ALBERT

Description: Abstract 3508 Background: Krüppel-like factor 5 (KLF5) is a transcription factor with regulatory roles in cell growth, survival, and differentiation. Our previous studies have indicated that KLF5 is required for myeloid blood lineage differentiation and is a target of genomic methylation in Acute Myeloid Leukaemia (AML) which is associated with reduced expression. Further, we have shown that treatment of leukaemia cell lines, displaying KLF5 hypermethylation, with demethylating agents is associated with induction of KLF5 expression, supporting an inhibitory role of methylation on expression (Diakiw et al, Leuk Res 36:110). Other recent studies have shown that KLF5 expression is reduced by the presence of a heterozygous genotype at a promoter SNP (rs3812852) and plays a role in pathogenic processes in heart and brain. We investigated both methylation and SNP genotype as mechanisms of KLF5 gene regulation in AML and their relationship to patient outcome. Methods: We analysed 232 diagnostic bone marrow mononuclear cell samples from a retrospective cohort of patients with AML. The genotype of the rs3812852 SNP was determined by Sequenom MassARRAY genotyping. DNA methylation of the KLF5 gene locus (intron 1) was assessed using Sequenom MassARRAY epityper. KLF5 gene expression was determined for 85 of these samples using quantitative real-time PCR. Of the 232 patients, 161 received treatment with induction chemotherapy and were included in survival analysis. Results: We found that 28 (12%) patients were heterozygous (G/A) for the SNP (rs3812852) in the KLF5 promoter, with one patient having the rare homozygous genotype (G/G) and 204 (88%) patients the common homozygous genotype (A/A). KLF5 gene expression correlated with the SNP genotype, with the heterozygous patients displaying significantly reduced KLF5 expression compared to normal controls (CD34 cells from normal donors, p=0.044). Consistent with our previous study on a smaller test cohort we observed that hypermethylation of this region of KLF5 is common and is present in 50% of samples in our larger cohort. AML patients with high methylation of KLF5 had significantly reduced KLF5 expression compared to normal controls and AML patients with low KLF5 methylation (p=0.032 and p=0.048 respectively). The correlation of gene expression with the SNP genotype and KLF5 methylation groups was also true when considered together, as patients with the heterozygous genotype (A/G) and high methylation had lower KLF5 expression than patients with the common homozygous genotype (A/A) and low KLF5 methylation (Figure 1A). Cox multi-variate analysis (variables of age, WCC and mutational status of FLT3-ITD, IDH1/2, NPM1) of survival of patients treated with induction chemotherapy showed that the presence of a G/A genotype at the KLF5 promoter SNP was not an independent predictor of overall survival (p= 0.502). In contrast hypermethylation of KLF5 intron 1 was a highly significant independent predictor of poorer overall survival (median survival 10.4 months versus 20.7 months, HR 1.5, p=0.001, Figure 1B). Importantly, when the heterozygous SNP genotype (A/G) and high methylation status were combined, this group of patients had even poorer OS (median survival 5.2 months, HR 3.6, Figure 1C) and this was significant compared to the group with homozygous SNP genotype (A/A) and low methylation (p=0.002). Conclusions: We have shown that both methylation of KLF5 intron 1 and a heterozygous genotype at a promoter polymorphism are associated with reduced KLF5 expression in AML. KLF5 hypermethylation is an independent predictor of poor prognosis and combined with the heterozygous SNP genotype defines a group with particularly poor survival indicating a connection with KLF5 gene expression and response to treatment. Importantly, as the majority of the effect is associated with hypermethylation of KLF5, these patients (50%) may benefit from treatment with demethylating agents. Disclosures: Wei: Celgene: Research Funding.

