ALBERT

Description: 〈div data-abstract-type="normal"〉〈p〉Galloping is a type of fluid-elastic instability phenomenon characterized by large-amplitude low-frequency oscillations of the structure. The aim of the present study is to reveal the underlying mechanisms of galloping of a square cylinder at low Reynolds numbers (〈span〉〈span〉〈img data-mimesubtype="gif" data-type="simple" src="http://static.cambridge.org/resource/id/urn:cambridge.org:id:binary:20190326101636644-0623:S0022112019001605:S0022112019001605_inline1.gif"〉 〈span data-mathjax-type="texmath"〉 〈/span〉 〈/span〉〈/span〉) via linear stability analysis (LSA) and direct numerical simulations. The LSA model is constructed by coupling a reduced-order fluid model with the structure motion equation. The relevant unstable modes are first yielded by LSA, and then the development and evolution of these modes are investigated using direct numerical simulations. It is found that, for certain combinations of 〈span〉〈span〉〈img data-mimesubtype="gif" data-type="simple" src="http://static.cambridge.org/resource/id/urn:cambridge.org:id:binary:20190326101636644-0623:S0022112019001605:S0022112019001605_inline2.gif"〉 〈span data-mathjax-type="texmath"〉 〈/span〉 〈/span〉〈/span〉 and mass ratio (〈span〉〈span〉〈img data-mimesubtype="gif" data-type="simple" src="http://static.cambridge.org/resource/id/urn:cambridge.org:id:binary:20190326101636644-0623:S0022112019001605:S0022112019001605_inline3.gif"〉 〈span data-mathjax-type="texmath"〉 〈/span〉 〈/span〉〈/span〉), the structure mode (SM) becomes unstable beyond a critical reduced velocity 〈span〉〈span〉〈img data-mimesubtype="gif" data-type="simple" src="http://static.cambridge.org/resource/id/urn:cambridge.org:id:binary:20190326101636644-0623:S0022112019001605:S0022112019001605_inline4.gif"〉 〈span data-mathjax-type="texmath"〉 〈/span〉 〈/span〉〈/span〉 due to the fluid–structure coupling effect. The galloping oscillation frequency matches exactly the eigenfrequency of the SM, suggesting that the instability of the SM is the primary cause of galloping phenomenon. Nevertheless, the 〈span〉〈span〉〈img data-mimesubtype="gif" data-type="simple" src="http://static.cambridge.org/resource/id/urn:cambridge.org:id:binary:20190326101636644-0623:S0022112019001605:S0022112019001605_inline5.gif"〉 〈span data-mathjax-type="texmath"〉 〈/span〉 〈/span〉〈/span〉 predicted by LSA is significantly lower than the galloping onset 〈span〉〈span〉〈img data-mimesubtype="gif" data-type="simple" src="http://static.cambridge.org/resource/id/urn:cambridge.org:id:binary:20190326101636644-0623:S0022112019001605:S0022112019001605_inline6.gif"〉 〈span data-mathjax-type="texmath"〉 〈/span〉 〈/span〉〈/span〉 obtained from numerical simulations. Further analysis indicates that the discrepancy is caused by the nonlinear competition between the leading fluid mode (FM) and the SM. In the pre-galloping region 〈span〉〈span〉〈img data-mimesubtype="gif" data-type="simple" src="http://static.cambridge.org/resource/id/urn:cambridge.org:id:binary:20190326101636644-0623:S0022112019001605:S0022112019001605_inline7.gif"〉 〈span data-mathjax-type="texmath"〉 〈/span〉 〈/span〉〈/span〉, the FM quickly reaches the nonlinear saturation state and then inhibits the development of the SM, thus postponing the occurrence of galloping. When 〈span〉〈span〉〈img data-mimesubtype="gif" data-type="simple" src="http://static.cambridge.org/resource/id/urn:cambridge.org:id:binary:20190326101636644-0623:S0022112019001605:S0022112019001605_inline8.gif"〉 〈span data-mathjax-type="texmath"〉 〈/span〉 〈/span〉〈/span〉, mode competition is weakened because of the large difference in mode frequencies, and thereby no mode lock-in can happen. Consequently, galloping occurs, with the responses determined by the joint action of SM and FM. The unstable SM leads to the low-frequency large-amplitude vibration of the cylinder, while the unstable FM results in the high-frequency vortex shedding in the wake. The dynamic mode decomposition (DMD) technique is successfully applied to extract the coherent flow structures corresponding to SM and FM, which we refer to as the galloping mode and the von Kármán mode, respectively. In addition, we show that, due to the