Originally published as JCO Early Release 10.1200/JCO.2006.08.7999 on June 11 2007

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kienle, D. Articles by Stilgenbauer, S. Search for Related Content PubMed PubMed Citation Articles by Kienle, D. Articles by Stilgenbauer, S.

Quantitative Gene Expression Deregulation in Mantle-Cell Lymphoma: Correlation With Clinical and Biologic Factors

Dirk Kienle, Tiemo Katzenberger, German Ott, Doreen Saupe, Axel Benner, Holger Kohlhammer, Thomas F.E. Barth, Sylvia Höller, Jörg Kalla, Andreas Rosenwald, Hans Konrad Müller-Hermelink, Peter Möller, Peter Lichter, Hartmut Döhner, Stephan Stilgenbauer

From the Department of Internal Medicine III, University of Ulm, Ulm; Department of Pathology, University of Würzburg, Würzburg; Department of Biostatistics and Department of Molecular Genetics, German Cancer Research Center (DKFZ) Heidelberg, Heidelberg; and the Department of Pathology, University of Ulm, Ulm, Germany

Address reprint requests to Stephan Stilgenbauer, MD, Internal Medicine III, University of Ulm, Robert-Koch-Straße 8, 89081 Ulm, Germany; e-mail: stephan.stilgenbauer{at}uniklinik-ulm.de

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
Purpose: There is evidence for a direct role of quantitative gene expression deregulation in mantle-cell lymphoma (MCL) pathogenesis. Our aim was to investigate gene expression associations with other pathogenic factors and the significance of gene expression in a multivariate survival analysis.

Patients and Methods: Quantitative expression of 20 genes of potential relevance for MCL prognosis and pathogenesis were analyzed using real-time reverse transcriptase polymerase chain reaction and correlated with clinical and genetic factors, tumor morphology, and Ki-67 index in 65 MCL samples.

Results: Genomic losses at the loci of TP53, RB1, and P16 were associated with reduced transcript levels of the respective genes, indicating a gene-dosage effect as the pathomechanism. Analysis of gene expression correlations between the candidate genes revealed a separation into two clusters, one dominated by proliferation activators, another by proliferation inhibitors and regulators of apoptosis. Whereas only weak associations were identified between gene expression and clinical parameters or blastoid morphology, several genes were correlated closely with the Ki-67 index, including the short CCND1 variant (positive correlation) and RB1, ATM, P27, and BMI (negative correlation). In multivariate survival analysis, expression levels of MYC, MDM2, EZH2, and CCND1 were the strongest prognostic factors independently of tumor proliferation and clinical factors.

Conclusion: These results indicate a pathogenic contribution of several gene transcript levels to the biology and clinical course of MCL. Genes can be differentiated into factors contributing to proliferation deregulation, either by enhancement or loss of inhibition, and proliferation-independent factors potentially contributing to MCL pathogenesis by apoptosis impairment.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
Mantle-cell lymphoma (MCL) is a distinct B-cell non-Hodgkin's lymphoma characterized by the chromosomal translocation t(11;14) leading to overexpression of cyclin D1 (CCND1), a regulator of the G₁ phase of the cell cycle.^1,2 The outcome is usually poor, but heterogeneity has been described, including patients surviving for more than 10 years.^3-5 Subsequently, attempts have been made to develop markers identifying disease risk groups. Several clinical characteristics have been associated with MCL prognosis such as "B" symptoms, bulky disease, and the International Prognostic Index (IPI).^3,4,6 Among the morphologic MCL subtypes, the blastoid variant has been associated with aggressive disease.^3,4,6-9 Tumor proliferation, determined by the immunohistochemical detection of Ki-67, was recognized as a strong prognostic marker.^4,6,9,10

More recently, the expression of several genes has been associated with MCL pathogenesis or prognosis. Many of these genes are closely related to cell-cycle control, particularly affecting the G₁/S phase checkpoint, such as CCND1,^1,2,5 CCND3,¹¹ P16,^12,13 P27,^14-16 CDK4,¹⁷ and RB1,¹⁸ suggesting a direct pathogenic relevance of this pathway in MCL. Using global cDNA microarray analysis, Rosenwald et al⁵ were able to construct a quantitative survival predictor based on the expression levels of 20 proliferation-associated genes (proliferation signature). A set of four genes from this signature was suggested as a substitute for the proliferation signature and was therefore proposed for future risk stratification.

