Search

Acute myeloid leukemias (AML) are comprised of multiple cell types with distinct capabilities to propagate the disease and resist therapy. Approximately 20% of AML patients carry gain- of-function mutations in IDH1 or IDH2 that result in over-production of the onco-metabolite 2-HG. Although IDH inhibitors can induce complete morphological remission, almost all patients eventually relapse. Analysis of clinical samples suggests that a population of IDH mutant cells is able to persist during treatment eventually acquiring 2-HG independence and drug resistance. Herein we characterized the molecular and cellular responses to the clinical IDH1 inhibitor AG-120 at high resolution using a novel multi-allelic mouse model of IDH1 mutant AML. We demonstrate that AG-120 exerts cell type-dependent effects on leukemic cells promoting delayed disease regression. Although IDH1 inhibition alone was not able to fully eradicate the disease, we uncovered that it increases cycling of rare leukemic stem cells and triggers transcriptional upregulation of the pyrimidine salvage pathway. Accordingly, AG-120 sensitized IDH1 mutant AML to azacitidine with the combination of AG-120 and azacitidine showing vastly improved efficacy in vivo. Our data highlight the impact of non- genetic heterogeneity on treatment response and provide mechanistic rationale for a drug combination that is being tested in clinical trials.STATEMENT OF SIGNIFICANCEInhibition of mutant IDH1 in AML is insufficient to eliminate the disease but promotes proliferation of quiescent leukemic stem cells. Our data provide a mechanistic explanation for the observed synergy between IDH inhibitors and azacitidine and suggest that IDH inhibitors may also synergize with other drugs that preferentially target actively dividing cells.

7019 Background: Acute myeloid leukemia (AML) is a disease with a dynamic mutational landscape; 6–10% of patients (pts) have somatic mutations in isocitrate dehydrogenase 1 ( IDH1), which can drive oncogenesis. Ivosidenib (IVO) is a potent oral targeted inhibitor of mutant IDH1 (mIDH1). IVO 500 mg QD + azacitidine (AZA) 75 mg/m2 SC or IV for 7 days in 28-day cycles was shown to significantly improve event-free survival (HR = 0.33 [95% CI 0.16, 0.69], p = 0.0011), median overall survival (24.0 vs 7.9 months), and complete remission + partial hematologic recovery rates (CR/CRh; 52.8% vs 17.6%) vs placebo (PBO) + AZA in the double-blind phase 3 AGILE study (NCT03173248) in pts with newly diagnosed IDH1-mutated AML (ND-AML). IDH1-mutation clearance ( IDH1-MC) and baseline co-mutation analysis from AGILE is reported. Methods: Genomic DNA from bone marrow mononuclear cells (BMMCs) or peripheral blood mononuclear cells (PBMCs), and/or bone marrow aspirate (BMA) were used for molecular studies. IDH1-MC analysis on BMMCs was performed by BEAMing digital PCR (limit of detection 0.02%-0.04%). BMA, BMMCs and PBMCs were utilized for co-mutational analysis by next-generation sequencing, ACE Extended Cancer Panel (detection limit 2%). Results: 146 pts were randomized: 72 to IVO+AZA; 74 to PBO+AZA. Median (range) baseline m IDH1 variant allele frequency in BMMCs was 36.7% (3.1–50.5) in the IVO+AZA arm and 35.5% (3.0–48.6) in the PBO+AZA arm. Updated IDH1-MC data (October 2021) from 47 IVO+AZA and 32 PBO+AZA treated pts with at least 1 on-treatment sample demonstrated IDH1-MC in 21/35 (60%) IVO+AZA pts achieving CR/CRh vs 4/11 (36%) PBO+AZA pts. In CR/CRh pts with time points available after IDH1-MC, suppression of the m IDH1 was durable and IDH1-MC maintained in all subsequent samples in 17/17 (100%) IVO+AZA treated pts and 1/3 (33%) PBO+AZA pts. Further analysis of baseline co-mutations on 120 pts (IVO+AZA: n = 58; PBO+AZA: n = 62) showed that DNMT3A, SRSF2, and RUNX1 were the most frequent in both treatment arms. Importantly, comparison of CR/CRh and non CR/CRh responses by cohort did not identify any single gene or pathway associated with an inferior outcome in IVO+AZA pts compared to PBO+AZA pts (p < 0.05, Fisher’s Exact test). Several genes ( DNMT3A, RUNX1, SRSF2, STAG2) and pathways (Differentiation, Epigenetics, Splicing) were associated with improved outcomes with IVO+AZA, including the RTK pathway, which was previously reported to be associated with primary resistance to IVO monotherapy. Further analysis of patient subgroups, including R132 variants (i.e., R132C vs R132S), will be presented. Conclusions: These data suggest that improved clinical outcomes with IVO+AZA are associated with sustained clearance of the m IDH1 clone including pts with disease that harbor mutations implicated in resistance to IVO monotherapy (e.g., with RTK pathway mutations). Clinical trial information: NCT03173248.

