Search

BACKGROUND: At a large academic teaching hospital, there are a variety of physicians and midlevel providers at the point of initial contact, and the extent of supervision of specifically trained oncology personnel may vary based on time of admission. Patients with acute leukemia may present with high risk disease processes that must be recognized and require prompt intervention to reduce both morbidity and short-term mortality. This is a retrospective review of the delivery of care at admission and key clinical outcomes for high risk patients presenting with acute myeloid leukemia (AML) or acute lymphoblastic leukemia (ALL) based on time of admission. The hypothesis of this study was that high risk patients with AML or ALL admitted overnight may have significant delays in management of the complications of acute leukemia with subsequent increases in morbidity and short-term mortality. METHODS: An institutional electronic database was queried to identify patients with ICD9 codes specific for AML/ALL. Inclusion criteria consisted of adults >18 years admitted to a single institution from 2010-2013. Key clinical data were then abstracted from the electronic medical records including lab values, time of admission (Daytime: 7am-8pm vs Nightime: 8pm-7am), and specific clinically important outcomes (time to specific therapy, time to chemotherapy, length of stay, ICU length of stay, organ failure, and mortality). Patients were categorized as high risk if they met established criteria requiring specific intervention [hyperleukocytosis defined as WBC >50 10^9/L, hyperuricemia defined as uric acid >8 mg/dL, and clinical suspicion for acute promyleocytic leukemia (APL)]. Variables with binary outcomes were tested for association with overnight admission using Fisher's exact test. All other variables were tested using the Wilcoxon two-group test. RESULTS: Between 2010 and 2013, 161 patients with AML/ALL were included in our analysis. Of those, 66 were classified as high risk (Table 1). In the high risk patients there were no significant differences in time to intervention based on time of admission including patients presenting with hyperleukocytosis and time to hydroxyurea administration (p=.32), patients presenting with hyperuricemia and time to allopurinol administration (p=.71) or rasburicase administration (p=.22), and in time to tretinoin (ATRA) administration in patients presenting with APL (p=.23). Time to definitive chemotherapy was significantly less for high risk patients admitted overnight (overnight median=48 hours, day median=56 hours, p=.042). However, rates of mechanical ventilation (p=.09), vasopressor usage (p=.37), and renal failure (p=.43) appeared similar between the groups. Additionally, length of stay (p=.83) and ICU length of stay (p=.44) was not significantly different for the two groups. 30-day mortality did not statistically differ between the two groups (overnight=19.4%, daytime=20%, p=.57). CONCLUSIONS: This is the first comprehensive analysis of the impact of the time of admission of acute leukemia patients at an academic tertiary cancer hospital, to our knowledge. Interestingly, nighttime admissions did not appear to significantly impact time to key clinical interventions or clinical outcomes in high risk patients admitted with acute leukemia. Although time to definitive chemotherapy was found to be significantly less in patients admitted overnight, confounding variables such as severity of illness at the time of admission may have impacted this analysis, and 30-day mortality rates were similar. Overall, this data supports the triage of patients with newly diagnosed or suspected acute leukemia to tertiary care centers as soon as possible. Table 1. Baseline Characteristics of High Risk Patients Age at Diagnosis Number % 1 High Risk Feature 66 100.0 Initial Point of Contact Number % Referring Hospital 45 68.2 Admission Time Number % Day Shift (7a-8p) 30 45.5 Night Shift (8p-7a) 36 54.5 Admission Location Number % Oncology Inpatient Service 53 80.3 Internal Medicine Inpatient Service 2 3.0 Medical ICU 11 16.7 Disclosures Foster: Celgene: Research Funding.