Alterations of genes from apoptosis and DNA damage response pathways were suggested to contribute to MCL pathogenesis,¹⁹ as reported for ATM,^20,21 TP53,^22-24 MDM2,^24,25 and MCL1.²⁶ Many of these genes are located in critical regions affected recurrently by genomic aberrations.^27-31

We performed a systematic expression analysis using real-time quantitative reverse transcriptase polymerase chain reaction (RQ-PCR) of genes that were suggested to be involved in MCL pathogenesis or prognosis based on published results. Main foci were regulators of the G₁/S phase checkpoint or of apoptosis, and the proliferation signature genes suggested by Rosenwald et al.⁵ Our aims were to elucidate interrelations with other pathogenic factors, such as tumor proliferation, genomic aberrations, or clinical characteristics, and to investigate the significance of quantitative gene expression for survival prediction in multivariate analysis.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
Patients
A total of 78 samples from untreated MCL patients were collected at the Universities of Würzburg (n = 63), Heidelberg (n = 8), and Ulm (n = 7), Germany, between October 1979 and November 2002 at the time of diagnosis. DNA and RNA were obtained for diagnostic procedures after informed consent according to the Declaration of Helsinki from frozen lymph node (LN) tissue blocks (65 patients) and, for expression comparison, from peripheral blood (PB) samples (13 patients). All samples showed CCND1 overexpression on protein and/or mRNA level. All LN samples were diagnosed histologically according to the WHO classification by reference pathology institutions at Würzburg and Ulm (by G.O., T.F.E.B., P.M., and H.K.M.-H.). Of the LN samples, 60 harbored a t(11;14) detected by fluorescent in situ hybridization (FISH) as described³²; in five samples, FISH analysis was not successful, but all of these showed CCND1 overexpression. For eight of the PB samples, histologic diagnosis was available (three by reference and five by nonreference pathology institutions); the remaining PB samples were included based on their t(11;14) positivity and CCND1 overexpression, typical blood smear morphology, and immunophenotype. Median tumor cell content was 85% (range, 50% to 99%), estimated by the percentage of cells harboring recurrent genomic aberrations by FISH, and was comparable between LN (median, 87%) and PB samples (median, 77%). Clinical, morphologic, and treatment characteristics are listed in Table 1. Therapy was not within a single clinical trial and therefore was heterogeneous. For standardization, three PB samples of healthy donors after CD19 purification (CD19⁺) were used.

RQ-PCR
Procedures were reported,^36,37 including a DNase digestion step to avoid contamination with genomic DNA. Average values of two independent endogenous control amplicons (PGK1, LMNB1) were used for expression normalization. Primers (Appendix Table A1, online only) were described previously: CCND3, CDK2, CDK4, E2F1, P27 (CDKN1B), LMNB1, MYC, PGK1 (all from Korz et al³⁶), MCL1, MDM2 (both from Kienle et al³⁷), CCND1 coding region,³⁸ or newly designed as described³⁶: ASPM, BMI, P16 (CDKN2A), CENPF, EZH2, KI67, RB1, TP53, Tubulin (TUBA1B), CCND1 3'-untranslated region (UTR); for CDC2, an Assay-on-Demand Gene Expression Probe was used (Applied Biosystems, Weiterstadt, Germany). Primers for ASPM, CDC2, CENPF, and Tubulin are based on the same gene sequences as reported in Rosenwald et al.⁵