Description: BACKGROUND: We have previously shown that one target of hyper-methylation in AML is the promoter of the tumour suppressor and stress-response mediator Growth Arrest and DNA Damage inducible 45A (GADD45A) (GADD45AmeHI; 42% of AML). In mice Gadd45a has recently been shown to play a critical role in HSC stress responses. Gadd45a deficiency leads to enhanced HSC self-renewal, DNA damage accumulation in HSC, increased susceptibility to leukemogenesis, and impairment in HSC apoptosis after genotoxic exposure (Chen et al, Blood 2014). These findings suggest that hypermethylation of the GADD45A gene may play an important role in the altered properties of HSC, leukaemic initiation and progression. Promoter hypermethylation of this gene defines a patient group with poor survival on standard therapy (Perugini et al, Leukaemia 2012). To explore further the molecular basis of the GADD45AmeHI patient group weperformed genetic profiling of diagnosis samples using a Sequenom multiplex mutation panel, or using whole exome sequencing for broader coverage (n=95 patients).Sequenom MassARRAY was used for quantitative detection of GADD45A promoter methylation in patient samples. For a cohort of matched diagnosis and relapse samples we used CpG methylation data for GADD45A determined by ERRBS (Akalin et al, PLoSGenetics 2012). Response to cytotoxic drugs and assessment of drug combinations with 5-Aza-deoxycytidine (decitabine, DAC) and anthracycline (Daunorubicin, DNR) was performed in AML cell lines, and with primary leukemic cell populations. RESULTS: The association of the GADD45AmeHI patient group with poor outcome was validated in an independent AML patient cohort of 48 patients from the Alfred Hospital, Melbourne, Australia (p=0.003; HR3.35). Whole exome sequencing and Sequenom multiplex analysis of 95 AML patients revealed a striking co-occurrence of the GADD45AmeHI phenotype with mutations in IDH1, IDH2, and TET2 (p

Description: Introduction - AML is a complex group of malignancies, with heterogeneity in morphology, cytogenetics, molecular characteristics, aggressiveness and importantly, in its response to treatment and survival outcomes. Next generation sequencing by the Cancer Genome Atlas Research Network analysed 200 primary AML cases and identified 23 genes that display recurrent somatic mutations at varying frequency in AML (NEJM 368(22):2059-2074). Defects in DNA repair are frequently identified in treatment-related AML and inherited mutations in genes of DNA repair pathways predispose patients to myeloid malignancies. For example, biallelic mutations in FANC genes, which cause the recessive heritable bone marrow failure syndrome Fanconi Anaemia (FA) are associated with high risk of progression to AML and other cancers (Kutler et al.Blood, 101:1249-1256), suggesting a potential involvement of FANC gene mutations in AML pathogenesis. Methods - In this study we present a two-stage approach to gene discovery in AML: initial unbiased whole genome sequence (WGS) and whole exome sequence (WES) analysis of tumour DNA from a cytogenetically normal AML case at diagnosis and relapse, and corresponding germ-line DNA (prepared from mesenchymal stromal cells). Potential oncogenic mutations and changes associated with disease progression were identified. WES of a further 96 diagnostic AML samples further defined recurrent mutations and allowed identification of affected functional groups and networks in AML. Results – WGS and WES were performed on diagnosis, non-haematopoietic and relapse samples from an index AML patient. Somatic SNVs and indels unique to the tumour samples include a number of variants in genes previously reported as recurrently somatically mutated in AML including FLT3, WT1 and IDH2. Somatic mutations in genes not previously associated with AML were also identified including a mutation in FANCD2 (p.S1412N) present in the index AML tumour DNA at diagnosis and at relapse. Variants in genes recurrently mutated at low frequency in AML can also be disease drivers, however separating such genes from the background level of mutation in AML requires analysis across multiple samples, and sequencing studies to determine recurrence and/or mutations in proteins involved in the same functional pathway or complex. STRING-db v9.05 (Franceschini et al. NAR, 2013(41), Database issue) was used to identify a larger network of proteins, including and associated with the FANC genes, involved in homologous recombination-mediated DNA repair. Known somatic mutations from other AML studies were mapped onto this network; as shown in Figure 1 multiple genes in this extended network are affected by somatic mutation in AML suggesting a potential role in pathogenesis. Analysis of our WES data from diagnosis samples from a further 96 Australian AML cases identified an additional two somatic mutations in genes from the extended STRING-db v9.05 FANC network. In total we identified 18 mutations in the 16 classified FANC genes and 8 variants in the BLM complex as shown in Figure 2. Two of the germline FANC gene mutations, FANCM-Q13333fs and FANCD2-R926X, are known pathogenic mutations in FA. Patients with mutations in the 8 FANC genes of the core complex form a distinct subset from those with mutations in the other 8 FANC genes. 5 of the 8 patients with mutations in the BLM complex also form a separate group while BLM complex mutations are present in 2 patients that also have FANC mutations. For the two patients with acquired changes the allele frequency for these FANC mutations is greater than 25% suggesting an early origin in disease. Discussion. Our findings suggest that germline and somatic mutations affecting function of the FANC DNA repair pathway may be a recurrent abnormality in AML, potentially contributing to leukaemogenesis. FANC/BLM gene mutations frequently co-exist with mutations in DNMT3A and DNMT1; 46% of the patients with DNMT3A/DNMT1 mutations are also mutant for FANC or BLM complex genes representing significant over-representation (p = 0.021). Within the group of FANC and BLM patients there is also significant under-representation of FLT3-ITD mutations and mutations in N-RAS and K-RAS (p = 0.051), raising the possibility that defects in homologous DNA repair may favour cooperation with alternative signalling pathways. Figure 1 Figure 1. Figure 2 Figure 2. Disclosures No relevant conflicts of interest to declare.

