ORIGINAL ARTICLE Prognostic Value of Chromosome 1Q21 Gain

Prognostic value of chromosome 1q21 gain by ﬂuorescent in situ hybridization and increase CKS1B expression in myeloma

R Fonseca1, SA Van Wier1, WJ Chng1, R Ketterling2, MQ Lacy3, A Dispenzieri3, PL Bergsagel1, SV Rajkumar3, PR Greipp3, MR Litzow3, T Price-Troska3, KJ Henderson3, GJ Ahmann1 and MA Gertz3

1Division of Hematology and Oncology, Mayo Clinic Comprehensive Cancer Center, Scottsdale, AZ, USA; 2Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA and 3Division of Hematology, Mayo Clinic, Rochester, MN, USA

A specific role for increased level of expression of CKS1B,asa including deletions and mutations of p53, and mutations of ras consequence of chromosome 1q21 copy number gain, has have been associated with disease progression, but only for a been postulated as both pathogenic, as well as a powerful small fraction of cases, and with yet to be proven pathogenic clinical prognostic factor in multiple myeloma (MM). The 1 purpose of this study is to determine the clinical associations consequences. and prognostic impact of copy number gain at chromosome Recently the group from the University of Arkansas (UAMS) 1q21 (with a bacteria artificial chromosome clone containing identified strong prognostic associations with an increased level CKS1B) and CKS1B gene level of expression in MM. We studied of gene expression of a cell cycle associated gene, CKS1B, the chromosome region 1q21 for copy number change in a located on chromosome 1q21.9,10 CKS1B favors cell cycle cohort of myeloma patients treated by high-dose therapy with progression by promoting degradation of p27 with release of the stem-cell rescue (HDT) (n ¼ 159). A separate cohort of patients, 11,12 treated by HDT was studied for CKS1B messenger RNA cyclin dependent kinases and entry into mitosis. This expression by gene expression profiling (n ¼ 67). 1q21 gain increased level of expression is associated with copy number was then correlated with clinical parameters and survival. Gain gain of chromosome 1 at band q21 in a subset of patients. of 1q21 copy number was detected in about a third of MM and Subsequently, the same group showed that 1q21 copy gain by was associated with more proliferative disease and poor-risk Fluorescent in situ hybridization (FISH) using a bacteria artificial cytogenetic categories such as t(4;14), and chromosome 13 chromosome (BAC) probe for CKS1B was also an important deletion. Both 1q21 gain and increase gene expression level 13 were significantly associated with reduced survival. However, prognostic factor in MM patients entered into total therapy II. neither is an independent prognostic marker in MM on multi- Previously comparative genomic hybridization studies have variate Cox proportional hazard analysis. shown that gains of chromosome 1q consistently involving the Leukemia (2006) 20, 2034–2040. doi:10.1038/sj.leu.2404403; 1q21 region is one of the most common genetic abnormalities in published online 5 October 2006 MM.14–16 Gain of 1q21 can occur as isochromosomes, Keywords: multiple myeloma; stem cell transplantation; gene 17,18 amplification; CKS1B; chromosome 1 duplications or jumping translocations in MM. Many of these chromosome changes are structural and do not involve copy number changes. However, the association of 1q21 copy gain and expression with proliferation and recurrent genetic abnormalities, and their relative impact on survival have not been clearly established. In Introduction this study, we assessed gain of chromosome 1q21 by FISH and increase CKS1B gene expression by gene expression profiling Molecular cytogenetic studies have revealed that, to a great (GEP), their relationship, prognostic impact and clinical and extent, the clinical heterogeneity of multiple myeloma (MM) is biologic correlations. dictated by the underlying genetic and cytogenetic aberrations present in the clonal plasma cells (PCs). This greater under- standing has defined several molecular subtypes of myeloma Patients and methods characterized primarily by the presence of specific chromosome translocations and associated chromosome ploidy changes.1–3 Patient cohorts This characterization is also associated, and ratified, by the Group 1: Mayo Clinic patients treated with HDT studied existence of specific transcription patterns that emerge from by FISH. We studied a cohort of 159 patients treated at the 4,5 these cytogenetically defined categories. It is also clear that Mayo Clinic that had been treated with melphalan-based high- these major cytogenetic categories are already present in the dose therapy (HDT). Patients were treated with melphalan at 6–8 early stages of the disease (i.e., MGUS), and that additional 200 mg/m2 (n ¼ 63) or melphalan at 140 mg/m2 plus total body genetic insults must participate in the progression to the more irradiation (n ¼ 83), and other conditioning regimens in 13. Of aggressive forms of the disease. Several candidate aberrations these patients 89 (56%) had been transplanted at the time of relapse. Of the 159 patients 114 have died. Median and Correspondence: Dr R Fonseca, Division of Hematology and minimum follow-up of the 45 survivors is 55 and 43 months, Oncology, Mayo Clinic Comprehensive Cancer Center, Johnson respectively. Research Building 356, Mayo Clinic, 13400 E Shea Boulevard, In all cases studied a research bone marrow (BM) sample had Scottsdale, AZ 85259, USA. E-mail: [email protected] been obtained before HDT. All patients gave written informed Received 11 July 2006; accepted 17 August 2006; published online consent for the collection and long-term storage of additional 5 October 2006 cells at the time of the BM examination including separate Chromosome 1q21 copy change and prognosis in myeloma R Fonseca et al 2035 authorization for genetic studies and separate consent for review a matrix to define the actual proportion of cells exhibiting a of their charts in accord with both federal regulations and possible amplification or copy gain pattern, and determine a Health Insurance Portability