Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Serial exome analysis of disease progression in premalignant gammopathies

Nearly all human cancers are preceded by precursor states far more prevalent than clinical malignancy itself. Understanding which premalignant lesions are truly biologically premalignant versus those with impending clinical progression (biologically malignant) is a central question in cancer biology. Tracking the progression of human premalignant states (or lack thereof) in vivo is not feasible in most tumors because precursor lesions are typically resected at diagnosis. Multiple myeloma (MM) is characterized by progressive growth of malignant plasma cells (PCs) in the bone marrow leading to organ dysfunction.1 Nearly all cases of clinical MM are preceded by asymptomatic monoclonal gammopathies (AMGs) further classified as either monoclonal gammopathy of undetermined significance (MGUS) or asymptomatic myeloma (AMM).1, 2 MM is a unique model to dissect genetic evolution of early cancer as the precursor state is well defined and not resectable. Several studies have characterized the genetic architecture of malignant PCs in both MM and its precursor states.3 MM tumor cells carry several cytogenetic abnormalities (most notably, IgH translocations and hyperdiploid karyotypes), and copy number abnormalities leading to genomic gains/losses and loss of heterozygosity (LOH). Comparisons of PCs in cohorts of MM versus MGUS/AMM show greater proportion of cytogenetically abnormal PCs in MM, suggesting the expansion of pre-existing and more proliferative clones during transition to MM.4, 5 However, nearly all of the cytogenetic changes described in MM tumor cells have also been observed in MGUS/AMM, and it has proven difficult to define the malignant phenotype on the basis of genetic changes alone.4, 5 Whole-genome and whole-exome-based sequencing strategies have recently been applied to human MM and shown that there are about 20–40 non-synonymous variants per MM tumor cell, but no single unifying variant has been identified.6, 7, 8 The most common targets of recurrent mutation were known cancer genes, such as KRAS and TP53, with KRAS mutations in 27% of the cases.6 Analysis of serial samples from MM patients undergoing therapy recently demonstrated the presence of subclones, with greater changes in patients with cytogenetically high-risk disease.9, 10 Similar findings were made via serial genome sequencing in a patient with t(4:14) myeloma evolving to PCs leukemia.8 In this study, we serially analyzed the exomes of patients with AMGs who progressed to clinical MM and compared the findings with those AMGs that did not progress to clinical MM.

Whole-exome sequencing was performed on tumor and germline DNA from 10 patients with AMGs (exome run summary in Supplementary Table 1, see Supplementary Methods online). Clinical characteristics of this cohort are shown in Supplementary Table 2 and these patients could be classified as MGUS (n=3) or AMM (n=7) by current criteria. Four of these patients (P1–P4) progressed to clinical MM requiring therapy, whereas 6 patients (NP1–NP6) that did not progress with at least 2 years of follow-up were studied as controls. Overall 261 somatic non-synonymous variants (SNVs) were seen in these 10 patients, with a mean of 26 variants per patient (range 10–74). The mean number of SNVs per sample was similar between patients with progressive or non-progressive disease (Figure 1a). In patients who progressed to clinical MM (P1–P4), the sequencing data from baseline was directly compared with that from the corresponding malignancy. In three of the paired samples, there were no new SNVs detected in the progression sample (Figure 1b). In the fourth patient (P2), a new SNV in a single gene (BBOX1) of unclear significance was detected in the progression sample. The lack of significant evolution from baseline to progression was also confirmed through analysis of minor/alternate-allele frequencies in the paired samples (Figure 1c). Variant calls from all progression samples were validated using Sanger sequencing (not shown). Serial samples (at 2 and 4 years of follow-up) were also available from a patient (NP6) without disease progression. These samples again demonstrated stable SNV profile, with balanced loss and gain of two new detectable SNVs at each time point (Figure 1b). Together these data indicated that nearly all of the SNVs found at the diagnosis of clinical MM were also present in the precursor state.

Figure 1

Analysis of somatic variants. (a) Number of somatic variants per sample. (b) Serial analysis of somatic variants in each patients. (c) Analysis of B-allele frequencies in paired samples.

