Search

INTRODUCTION: Flow cytometry has been used extensively to detect MM cells in the bone marrow (BM), micro-residual disease and circulating myeloma cells. Accumulating literature defines MM cells as CD138+/CD38+ for the primary gating of plasma cells; however, several studies demonstrated the presence of clonogenic CD138-negative MM cells, and that hypoxia can decrease the expression of CD138 in MM cells. We propose a novel set of biomarkers to detect MM cells regardless of their CD138 expression or hypoxic status. METHODS and RESULTS: We have tested the effect of hypoxia on the expression of the MM markers, CD138 and CD38, and found that hypoxia decreased the expression of CD138, therefore it cannot be used as a universal marker for MM cells. Hypoxia did not alter the expression of CD38 in MM cells; however, CD38 is a general leukocyte marker and cannot be used alone to identify MM cells, since it is expressed on multiple other leukocytes including T-cells, B-cells, monocytes, NK cells, and dendritic cells. Therefore, we negatively selected these cells by flow cytometry using the specific markers for each of these populations including CD3, CD19, CD14, CD16, and CD123; respectively. Therefore, we detected MM cells as CD38+/CD3-/CD19-/CD14-/CD16-/CD123-. We used CD38 antibody conjugated with APC and FITC-conjugated antibodies for all the other markers, thus MM cells were defined as APC+/FITC- population. To compare traditional method (CD138-based) with our strategy to detect hypoxic and normoxic MM cells, MM cell lines were stained with a cocktail of CD38-APC, FITC-antibodies, and CD138-V450, and analyzed by flow cytometry. The use of CD138+ as a universal marker for MM cells detected 85-100% of the normoxic cells, and only 60-75% of the hypoxic MM cells. While APC+/FITC- strategy detected close to a 100% of the MM cells independent of the cells’ normoxic/hypoxic status or expression of CD138. The ability of the new strategy to detect hypoxic and normoxic MM cells in the peripheral blood was tested by staining hypoxic and normoxic MM cells with cell-tracker Calcein-Red-Orange (as a positive control), spiked 104 MM cells into 106 mononuclear cells from a healthy donor (1% MM in total), and the percentile of Calcein-Red-Orange+ (as a positive control), APC+/FITC-, and CD138+ populations, were analyzed by flow cytometry. Calcein-Red-Orange staining showed exact 1% of MM cells detected in total mononuclear cells for both hypoxic and normoxic MM cells; detection with CD138+ showed 0.95% for normoxic and 0.45% for hypoxic cells; and detection of MM cells using the APC+/FITC- strategy showed 1.05% for normoxic and 1.1% for hypoxic MM cells. Hence, the new strategy detects MM cells selectively and independently of their CD138 expression or hypoxic status. We have used the APC+/FITC- strategy to detect MM cells in BM CD138-negative fractions of 20 MM patients. The APC+/FITC- strategy was able to detect a range of 1.6-44% myeloma cells in the CD138-negative population of BM mononuclear cells isolated from MM patients. Moreover, we have assessed the clonality of this population using APC+/FITC- in the CD138-negative fractions of MM patients. We found that the clonality of the APC+/FITC- population was similar to the clonality of the original disease (CD138+ cells) in 70% of the cases. The other 30% of the cases showed very low involvement of myeloma population in the BM and showed Kappa/Lambda ratio within normal range. Analyzing the prevalence of circulating MM cells in the peripheral blood from 12 MM patients showed that all patients had a higher number of circulating MM cells as detected by APC+/FITC- strategy compared to CD138+, and the fold change ranged from 1.5 to 86 times. CONCLUSION: We found that CD138 cannot be used as a universal marker to detect MM cells. Moreover, we developed a novel strategy to detect MM cells independent of their CD138 expression or hypoxic status; and we used CD38+/CD3-/CD19-/CD14-/CD16-/CD123- population as an alternative set of biomarkers to detect MM cells. This strategy was able to detect a clonal MM cell population in the CD138-negative fraction of BM mononuclear cells isolated from MM patients, as well as in the peripheral blood. Currently, we are exploring the ability of this strategy to predict relapse in MM patients whose BM was defined a CD138-negative. More investigation to characterize this population and the role in tumor recurrence and drug resistance in MM is warranted. Disclosures No relevant conflicts of interest to declare.

