Tobias Meißner

Growth factors in multiple myeloma: a comprehensive analysis of their expression in tumor cells and bone marrow environment using Affymetrix microarrays

BackgroundMultiple myeloma (MM) is characterized by a strong dependence of the tumor cells on their microenvironment, which produces growth factors supporting survival and proliferation of myeloma cells (MMC). In the past few years, many myeloma growth factors (MGF) have been described in the literature. However, their relative importance and the nature of the cells producing MGF remain unidentified for many of them.MethodsWe have analysed the expression of 51 MGF and 36 MGF receptors (MGFR) using Affymetrix microarrays throughout normal plasma cell differentiation, in MMC and in cells from the bone marrow (BM) microenvironment (CD14, CD3, polymorphonuclear neutrophils, stromal cells and osteoclasts).Results4/51 MGF and 9/36 MGF-receptors genes were significantly overexpressed in plasmablasts (PPC) and BM plasma cell (BMPC) compared to B cells whereas 11 MGF and 11 MGFR genes were overexpressed in BMPC compared to PPC. 3 MGF genes (AREG, NRG3, Wnt5A) and none of the receptors were significantly overexpressed in MMC versus BMPC. Furthermore, 3/51 MGF genes were overexpressed in MMC compared to the the BM microenvironment whereas 22/51 MGF genes were overexpressed in one environment subpopulation compared to MMC.ConclusionsTwo major messages arise from this analysis 1) The majority of MGF genes is expressed by the bone marrow environment. 2) Several MGF and their receptors are overexpressed throughout normal plasma cell differentiation. This study provides an extensive and comparative analysis of MGF expression in plasma cell differentiation and in MM and gives new insights in the understanding of intercellular communication signals in MM.

Tumor cells in multiple myeloma patients inhibit myeloma-reactive T cells through carcinoembryonic antigen-related cell adhesion molecule-6

Although functionally competent cytotoxic, T cells are frequently observed in malignant diseases, they possess little ability to react against tumor cells. This phenomenon is particularly apparent in multiple myeloma. We here demonstrate that cytotoxic T cells reacted against myeloma antigens when presented by autologous dendritic cells, but not by myeloma cells. We further show by gene expression profiling and flow cytometry that, similar to many other malignant tumors, freshly isolated myeloma cells expressed several carcinoembryonic antigen-related cell adhesion molecules (CEACAMs) at varying proportions. Binding and crosslinking of CEACAM-6 by cytotoxic T cells inhibited their activation and resulted in T-cell unresponsiveness. Blocking of CEACAM-6 on the surface of myeloma cells by specific monoclonal antibodies or CEACAM-6 gene knock down by short interfering RNA restored T-cell reactivity against malignant plasma cells. These findings suggest that CEACAM-6 plays an important role in the regulation of CD8+ T-cell responses against multiple myeloma; therefore, therapeutic targeting of CEACAM-6 may be a promising strategy to improve myeloma immunotherapy.

miRNAs in multiple myeloma – a survival relevant complex regulator of gene expression

Purpose microRNAs regulate gene-expression in biological and pathophysiological processes, including multiple myeloma. Here we address i) What are the number and magnitude of changes in miRNA-expression between normal plasma cells and myeloma- or MGUS-samples, and the latter two? ii) What is the biological relevance and how does miRNA-expression impact on gene-expression? iii) Is there a prognostic significance, and what is its background? Experimental design Ninety-two purified myeloma-, MGUS-, normal plasma cell- and myeloma cell line-samples were investigated using miChip-arrays interrogating 559 human miRNAs. Impact on gene-expression was assessed by Affymetrix DNA-microarrays in two cohorts of myeloma patients (n = 677); chromosomal aberrations were assessed by iFISH, survival for 592 patients undergoing up-front high-dose chemotherapy. Results Compared to normal plasma cells, 67/559 miRNAs (12%) with fold changes of 4.6 to −3.1 are differentially expressed in myeloma-, 20 (3.6%) in MGUS-samples, and three (0.5%) between MGUS and myeloma. Expression of miRNAs is associated with proliferation, chromosomal aberrations, tumor mass, and gene expression-based risk-scores. This holds true for target-gene signatures of regulated mRNAs. miRNA-expression confers prognostic significance for event-free and overall survival, as do respective target-gene signatures. Conclusions The myeloma-miRNome confers a pattern of small changes of individual miRNAs impacting on gene-expression, biological functions, and survival.