7042 Background: Ivosidenib (IVO) is a potent oral targeted inhibitor of mutant isocitrate dehydrogenase 1 (mIDH1). IVO plus azacitidine (AZA) significantly improved event-free survival (EFS), overall survival and complete remission + partial hematologic recovery rates compared with placebo (PBO) + AZA, in patients (pts) with newly diagnosed IDH1-mutant acute myeloid leukemia (AML) in the Phase 3 AGILE trial (NCT03173248). Here we report blood count recovery results from the AGILE trial. Methods: Pts were randomized 1:1 to IVO 500 mg QD + AZA 75 mg/m2 SC or IV for 7 days in 28-day cycles (n = 72), or PBO+AZA (n = 74). Red blood cell (RBC)/platelet transfusion history were assessed at screening and follow-up. Bone marrow (BM) and peripheral blood samples were obtained at screening, and during weeks 9, 17, 25, 33, 41, 53, and every 24 weeks thereafter, and at end of treatment and during EFS follow up. Samples were analyzed at each local site according to ICSH guidelines. Results: In the IVO+AZA and PBO+AZA arms, 4.2% and 5.5% of pts, respectively, received concomitant granulocyte colony-stimulating factor. Hemoglobin levels steadily increased from baseline at a similar rate in both treatment arms. Mean platelet count recovered from baseline values in the IVO+AZA and PBO+AZA arms (71.0 and 92.6 x 109/L, respectively) as early as week 9 of treatment (171.1 and 155.1 x 109/L, respectively) and continued to steadily increase thereafter in the treated population. In pts receiving IVO+AZA, mean neutrophil counts rapidly increased from baseline (0.99 x 109/L) to week 2 (2.05 x 109/L) and week 5 (4.07 x 109/L), and then generally stabilized to within the normal range to study end (last available cycle value; ̃2.0 x 109/L). Mean neutrophil counts initially declined with PBO+AZA before slowly recovering to near-normal levels after 36-40 weeks. The increased blood counts were accompanied by a rapid decrease in the mean BM blast percentage from 54.8% at baseline to 12.0% and 7.2% at week 9 and 17, respectively, in IVO+AZA treated patients and were maintained for 149 weeks. The decline in BM blasts was slower in the PBO+AZA arm (53.7%, 34.6% and 19.6% at baseline, week 9 and week 17, respectively). Among patients who were RBC/platelet transfusion-dependent at baseline (̃54.0% in both groups), 46.2% in the IVO+AZA group achieved RBC/platelet transfusion independence compared with 17.5% in the PBO+AZA arm (1-sided p = 0.0032). Additionally, fewer adverse events of febrile neutropenia (28.2% vs 34.2%) and infections (28.2% vs 49.3%) were reported in the IVO+AZA arm compared to the PBO+AZA arm. Conclusions: IVO+AZA demonstrated a significant clinical benefit compared with PBO+AZA and this sub-analysis demonstrated a rapidly improved recovery of blood counts and a reduced dependence on RBC and/or platelet transfusion. Moreover, rates of febrile neutropenia and infections were reduced with IVO+AZA. Clinical trial information: NCT03173248.

ObjectiveARID1A is commonly mutated in pancreatic ductal adenocarcinoma (PDAC), but the functional effects of ARID1A mutations in the pancreas are unclear. Understanding the molecular mechanisms that drive PDAC formation may lead to novel therapies.DesignConcurrent conditional Arid1a deletion and Kras activation mutations were modelled in mice. Small-interfering RNA (siRNA) and CRISPR/Cas9 were used to abrogate ARID1A in human pancreatic ductal epithelial cells.ResultsWe found that pancreas-specific Arid1a loss in mice was sufficient to induce inflammation, pancreatic intraepithelial neoplasia (PanIN) and mucinous cysts. Concurrent Kras activation accelerated the development of cysts that resembled intraductal papillary mucinous neoplasm. Lineage-specific Arid1a deletion confirmed compartment-specific tumour-suppressive effects. Duct-specific Arid1a loss promoted dilated ducts with occasional cyst and PDAC formation. Heterozygous acinar-specific Arid1a loss resulted in accelerated PanIN and PDAC formation with worse survival. RNA-seq showed that Arid1a loss induced gene networks associated with Myc activity and protein translation. ARID1A knockdown in human pancreatic ductal epithelial cells induced increased MYC expression and protein synthesis that was abrogated with MYC knockdown. ChIP-seq against H3K27ac demonstrated an increase in activated enhancers/promoters.ConclusionsArid1a suppresses pancreatic neoplasia in a compartment-specific manner. In duct cells, this process appears to be associated with MYC-facilitated protein synthesis.

Vibrio parahaemolyticus is a foodborne pathogen isolated from coastal waters of the United States, and from seafoods including fish. No information is available on the viability of V. parahaemolyticus on raw, chilled and frozen fish. A three-strain mixture of V. parahaemolyticus was inoculated on fish fillets (pH 6.4) to obtain a bacterial load of 10(4) (high) or 10(3) (low) CFU/fillet, and stored at 4C or 8C for 9 days or at -18C for seven weeks. At 4C and 8C, and at both levels of inoculation, V. parahaemolyticus survived on the fillets for the entire duration of the study. However, there was a significant reduction (P < 0.01) in V. parahaemolyticus population on the fillets by 9 days of storage. In the frozen fillets, there was a sharp decline (P < 0.01) in the population of V. parahaemolyticus by day 5 of storage. Although chilling and freezing significantly (P < 0.01) inactivated high numbers of V. parahaemolyticus on fish, they cannot be relied upon as a method to reduce V. parahaemolyticus on fish, since the time and magnitude of reduction depends on the initial load of the pathogen and the storage temperature.