mode competition mechanism, the galloping-type oscillation completely disappears below a critical mass ratio. From these results, we conclude that transverse galloping of a square cylinder at low 〈span〉〈span〉〈img data-mimesubtype="gif" data-type="simple" src="http://static.cambridge.org/resource/id/urn:cambridge.org:id:binary:20190326101636644-0623:S0022112019001605:S0022112019001605_inline9.gif"〉 〈span data-mathjax-type="texmath"〉 〈/span〉 〈/span〉〈/span〉 is essentially a kind of single-degree-of-freedom (SDOF) flutter, superimposed by a forced vibration induced by the natural vortex shedding. Mode competition between SM and FM in the nonlinear stage can put off the onset of galloping, and can completely suppress the galloping phenomenon at relatively low 〈span〉〈span〉〈img data-mimesubtype="gif" data-type="simple" src="http://static.cambridge.org/resource/id/urn:cambridge.org:id:binary:20190326101636644-0623:S0022112019001605:S0022112019001605_inline10.gif"〉 〈span data-mathjax-type="texmath"〉 〈/span〉 〈/span〉〈/span〉 and low 〈span〉〈span〉〈img data-mimesubtype="gif" data-type="simple" src="http://static.cambridge.org/resource/id/urn:cambridge.org:id:binary:20190326101636644-0623:S0022112019001605:S0022112019001605_inline11.gif"〉 〈span data-mathjax-type="texmath"〉 〈/span〉 〈/span〉〈/span〉 conditions.〈/p〉〈/div〉

Description: Burkitt lymphoma (BL) is the most common non-Hodgkin lymphoma (NHL) in children. Although it accounts for only 1-5% of NHL in adults, approximately 60% of the BL cases diagnosed each year in western countries occur in patients 〉40 years of age. Although adult and pediatric BL cases are indistinguishable by molecular classification, pediatric patients have a significantly better outcome than adults. While translocation of MYC to the immunoglobulin heavy or light chain genes is characteristic of pediatric and adult BL, genetic differences may contribute to the superior clinical outcome of childhood cases. Therefore, we aimed to identify the spectrum of additional genetic abnormalities that occur in adult and pediatric BL. Copy number analysis, gene expression profiling (GEP), and targeted sequencing of ~400 genes known to be mutated in NHLs were performed on a cohort of molecularly defined BL samples. Copy number abnormalities (CNAs) were identified by the Affymetrix 250k NspI SNP array in 73 BL tumors (28 adult, 45 pediatric), and sequencing was performed on 52 BLs (21 adult, 31 pediatric). Pediatric cases had fewer CNAs than adults. The most common focal abnormality identified was a gain on 13q31.3 encompassing MIR17HG. It was more frequent in adult compared to pediatric cases (35% vs 16%, p=0.085) and was associated with increased expression of miR-17~92 cluster members; and among adults, patients with this gain trended towards worse overall survival, though the number of cases with available information was small. Gain of 8q was found in ~20% of adult cases, but in no pediatric cases. Surprisingly, cases with 8q gain had significantly lower MYC mRNA expression (p〈 0.001) and lower protein expression. In cases with MYC gain 0/4 cases were positive for MYC protein expression by immunohistochemistry; in contrast,6/10 cases with no MYC gain were positive for MYC expression. This suggests that gain of 8q is driven by another gene in the region. Additional genetic alterations included gains of genomic loci encompassing MCL1 and MDM4 (1q21-24) and losses encompassing RB1, p53 and CDKN2A/CDKN2B. Pathway analysis of genes differentially expressed by CN status showed an enrichment of genes involved in cell cycle regulation, the p53 signaling pathway, and the ubiquitin proteasome pathway. The frequencies of mutations in commonly mutated genes including MYC, ID3, TP53, CCND3, DDX3X, ARID1A, and TCF3 were not significantly different in adult and pediatric BL. However, BCL2, (43%, p

