Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Leukemia. Author manuscript; available in PMC 2014 April 1.
Published in final edited form as:
Published online 2012 October 3. doi:  10.1038/leu.2012.282
PMCID: PMC3972006

Combining fluorescent in situ hybridization data with ISS staging improves risk assessment in myeloma: an International Myeloma Working Group collaborative project

H Avet-Loiseau,1 BGM Durie,2 M Cavo,3 M Attal,4 N Gutierrez,5 J Haessler,6 H Goldschmidt,7 R Hajek,8 JH Lee,9 O Sezer,10 B Barlogie,11 J Crowley,6 R Fonseca,12 N Testoni,13 F Ross,14 SV Rajkumar,15 P Sonneveld,16 J Lahuerta,17 P Moreau,18 G Morgan,19 and on behalf of the International Myeloma Working Group


The combination of serum β2-microglobulin and albumin levels has been shown to be highly prognostic in myeloma as the International Staging System (ISS). The aim of this study was to assess the independent contributions of ISS stage and cytogenetic abnormalities in predicting outcomes. A retrospective analysis of international studies looking at both ISS and cytogenetic abnormalities was performed in order to assess the potential role of combining ISS stage and cytogenetics to predict survival. This international effort used the International Myeloma Working Group database of 12 137 patients treated worldwide for myeloma at diagnosis, of whom 2309 had cytogenetic studies and 5387 had analyses by fluorescent in situ hybridization (iFISH). Comprehensive analyses used 2642 patients with sufficient iFISH data available. Using the comprehensive iFISH data, combining both t(4;14) and deletion (17p), along with ISS stage, significantly improved the prognostic assessment in terms of progression-free survival and overall survival. The additional impact of patient age and use of high-dose therapy was also demonstrated. In conclusion, the combination of iFISH data with ISS staging significantly improves risk assessment in myeloma.

Keywords: myeloma, cytogenetics, prognostic


There have been significant improvements in the outcomes for patients with multiple myeloma over the past 10 years. These improvements are related to the systematic use of high-dose melphalan in the youngest patients, and the availability of novel drugs such as thalidomide, bortezomib and lenalidomide. However, despite these improvements, the course of the disease remains highly variable. While some patients experience 10–15-year survivals, others succumb to highly refractory disease within a few months. Many studies have focused on the description of prognostic factors capable of predicting this heterogeneity in survival.16 Among the large number of predictive parameters, a major advance was achieved with the description of the International Staging System (ISS), which combines serum β2-microglobulin and albumin levels.7 Using this model, patients are classified into three groups with very different overall survival (OS). However, an important limitation of this system is that it does not directly incorporate the role of intrinsic myeloma cell variability at the molecular level. In the past decade, it has been shown that recurrent genetic abnormalities present in the malignant plasma cells displayed a strong prognostic power.815 Among them, the most important are the t(4;14), del(17p) and 1q gains.

To address this issue, we conducted an international retrospective study involving a total population of 12 137 patients from sites used for the ISS analyses.7 From this population, 2642 patients with multiple myeloma analyzed at diagnosis for β2-microglobulin and albumin had complete fluorescent in situ hybridization (iFISH) data available and were used for combined ISS staging and iFISH analyses.


Patients with newly diagnosed multiple myeloma were treated in several institutions worldwide (Table 1), and full details have been published previously.7 The median age was 60 years (range = 23–93 years). In all, 59% of the patients received an intensive approach based on single or double high-dose melphalan courses, whereas 41% received more conventional treatment. The results of these clinical trials have been either fully published or reported in international meetings. None of these patients received bortezomib or lenalidomide as frontline therapy. Conventional cytogenetics were performed in 2309 patients and iFISH in 5387 patients. The iFISH studies were performed on sorted or immunologically recognized plasma cells. Most of the iFISH analyses were focused on del(13), t(4;14), del(17p), t(11;14) and t(14;16). The ISS stages were calculated as published previously: stage I was defined by a β2-microglobulin level lower than 3.5 mg/l and an albumin level above 35 g/l; stage III was defined by a β2-microglobulin level above 5.5 mg/l; and all other patients were classified as stage II.

