Approximately 10% to 15% of patients with essential thrombocythemia (ET) lack the common driver mutations, so-called “triple-negative” (TN) disease. We undertook a systematic approach to investigate for somatic mutations and delineate gene expression signatures in 46 TN patients and compared the results to those with known driver mutations and healthy volunteers. Deep, error-corrected, next-generation sequencing of peripheral blood mononuclear cells using the HaloPlexHS platform and whole-exome sequencing was performed. Using this platform, 10 (22%) of 46 patients had detectable mutations (MPL, n = 6; JAK2V617F, n = 4) with 3 of 10 cases harboring germline MPL mutations. RNA-sequencing and DNA methylation analysis were also performed by using peripheral blood mononuclear cells. Pathway analysis comparing healthy volunteers and ET patients (regardless of mutational status) identified significant enrichment for genes in the tumor necrosis factor, NFκB, and MAPK pathways and upregulation of platelet proliferative drivers such as ITGA2B and ITGB3. Correlation with DNA methylation showed a consistent pattern of hypomethylation at upregulated gene promoters. Interrogation of these promoter regions highlighted enrichment of transcriptional regulators, which were significantly upregulated in patients with ET regardless of mutation status, including CEBPβ and NFκB. For “true” TN ET, patterns of gene expression and DNA methylation were similar to those in ET patients with known driver mutations. These observations suggest that the resultant ET phenotype may, at least in part and regardless of mutation type, be driven by transcriptional misregulation and may propagate downstream via the MAPK, tumor necrosis factor, and NFκB pathways with resultant JAK-STAT activation. These findings identify potential novel mechanisms of disease initiation that require further evaluation.

Liposomal daunorubicin and cytarabine (CPX-351) improves overall survival (OS) compared to 7+3 chemotherapy in older patients with secondary acute myeloid leukaemia (AML); to date there have been no randomized studies in younger patients. The high-risk cohort of the UK NCRI AML19 trial (ISRCTN78449203) compared CPX-351 with FLAG-Ida in younger adults with newly-diagnosed adverse cytogenetic AML or high-risk myelodysplastic syndromes (MDS). 189 patients were randomized (median age 56y). By clinical criteria 49% had de novo AML, 20% secondary AML and 30% high risk MDS. MDS-related cytogenetics were present in 73% of patients, with complex karyotype in 49%. TP53 was the most commonly mutated gene, in 43%. Myelodysplasia-related gene mutations were present in 75 patients (44%). The overall response rate (CR + CRi) after course two was 64% and 76% for CPX-351 and FLAG-Ida (OR:0.54, 95%CI 0.28-1.04, p=0.06). There was no difference in OS (13.3 months vs 11.4 months, HR:0.78, 95%CI 0.55-1.12, p=0.17) or event-free survival (HR:0.90, 95%CI 0.64-1.27, p=0.55) in multivariable analyses. However, relapse-free survival was significantly longer with CPX-351 (median 22.1 vs 8.35 months, HR:0.58, 95% CI 0.36-0.95, p=0.03). There was no difference between the treatment arms in patients with clinically defined secondary AML (HR:1.1, 95%CI 0.52-2.30) or those with MDS-related cytogenetic abnormalities (HR:0.94, 95%CI 0.63-1.40), however an exploratory sub-group of patients with MDS-related gene mutations had significantly longer OS with CPX-351 (median 38.4 vs 16.3 months, HR:0.42, 95%CI 0.21-0.84, heterogeneity p=0.05). In conclusion, OS in younger patients with adverse risk AML/MDS was not significantly different between CPX-351 and FLAG-Ida.

The PML::RARA fusion protein is the hallmark driver of Acute Promyelocytic Leukemia (APL) and disrupts retinoic acid signaling, leading to wide-scale gene expression changes and uncontrolled proliferation of myeloid precursor cells. While known to be recruited to binding sites across the genome, its impact on gene regulation and expression is under-explored. Using integrated multi-omics datasets, we characterize the influence of PML::RARA binding on gene expression and regulation in an inducible PML::RARA cell line model and APL patient ex vivo samples. We find that genes whose regulatory elements recruit PML::RARA are not uniformly transcriptionally repressed, as commonly suggested, but also may be upregulated or remain unchanged. We develop a computational machine learning implementation called Regulatory Element Behavior Extraction Learning to deconvolute the complex, local transcription factor binding site environment at PML::RARA bound positions to reveal distinct signatures that modulate how PML::RARA directs the transcriptional response.