Description: Background: c-MYC (MYC) is commonly dysregulated in aggressive B cell lymphomas. MYC associated lymphoma, especially Double Hit lymphoma (DHL) and Double-Expression Lymphoma (DEL) which are characterized by MYC and BCL2 dual overexpression usually present with the inferior outcome as rapid disease progression and poor response to standard chemotherapy regimen. Nevertheless, MYC is considered as an "undruggable" target and targeting strategies such as suppressing MYC transcription by bromodomain (BRD)-4 inhibitors have been widely investigated in both preclinical models and clinical trials. However, increasing evidence has shown that lymphoma cells displayed a wide range of resistance to BRD-4 inhibition, due to transcription adaptation or kinome reprogramming. Hence, alternative approaches for suppressing MYC or its function are urgently needed. Strategies directed against oncoprotein translation may efficiently repress key oncoproteins regardless of the abundant MYC mRNA due to genetic aberrations and secondary transcription up-regulated by MYC. Rocaglate is a class of natural products derived from plants of the Aglaia genus that have been demonstrated to potently inhibit protein translation initiation via eIF4A. The use of rocaglates for anti-cancer treatment was limited due to the scarcity and instability of these natural products e.g. Silvestrol. Recent chemical modification and screening studies have unveiled a few synthetic rocaglates that are more potent than Silvestrol such as SDS-1-021-(−), which unlocked the potential use of rocaglates for clinical applications. Methods and Results: To probe effective reagents for MYC-driven lymphoma, a screening of 50 drugs targeting common oncogenic pathways and tumorigenic machinery was performed in two isogenic B-lymphoma lines. Several protein translation inhibitors, such as mTOR kinase inhibitors (TORKi), and eIF4A inhibitor Silvestrol were identified as the most potent drugs in all of the tested cells. Further, we found that Silvestrol but not TORKi efficiently repressed MYC protein translation in MYC-driven B lymphoma cells, whereas neither of them inhibited BCL2 expression. Moreover, we demonstrated that eIF4E knockdown or eIF4E/eIF4G disruptor Briciclib did not significantly affect MYC expression, whereas eIF4A inhibitor hippuristanol and rocaglates derivate SDS-1-021-(−) diminished MYC expression similar to that observed in Silvestrol treated cells. By using dual luciferase assay, we demonstrated that rocaglates had stronger cap-dependent and IRES translation inhibition than TORKi in lymphoma cells. Next, by native RNA immunoprecipitation and siRNA knocking down, we found that rocaglates repressed MYC translation not only via eIF4A1 but also via eIF4A2, however, with different underlying mechanisms. Furthermore, to explore the molecular targets of rocaglates treatment in B cell lymphoma, we performed TMT-Mass Spectrometry which identified multiple oncoproteins including NEK2, MYC, MCL1, TCF3, BCL6, PLK1, AURKA, and WEE1 were significantly down-regulated by SDS-1-021-(−) treatment. Finally, we demonstrated that SDS-1-021-(−) is highly potent as a single agent and synergized with ABT199 at a low dose (0.2mg/kg) in PDX models with DHL/DEL. Brief summary : Our pre-clinical study provided strong evidence that rocaglates but not TORKi efficiently suppress MYC protein translation because 1). Rocaglates exhibit strong inhibition on both Cap- and IRES-dependent translation, 2). Rocaglates decrease PLK1 and AURKA/B thus destabilizing MYC protein. The synthetic rocaglate SDS-1-021-(−) is a potent agent that exhibits significant synergistic killing effect with ABT199 on DHL/DEL cells in the pre-clinical animal study. Figure Figure. Disclosures Lunning: TG Therapeutics: Consultancy; AbbVie: Consultancy; Genentech: Consultancy; Astra-Zeneca: Consultancy; Genzyme: Consultancy; Celgene: Consultancy; Bayer: Consultancy; Gilead: Consultancy; Spectrum: Consultancy; Genentech: Consultancy; Seattle Genetics: Consultancy; Portola: Consultancy; Kite: Consultancy; Juno: Consultancy; Janssen: Consultancy; Verastem: Consultancy. Vose:Novartis: Honoraria, Research Funding; Epizyme: Honoraria; Incyte Corp.: Research Funding; Bristol Myers Squibb: Research Funding; Kite Pharma: Research Funding; Merck Sharp & Dohme Corp.: Research Funding; Abbvie: Honoraria; Legend Pharmaceuticals: Honoraria; Seattle Genetics, Inc.: Research Funding; Acerta Pharma: Research Funding; Celgene: Research Funding; Roche: Honoraria.