Immunomodulatory drugs thalidomide and lenalidomide affect osteoblast differentiation of human bone marrow stromal cells in vitro

Osteoblastic activity is severely impaired in active myeloma, contributing to the development of myeloma bone disease. Although several drugs reducing osteoclast-mediated bone degradation are in clinical use, approaches to specifically augment bone formation are at an early stage of development. Novel antimyeloma drugs not only directly act on myeloma cells, but impact on the microenvironment as well. Proteasome inhibitors were previously shown to have bone anabolic properties. Here we investigated the impact of immunomodulatory drugs (IMiDs) on bone formation. Treatment with thalidomide and lenalidomide significantly inhibited osteoblast development in vitro, as reflected by a reduction of alkaline phosphatase activity and matrix mineralization. The effects were upheld in combination with bortezomib. The IMiDs upregulated Dickkopf-1 (DKK1) and inhibin beta A, but blocking these molecules was not able to restore regular osteoblast development. We therefore performed gene expression profiling to reveal other osteoblast regulatory factors that might be involved in the IMiD-mediated effect on osteoblast development. Our data indicate that osteoblast inhibition is possibly an IMiD-class effect mediated by downregulation of major osteoblast regulators (e.g., runt-related transcription factor 2, distal-less homeobox 5, pleiotrophin) and concurrent induction of secreted inhibitors of osteoblast formation (e.g. DKK1, activin A, gremlin 1). Our results highlight the need for bone anabolic therapeutics in myeloma, counteracting the negative impact of prolonged IMiD exposure on bone metabolism.

ASSOCIATION OF ANTIGEN-SPECIFIC T-CELL RESPONSES WITH ANTIGEN- EXPRESSION AND IMMUNOPARALYSIS IN MULTIPLE MYELOMA

Purpose: Cancer testis antigens (CTA) are immunotherapeutical targets aberrantly expressed on multiple myeloma cells, especially at later stages, when a concomitant immunoparesis hampers vaccination approaches. Experimental Design: We assessed the expression of the multiple myeloma antigen HM1.24 (reported present in all malignant plasma cells) and the CTAs MAGE-A2/A3 and NY-ESO-1 (aberrantly expressed in a subset of patients with myeloma), in CD138-purified myeloma cells by qRT-PCR (n = 149). In a next step, we analyzed the antigen-specific T-cell responses against these antigens by IFNγ EliSpot assay (n = 145) and granzymeB ELISA (n = 62) in relation to stage (tumor load) and expression of the respective antigen. Results: HM1.24 is expressed in all plasma-cell samples, whereas CTAs are significantly more frequent in later stages. HM1.24-specific T-cell responses, representing the immunologic status, significantly decreased from healthy donors to advanced disease. For the CTAs, the probability of T-cell responses increased in early and advanced stages compared with healthy donors, paralleling increased probability of expression. In advanced stages, T-cell responses decreased because of immunoparesis. Conclusion: In conclusion, specific T-cell responses in myeloma are triggered by antigen expression but suppressed by tumor load. Future CTA-based immunotherapeutical approaches might target early plasma-cell diseases to establish prophylactically a specific T-cell response against late-stage antigens in immunocompetent patients. Clin Cancer Res; 21(7); 1712–21. ©2015 AACR.

Parametric Histogram Analysis of Dynamic Contrast-enhanced MRI in Multiple Myeloma: A Technique to Evaluate Angiogenic Response to Therapy?

RATIONALE AND OBJECTIVES From dynamic contrast-enhanced magnetic resonance imaging, it is known that microcirculation patterns in multiple myeloma differ depending on the infiltration pattern. The purpose of this study was to evaluate histogram analysis of dynamic contrast-enhanced magnetic resonance imaging in MM to monitor early treatment response on the basis of microcirculation patterns. MATERIALS AND METHODS A total of 51 patients with multiple myeloma requiring therapy were examined. Dynamic contrast-enhanced magnetic resonance imaging of the lumbar spine was performed before and after conventional or high-dose chemotherapy with autologous stem cell transplantation. Statistical analysis included 245 vertebrae and dynamic microcirculation parameters as displayed in histograms. Resulting parameters (amplitude, exchange rate constant, skewness, kurtosis, and left shift) were correlated with therapeutic response. RESULTS More than 70% of histograms derived from the microcirculation parameters showed a difference between the maximum peak before and after therapy (left shift). However, there was no significant difference between the particular treatment. Significantly different skewness of amplitude in 98% and kurtosis of exchange rate constant (94.1% and 98%) were seen in the patients who responded to treatment (P for each < .05). CONCLUSIONS Histogram analysis revealed early changes after therapy resulting in a shift toward more (kurtosis) and lower values (skewness) of microcirculation parameters. Therefore, histogram analysis can determine and describe if a chosen therapy works at all. However, there were no differences between the chosen therapies. This needs to be reevaluated in a larger number of treated patients. Histogram analysis can also be an adjunct to a subjective visual analysis but is hampered by heterogeneous infiltration pattern seen in multiple myeloma.