and Accountability Act (HIPAA) final ratio (Supplementary Figure 1). A ratio greater than 1 guidelines. The consent was approved by the Institutional represents copy gain of 1q21. This strategy will allow the Review Board of the Mayo Clinic College of Medicine. Cytospin assessment of the effect of 1q21 gain independent of ploidy slides were made in all cases and the remaining sample was changes. The second strategy is similar to that used by purified using CD138 þ magnetic beads when possible and Hanamura et al.,13 where the absolute copy of 1q21 is counted. stored in liquid nitrogen for future use. All patients in whom We considered patients who have three or more signals in more available samples were obtained were used in this analysis. than 20% of cells counted as having gain of chromosome 1q21. One hundred cIg-positive cells were counted. Group 2: Mayo Clinic patients studied by GEP. Gene expression analysis was performed on CD138 þ selected PCs from pretreatment BM samples of 67 newly diagnosed MM GEP patients using the Affymetrix U133A chip (Affymetrix, Santa Gene expression analysis was performed on CD138 þ cells Clara, CA, USA). This group of patients was subsequently treated using the Affymetrix U133A chip (Affymetrix, Santa Clara, CA, with various forms of therapy including HDT and stem cell USA). RNA isolation, purification and microarray hybridization rescue in 63% (none of these patients are in group 1), has been reported previously.23 Gene expression intensity melphalan-based oral chemotherapy in 21%, dexamethasone- values were log transformed, normalized to the median and based chemotherapy in 13% and novel agents in 3%. analyzed using GeneSpring 7 (Agilent Technologies, Palo Alto, CA, USA). Group 3: UAMS patients treated with HDT (Total Therapy II). These are newly diagnosed patients entered into Total Therapy II studies. These patients have FISH analysis for Statistics 1q21 copy number using the same probe as that used in our To describe the prevalence of abnormalities we used simple þ study, as well as GEP performed on CD138 selected PCs from descriptive statistics. The Wilcoxon rank sum test was used to pretreatment BMs using the Affymetrix U133plus chip. The FISH compare the difference in PC labeling index (PCLI) between 13,19 and GEP of these patients have been published. For our patients with and without 1q21 copy change. Correlation analysis we utilized a subset of patients (n ¼ 248) with both FISH between the level of CKS1B gene expression and the PCLI was and GEP results available (These data are available from the NIH assessed using the Pearson’s correlation coefficient. The Gene Expression Omnibus, http://www.ncbi.nlm.nih.gov/geo/, relationship between 1q21 copy change and CKS1B gene under accession number GSE2658). These patients were expression and cytogenetic categories was assessed by the w2 assigned Translocations and Cyclin (TC) class and a gene test. Two-tailed p-values were reported with the significance expression-based proliferation index (PI) based on published level set at 0.05. Survival (overall and progression free) 4 methods. probabilities were determined using the methods of Kaplan– Meier. Cox proportional hazards models were used to assess the association of several prognostic factors with survival. Methods FISH. The cytoplasmic immunoglobulin light chain (cIg-FISH) technique20 that combined the interphase FISH strategy with fluorescent detection of cIg was used. For the detection of the Results common chromosome abnormalities (t(11;14)(q13;q32), t(4;14)(p16;q32), þ 14q32, chromosome 13 abnormalities Clinical features of patients studied for DNA copy gain (D13) and À17p13), we used the same probes and conditions A total of 159 patients, in whom we had stored research samples published previously by us.21 and had been treated with HDT and stem cell transplant To score for 1q21 copy gain by FISH we used a BAC RPCI between January 1990 and January 2001, were studied. The 11.C 307C12 whose insert contains CKS1B. This is the same characteristics of these patients are given in Supplementary probe used to show the prognostic importance of chromosome Table 1. 1q21 gain in a recent study.13 Briefly, the BAC was grown in selective media, DNA was extracted and directly labeled with a SpectrumGreen fluorophore by standard nick translation. The Prevalence of 1q21/CKS1B gain among patients in following PCR primers were designed to confirm the BAC DNA group 1 contained CKS1B, 50-GGGCGTGGTTAGGGTACTGA-30 and Using the ratio method, 46 (30%) of 159 patients studied had 50-AGCGCTACTCTCCTGTTAGGG-30. The labeled probe was 1q21 gain (ratio41). The ratio between CKS1B and control then tested on normal metaphases to reaffirm localization of the probe was greater than 2.0 in only two cases and between 1.5 probe to chromosome 1 at band q21, and exclude cross and 2.0 in 30 cases. When absolute copy number was used, 59 hybridization. A commercially available chromosome enumera- (37%) patients had 1q21 gain. The copy number ranges from 3 tion probe (CEP) for the alpha satellite DNA of chromosome 1 to 8 with the median being 5. The level of gene copy number labeled in SpectrumOrange (Vysis, Downer Grove, IL, USA) was change observed is much lower than that observed with true used as ploidy control. amplifications, such as c-myc in PC derived from patients with To ascertain for the presence of gene copy gain we used two PC leukemia or the human MM cell lines (Figure 1) or for Her-2/ different strategies. In the first, we use a methodology designed neu where the ratio and the absolute number of copy of the for the study of HER-2/neu.22 In brief, 100 cells per sample were amplified gene are typically greater than 2.0 and 10, respec- initially scored to determine concordance of signals derived tively. Therefore, we have decided to use copy number gain from the CEP and the 1q21 probes. If the ratio was determined to instead of amplification to describe the increase in 1q21 in this be greater than 1, the results of 30 clonal cells were entered into study.

