in FLT3 dependent signaling in ALL. Since FLT3 overexpression has been identified as a potential target in a subset of B-other ALL and since FLT3-inhibitors have shown a good efficacy in the Ph+ treatment, our found signature could be a new way to identify FLT3-inhibitors (Ponatinib) in combination with chemotherapy. Supported by: ELN, AIL, AIRC, project Regione-Università 2010-12 (L. Bolondi), FP7 NGS-PTL project, HARMONY project, Fondazione del Monte BO e RA project.
Martinelli: Celgene: Consultancy; Amgen: Consultancy; Johnson&Johnson: Consultancy; Pfizer: Consultancy; Roche: Consultancy; Ariad/Incyte: Consultancy.