Description: Peripheral T-cell lymphoma (PTCL) is a group of complex clinicopathological entities, often associated with an aggressive clinical course. Angioimmunoblastic T-cell lymphoma (AITL) and PTCL-not otherwise specified (PTCL-NOS) are the 2 most frequent categories, accounting for 〉50% of PTCLs. Gene expression profiling (GEP) defined molecular signatures for AITL and delineated biological and prognostic subgroups within PTCL-NOS (PTCL-GATA3 and PTCL-TBX21). Genomic copy number (CN) analysis and targeted sequencing of these molecular subgroups revealed unique CN abnormalities (CNAs) and oncogenic pathways, indicating distinct oncogenic evolution. PTCL-GATA3 exhibited greater genomic complexity that was characterized by frequent loss or mutation of tumor suppressor genes targeting the CDKN2A/B-TP53 axis and PTEN-PI3K pathways. Co-occurring gains/amplifications of STAT3 and MYC occurred in PTCL-GATA3. Several CNAs, in particular loss of CDKN2A, exhibited prognostic significance in PTCL-NOS as a single entity and in the PTCL-GATA3 subgroup. The PTCL-TBX21 subgroup had fewer CNAs, primarily targeting cytotoxic effector genes, and was enriched in mutations of genes regulating DNA methylation. CNAs affecting metabolic processes regulating RNA/protein degradation and T-cell receptor signaling were common in both subgroups. AITL showed lower genomic complexity compared with other PTCL entities, with frequent co-occurring gains of chromosome 5 (chr5) and chr21 that were significantly associated with IDH2R172 mutation. CN losses were enriched in genes regulating PI3K–AKT–mTOR signaling in cases without IDH2 mutation. Overall, we demonstrated that novel GEP-defined PTCL subgroups likely evolve by distinct genetic pathways and provided biological rationale for therapies that may be investigated in future clinical trials.

Description: Key Points Adult-mBLs have distinct and more frequent DNA copy number abnormalities compared with pediatric-mBL. Comprehensive genomic analysis revealed that the BCR signaling pathway is a potential therapeutic target in adult-mBL.

Description: Key Points IDH2 R172 mutations define a unique subgroup with distinct TFH-like gene expression signatures in AITL. IDH2 R172 mutations can induce DNA and repressive histone hypermethylation in AITL.

Description: Introduction: Diversity of the T-cell receptor (TCR) repertoire reflects the initial V(D)J recombination events as shaped by selection by self and foreign antigens. Next generation sequencing is a powerful method for profiling the TCR repertoire, including sequences encoding complementarity-determining region 3 (CDR3). Peripheral T-cell lymphoma (PTCL) is a group of malignancies that originate from mature T-cells. T-cell clonality of PTCL is routinely evaluated with a PCR-based method to detect TCR gamma and less frequently beta chain rearrangements using genomic DNA. However, there are limitations with this approach, chief among which is the lack of sequence information. To date, the TCR repertoire of different subtypes of PTCL remains poorly defined. Objective: The purpose of this study was to determine the utility of RNA-seq for assessing T-cell clonality and analyzing the TCR usage in PTCL samples. Methods: We analyzed RNA-seq data from 30 angioimmunoblastic T-cell lymphoma (AITL), 23 Anaplastic large cell lymphoma (ALCL), 10 PTCL-NOS, and 17 NKCL. Data from naïve T cells, TFH cells, and T-effector cells (CD4+ CD45RA− TCRβ+ PD-1lo CXCR5lo PSGL-1hi) were obtained from publicly available resources. Referenced TCR and immunoglobulin transcripts according to the International ImMunoGeneTics Information System (IMGT) database were quantified by Kallisto software. We divided the pattern of Vβ (T-cell receptor beta variable region) into three categories: monoclonal (mono- or bi-allelic), oligoclonal (3-4 dominant clones), and polyclonal. CDR3 sequences were extracted by MiXCR program. PCR of the gamma chain using genomic DNA was utilized to validate the clonality of selected cases. Single nucleotide variants (SNVs) were called from aligned RNA-seq data using Samtools and VarScan 2 programs. Results: Analysis of RNA-seq data identified preferential usage of TCR-Vβ, Dβ (diversity region), and Jβ (joining region), length diversity of CDR3, and usage of nontemplated bases. Dominant clones could be identified by transcriptome sequencing in most cases of AITL (21/30), ALCL (14/23), and PTCL-NOS (7/10). Median CDR3 length is 42 nucleotides (nt) in normal T cells, 41 nt in ALCL, 48 nt in PTCL-NOS, and 44 nt in AITL. In 30 AITL samples, 20 showed monoclonal Vβ with a single peak, and 9 showed polyclonal Vβ. One case had two dominant clones with different CDR3, only one of which was in frame, implying biallelic rearrangements. As many as 3511 clones supported by at least four reads could be detected in polyclonal cases. In monoclonal cases, the dominant clone varied between 11.8% and 92.8% of TCR with Vβ rearrangements. TRBV 20-1, which is the most commonly used segment in normal T cells, is also frequently used in the dominant clones in AITL. The monoclonal AITL cases showed mutation of TET2, RHOA, DNMT3A or IDH2 whereas most of the polyclonal cases were negative or had low VAF mutation suggesting low or absent of tumor infiltrate in the specimen sequenced. There is no obvious correlation of any of the mutations with Vβ usage. Clonal B cell expansion was noted in some AITL samples. The occurrence of a preferential TRBV9 expansion in PTCL-NOS was striking. More than half of ALCL samples (14/23) showed expression of clonal Vβ, but 3/14 dominant clones were out-of-frame. γ chain expression was very low in cells expressing TCRαβ, but both expression levels and clonality were higher in TCRγδ expressing tumors. NKCL did not express significant levels of TCR Vβ or Vγ genes. Discussion/Interpretation: Transcriptome sequencing is a useful tool for understanding the TCR repertoire in T cell lymphoma and for detecting clonality for diagnosis. Clonal, often out-of-frame, Vβ transcripts are detectable in most ALCL cases and preferential TRBV9 usage is found in PTCL-NOS. Disclosures No relevant conflicts of interest to declare.