Leukemia Chromosome 1q21 copy change and prognosis in myeloma R Fonseca et al 2036 Table 1 Correlation between 1q21 gain and cytogenetic category (group 1)

Patients with abnormalities Patients with 1q21 gaina Patients without 1q21 gain n/N (%) n/N (%) n/N (%) P-value

t(11;14) 26/154 (17) 6/45 (13) 20/109 (18) 0.45 17p13.1À(p53) 17/154 (11) 2/45 (4) 15/109 (14) 0.09 t(4;14) 22/124 (18) 13/38 (34) 9/86 (11) 0.001 Del 13 87/158 (55) 33/46 (72) 54/112 (48) 0.007 t(4;14)+D13 19/124 (15) 11/38 (29) 8/86 (9) 0.005 aRatio of 1.0 or greater.

Figure 1 FISH patterns consistent with copy gain of the 1q21 genomic region that contains CKS1B (a, b and d). (a) Depicts a PC from a patient with 1q21 gain with an overall (30 clonal cells) ratio of CKS1B probe (green) to CEP1 (red) signal of 1.83. (b) Depicts a metaphase spread from a patient with 1q21 gain that shows duplication of the site within a copy of chromosome. In comparison (c) depicts a pattern of true gene ampliﬁcation for c-myc (red) as seen in a patient with PC leukemia. The probe CL (green) refers to a BAC clone that localizes to the constant light chain region 22q11.2. (d) Shows how some of the abnormal FISH patterns could arise from ‘jumping translocations’ of chromosome 1 (described in detail in Sawyer et al.38) as shown by dissociation between the CEP1 signal and the CKS1B probe.