Table 1
Distribution of iFISH data by country: numbers of patients with data available


The Kaplan–Meier method16 was used to estimate progression-free survival (PFS) and OS. PFS was defined as the maximum time from either the start of diagnosis or the start of treatment date to the occurrence of death from any cause, disease progression or relapse, or censored at the date of last contact. OS was defined as the maximum time from either the diagnosis or the treatment date to the date of death from any cause, or censored at the date of last contact. PFS and OS curves were compared by the log-rank test.17 Groupings based on combinations of ISS stage and cytogenetic abnormalities were explored using recursive partitioning.18


By definition, β2-microglobulin and albumin levels were available for all the patients. We first considered conventional cytogenetics that were performed in 2309 patients. An abnormal karyotype was observed in 33% of patients, a percentage in agreement with those usually reported in multiple myeloma. Karyotypes, when abnormal, were usually complex, which was in agreement with previously reported studies. Specific cytogenetic features evaluated included the following: the presence or absence of any chromosomal cytogenetic abnormalities; 13q – by cytogenetics; and hypodiploidy and hyperdiploidy. The presence of any abnormal karyotype, cytogenetic 13q – and hypodiploidy were all associated with a shorter PFS and OS (P<0.0001). Unfortunately, it was not possible to retain conventional cytogenetics in full analysis, as results were available from only one center, which did not have iFISH data available.

It was therefore elected to proceed with detailed analyses using data from 2642 patients with full iFISH information available. In this population, del(13) by iFISH was observed in 45% of the patients. The t(4;14) translocation was identified in 12.8% of the patients and del(17p) was found in 13.6% of the patients. Translocations t(11;14) and t(14;16) were observed in 20.5% and 2.9% of the patients, respectively.

Table 2 summarizes cross-correlations between iFISH findings, the use of high-dose therapy (HDTx) (or not), age and the ISS stage. It can be seen that iFISH findings are not highly cross-correlated with ISS stage; however, the use of HDTx was more common with lower ISS stage, and elderly patients (≥65 years) were more frequently stage II or III (P-values <0.001).

Table 2
Cross-correlations between iFISH, high-dose therapy use, age and ISS stage

Regarding prognosis, the 4-year PFS and OS estimates were, respectively, 26% and 52% for patients with del(13) versus 39% and 66% in patients lacking the abnormality (P<0.0001 for PFS and OS). IFISH t(4;14) and iFISH del (17p) were assessed separately for ISS stages I, II and III, and the results are displayed in Figures 1a–c (t(4:14)) and d–f (del(17p)). It can be seen that for both PFS and OS there are significant differences related to the presence of t(4:14) or del (17p), which reduce PFS and OS for each ISS stage: most strikingly for stage I, but also significantly for stages II and III. Conversely, for stage I patients negative for either t(4;14) or del (17p), the outcomes are excellent with 4-year OS estimates of 79% and 77%, respectively.

Figure 1
OS and PFS by ISS stages I, II and III with (+) and without (–) (ac) t(4;14) and (d–f) del(17p). OS and PFS for (a) ISS stage I with (+) or without (–) t(4;14); (b) ISS stage II with (+) or without (–) t(4;14); ...

To further illustrate the impact of combining ISS stage and iFISH, we derived a model via recursive partitioning illustrated in Figure 2. For this model, ISS-iFISH group I was defined by patients with ISS stage I or II with neither t(4;14) nor del(17p); ISS-iFISH group II was defined by either ISS stage III, with neither t(4;14) nor del(17p), or ISS stage I, with either t(4;14) or del(17p). Finally, ISS-iFISH group III was defined by ISS stage II or III with either t(4;14) or del(17p). In all, 51% of the patients were classified in group I, and 29% and 20% were in groups II and III, respectively. The 4-year PFS estimates were 39%, 20% and 11% for ISS-iFISH groups I, II and III, respectively (Figure 3a). Figure 3b shows the 4-year OS estimates, which were 71%, 45% and 33% for ISS-FISH groups I, II and III, respectively.

Figure 2
Recursive partitioning tree.
Figure 3
PFS and OS estimates (4-year) for ISS-iFISH categories group I, group II and group III. (a) PFS and (b) OS for the three groups derived from recursive partitioning.