Multiple Myeloma (MM) is a heterogeneous disease but the hallmark genetic changes involve large numbers of genomic rearrangements. Recent studies have focused on attempts to identify individual driver mutations that might provide both prognostic information and unique therapeutic targets. Whole genome and exome sequencing of increasingly large numbers of patient samples have identified a number of commonly mutated genes in MM patients. However, none of these mutations are found in more than one quarter of patients and most are found in less than 10% of samples sequenced. We recently reported a large cohort of MM exome sequences involving 84 samples from 67 patients (Nat Commun. 2014;5:2997). We defined a diverse set of gene mutations with significant heterogeneity across our cohort with a median of 52 (range 21-488) mutations identified per sample. Although computational approaches can be used to prioritize mutations that are expected to alter protein structure and function, it is more challenging to determine which mutations are likely to be clinically meaningful. As a first step towards that understanding, here we report the frequency of expression of mutant alleles in Multiple Myeloma. In this study we report RNA-seq (100 million paired end reads on Illumina HiSeq) data on 14 samples from 10 MM patients for which we have previously performed exome sequencing and correlate allele-specific expression to the DNA mutant allele frequency. We find that a minority, average 27% (range 11-48%), of previously identified DNA mutations are expressed at detectable levels in MM patients. We also compared the allele frequency found in the RNA-seq to that from our exome sequencing to identify genes that demonstrate differential allelic expression and show that this is a common phenomenon in MM patients. We identified 42 such mutations in our analysis supported by at least 10 RNA-seq reads that showed a significant difference as determined by Bayesian hypothesis testing. For instance, the CCND1 mutant allele is expressed at a higher level than would be predicted based on exome-seq frequencies. Another gene showing a similar pattern of increased expression of the mutant allele in one patient was PARP4 (87% in RNA-seq vs 49% in exome-seq). Conversely, the mutant allele frequency of EIF1AX was lower than would be expected suggesting that the mutant allele may be suppressed in our patient (15% in RNA-seq vs 67% in exome-seq). Moreover, among a subset of genes previously identified as recurrently mutated within our patient samples we see that 8/11 (73%) express the mutant allele, providing further evidence that these genes may in fact be important in disease pathogenesis. Therefore, while a large number of mutations have been described in MM, only a small fraction of the mutant alleles have detectable expression and are likely to be biologically relevant. Unbalanced allelic expression of mutant alleles appears to be a relatively common occurrence in MM patients and may help explain why patients with the same identified mutation do not always behave in a similar fashion. This analysis for the first time highlights the important issue that DNA-based reporting of mutations may have significant limitations. It will be important in the future to study expression of mutant alleles in order to understand the biology, generate prognostic models and develop targeted therapies. Disclosures No relevant conflicts of interest to declare.

RNA has a diverse sets of regulatory functions besides being a messenger between DNA and protein. Recent analysis of RNA repertoire has identified a large numbers of non-coding transcripts. One of which, long intergenic non-coding RNA (lincRNA) with transcripts longer than 200 nucleotides, are located between the protein coding genes and do not overlap exons of either protein-coding or other non-lincRNA genes. lincRNAs have been considered to provide regulatory functions, however, their precise role in cellular biology remains unclear. Here, we have evaluated the lincRNA profile and their clinical role in MM. We performed RNA-seq on CD138+ MM cells from 320 patients and 18 normal bone marrow plasma cells (NBM) and analyzed for lincRNA. Data from Unstranded 50 bp paired-end RNAseq reads were mapped to the human genome and evaluated for frequency and type of lncRNA. Patient data for MM characteristics, cytogenetic and FISH as well as clinical survival outcomes were also analyzed and correlated with lncRNA data. We compared differentially expressed lincRNAs and protein coding genes in MM versus NBM samples. lincRNA and protein coding genes that have more than 2 reads/million reads for at least 50 samples (~15%) were included in the analysis. We identified 192 significantly expressed lincRNA (adj p value In summary, we report the first differential lincRNA expression in MM showing a significant role in disease biology as well as clinical outcome. lincRNAs are still functionally poorly characterized and our ongoing integrative approach will provide a link between lincRNAs and protein coding genes in MM. Disclosures Anderson: Celgene: Consultancy; Sanofi-Aventis: Consultancy; Onyx: Consultancy; Acetylon: Scientific Founder, Scientific Founder Other; Oncoprep: Scientific Founder Other; Gilead Sciences: Consultancy.