Description: Background: Myeloproliferative neoplasia (MPN) and angioimmunoblastic T-cell lymphoma (AITL) are two distinct hematologic cancers. However, we have shown that patients with both malignancies probably occur with frequencies higher than expected in the background population (1,2). AITL tumors show a high frequency (30-70%) of mutations in RHOA and epigenetic modifier genes such as DNMT3A, TET2 and IDH2. The latter genomic alterations are similar to those found in myeloid diseases such as MPN. This observation has raised questions on whether the two hematologic malignancies may share genomic aberrations (e.g. at progenitor level) suggesting a possible pathogenetic relationship between these diseases. Aim: To further investigate these questions, we explored the mutational landscape in MPN and AITL tumors occurring in the same host. Methods: From a Danish cohort of 97 patients with co-existing MPN and lymphoma identified through the Danish Lymphoma and Pathology Registries and immunohistochemically validated by tertiary center hematopathologists, five patients with MPN and concurrent and/or subsequent AITL were included. Paired DNA samples obtained from lymph node biopsies (AITL component), bone marrow aspirates (MPN component) and saliva specimens (normal reference DNA) were analyzed by whole exome sequencing. A minimum variant allele frequency threshold of 10% was applied. However, mutations in epigenetic modifier genes, RHOA and JAK2, were also included, if found at slightly lower frequencies. Results: Figure 1 shows an overview of the identified mutations. At a VAF% cut-off value of 10 and a median coverage of 30, mutations shared between the MPN and AITL specimens were found within the DNMT3A, TET2, and IDH2 genes. Mutations in the JAK2 gene were identified in all MPN and in some of the AITL samples. NOTCH2 mutations were frequent and some were shared between the MPN and AITL samples. Two patients with essential thrombocytosis and AITL had the same NOTCH2 (C19W) and NCOR1 (K178N) mutation. Most of the identified genomic alterations were inactivating missense mutations. Conclusion: Co-existence of MPN and AITL occurs and may not be a random event. We studied tumor tissues from patients diagnosed with both diseases, and identified recurrent shared and non-shared mutations in samples from both types of cancers. These mutations may contribute to the development of the two malignancies in the same host, partly through common steps at precursor level (e.g. DNMT3A) and partly more downstream at a more lineage-specific level (e.g. IDH2). The functional role of the observed NOTCH2 and NCOR1 mutations is currently under investigation. Holst J et al. Blood 2017, 130 (S1): 1525 Frederiksen H et al. Blood 2011;118(25):6515-6520 Disclosures Tam: Takeda: Consultancy; Paragon Genomics: Consultancy.