Introduction The incidence of multiple myeloma (MM) increases with age, yet some cytogenetic changes are actually more common in younger patients with MM (Avet-Loiseau J Clin Oncol 2013). This suggests that a mechanism other than chromosomal changes underlies the increased incidence with age. Senescent cells secrete a number of proinflammatory cytokines, chemokines, growth factors and proteases resulting in the senescence-associated secretory phenotype (SASP), which can promote tumor growth. Preclinical data suggests that myeloma bone marrow stromal cells express the SASP (Andre PLOS ONE 2013). We hypothesized that SASP factors correlate with age in patients with MM. Methods Peripheral blood serum and matched bone marrow aspirate plasma from patients with multiple myeloma were evaluated for selected factors associated with the SASP using quantitative multiplex immunoassay (Rules Based Medicine, Austin TX USA). SASP factors with a known role in MM [interleukin-6 (IL-6), interleukin-8 (IL-8), interleukin-15 (IL-15), granulocyte-macrophage colony-stimulating factor (GMCSF), intercellular adhesion molecule 1(ICAM1), osteoprotegerin (OPG), hepatocyte growth factor (HGF), insulin-like growth factor-binding protein(IGFBP-1), interleukin-1 beta (IL1b), monocyte chemotactic protein 1(MCP-1), macrophage inflammatory protein-1 alpha(MIP-1a), angiogenin, leptin, vascular endothelial growth factor receptor 1(VEGFR1) and stem cell factor(SCF)] were selected. The relationship between age and SASP factors were analyzed using Kendall tau rank correlation coefficient. Results Samples from 25 patients (each with peripheral blood serum and matched bone marrow aspirate plasma) were analyzed. The median age was 62 (range 47 - 74). Disease states were as follows: 36% newly diagnosed/untreated, 24% pretransplant and 40% relapsed. ISS stage included 40% stage I, 28% stage II and 32% stage III. Three of the selected SASP factors in the peripheral blood correlated with age: IL-8 (Kendall Tau 0.334, p=0.027), OPG (Kendall Tau 0.289, p=0.046) and MCP-1 (Kendall Tau 0.332, p=0.022). No SASP factors tested in the bone marrow plasma were significantly correlated with age. Conclusions We demonstrated age-associated differences in the SASP factors IL-8, OPG and MCP-1 in the peripheral blood of myeloma patients. Future research will examine differences between patients with myeloma and age-matched controls without cancer. Disclosures: Vij: Celgene : Honoraria, Research Funding, Speakers Bureau; Millennium: Honoraria, Speakers Bureau; Onyx: Honoraria, Research Funding, Speakers Bureau. Stockerl-Goldstein:Celgene : Speakers Bureau; Celgene : Speakers Bureau; Millennium: Speakers Bureau; Millennium: Speakers Bureau.

Abstract 1830 Multiple myeloma (MM) is a fatal disease characterized by clonal expansion of malignant plasma cells. The etiopathogenesis of MM is not fully understood. Several numerical and structural chromosomal aberrations have been identified as diagnostic markers and predictors of evolution in MM. Cytogenetic studies in MM patients are often not informative due to technical difficulties related to low proliferation of malignant plasma cells and outgrowth of non-malignant cells. Fluorescence in-situ hybridization (FISH) on CD138+ sorted plasma cells is probably the best method for maximizing diagnostic yield in MM, but is limited to the genomic regions queried. To overcome the limitations of the amount of clinical material available and to be able to interrogate large number of MM specific genomic aberrations, we developed and validated a MM genomic copy number signature. This signature comprised of 183 MM specific genes, was developed by pooling data from extensive meta-analyses on publically available raw data from ∼450 MM patients and copy number data generated by high-resolution SNP arrays (Affymetrix) from 39 MM patients in our cohort. To validate this signature of a large number of genes, we tested a recently developed innovative high throughput digital technology NanoString - nCounter assay. This technology captures and counts individual DNA molecules without enzymatic reactions or bias and is notable for its high levels of sensitivity, linearity, multiplex capability, and digital readout. It requires minimal input of DNA (∼300ng) making it a valuable tool for genomic copy number signature validation, diagnostic testing, and large translational studies, all of which often are limited by the very small amounts of clinical material available. Digital data was generated using nCounter analysis in 42 newly diagnosed, untreated MM patients. To identify the true acquired somatic copy number changes matched germline (skin) and tumor (sorted CD138+ cells) were analyzed from each of these MM patients. All of the genes tested demonstrated highly significant concordance with our microarray data (P < 0.05). The dynamic range in copy number calls with this assay is very large since there are no saturation issues and there is very low background. In this study, we were able to detect a maximum of 9 copies in some of the targets. We observed amplification of chromosomes 1q(51%), 3(65%), 5(65%), 7(70%), 9(56%), 11(72%), 15(56%), 19(53%), 21(42%), and deletion of chromosomes 1p(25%), 6q(28%), 8p(42%), 12p(40%), 13(47%), 14(26%) and 16q(49%). Interestingly, cytoband 2p11.2 and 14q32.33 consisting IGK and IGH genes were deleted in 75% and 93% of the patient population respectively. Overall, our results correlate well with the known pattern of genomic aberrations in MM. Additional analysis in an extended panel with clinically categorized samples is carried on to test the utility of this myeloma specific gene signature. To the best of our knowledge this is the first application of a high-throughput digital system to validate genomic copy number signature in cancer. Disclosures: No relevant conflicts of interest to declare.