The ISS-iFISH groups were assessed by country as detailed in Table 1. For the Czech Republic, Germany (Berlin), Korea, the United Kingdom and the United States, the individual numbers were too small to assess the statistical validity of the ISS-iFISH groups. However, with the higher patient numbers, the ISS-iFISH groupings were individually, statistically validated for France (P<0.001), Italy (Bologna) (P<0.001), Germany (Heidelberg) (P = 0.008) and Spain (P<0.001).

The ISS-iFISH model was further assessed by stratification by age (<65 years; ≥65 years) and with or without the use of HDTx (Figure 4). There is a clear impact of age. As can be seen in Figure 4c, the very best OS is for patients under the age of 65 years in group I (ISS stage I or II with neither t(4:14) nor del(17p)) with a 4-year OS estimate of 75%. Conversely, the poorest outcome is for patients ≥65 years in group III (ISS stage III with either t(4;14) or del (17p)) with a 4-year OS estimate of 24% (ISS-iFISH).

Figure 4
OS by ISS stage, iFISH, age (<65 years; ≥65 years) and use or not of HDTx. OS for the three groups derived from recursive partitioning (a) for age o65 years, (b) for age X65 years, (c) for both age <65 years and ≥65 years, ...

Similarly, the use of HDTx (or not) has an impact (Figures 4d and e). The very best outcome is for patients in group I who received HDTx with a 4-year OS estimate of 77%. Conversely, the poorest outcome is for patients in ISS-iFISH group III without HDTx with a 4-year OS estimate of 18%. It can also be noted that for both groups II and III (the poorest risk groups), there is a beneficial impact with the use of HDTx. For group II, the OS is 54% with HDTx and 25% without (Figure 4e). For group III, the OS is 41% with HDTx and 18% without HDTx (Figure 4e). Both age and HDTx were assessed along with ISS-iFISH in multivariate models. Both did contribute to P-values <0.001 and maximum R2 values up to 24.6% for combined ISS-iFISH models.


The main objective of this retrospective international study was to evaluate the role of genetic changes when combined with ISS in risk assessment. The hypothesis was that ISS was especially reflecting the tumor mass and patient condition, but did not take into account intrinsic plasma cell characteristics, which may have a major role in resistance to therapy and disease evolution. The two main intrinsic prognostic parameters are plasma cell proliferation and genetic changes. Plasma cell proliferation can be evaluated not only by labeling index but also by conventional karyotyping. Actually, it is now demonstrated that chromosomal changes are present in virtually 100% of the patients.19 The fact that karyotypes are abnormal in only 20–30% of the patients is related to the low proliferative index observed in most of the patients. Patients with a low proliferation display a normal karyotype that corresponds to normal bone marrow myeloid cells. Thus, abnormal cytogenetics can be analyzed as a high proliferative index. As labeling index was not available in this series, we focused our analyses upon cytogenetics. As expected, patients with an abnormal karyotype (whatever the chromosomal abnormalities) presented a shorter PFS and OS, independently of ISS stages. Unfortunately, we were not able to fully address the issue of karyotype information in a multivariable model, as patients with karyotype information did not have iFISH data.