Background:The PML;RARA fusion protein is the hallmark driver of Acute Promyelocytic Leukemia (APL). The fusion disrupts retinoic acid signalling, leading to the proliferation of myeloid precursors halted at the promyelocyte stage of maturation. Chromatin conformation plays a fundamental role in controlling regulatory networks driving cell differentiation, however the genome wide organisational changes that occur during the PML;RARA differentiation block remains to be elucidated. Aims: We have aimed to characterise three hallmarks of PML;RARA driven transcriptional mis-regulation: transcription factor binding, epigenetic remodelling and higher order chromatin organisation. Methods: To a PML;RARA inducible U937-PR9 cell line model (Figure 1a), we have applied 1) Promoter Capture Hi-C (PCHi-C) to characterise long-range regulatory interactions, 2) Cut&Run, a newly described chromatin profiling technique, and 3) RNA-seq. All datasets are paired and have been modelled as induced vs non-induced. Results: PML;RARA induction resulted in 855 significantly differentially expressed genes (DEGs, adjusted pvalue < 0.01). Up regulated genes (457) were enriched for genes involved in proliferative and oncogenic pathways. Down regulated genes (398) were enriched for drivers of cell differentiation, including the master regulators of myeloid differentiation: CEBPE, CEBPB and CEBPA. We identified 15,412 PML;RARA binding sites using Cut&Run (n=2). These sites encompassed 95% of previously uncovered PML;RARA binding sites, in addition to ~12,000 novel sites. 53% of these regions were distributed at gene promoters, 23% were intergenic and 24% at gene bodies. We observed a global decrease in H3k27ac after PML;RARA induction (86% regional loss), a significant proportion of these regions coincide with PML;RARA binding (Fisher's test 2.2-16). 1,900 PML;RARA peaks were associated with 66% DEGs. Interestingly, a significant proportion of PML;RARA peaks were not associated with DEGs. Applying PCHi-C (n=3) we identified >60,000 consistent differential interactions (DIs, ihw < 0.01), with a mean interaction distance of 100kb. 30,039 interactions increased and 30,403 interactions decreased after PML;RARA induction. Over half of DIs directly involved 8,066 PML;RARA binding sites (Fisher's test 2.2-16). We observed that genes losing long-range interactions were more likely to have decreased expression (59%) and this observation increased if a gene is also bound by PML;RARA (65%). SPI1, ID2, CEBPA, CEBPB, CEBPE, EGR1, and TFEB, key regulators of myeloid differentiation, display this pattern of PML;RARA driven negative regulation. We also observed that genes with increased expression were more likely to gain long-range interactions (60%). The top ranked gene across all datasets was the prostaglandin receptor 4 (PTGER4), a G-Protein coupled receptor highly expressed in AML and upregulated in our datasets (3.1 fold) (figure 1b). PTGER4 gained long-range interactions with an intergenic region ~300kb downstream of its promoter. This intergenic region is highly enriched for both H3k27ac and PML;RARA. This suggests that PML;RARA facilitates the long-range activation of PTGER4. We see a similar pattern to PTGER4 in 15% of DEGs where PML;RARA is anchored to both ends of the DI. Summary/Conclusion: Applying novel NGS techniques to a simple model, we have highlighted that specific chromosome conformations are pivotal to the transcriptional mis-regulation driven by PML;RARA. We show PML;RARA may be directly involved in the re-organisation of the genome and this 3D architecture is pivotal in driving the Leukemia differentiation block. Disclosures Dillon: Abbvie: Consultancy, Honoraria; Novartis: Consultancy, Honoraria; Pfizer: Consultancy, Honoraria; TEVA: Consultancy, Honoraria.

No abstract