Association of 1q21 gain and increased CKS1B Chromosome 1q21 gain and elevated CKS1B gene expression with other genetic abnormalities (groups 1 expression was associated with more proliferative MM and 3) In patients studied by FISH (group 1), 1q21 gain (ratio41) was Of the 46 cases with 1q21 to CEP ratio greater than 1, most had associated with higher PCLI, a validated marker of MM determinations of D13, deletions of 17p13, t(11;14) and t(4;14) proliferation24 (P ¼ 0.02). In patients studied by GEP (Group (Table 1). Patients with 1q21 copy number gain had a higher 2), level of CKS1B gene expression was correlated positively prevalence of D13 (33/46, 72%, P ¼ 0.007), and t(4;14) (13/45, with the PCLI (Pearson coefﬁcient, r ¼ 0.67, Po0.001). 29%, w2 P ¼ 0.001). In contrast these patients had a prevalence of t(11;14) as expected (6/45, 13%) and surprisingly a low or normal prevalence of p53 deletions (2/45, 4%). However, Effects of 1q21 gain on overall survival (group 1) CKS1B gene expression level was similar between patients with The median survival of patients (n ¼ 159) with 1q21 gain spike expression of FGFR3/MMSET, CCND1, and MAF (normal- (ratio41.0, n ¼ 46) was 29.9 months (95% conﬁdence interval ized expression 1.8671.62 versus 1.6971.38 versus 2.587 (CI) 16.4–40 months) versus 38 months (95% CI 28.7–50.4 1.88, analysis of variance P ¼ 0.28). months) for those with a normal ratio (n ¼ 113) (log-rank,

Leukemia Chromosome 1q21 copy change and prognosis in myeloma R Fonseca et al 2037 P ¼ 0.1224). When a more stringent cutoff ratio of 1.5 or greater expression between those with and without 1q21 copy gain. was used, patients with elevated ratio (n ¼ 31) had a signiﬁcantly Using the median expression level (normalized expression of 1) shorter survival (median survival of 21.9 months (95% CI 13.6– as a cutoff for high expression level, 76 patients had gain of 37 months) versus 38 months (95% CI 29.8–50.4 months, log- 1q21 and increase CKS1B expression, 44 patients had normal rank P ¼ 0.04) (Figure 2a). However, when 1q21 gain was 1q21 but had CKS1B increased expression, 38 patients had gain included in the multivariate model, it did not remain an of 1q21 but normal CKS1B gene expression and 126 patients independent prognostic factor as it was superseded by the had normal 1q21 and no increase in CKS1B expression. negative prognostic effect of the t(4;14) and a higher PCLI Interestingly, when the survival of patients with these combina- (Table 2). tions of 1q21 copy and CKS1B expression were compared, When an absolute increase in 1q21 copy regardless of patients with increase 1q21 had the worst prognosis regardless chromosome 1 copy was used as the criterion, 1q21 gain is of expression levels (Figure 2c). In a multivariate Cox propor- again signiﬁcantly associated with reduced overall survival but tional hazard analysis including the following variables: gain of was not an independent prognostic factor when included in a 1q21, increase CKS1B expression (above median), 4p16 TC Cox proportional hazard analysis including PCLI and t(4;14) class (corresponding to t(4;14)) and a PI based on median (data not shown). expression of 12 genes (this index correlated closely with PCLI;

Effects of increased CKS1B gene expression on survival (group 2) a 1.0 Survival analysis was performed on the 67 patients with new 0.9 diagnosis MM in whom gene expression data was available. Although this group of patients is small we hypothesized that a 0.8 powerful prognostic factor would still be able to discern 0.7 prognostic subgroups. When the groups were dichotomized by 0.6 the median value of expression for CKS1B there was no 0.5 ¼ difference in overall survival (P 0.13). When we applied a 0.4 more stringent criterion (normalized CKS1B expression level 42.0), those having the high level of expression (n ¼ 12) had a 0.3 No 1q21 gain significant shorter survival than those with lower levels (n ¼ 55): Surviving Percent 0.2 12.8 months versus not yet reached (log-rank, P ¼ 0.007) 0.1 1q21 gain (Figure 2b). Interestingly, when we performed FISH in 23 of 0.0 these patients who had materials for FISH studies, the 12 01020 30 40 50 60 70 80 90 100110120130140 patients with 1q21 copy gain were enriched for cases with OS (mths) normalized CKS1B gene expression of greater than 2 (10 of 15 (67%) versus 2 of 8 (25%), P ¼ 0.05). In a multivariate analysis, b 100 limited by the low number of cases, and including only the following variables: PCLI greater than 1, 4p16 TC class and 90 CKS1B expression greater than 2, only PCLI emerged as an 80 independent prognostic factor. 70 CKS1B > 2 60 CKS1B < 2 Interactions between gain of 1q21 and increase CKS1B 50 expression and their effect on survival (group 3) 40 CKS1B expression increased with 1q21 copy number (Figure 3). 30 Percent Survival Percent However, there was significant overlap in CKS1B gene 20 10 Figure 2 Impact of 1q21 gain and increase CKS1B expression on 0 survival. (a) Kaplan–Meier analysis for overall survival according to the 0 5 101520253035404550 level of CKS1B copy number change (group 1 patients). Patients with OS (mths) evidence of gain of 1q21 at a cut-off ratio of greater than 1.5 (n ¼ 31) have shorter survival than others (n ¼ 128) (log-rank P ¼ 0.042). c 100 (b) Kaplan–Meier analysis for overall survival according to the level of CKS1B gene expression (67 group two patients). Patients with 90 normalized CKS1B expression level of greater than 2 (n ¼ 12) had 80 significantly shorter survival than those with lower expression (median overall survival 13.1 months versus not yet reached, log-rank 70 P ¼ 0.007). (c) Kaplan–Meier analysis for overall survival according 60 to the both 1q21 copy gain and the level of CKS1B gene expression 50 (248 group three patients). The survival of the four groups of patients was significantly different (log-rank P ¼ 0.03). On pair-wise compar- 40 Gain of CKS1B with increase expression isons, patients with copy number gain had similar survival irrespective 30 of the level of expression. These patients had significantly worse Survival Percent Normal CKS1B copy with increase expression survival than those with elevated gene expression alone. In turn, each 20 Gain of CKS1B with normal expression of these prior groups had significant worse survival compared to 10 Normal CKS1B copy and expression patients with normal copy and expression. The limitation of this analysis is the relatively short follow-up such that of the four groups, 0 only the median survival for those with increase copy and gene 01020304050607080 expression has been reached (49.4 months). OS (mths)