Description: Abstract LBA-3 Mutations in the transcription factor genes, RUNX1 and CEBPA, can lead to an autosomal dominant familial predisposition to MDS/AML. Using a candidate gene approach, we have detected domain specific heterozygous mutations in the GATA2 gene in 4 MDS/AML families which predispose to MDS/AML. The same novel heterozygous T354M missense mutation was observed in 3 families and a 355delT mutation in 1 family, all with multigenerational transmission of MDS and/or MDS/AML. Importantly, these genetic variants segregate with all affected members in each of the families. The 2 mutated threonine residues are in 5 consecutive highly conserved threonine residues at the DNA-binding, protein-protein interacting second zinc finger (ZF2) of GATA2. Neither these mutations, nor any other variants in the GATA2 coding sequence, were seen in a population screen of 695 normal individuals. Haplotype analysis suggests that the T354M mutation has multiple ancestral origins. While mutations in RUNX1 and CEBPA, can also lead to familial predisposition to MDS/AML, these patients with GATA2 mutations are unique in that there is no obvious pre-MDS or pre-leukaemic phenotype such as thrombocytopenia (RUNX1) and eosinophilia (CEBPA) in predisposed carriers. Most patients in these families have had a rapid disease course “appearing out of the blue” leading to death, with a variety of ages of onset from teenagers to early 40s. Yet remarkably, there are still asymptomatic carriers in their 60s. One of these carriers, and his 2 children, has had bone marrow prophylactically stored over 15 years ago in case of disease onset. No pathogenic GATA2 coding sequence changes were found in 268 sporadic MDS/AML patient samples. Additionally, GATA2 mutations were not found in germline samples from 35 other families predisposed to AML and various other hematological malignancies. Both the T354M and 355delT mutants appear to localize appropriately to the nucleus and maintain at least some DNA binding in electrophoretic mobility shift assays. We used the known murine Gata3 ZF2 structure bound to DNA to model the effects of the observed mutations and demonstrated that the T354 residue does not contact DNA but makes polar contact with the adjacent threonines, and via its amino group, with C349 which coordinates the zinc atom. Replacement of the T354 side-chain with the bulky methionine moiety may affect the zinc contacts and is predicted to alter the overall structure of this ZF2. In contrast, 355delT will shorten the conserved threonine string which is predicted to impact on the orientation and position of L359 which directly contacts DNA. Thus, 355delT is likely to have an effect on DNA binding. Luciferase reporter assays indicate that T354M and 355delT greatly reduce the transactivation ability of GATA2 on multiple response elements, impacting on downstream target genes such as RUNX1 and CD34. Of note, T354M shows a markedly lesser synergistic effect than wildtype (WT) GATA2 with PU.1 on the CSF1R promoter. Competition assays show that these mutations may be acting in a dominant negative fashion in some biological contexts. In stable promyelocytic HL-60 cell lines expressing regulatable GATA2 (WT or T354M), T354M allows proliferation to proceed even under stimulus to differentiate with all-trans retinoic acid. Microarray studies indicated that the down regulation of proapoptotic BCL-xS by T354M, but not WT, may be responsible for this phenotype. GATA2 is considered to be a hematopoietic “stemness” gene, highly expressed in haematopoietic stem cells and is required for megakaryocyte and mast cell production. GATA2 is down regulated during myeloid differentiation and forced overexpression prevents such differentiation. Discovery of GATA2 mutants in MDS/AML predisposed families provides new tools for probing the mechanism of GATA2 induced leukemogenesis, and possibly also for clarifying its role in maintenance of stemness. Our findings highlight the power of investigating familial predispositions to cancer identifying specific mutations with unique biological effects. They have immediate implications for diagnostic genetic testing, and longer term therapeutic implications through identification of drugable biological pathways such as apoptosis. The poor outcome associated with these mutations may suggest that an aggressive strategy is appropriate in the treatment of affected individuals in families found to be carrying GATA2 mutations. Disclosures: No relevant conflicts of interest to declare.