NPM1 is overexpressed in hyperdiploid multiple myeloma due to a gain of chromosome 5 but is not delocalized to the cytoplasm

Multiple myeloma (MM) is proposed to consist of two main pathogenetic groups. Although hyperdiploid MM (HD) is characterized by multiple trisomies of odd chromosomes, in nonhyperdiploid MM (NHD), one of the recurrent primary immunoglobulin heavy chain (IGH) translocations and deletion of chromosome 13 can frequently be found. In this study, we analyzed gene‐expression profiles of patients with previously untreated MM. Fifty‐four genes were significantly differentially expressed between the two groups. NPM1 was upregulated in HD. The differential expression of 25 genes, including NPM1 and 13 ribosomal protein genes, was validated using a published gene expression data set. The overexpression of NPM1 in HD was further confirmed by quantitative real‐time PCR and Western blotting. NPM1 was significantly overexpressed in HD as the result of a gain of chromosome 5. Insertions into exon 12 of NPM1 were not detected. NPM1 was localized to the nucleoli of MM cells. Furthermore, HD was associated with an overexpression of ribosomal protein genes, independent of their localization on the trisomic or other chromosomes. Our results indicate that the gain of chromosome 5 might play an important role in the pathogenesis of HD.

Profound impact of sample processing delay on gene expression of multiple myeloma plasma cells

BackgroundGene expression profiling (GEP) has significantly contributed to the elucidation of the molecular heterogeneity of multiple myeloma plasma cells (MMPC) and only recently it has been recommended for risk stratification. Prior to GEP MMPC need to be enriched resulting in an inability to immediately freeze bone marrow aspirates or use RNA stabilization reagents. As a result in multi-center MM trials sample processing delay due to shipping may be an important confounder of molecular analyses and risk stratification based on GEP data.ResultsWe compared GEP data of 145 in-house and 246 shipped samples and detected 3301 down-regulated and 3501 up-regulated genes in shipped samples. For 3994 genes we confirmed differential expression in an independent set of 85 in-house and 97 shipped samples. Differentially expressed genes were enriched in processes like ribosome biogenesis, cell cycle, and apoptosis. Among GEP based risk predictors the IFM-15 seemed to underestimate high risk in shipped samples, whereas the GEP70 and the EMC-92 gene signatures were more robust. In order to provide a tool to assess the “shipping effect” in public repositories, we generated a 17-gene predictor for shipped samples with a 10-fold cross validation error rate of 0.06 for the training set and an error rate of 0.15 for the validation set.ConclusionSample processing delay significantly influences GEP of MMPC, implying it should be avoided if samples were used for risk stratification.

RNA based individualized drug selection in breast cancer patients without patient-matched normal tissue

Background While standard RNA expression tests stratify patients into risk groups, RNA-Seq can guide personalized drug selection based on expressed mutations, fusion genes, and differential expression (DE) between tumor and normal tissue. However, patient-matched normal tissue may be unavailable. Additionally, biological variability in normal tissue and technological biases may confound results. Therefore, we present normal expression reference data for two sequencing methods that are suitable for breast biopsies. Results We identified breast cancer related and drug related genes that are expressed uniformly across our normal samples. Large subsets of these genes are identical for formalin fixed paraffin embedded samples and fresh frozen samples. Adipocyte signatures were detected in frozen compared to formalin samples, prepared by surgeons and pathologists, respectively. Gene expression confounded by adipocytes was identified using fat tissue samples. Finally, immune repertoire statistics were obtained for healthy breast, tumor and fat tissues. Conclusions Our reference data can be used with patient tumor samples that are asservated and sequenced with a matching aforementioned method. Coefficients of variation are given for normal gene expression. Thus, potential drug selection can be based on confidently overexpressed genes and immune repertoire statistics. Materials and Methods Normal expression from formalin and frozen healthy breast tissue samples using Roche Kapa RiboErase (total RNA) (19 formalin, 9 frozen) and Illumina TruSeq RNA Access (targeted RNA-Seq, aka TruSeq RNA Exome) (11 formalin, 1 frozen), and fat tissue (6 frozen Access). Tumor DE using 10 formalin total RNA tumor samples and 1 frozen targeted RNA tumor sample.

Detection of Cross-Sample Contamination in Multiple Myeloma Samples and Sequencing Data

The increasing applicability and sensitivity of next generation sequencing methods exacerbate one of the main issues in the molecular biology laboratory, namely cross-sample contamination. This type of contamination, which could massively increase the rate of false-positive calls in sequencing experiments, can originate at each step during the processing of multiple myeloma samples, such as CD138-selection of tumor cells, RNA and DNA isolation or the processing of sequencing libraries. Here we describe a Droplet Digital PCR (ddPCR) method and a simple bioinformatic solution for the detection of contamination in patients samples and derived sequencing data, which are based on the same principle: detection of alternative alleles for single-nucleotide polymorphisms (SNPs) that are homozygous according to the control (germ line) sample.