We then looked at iFISH data. The major advantage of iFISH is that it allows the identification of genetic changes independently of the proliferative index. However, it has to be stressed here that iFISH in myeloma requires plasma cell identification or sorting because of the frequent low percentage of plasma cells in diagnostic samples. Many reports in the past decade have shown that some recurrent abnormalities can have a prognostic impact. After the demonstration of the prognostic impact of del(13) (actually monosomy 13),8,9 several reports showed the major influence of t(4;14) and del(17p) on both PFS and OS.1113 In this study, analyses were focused on abnormalities analyzed in the majority of the patients, that is, del(13), t(4;14), del(17p) and t(11;14). As reported previously, del(13), t(4;14) and del(17p) were associated with a shorter PFS and OS. In contrast, t(11;14) was prognostically neutral. Deletion 13 was present in 45% of the patients. The 4-year PFS and OS estimates were, respectively, 26% and 52% for patients with del(13) versus 39% and 66% in patients lacking the abnormality (P<0.0001 for PFS and OS). Translocation t(4;14) was observed in 12.8% of the patients, in agreement with previous reports. The 4-year PFS and OS estimates were, respectively, 11% and 35% for patients with t(4;14) versus 32% and 60% for patients lacking t(4;14) (P<0.0001 for PFS and OS). This strong prognostic impact is in agreement with the data reported previously. Finally, del(17p) was observed in 13.6% of the patients analyzed for this abnormality. The 4-year PFS and OS estimates were, respectively, 18% and 46% for patients with del(17p) versus 36% and 65% for patients with no deletion (P<0.0001 for PFS and OS). Because del(13) has been previously related to t(4;14) and del(17p), and because its prognostic value has been shown to be mainly related to these latter abnormalities,20 we looked at the outcome of patients with del(13), but lacking both t(4;14) and del(17p). These del(13) patients displayed a poorer prognosis than patients lacking del(13), but with a lower impact (4-year PFS estimates of 28% versus 36%, and 4-year OS estimates of 59% and 65%, respectively). Thus, the final analyses incorporated ISS stages and t(4;14) and del(17p) only as the dominant genetic features.

The key issue is whether consideration of iFISH findings significantly improves the basic ISS model. To address this question, we examined the median PFS and OS of patients presenting with either t(4;14) or del(17p) in each ISS stage. We found that consideration of both t(4;14)-positive and del(17p)-positive patients by ISS stage significantly affected outcomes. The 4-year OS estimates are reduced from 79 to 24% (t(4;14)) or 31% (del(17p)) for ISS stage I versus stage 3 (P<0.0001; Figure 1). Thus, risk assessment is definitely improved by considering ISS plus iFISH results versus ISS staging alone.

As the prognostic impacts of t(4;14) and del(17p) were almost identical, and as both abnormalities were rarely associated, we combined the two abnormalities to look at their prognostic impact in each ISS stage. A strong effect of chromosomal changes was observed in each stage, the 4-year PFS going from 53%, 36% and 24% for patients with no poor-risk genetic changes and ISS stages 1, 2 and 3, respectively, to 22%, 14% and 9% for patients with abnormalities (P<0.0001 in each case). Similar results were obtained for OS: the 4-year OS estimates dropped down from 83%, 68% and 49% to 57%, 37% and 28%, respectively (P<0.0001 in each case). To produce an effective and simple prognostic model, we tested different combinations. These analyses resulted in three different iFISH-ISS groups: group I was defined by ISS stages I or II without chromosomal change, group II combined ISS stage III with no genetic abnormality and ISS stage I with abnormality, and finally group III was defined by ISS stages II or III with genetic abnormality. The 4-year PFS and OS estimates were very different: 44%, 23% and 12% for PFS and 76%, 52% and 33% for OS, respectively (P<0.0001 in each case). Regarding the distribution of patients in each iFISH-ISS group, most patients were in group I (51%), and equally distributed in groups II and III (29% and 20%, respectively). The analysis in subgroups of patients, such as according to age or treatment strategy, confirmed the accuracy of the model (Figure 4).

Two additional important points can be noted. First, both age and the use of HDTx (or not) should be considered in risk assessment. Second, outcomes in the highest risk groups (ISS-iFISH groups II and III) were better with the use of HDTx than without (Figures 4d and e).

Combining iFISH with other prognostic factors has been examined by several groups, including smaller numbers of patients. In the Intergroupe Francophone du Myélome experience,20 the combination of t(4;14), del(17p) (defined by the presence in at least 60% of the plasma cells) and a high β2-microglobulin level produced similar results. Recently, one of the German groups reported similar data.21 Finally, a recent update of the Mayo Clinic mSMART model showed that this stratification produced a good separation of the patients, based on outcome.22

Ongoing evaluation is required for patients treated with bortezomib and/or lenalidomide. Bortezomib seems to overcome (at least partly) the prognostic impact of t(4;14).2326 The use of both ISS and molecular changes by iFISH should be evaluated in upcoming trials using these novel agents. Finally, recent reports suggested that novel approaches based on microarray technology should be used to achieve a more powerful prediction.27,28 Even though single institution reports reveal helpful correlations, they need to be validated. Collective analyses, similar to those conducted in this article for iFISH, are now underway to assess the impact of gene expression profiling results. However, it should be taken into account that gene expression profiling approaches are probably not feasible in all the patients because of sample quality, and especially in a multicenter setting. In contrast, ISS and iFISH approaches are feasible routinely in most specialized laboratories, and thus their clinical applicability is more realistic.29



The authors declare no conflict of interest.