Description: Background: Majority of MDS cases appear to be sporadic in nature, but 10-15% have clear familial basis due to predisposing mutations in genes such as RUNX1, GATA2, CEBPA and DDX41. Contribution of germline variants in sporadic MDS is not studied. This study attempts to address the contribution of germline variants in MDS pathogenesis. Methods: We performed amplicon-based massively parallel sequencing (AmpliSeq custom panel adapted for Illumina HiSeq2500 sequencing) on all coding regions of 29 myeloid genes for 144 MDS samples. After identifying the variants in five genes (TET2, MET, GATA2, ASXL1, NOTCH1), we tested an additional 96 MDS samples including therapy-related myeloid neoplasm (T-MN) using a Sequenom assay. We also analyzed WES data for these variants in 178 AML samples and 758 normal controls and AmpliSeq data for ASXL1 and TET2 variants in 655 CML samples. Results: Collation of all coding variants in the 29 myeloid genes sequenced identified germline variants occurring in primary MDS at frequencies significantly higher than expected when compared to the normal population (ExAC and matched cohort were similar) (Table 1). These variants occurred in 5 genes (TET2, MET, GATA2, ASXL1 and NOTCH1) at increased frequencies of 1.5-16.6 fold. Numerous MDS samples had multiple variants (4 with 4 variants, 4 with 3 variants, 18 with 2 variants) while 70 had 1 variant. The 3 germline MET variants have been previously investigated in solid tumorigenesis and likely generate MET variant proteins that contribute to numerous cancer types including MDS. Interestingly, 7/17 (41%) MDS cases with germline MET variants also had other cancers including pancreatic, gastric and laryngeal cancers. Of the TET2 variants, Y867H and P1723S were concurrent in 5 MDS, 5 AML and 6 CML samples indicative of them being on the same allele (i.e. a haplotype). They were seen at higher than normal frequency in MDS and AML, but were not significantly enriched in CML. We are currently confirming their coexistence on the same allele and assaying for decreased TET2 activity to determine whether one or both variants contribute to the phenotype. Other variants identified in MDS include the rare GATA2 (P161A) variant which is present in 1% of the population and the nearby common GATA2 (A164T) allele (~20%). These were mutually exclusive in our cohort and were seen at 3.9 and 1.5-fold, respectively, above the expected population frequency. We generated the P161A variant using site-directed mutagenesis and assayed for GATA2 transactivation activity in HEK293 cells with a GATA2-responsive LYL1 promoter-Luciferase construct (Figure 1). We also included empty vector (EV), wildtype (WT) GATA2 and T354M which is the most common highly penetrant autosomal dominant mutation leading to familial MDS/AML. As expected, T354M displayed a marked decrease in transactivation ability when compared to WT. The P161A variant similarly displayed loss-of-function in this assay, but not to the same magnitude as T354M. This is consistent with the hypothesis that reduced GATA2 function predisposes to myeloid malignancy where decreasing GATA2 activity correlates with increasing risk of developing malignancy. In our study 10/36 (28%) cases harboring these variants were T-MN cases. Apart from MET (E168D) (11.4-fold), the 2 rare variants with highest frequency in MDS versus controls were ASXL1 (N986S) (16.6-fold) and NOTCH1 (R912W) (6.5-fold). ASXL1 is an epigenetic regulator often mutated in hematopoietic malignancy and aberrant NOTCH1 function has been associated with myeloid and lymphoid malignancies. Conclusions: We have identified common and rare germline variants in genes involved in myeloid malignancy that may contribute to MDS pathogenesis. It remains to be seen whether they contribute to initiation, maintenance and/or progression of MDS and other hematopoietic malignancies. This is the first study reporting higher frequency of germline variants in sporadic MDS cases. Table 1. Frequency of germline variants in MDS, AML and CML in comparison to ExAC. Table 1. Frequency of germline variants in MDS, AML and CML in comparison to ExAC. Disclosures Hiwase: Celgene Corporation: Research Funding.