Leukemia Chromosome 1q21 copy change and prognosis in myeloma R Fonseca et al 2038 Table 2 Cox (multivariate) analysis of factors impacting time to progression and overall survival in group 1 patients

Time to progression Overall survival

Parameter Risk ratio P-value Risk ratio P-value

Status at stem cell transplant 2.3 0.002 2.3 0.001 t(4;14) 2.1 0.0006 1.4 0.02 17p13.1À(p53) 1.4 0.17 1.1 0.75 D13 1.2 0.14 1.1 0.53 Labeling index 1.2 0.01 1.2 0.003 Beta 2 microglobulin 1.0 0.44 1.1 0.20 Bone marrow plasmacytosis 1.01 0.06 1.0 0.68 1q21/CEP1 ratio 1.5 or greater 1.1 0.54 1.3 0.76 Abbreviation: CEP, chromosome enumeration probe.

8 Table 3 Cox (multivariate) analysis of factors impacting overall survival in group 3 patients

4 Variable Risk ratio P-value

Proliferation index 1.41 0.018 2 4p16 TC class 2.10 0.020 1q21 gain 1.73 0.071 Increase CKS1B expression 1.03 0.935 1 Abbreviation: TC, translocations and cyclin. Expression 0.5 Both 1q21 copy gain and increase CKS1B expression are CKS1B 0.25 associated with shortened survival in all three datasets analyzed. However, the net effect of these markers on survival seem to be modest and does not retain independence when entered into 0.125 multivariate models that incorporates the high-risk cytogenetic categories of MM, particularly the t(4;14)(p16;q32), and proliferation markers. In the study by Hanamura et al.,13 gain 0.0625 2 Copies 3 Copies > 4 Copies of 1q21 emerged as an independent prognostic factor. However, in that study, proliferation markers and t(4;14) were not included 1q21/CKS1B Copy Number in the multivariate analysis. Our analysis would suggest that possibly, the association between 1q21 gain and prognosis Figure 3 Comparison of CKS1B expression level across different relates to its associations with other known high-risk factors such 1q21 copy number change. CKS1B expression increases from two copy of 1q21 (n ¼ 134, normalized expression 0.9570.69) to three as the t(4;14) and PCLI. copies (n ¼ 70, normalized expression 1.5771.34) to four or more One can think of four groups of patients according to 1q21 copies (n ¼ 44, normalized expression 2.671.91) (Po0.00001). copy and CKS1B gene level status; no copy number change and no increased level of expression, copy number gain but no increased transcription, no copy number gain but increased unpublished data), only 4p16 TC class and the PI emerged as gene expression and lastly increased copy number and independent prognostic factors (Table 3). increased level of expression. Indeed, one-third of patients from the UAMS dataset have discordant CKS1B copy number and expression. When the relative prognostic strength of 1q21 copy Discussion gain and increase CKS1B expression is analyzed in a multivariate model, 1q21 copy gain is the more significant prognostic In this study, we show that gain of 1q21 is seen in about 30–37% factor. of patients depending on the method of detection. Gain of 1q21 There is evidence that 1q21 amplification is associated with is also associated with t(4;14) and chromosome 13 deletion but disease progression in MM.13,25–27 Chromosome 1q21 abnorm- not t(11;14). This is consistent with the findings of a recently alities have also been associated with genomic instability in published study which showed that gain of 1q21 is associated MM. Chromosome 1q21 gain may therefore be a marker of with t(4;14) and MAF translocations, as well as chromosome 13 more clonally advance and genomically unstable tumor. Indeed deletion but not with t(11;14).13 Interestingly, we do not see the patients with 1q21 gain without elevated CKS1B expression same association between CKS1B gene expression and these have equally poor survival as those with elevated CKS1B primary IgH translocations. As expected, increase CKS1B expression. In addition, the subsequent overexpression of expression is associated with more proliferative disease, as it candidate oncogenes or genes regulating cell cycle progression is one of the genes associated with cell cycle activity and used and proliferation, including CKS1B, located on the genome by us to calculate the PI. We show for the first time that 1q21 region of interest may explain the greater prognostic significance copy gain is also associated with more proliferative disease. associated with 1q21 gain. Consistent with this, a large number