Background: Significant heterogeneity has been described in Multiple Myloma (MM), especially at the genomic level. Frequent gains and losses of DNA along with various mutations have been observed, and differential allelic expression is being characterized. Fusion proteins are common and maybe associated with cell transformation, growth and lethality of tumor cells. However, unlike in leukemia, no consistent fusion gene product has been consistently identified. As IgH-related translocations have an important role in myeloma, we have investigated fusion genes involving IgH, to understand their biology and explore a possible effect on survival in MM. Methods: We performed deep RNA-Seq on purified MM cells from 430 newly-diagnosed MM patients and analyzed gene expression profiles, isoform signatures and both novel and known fusion genes using two common algorithms: TopHat and MapSplice. We also correlated genomic data with patient data including cytogenetics and FISH, as well as survival. Results: We primarily focused on fusions involving the IGH gene (chromosome 14) and found that about one fourth of the patients (57 out of 430) presented an IGH fusion gene (97 patients according to TopHat, 303 according to Mapsplice, 57 according to both). These included the well described t(4,14) fusion involving the MMSET gene (found in 47 patients by both algorithms, 49 by Tophat, 54 by Mapsplice). Additionally we observed fusions involving chromosomes 14 and chromosomes 1, 4, 11, 12, and 16. The counterpart genes involved in the IgH fusions included PDE3A (chromosome 12 - 4 patients); HFM1 (chromosome 1, 2 patients); NFKB1, FGFR3, CIITA, WWOX and MRPL21 (chromosomes 4, 16, and 11, 1 patient). As RNA-Seq data allows the precise localization of the breakpoints, we were able to identify that out of the 47 t(4,14) patients, 62% were MB4-1, 9.5% were MB4-2 and 28.5% were MB4-3. Interestingly, we did not see fusion products involving IgH and other known parts on Chromosomes 8 and 20. We studied event-free survival in a subset of 265 patients with available survival data and found that, as predicted, patients with an IgH-MMSET fusion had significantly lower survival than others. However, patients with a fusion gene involving IGH and any other partner have a significantly better prognosis as a group. Moreover, the poor prognosis of IgH-MMSET fusion appears to be driven by MB4-3 patients. Importantly, the fusions identified using RNA-seq were also validated by FISH analysis. All t(4,14) fusions were characterized by a very high MMSET expression (FPKM greater than 20) while patients with other fusions presented a lower MMSET expression (FPKM lower than 10). Conclusion: Our study suggests that IgH-related translocations in myeloma may impact tumor biology by a number of mechanism, one of which is the generation of fusion proteins with functional consequences. It also highlights a possible clinical impact that requires validation in larger cohorts. Disclosures No relevant conflicts of interest to declare.

Alternative splicing is a crucial mechanism for gene regulation, which enhances the diversity of transcriptome and proteome. Misregulation of alternative splicing has been implicated in number of disease processes including cancer. Our data utilizing exon array profile from 170 uniformly treated newly diagnosed patients with MM confirms clinical relevance of splicing as demonstrated by impact of level and extent of alternate splicing on both progression free and overall survival. Fox2, a RNA splicing factor, is one of the most important genes predicting clinical outcome in these patients. We confirmed Fox2 expression in 10 MM cell lines at both RNA and protein levels. Immunohistochemistry staining showed a predominant nuclear localization of Fox2. Importantly, we also observed that MM cell - bone marrow stromal cells (BMSC) interaction led to significant inhibition of Fox2 expression in MM cells. Similar response was also observed using BMSC supernatants. While IL6 treatment significantly downregulated the expression of Fox2 in MM1S and RPMI8226 cells in a dose-dependent manner, IGF-1 treatment had no significant impact on Fox2 expression in MM cell lines. Since Fox2 has been described to plays a role in the maintenance of cell cytoskeleton, we therefore evaluated whether Fox2 might influence the migration and adhesion in MM cells. Transwell migration assay showed enhanced migration rate of Fox2-knocking down- MM1S and RPMI8226 cells versus controls. We also observed the increased cell adhesion to fibronetin in both cell lines upon Fox2 knockdown. Actin polymerization evaluated by Alexa488-conjugated phalloidin staining and confocal microscope analysis showed Fox2 knocking down cells with increased actin polymerization in both MM1S and RPMI8226 cell lines. Interstingly, we observed that Fox2 knockdown in MM cell lines did not affect the cell proliferation and survival. As Fox-2 is a splicing factor, we further evaluated the molecular impact of Fox2 expression in multiple myeloma by RNA-seq analysis. Our data revealed that Fox2 functions in regulating both protein-coding and non-coding RNA alternative splicing. Knockdown of Fox2 resulted in significant isoform up-regulation (60 in MM1S and 151 in RPMI8226) and down-regulation (70 in MM1S and 69 in RPMI8226). Gene enrichment analysis showed these genes are clustered in cell cytoskeleton regulation, microtubule-based movement, ATP binding, amongst others. Our study then focused on Fox2 knockdown-induced significant isoform switch in MM cell lines. We designed the primers testing the spliced exons and confirm the isoform switch in MM cells by PCR analysis (e.g. Pyk2 and PFDN6). Importantly, our RNA seq data showed that Fox2 regulates the expression of a series of microRNAs and long-noncoding RNAs (e.g. MALAT1 RPMI8226 CNT (58) vs Fox2 knockdown (108)), which provides us a new insight into impact of Fox2 on non-protein coding RNAs. We have also validated the RNA-seq data by Q-PCR analysis. In summary, our results identify Fox2 as a biologically important RNA binding protein that is regulated by bone marrow microenvironment interaction and with essential function and potential clinical implications in multiple myeloma. Disclosures No relevant conflicts of interest to declare.