In view of lack of consistent progression-specific SNVs, we turned our attention to regions of somatic copy number alterations (CNAs), intra-clonal variation and pattern of somatic mutations. Analysis of LOH patterns revealed that the degree of LOH at baseline was much greater in patients with Prog-AMG than in NonProg-AMG, although there was no consistent pattern shared by all samples (Figure 2a). In the patients who progressed to MM, the LOH regions at MM stage matched those at baseline with little change. However, in two patients (P2 and P3), new regions of MM-specific LOH were identified (Figures 2a–c). Interestingly, in one of these patients (P3), the region of new LOH correlated with the site of p53 on chromosome 17 (Figure 2b) and possibly involved a dominant subclone as the contribution of the cells carrying the deletion was 10% lower than the estimated fraction of tumor cells in this sample. Profile of somatic CNAs again revealed several regions of alterations throughout the genome, which were again largely maintained between baseline and progression samples (Figure 2d). Together these data demonstrate that patients with progressive lesions may have a greater degree of genomic gains/losses compared with those that remain clinically asymptomatic.

Figure 2

Analysis of genomic gains/losses with disease progression. (a) Analysis of LOH patterns. (b) Analysis of copy number variation (CNV) pattern in chromosome 17 in patient P3. (c) Analysis of CNV pattern in chromosome 8 in patient P2. (d) Analysis of CNVs. (e) Comparison of mutational profile with published data in patients with clinical myeloma.

An important and unique aspect of this study is the concurrent analysis of samples from patients with AMG that did not progress to clinical MM. Therefore, we compared the targets of somatic mutations in this cohort with those published for patients with MM.6, 7, 8 Targets of somatic mutations in Prog-AMGs had a greater overlap than NonProg-AMGs with the MM dataset (19 of 74 (26%) in Prog-AMG versus 20 of 181 (11%) in NonProg-AMG, P=0.01) (Figure 2e). This was also true when mutations in known oncogenes in the COSMIC database overlapping between the current AMG cohorts and the MM datasets were compared (3 of 5 in Prog-AMG versus 0 of 10 in NonProg-AMG) (Figure 2e). Notably, this included KRAS and TP53 wherein mutations were only detected in Prog-AMG, but not in patients with NonProg-AMG. Therefore, the pattern of mutations at baseline may correlate with the risk of progression to MM.

These data demonstrate that the vast majority of somatic mutations in MM are present at least several months and possibly years before the onset of symptoms indicative of the diagnosis of clinical malignancy. Similar findings were recently reported by Walker et al.11 who compared exomes from four patients with AMM and corresponding MM. Interestingly, the net mutation load was similar between progressive and non-progressive lesions, demonstrating that long periods of clinical stability are feasible in AMGs in spite of the level of mutational load comparable to that seen in clinical MM. Instead of the number of mutations, the pattern of mutations and genomic changes at baseline may impact the risk of developing clinical malignancy. In such a model, lesions that lack mutations in critical genes are unable to mediate progressive growth needed for malignant phenotype. The list of targets preferentially mutated in patients with Prog-AMGs consisted of well-known cancer genes, such as KRAS and TP53.6 The presence of mutations in KRAS in myeloma has been previously correlated with an aggressive course and disease progression.12 Cytogenetic deletion of chromosome 17p (carrying TP53) has also been well-documented as a feature of high-risk MM. Mutations in TP53 have also been described in MM and correlate with the presence of del17p.13, 14 Further studies in asymptomatic patients are needed to identify additional mutations that may predict the increased risk of malignancy.

Somatic LOH and CNA analyses indicated that the genomes in Prog-AMG also carried greater regions of genomic loss/gains than in patients with NonProg-AMG. These data suggest that the degree of genomic instability may itself be a marker of the risk for malignant transformation. In some instances, it is possible that these secondary changes may contribute more directly to malignant transformation through evolution of a subclone. For example, in patient P3, progression to MM was accompanied with the appearance of new LOH interval in chromosome 17p, including TP53.

To our knowledge, these data provide the first direct comparison of serial exomes in progressive versus non-progressive precursor states in human cancer. In contrast to recent studies with myelodysplasia and acute leukemia, the precursor state in this setting was observed without any specific therapy, therefore, the findings are not impacted by possible therapy-induced effects.15 Our findings demonstrate that most somatic mutations and genome alterations predate clinical malignancy and that mutational loads comparable to clinical malignancy could be associated with long periods of dormancy in premalignant states. Instead of net mutational load, the pattern of mutations and the degree of genomic instability may predict malignant fate. Our findings should be considered as preliminary due to the small numbers of AMG exomes sequenced thus far. Analysis of spectrum of mutations in precursor states may identify lesions that are more likely to progress to clinical cancer.


  1. 1

    Landgren O, Kyle RA, Rajkumar SV . From myeloma precursor disease to multiple myeloma: new diagnostic concepts and opportunities for early intervention. Clin Cancer Res 2011; 17: 1243–1252.