Leukemia Chromosome 1q21 copy change and prognosis in myeloma R Fonseca et al 2039 of candidate genes in the 1q21 region (e.g., BCL9, MCL1, are early events in monoclonal gammopathy of undetermined CKS1B and MUC1) have been pathogenically implicated in MM significance and do not evolve during transition to multiple and other malignancies.9,28–37 More recently, integrated analy- myeloma. Leukemia 2004; 18: 1879–1882. sis of high-resolution array comparative genomic hybridization 9 Shaughnessy J. Amplification and overexpression of CKS1B at chromosome band 1q21 is associated with reduced levels of and GEP identified a high priority minimal region of DNA copy p27Kip1 and an aggressive clinical course in multiple myeloma. number change on 1q21. This 10 Mb region contain several Hematology 2005; 10 (Suppl 1): 117–126. genes including the aforementioned CSK1B, BCL9 and MCL1.37 10 Shaughnessy Jr JD, Barlogie B. Using genomics to identify high-risk FISH detection of 1q21 gain may therefore provide more myeloma after autologous stem cell transplantation. Biol Blood significant prognostic information. This strategy also indirectly Marrow Transplant 2006; 12: 77–80. tests the ability of surrounding genes to be associated with a 11 Ganoth D, Bornstein G, Ko TK, Larsen B, Tyers M, Pagano M et al. The cell-cycle regulatory protein Cks1 is required for negative outcome in MM. SCF(Skp2)-mediated ubiquitinylation of p27. Nat Cell Biol 2001; In those cases with only increased CKS1B gene expression 3: 321–324. and no copy gain, the elevated gene expression is likely owing 12 Spruck C, Strohmaier H, Watson M, Smith AP, Ryan A, Krek TW to downstream effects of other transcription factors that result in et al. A CDK-independent function of mammalian Cks1: targeting increased expression of a gene integral to the control of cell of SCF(Skp2) to the CDK inhibitor p27Kip1. Mol Cell 2001; 7: cycle, and is thus a proliferation marker. 639–650. 13 Hanamura I, Stewart JP, Huang Y, Zhan F, Santra M, Sawyer JR In summary, our study shows that neither 1q21 gain as et al. Frequent gain of chromosome band 1q21 in plasma cell detected by FISH nor increased CKS1B expression are indepen- dyscrasias detected by fluorescence in situ hybridization: Inci- dent prognostic factors in MM and their impact on survival are dence increases from MGUS to relapsed myeloma and is related to probably mediated by their close associations with high-risk prognosis and disease progression following tandem stem cell cytogenetic such as t(4;14), and more proliferative disease. transplantation. Blood, (E-pub ahead of print 16 May 2006) 2006, Chromosome 1q21 gain has a more significant impact on (DOI 10.1182/blood-2006-03-009910). survival than increase CKS1B expression and probably reflects 14 Avet-Loiseau H, Andree-Ashley LE, Moore 2nd D, Mellerin MP, Feusner J, Bataille R et al. Molecular cytogenetic abnormalities in an underlying more clonally advance and genetically unstable multiple myeloma and plasma cell leukemia measured using tumor. comparative genomic hybridization. Genes Chromosomes Cancer 1997; 19: 124–133. 15 Gutierrez NC, Hernandez JM, Garcia JL, Canizo MC, Gonzalez M, Acknowledgements Hernandez J et al. Differences in genetic changes between multiple myeloma and plasma cell leukemia demonstrated by RF is a Clinical Investigator of the Damon Runyon Cancer comparative genomic hybridization. Leukemia 2001; 15: 840– Research Fund. This work was supported by the International 845. Waldenstro¨m Macroglobulinemia Foundation, and Grants R01 16 Cigudosa JC, Rao PH, Calasanz MJ, Odero MD, Michaeli J, Jhanwar SC et al. Characterization of nonrandom chromosomal CA83724-01, SPORE P50 CA100707-01, the Donaldson Chari- gains and losses in multiple myeloma by comparative genomic table Trust fund and P01 CA62242 from the National Cancer hybridization. Blood 1998; 91: 3007–3010. Institute, and the Fund to Cure Myeloma. WJC is funded by an 17 Sawyer JR, Lukacs JL, Munshi N, Desikan KR, Singhal S, Mehta J International Fellowship from the Agency for Science, Technology et al. Identification of new nonrandom translocations in multiple and Research, Singapore (A*STAR). myeloma with multicolor spectral karyotyping. Blood 1998; 92: 4269–4278. 