Multiple Myeloma (MM) is characterized by genomic heterogeneity that contributes to differences in clinical outcome. Mis - or unrepaired DNA damage poses a serious threat to genomic stability, potentially leading to the formation of oncogenic mutations, including translocations, deletions and amplifications. Spontaneous endogenous DNA damage represents an essential portion of DNA lesions; therefore, a thorough knowledge of its types and prevalence is of high importance for its impact on myelomagenesis, its interaction with exogenous DNA damaging sources, and consequently, at improving the clinical outcome of myeloma patients. First, we evaluated the occurrence of spontaneous DNA damage in a panel of 4 MM cell lines, 4 CD138+ primary MM patients samples and in 4 solid tumor cell lines. Using single-cell gel electrophoresis (comet assay) under neutral and alkaline conditions we observed high levels of olive tails moments indicative of DNA damage in all MM cell lines (mean±SD, MM1S 47.56±7.1, OPM2 60.92±7.9) and primary MM cells as well as solid tumor cell lines, compared to normal cell controls (PBMCs 6.283±3.56, HEEC 1.448±0.2, BJ 0.64±0.58) (P=.0001). Significantly higher signal of the most widely used marker of DNA damage γH2ax phosphorylated at serine139, 53BP1 and RPA32 as indicator of Single strand breaks (SSB) was observed by immunoblotting in all MM cell lines compared to normal controls confirming the high occurrence of DNA damage in MM in absence of any exogenous genotoxic insult. These results were further confirmed by immunocytochemistry with γH2Ax, ku70/80, 53BP1, Rad51. Remarkably, the Non homologous end joining (NHEJ) marker ku70/80 was co-localized with γH2Ax and was present in ~80% of cells, indicating the activation of NHEJ throughout the cell cycle. To further elucidate the differential spontaneous DNA damage in MM, we detected and quantified the abasic sites by using an ELISA-based assay. The distribution of abasic sites showed same pattern as the DSBs in all the MM cell types analyzed, indicating that abasic sites constitute an important portion of spontaneous DNA damage (OPM2 57.3/105, PBMCs 12/105bp). On the basis of these findings, we hypothesized that the increased stimulation by endogenous oxidative stress or possible inactivation of DNA damage repair mechanisms might be implicated to this observed high occurrence of DSBs in MM. By using our chromosomally integrated green fluorescent protein reporter construct-based assay, we observed that NHEJ and homologous recombination (HR) were significantly more active in all MM cells compared to normal controls. Moreover, we found that increased stimulation by endogenous oxidative stress was present in all MM cells evaluated, while a strong correlation between the levels of oxidative stress and the spontaneous DSBs was observed (r=0.85, P As the information conveyed by epigenetic modifications play a key role in the regulation of DNA processes including DNA damage and repair, we performed Chip-seq analysis for γH2ax. After performing peak calling on γH2ax ChIP-seq data, we sought to determine whether γH2ax enrichment regions tend to occur in previously reported common fragile sites or early replicating sites. Interestingly, we found that spontaneous DNA damage is equally representative in replicating fragile sites and random sites throughout genome. Next, we performed sequential double chip-seq analysis for γH2ax for the first Chip and a total 5 different epigenetic marks for the re-Chip. We performed peak-calling analysis on each data type, and strikingly, we found that similar numbers of peaks can be found across re-ChIP dataset with H3K27ac having the greatest number of peaks (653). Of the 5 marks evaluated, we observed that only H4K20me2 showed significant enrichment in promoter regions relative to random chance (P=.039) within 1kb of transcription start sites. A similar trend is also observed for the region within 3kb of transcription start sites (P=.026), and 2kb of transcription start sites (P=.084, weakly significant), indicating a possible role of methylation of K20 in proper chromatin organization in proximity of DSBs. In conclusion, our study demonstrates the higher occurrence of ongoing spontaneous DNA damage in MM and provides insights into possible relationship between the aberrant epigenome and spontaneous DNA damage revealing mechanism that might be dysregulated promoting genomic instability. Disclosures Anderson: Celgene: Consultancy; Sanofi-Aventis: Consultancy; Onyx: Consultancy; Acetylon: Scientific Founder, Scientific Founder Other; Oncoprep: Scientific Founder Other; Gilead Sciences: Consultancy. Dimopoulos:Celgene: Consultancy, Honoraria.