Description: Introduction: Transformed nodular lymphocyte predominant Hodgkin lymphoma (tNLPHL) with a typical diffuse large B-cell lymphoma (DLBCL) pattern is rare and not well studied by genomic analysis. We employed next generation sequencing and copy number analysis (CNA) to examine the pathogenesis of these tumors. Methods: We identified 19 cases of tNLPHL with DLBCL morphology and sheet-like growth from three institutions. NLPHL preceded transformation in 5 patients and was concurrent with transformation in 11. All cases of tNLPHL were sequenced using a targeted sequencing panel of 356 genes that included commonly mutated genes associated with lymphoma. We had 8 cases with matched germline DNA. We also performed CNA using Oncoscan on 18 cases of tNLPHL. Library preparation with paired end 100 bp sequencing and 6-10 million reads/case was performed on an Illumina HiSeq 2500. Fisher's exact test was used to compare the role of mutations in tNLPHL to three large series of de novo DLBCL. Results: The CNA showed frequent gains in REL and loss of CDKN2A. Mutation analysis showed frequent mutations of genes associated with the PI3K pathway such as SGK1 (26%), ZFP36L1 (16%), PIK3R1 (11%), and IL7R (11%), the NF-kB pathway such as CARD11 (21%), JUNB (21%), BCL10 (11%), NFKBIA (11%), TNFAIP3 (11%), histone/DNA modification such as KMT2D (26%), EP300 (21%), TET2 (11%), TET3 (11%), and the NOTCH pathway such as NOTCH2 (16%), NOTCH1 (1 case), CTBP2 (11%). Mutations in genes involved in immune surveillance and TP53 abnormalities were infrequent. Compared to de novo DLBCL, mutations in IL7R (10.5% vs 0.6%, p=0.03), JUNB (21% vs 4.2%, p=0.01), and SMARCAL1 (11% vs 0%, p=0.01) were significantly higher in tNLPHL than in germinal center B-cell (GCB) subtype of DLBCL. Conclusion: The mutational spectrum of tNLPHL resembles the DLBCL Cluster 4 of Chapuy et al (Nat Med, 2018), which were primarily GCB-DLBCL with frequent mutations in the PI3K pathway (SGK1), NF-kB pathway (CARD11, JUNB), and histone modification. The mutational spectrum is also distinctive in having frequent mutations that are not often seen together in DLBCL, such as TET2, JUNB and NOTCH2. Distinct from transformed follicular lymphoma, TP53 abnormalities and mutations affecting immune surveillance are uncommonly observed. This study provides new insights into the biology of tNLPHL and may highlight potential targets for therapy in the future. Disclosures Herrera: Adaptive Biotechnologies: Consultancy; Bristol-Myers Squibb: Consultancy, Research Funding; Gilead Sciences: Consultancy, Research Funding; Seattle Genetics: Consultancy, Research Funding; AstraZeneca: Research Funding; Merck: Consultancy, Research Funding; Genentech, Inc.: Consultancy, Research Funding; Pharmacyclics: Research Funding; Immune Design: Research Funding; Kite Pharma: Consultancy, Research Funding.

Description: In this study, a CFD-based linear dynamics model combined with the direct Computational Fluid Dynamics/Computational Structural Dynamics (CFD/CSD) simulation method is utilized to study the physical mechanisms underlying frequency lock-in in vortex-induced vibrations (VIVs). An identification method is employed to construct the reduced-order models (ROMs) of unsteady aerodynamics for the incompressible flow past a vibrating cylinder at low Reynolds numbers . Reduced-order-model-based fluid-structure interaction models for VIV are also constructed by coupling ROMs and structural motion equations. The effects of the natural frequency of the cylinder, mass ratio and structural damping coefficient on the dynamics of the coupled system at are investigated. The results show that the frequency lock-in phenomenon at low Reynolds numbers can be divided into two patterns according to different induced mechanisms. The two patterns are 'resonance-induced lock-in' and 'flutter-induced lock-in'. When the natural frequency of the cylinder is in the vicinity of the eigenfrequency of the uncoupled wake mode (WM), only the WM is unstable. The dynamics of the coupled system is dominated by resonance. Meanwhile, for relatively high natural frequencies (i.e. greater than the eigenfrequency of the uncoupled WM), the structure mode becomes unstable, and the coupling between the two unstable modes eventually leads to flutter. Flutter is the root cause of frequency lock-in and the higher vibration amplitude of the cylinder than that of the resonance region. This result provides evidence for the finding of De Langre (J. Fluids Struct., vol. 22, 2006, pp. 783-791) that frequency lock-in is caused by coupled-mode flutter. The linear model exactly predicts the onset reduced velocity of frequency lock-in compared with that of direct numerical simulations. In addition, the transition frequency predicted by the linear model is in close coincidence with the amplitude of the lift coefficient of a fixed cylinder for high mass ratios. Therefore, it confirms that linear models can capture a significant part of the inherent physics of the frequency lock-in phenomenon. © 2015 Cambridge University Press.