1. Durie BG, Salmon SE, Moon TE. Pretreatment tumor mass, cell kinetics, and prognosis in multiple myeloma. Blood. 1980;55:364–372. [PubMed]
2. Bataille R, Durie BG, Grenier J, Sany J. Prognostic factors and staging in multiple myeloma: a reappraisal. J Clin Oncol. 1986;4:80–87. [PubMed]
3. Bataille R, Durie BG, Grenier J. Serum beta2 microglobulin and survival duration in multiple myeloma: a simple reliable marker for staging. Br J Haematol. 1983;55:439–447. [PubMed]
4. Bataille R, Boccadoro M, Klein B, Durie B, Pileri A. C-reactive protein and beta-2 microglobulin produce a simple and powerful myeloma staging system. Blood. 1992;80:733–737. [PubMed]
5. Bataille R, Jourdan M, Zhang XG, Klein B. Serum levels of interleukin 6, a potent myeloma cell growth factor, as a reflect of disease severity in plasma cell dyscrasias. J Clin Invest. 1989;84:2008–2011. [PMC free article] [PubMed]
6. Greipp PR, Leong T, Bennett JM, Gaillard JP, Klein B, Stewart JA, et al. Plasmablastic morphology—an independent prognostic factor with clinical and laboratory correlates: Eastern Cooperative Oncology Group (ECOG) myeloma trial E9486 report by the ECOG Myeloma Laboratory Group. Blood. 1998;91:2501–2507. [PubMed]
7. Greipp PR, San Miguel J, Durie BG, Crowley JJ, Barlogie B, Bladé J, et al. International staging system for multiple myeloma. J Clin Oncol. 2005;23:3412–3420. [PubMed]
8. Zojer N, Königsberg R, Ackermann J, Fritz E, Dallinger S, Krömer E, et al. Deletion of 13q14 remains an independent adverse prognostic variable in multiple myeloma despite its frequent detection by interphase fluorescence in situ hybridization. Blood. 2000;95:1925–1930. [PubMed]
9. Facon T, Avet-Loiseau H, Guillerm G, Moreau P, Geneviève F, Zandecki M, et al. Chromosome 13 abnormalities identified by FISH analysis and serum beta2-microglobulin produce a powerful myeloma staging system for patients receiving high-dose therapy. Blood. 2001;97:1566–1571. [PubMed]
10. Moreau P, Facon T, Leleu X, Morineau N, Huyghe P, Harousseau JL, et al. Recurrent 14q32 translocations determine the prognosis of multiple myeloma, especially in patients receiving intensive chemotherapy. Blood. 2002;100:1579–1583. [PubMed]
11. Fonseca R, Blood E, Rue M, Harrington D, Oken MM, Kyle RA, et al. Clinical and biologic implications of recurrent genomic aberrations in myeloma. Blood. 2003;101:4569–4575. [PubMed]
12. Jaksic W, Trudel S, Chang H, Trieu Y, Qi X, Mikhael J, et al. Clinical outcomes in t(4;14) multiple myeloma: a chemotherapy-sensitive disease characterized by rapid relapse and alkylating agent resistance. J Clin Oncol. 2005;23:7069–7073. [PubMed]
13. Gertz MA, Lacy MQ, Dispenzieri A, Greipp PR, Litzow MR, Henderson KJ, et al. Clinical implications of t(11;14)(q13;q32), t(4;14)(p16.3;q32), and – 17p13 in myeloma patients treated with high-dose therapy. Blood. 2005;106:2837–2840. [PubMed]
14. Chng WJ, Kuehl WM, Bergsagel PL, Fonseca R. Translocation t(4;14) retains prognostic significance even in the setting of high-risk molecular signature. Leukemia. 2008;22:459–461. [PubMed]
15. Hanamura I, Stewart JP, Huang Y, Zhan F, Santra M, Sawyer JR, et al. Frequent gain of chromosome band 1q21 in plasma-cell dyscrasias detected by fluorescence in situ hybridization: incidence increases from MGUS to relapsed myeloma and is related to prognosis and disease progression following tandem stem-cell transplantation. Blood. 2006;108:1724–1732. [PubMed]
16. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53:457–458.
17. Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep. 1966;50:163–170. [PubMed]
18. LeBlanc M, Crowley J. Survival trees by goodness of split. JASA. 1993;88:457–467.
19. Avet-Loiseau H, Li C, Magrangeas F, Gouraud W, Charbonnel C, Harousseau JL, et al. Prognostic significance of copy-number alterations in multiple myeloma. J Clin Oncol. 2009;27:4585–4590. [PMC free article] [PubMed]
20. Avet-Loiseau H, Attal M, Moreau P, Charbonnel C, Garban F, Hulin C, et al. Genetic abnormalities and survival in multiple myeloma: the experience of the Intergroupe Francophone du Myélome. Blood. 2007;109:3489–3495. [PubMed]
21. Neben K, Jauch A, Bertsch U, Heiss C, Hielscher T, Seckinger A, et al. Combining information regarding chromosomal aberrations t(4;14) and del(17p13) with the International Staging System classification allows stratification of myeloma patients undergoing autologous stem cell transplantation. Haematologica. 2010;95:1150–1157. [PubMed]
22. Kapoor P, Fonseca R, Rajkumar SV, Sinha S, Gertz MA, Stewart AK, et al. Evidence for cytogenetic and fluorescence in situ hybridization risk stratification of newly diagnosed multiple myeloma in the era of novel therapie. Mayo Clin Proc. 2010;85:532–537. [PMC free article] [PubMed]
23. San Miguel JF, Schlag R, Khuageva NK, Dimopoulos MA, Shpilberg O, Kropff M, et al. Bortezomib plus melphalan and prednisone for initial treatment of multiple myeloma. N Engl J Med. 2008;359:906–917. [PubMed]
24. Barlogie B, Anaissie E, van Rhee F, Haessler J, Hollmig K, Pineda-Roman M, et al. Incorporating bortezomib into upfront treatment for multiple myeloma: early results of total therapy 3. Br J Haematol. 2007;138:176–185. [PubMed]
25. Avet-Loiseau H, Leleu X, Roussel M, Moreau P, Guerin-Charbonnel C, Caillot D, et al. Bortezomib plus dexamethasone induction improves outcome of patients with t(4;14) myeloma but not outcome of patients with del(17p). J Clin Oncol. 2010;28:4630–4634. [PubMed]
26. Cavo M, Tacchetti P, Patriarca F, Petrucci MT, Pantani L, Galli M, et al. Bortezomib, thalidomide and dexamethasone compared with thalidomide and dexamethasone as induction before, and consolidation after, double autologous stem cell transplantation in newly diagnosed multiple myeloma: results from a randomized phase 3 study. Lancet. 2010;376:2075–2085. [PubMed]
27. Shaughnessy JD, Jr, Zhan F, Burington BE, Huang Y, Colla S, Hanamura I, et al. A validated gene expression model of high-risk multiple myeloma is defined by deregulated expression of genes mapping to chromosome 1. Blood. 2007;109:2276–2284. [PubMed]
28. Decaux O, Lodé L, Magrangeas F, Charbonnel C, Gouraud W, Jézéquel P, et al. Prediction of survival in multiple myeloma based on gene expression profiles reveals cell cycle and chromosomal instability signatures in high-risk patients and hyperdiploid signatures in low-risk patients: a study of the Intergroupe Franco-phone du Myélome. J Clin Oncol. 2008;26:4798–4805. [PubMed]
29. Avet-Loiseau H, Attal M, Campion L, Caillot D, Hulin C, Marit G, et al. Long-term analysis of the IFM 99 trials for myeloma: cytogenetic abnormalities [t(4;14), del(17p), 1q gains] play a major role in defining long-term survival. J Clin Oncol. 2012;30:1949–1952. [PubMed]