18 Sawyer JR, Tricot G, Lukacs JL, Binz RL, Tian E, Barlogie B et al. References Genomic instability in multiple myeloma: evidence for jumping segmental duplications of chromosome arm 1q. Genes Chromo- 1 Fonseca R, Barlogie B, Bataille R, Bastard C, Bergsagel PL, Chesi M somes Cancer 2005; 42: 95–106. et al. Genetics and cytogenetics of multiple myeloma: a workshop 19 Zhan F, Huang Y, Colla S, Stewart JP, Hanamura I, Gupta S et al. report. Cancer Res 2004; 64: 1546–1558. The molecular classification of multiple myeloma. Blood, (E-pub 2 Fonseca R, Debes-Marun CS, Picken EB, Dewald GW, Bryant SC, ahead of print 25 May 2006) 2006, (DOI 10.1182/blood-2005-11- Winkler JM et al. The recurrent IgH translocations are highly 013458). associated with nonhyperdiploid variant multiple myeloma. Blood 20 Ahmann GJ, Jalal SM, Juneau AL, Christensen ER, Hanson CA, 2003; 102: 2562–2567. Dewald GW et al. A novel three-color, clone-specific fluorescence 3 Smadja NV, Leroux D, Soulier J, Dumont S, Arnould C, Taviaux S in situ hybridization procedure for monoclonal gammopathies. et al. Further cytogenetic characterization of multiple myeloma Cancer Genet Cytogenet 1998; 101: 7–11. confirms that 14q32 translocations are a very rare event in 21 Fonseca R, Blood E, Rue M, Harrington D, Oken MM, Kyle RA hyperdiploid cases. Genes Chromosomes Cancer 2003; 38: et al. Clinical and biologic implications of recurrent genomic 234–239. aberrations in myeloma. Blood 2003; 101: 4569–4575. 4 Bergsagel PL, Kuehl WM, Zhan F, Sawyer J, Barlogie B, 22 Perez EA, Roche PC, Jenkins RB, Reynolds CA, Halling KC, Ingle Shaughnessy Jr J. Cyclin D dysregulation: an early and unifying JN et al. HER2 testing in patients with breast cancer: poor pathogenic event in multiple myeloma. Blood 2005; 106: correlation between weak positivity by immunohistochemistry and 296–303. gene amplification by fluorescence in situ hybridization. Mayo 5 Mattioli M, Agnelli L, Fabris S, Baldini L, Morabito F, Bicciato S Clin Proc 2002; 77: 148–154. et al. Gene expression profiling of plasma cell dyscrasias reveals 23 Abraham RS, Ballman KV, Dispenzieri A, Grill DE, Manske MK, molecular patterns associated with distinct IGH translocations in Price-Troska TL et al. Functional gene expression analysis of clonal multiple myeloma. Oncogene 2005; 24: 2461–2473. plasma cells identifies a unique molecular profile for light chain 6 Chng WJ, Van Wier SA, Ahmann GJ, Winkler JM, Jalal SM, amyloidosis. Blood 2005; 105: 794–803. Bergsagel PL et al. A validated FISH trisomy index demonstrates 24 Greipp PR, Lust JA, O’Fallon WM, Katzmann JA, Witzig TE, Kyle the hyperdiploid and nonhyperdiploid dichotomy in MGUS. Blood RA. Plasma cell labeling index and beta 2-microglobulin predict 2005; 106: 2156–2161. survival independent of thymidine kinase and C-reactive protein in 7 Fonseca R, Bailey RJ, Ahmann GJ, Rajkumar SV, Hoyer JD, Lust JA multiple myeloma. Blood 1993; 81: 3382–3387. et al. Genomic abnormalities in monoclonal gammopathy of 25 Gutierrez NC, Garcia JL, Hernandez JM, Lumbreras E, Castellanos undetermined significance. Blood 2002; 100: 1417–1424. M, Rasillo A et al. Prognostic and biologic significance of 8 Kaufmann H, Ackermann J, Baldia C, Nosslinger T, Wieser R, Seidl chromosomal imbalances assessed by comparative genomic S et al. Both IGH translocations and chromosome 13q deletions hybridization in multiple myeloma. Blood 2004; 104: 2661–2666.