Abstract 809 Immunoglobulin light chain amlyloidosis (AL) is a rare plasma cell disorder characterized by deposition of misfolded light chains in various organ systems with an average survival of 1–2 years. AL is also the most common form of systemic amyloidosis with 1200–3200 newly diagnosed cases reported annually in the United States. Very little is known regarding specific genomic aberrations associated with AL-amyloidosis. Aside from the light chain selection, no phenotypic or genetic features have been identified that distinguish AL amyloidosis from other plasma cell dyscrasias. Understanding the genetics of AL and the molecular mechanisms involved in amyloid formation may lead to early diagnosis and the identification of novel drug targets and therapies. We therefore have attempted to study the genomic landscape of AL patients and MM for comparison. Genomic copy number and loss of heterozygosity (LOH) analyses were performed on DNA derived from tumor (CD138 sorted cells) and matched germline (skin) from biopsy proven AL patients using Affymetrix single nucleotide polymorphism (SNP) 6.0 arrays. Numerous genomic changes with gains in chromosome 1q, 6, 9, 11q, 15, 19 and 21 and loss on chromosome 1p, 2q, 8, 10, 12, 13, 14, 16, 18, 20 and 22 were observed in more than 10% of the patients. Recurrent genomic changes in about 249 segments involving 457 genes were present in about 1/3 of AL patients. In particular, deletion of IGK, IGH, PIK3CA, FLT3, RB1, PCDH9, GPC6, RASA3, ADAM6 genes and amplification of CFHR1, JAK2, GCNT1, TSC1, PGR genes were observed. Gene network analysis showed five distinct major modules consisting of 51 distinct elements and involving PDGF, TP53, interleukin signaling, TRKA signaling, cell cycle and mitotic pathways were enriched. Allele specific copy number analysis in tumor (ASCAT) profile showed increased ploidy status of the AL genome in 47% of the assessed patients. LOH was observed in chromosomes 4, 5, 6, 8, 9, 12, 13, 18 and 22 in 30% of patients, ranging from 5Mb to entire chromosome. Furthermore, genomic comparisons of AL with multiple myeloma (MM) showed the typical archetype of myeloma's signature with exception of gain of chromosomes 3, 5, 7 and loss of chromosome 6q and 8p. Interestingly deletion of IGH, IGK locus and PIK3CA gene were observed at a higher frequency in AL patients. Categorical analysis using isotype specific classification in AL showed a significantly higher frequency of deletion in chromosome 14, 13, 8 and amplification of chromosome 9q in the kappa type than lambda isotype. To the best of our knowledge, this is the first ultra-high resolution study of the genomic landscape of AL amyloidosis. In this study, we have found several novel genes and pathways associated with this rare disease. The numerous copy number alterations of AL thus reflect the genomic complexity and the heterogeneity of this disease. Additional genome-wide analysis in a larger panel with target organ stratified patients is under way and may further our understanding of genetic changes specifically associated with AL. Disclosures: No relevant conflicts of interest to declare.