    Article  PubMed  PubMed Central  Google Scholar 

  2. 2

    Dhodapkar MV, Sexton R, Waheed S, Usmani S, Papanikolaou X, Nair B et al. Clinical, genomic and imaging predictors of myeloma progression from asymptomatic monoclonal gammopathies (SWOG S0120). Blood 2013; 123: 78–85.

    Article  PubMed  Google Scholar 

  3. 3

    Morgan GJ, Walker BA, Davies FE . The genetic architecture of multiple myeloma. Nat Rev Cancer 2012; 12: 335–348.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  4. 4

    Lopez-Corral L, Gutierrez NC, Vidriales MB, Mateos MV, Rasillo A, Garcia-Sanz R et al. The progression from MGUS to smoldering myeloma and eventually to multiple myeloma involves a clonal expansion of genetically abnormal plasma cells. Clin Cancer Res 2011; 17: 1692–1700.

    CAS  Article  PubMed  Google Scholar 

  5. 5

    Lopez-Corral L, Sarasquete ME, Bea S, Garcia-Sanz R, Mateos MV, Corchete LA et al. SNP-based mapping arrays reveal high genomic complexity in monoclonal gammopathies, from MGUS to myeloma status. Leukemia 2012; 26: 2521–2529.

    CAS  Article  PubMed  Google Scholar 

  6. 6

    Chapman MA, Lawrence MS, Keats JJ, Cibulskis K, Sougnez C, Schinzel AC et al. Initial genome sequencing and analysis of multiple myeloma. Nature 2011; 471: 467–472.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  7. 7

    Walker BA, Wardell CP, Melchor L, Hulkki S, Potter NE, Johnson DC et al. Intraclonal heterogeneity and distinct molecular mechanisms characterize the development of t(4;14) and t(11;14) myeloma. Blood 2012; 120: 1077–1078.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  8. 8

    Egan JB, Shi CX, Tembe W, Christoforides A, Kurdoglu A, Sinari S et al. Whole genome sequencing of multiple myeloma from diagnosis to plasma cell leukemia reveals genomic initiating events, evolution and clonal tides. Blood 2012; 120: 1060–1066.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  9. 9

    Keats JJ, Chesi M, Egan JB, Garbitt VM, Palmer SE, Braggio E et al. Clonal competition with alternating dominance in multiple myeloma. Blood 2012; 120: 1067–1076.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  10. 10

    Weston-Bell N, Gibson J, John M, Ennis S, Pfeifer S, Cezard T et al. Exome sequencing in tracking clonal evolution in multiple myeloma following therapy. Leukemia 2013; 27: 1188–1191.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  11. 11

    Walker BA, Wardell CP, Melchor L, Brioli A, Johnson DC, Kaiser MF et al. Intraclonal heterogeneity is a critical early event in the development of myeloma and precedes the development of clinical symptoms. Leukemia 2013; 28: 384–90.

    Article  PubMed  PubMed Central  Google Scholar 

  12. 12

    Neri A, Murphy JP, Cro L, Ferrero D, Tarella C, Baldini L et al. Ras oncogene mutation in multiple myeloma. J Exp Med 1989; 170: 1715–1725.

    CAS  Article  PubMed  Google Scholar 

  13. 13

    Chng WJ, Price-Troska T, Gonzalez-Paz N, Van Wier S, Jacobus S, Blood E et al. Clinical significance of TP53 mutation in myeloma. Leukemia 2007; 21: 582–584.

    CAS  Article  PubMed  Google Scholar 

  14. 14

    Xiong W, Wu X, Starnes S, Johnson SK, Haessler J, Wang S et al. An analysis of the clinical and biologic significance of TP53 loss and the identification of potential novel transcriptional targets of TP53 in multiple myeloma. Blood 2008; 112: 4235–4246.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  15. 15

    Walter MJ, Shen D, Ding L, Shao J, Koboldt DC, Chen K et al. Clonal architecture of secondary acute myeloid leukemia. N Engl J Med 2012; 366: 1090–1098.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to M V Dhodapkar.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies this paper on the Leukemia website

Supplementary information

Rights and permissions

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. To view a copy of this license, visit

Reprints and Permissions

About this article

Cite this article

Zhao, S., Choi, M., Heuck, C. et al. Serial exome analysis of disease progression in premalignant gammopathies. Leukemia 28, 1548–1552 (2014).

Download citation

Further reading


Quick links