Leukemia Chromosome 1q21 copy change and prognosis in myeloma R Fonseca et al 2040 26 Liebisch P, Wendl C, Wellmann A, Krober A, Schilling G, 32 Gilles F, Goy A, Remache Y, Shue P, Zelenetz AD. MUC1 Goldschmidt H et al. High incidence of trisomies 1q, 9q, dysregulation as the consequence of a t(1;14)(q21;q32) transloca- and 11q in multiple myeloma: results from a comprehensive tion in an extranodal lymphoma. Blood 2000; 95: 2930–2936. molecular cytogenetic analysis. Leukemia 2003; 17: 33 Hatakeyama S, Osawa M, Omine M, Ishikawa F. JTB: a novel 2535–2537. membrane protein gene at 1q21 rearranged in a jumping 27 Rosinol L, Carrio A, Blade J, Queralt R, Aymerich M, Cibeira MT translocation. Oncogene 1999; 18: 2085–2090. et al. Comparative genomic hybridisation identifies two variants 34 Hatzivassiliou G, Miller I, Takizawa J, Palanisamy N, Rao PH, Iida of smoldering multiple myeloma. Br J Haematol 2005; 130: S et al. IRTA1 and IRTA2, novel immunoglobulin superfamily 729–732. receptors expressed in B cells and involved in chromosome 1q21 28 Chen W, Palanisamy N, Schmidt H, Teruya-Feldstein J, Jhanwar abnormalities in B cell malignancy. Immunity 2001; 14: 277–289. SC, Zelenetz AD et al. Deregulation of FCGR2B expression by 35 Inoue J, Otsuki T, Hirasawa A, Imoto I, Matsuo Y, Shimizu S et al. 1q21 rearrangements in follicular lymphomas. Oncogene 2001; Overexpression of PDZK1 within the 1q12-q22 amplicon is likely 20: 7686–7693. to be associated with drug-resistance phenotype in multiple 29 Cheng KW, Lahad JP, Kuo WL, Lapuk A, Yamada K, Auersperg N myeloma. Am J Pathol 2004; 165: 71–81. et al. The RAB25 small GTPase determines aggressiveness of 36 Willis TG, Zalcberg IR, Coignet LJ, Wlodarska I, Stul M, Jadayel ovarian and breast cancers. Nat Med 2004; 10: 1251–1256. DM et al. Molecular cloning of translocation t(1;14)(q21;q32) 30 Craig RW, Jabs EW, Zhou P, Kozopas KM, Hawkins AL, Rochelle defines a novel gene (BCL9) at chromosome 1q21. Blood 1998; JM et al. Human and mouse chromosomal mapping of the myeloid 91: 1873–1881. cell leukemia-1 gene: MCL1 maps to human chromosome 1q21, a 37 Carrasco DR, Tonon G, Huang Y, Zhang Y, Sinha R, Feng B et al. region that is frequently altered in preneoplastic and neoplastic High-resolution genomic profiles define distinct clinico-pathoge- disease. Genomics 1994; 23: 457–463. netic subgroups of multiple myeloma patients. Cancer Cell 2006; 31 Dyomin VG, Palanisamy N, Lloyd KO, Dyomina K, Jhanwar SC, 9: 313–325. Houldsworth J et al. MUC1 is activated in a B-cell lymphoma 38 Sawyer JR, Tricot G, Mattox S, Jagannath S, Barlogie B. Jumping by the t(1;14)(q21;q32) translocation and is rearranged and translocations of chromosome 1q in multiple myeloma: evidence amplified in B-cell lymphoma subsets. Blood 2000; 95: for a mechanism involving decondensation of pericentromeric 2666–2671. heterochromatin. Blood 1998; 91: 1732–1741.

Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)

Leukemia