Statistical Analysis
Correlations of gene expression levels (log2 scale) with dichotomous factors were performed by a multiple testing procedure³⁹ using two-sample Welch t statistics with P value adjustment according to the step-down maxT procedure (1,000 bootstrap iterations to control the family-wise error rate). Correlations between gene expression levels were estimated by Spearman's rank correlation coefficient. Seriation was done by optimal leaf ordering for hierarchical clustering.⁴⁰ To identify short CCND1 variant expressers, Kernel estimation of the two-dimensional density of the CCND1 coding region and the CCND1 3'UTR expression was performed.⁴¹ To test for the effect of gene groups on Ki-67 index or survival, Goeman's global test was performed.^42,43 Survival distributions were estimated according to Kaplan and Meier.⁴⁴ Pairwise comparisons of survival curves were done by log-rank tests, and multivariate survival analysis was by Cox proportional hazards regression. Backward-selection (level 0.5) was used to exclude redundant variables. Missing values were estimated by applying a multiple-imputation technique.⁴⁵ The Mann-Whitney test statistic was used for subgroup comparisons between samples with versus without genomic abnormalities. Statistical analyses were performed with the R software environment, version 2.3.1⁴⁶ together with Bioconductor, version 1.8.⁴⁷

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
Gene Expression in Nodal and PB MCL Cells
Of 78 MCL samples under study, 65 were from LN tissue and 13 were from PB samples. Despite a similar tumor cell content as detected by FISH and comparable CCND1 levels (Appendix Table A2, online only) in PB and LN samples, several candidate genes showed expression differences between these cohorts (Appendix Table A2) including CDC2, CENPF, ASPM, and Tubulin (overexpressed in LN samples), and P27, CCND3, MCL1, BMI, and ATM (lower expression in LN samples). Similar expression alterations occurred when the PB samples were compared with leukemic LN samples only (Appendix Table A2). Therefore, additional analyses were restricted to LN samples only.

Gene Expression, Genomic Aberrations, and VH Mutation Status
Several candidate genes are located within critical regions of recurrent genomic abnormalities in MCL (Appendix Table A3, online only). Frequent genomic losses in MCL involve the loci of P16, RB1, ATM, and TP53.^27-31 For these genes, a transcript downregulation was detected in cases with loss of the respective locus (Fig 1). The correlation was highly significant for TP53 and RB1 but inconsistent for P16 or ATM.

View larger version (11K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig 1. Expression of candidate genes from critical genomic regions. (A) ATM, (B) p16, (C) RB1, and (D) TP53. Gene expression is shown after normalization on CD19+ controls. No loss: samples with a normal gene copy number; Loss: loss of the respective gene locus. P values: Mann-Whitney test statistics for subgroup comparisons. Median expression levels: (A) 0.50 (no loss), 0.40 (loss); (B) 0.19 (no loss), 0.08 (loss); (C) 0.70 (no loss), 0.30 (loss); (D) 2.86 (no loss), 0.90 (loss).

MCL patient samples are divided into a VH-unmutated and a VH-mutated subgroup.^35,49-52 Expression of the candidate genes was not significantly different between the VH-mutation subgroups of MCL (data not shown).

CCND1 Variant Expression
To differentiate expression of the short (1.7 kb) and long (4.5 kb) CCND1 variant differing in the length of the 3'-UTR,^53,54 we quantified transcripts specific for the CCND1 coding region and the CCND1 3'-UTR. We used a cutoff ratio of 15 between these transcripts for subgroup separation (Appendix Fig A1, online only) resulting in nine short variant and 56 long variant expressers. Short variant expression was associated with poor survival (median, 8.4 v 45.1 months; P <.001; Appendix Fig A2, online only) and high proliferation rates (median Ki-67 index, 60% v 30%; P = .02). Short variant expressers exhibited significantly higher CCND1 coding region levels compared with long variant samples (median, 70 v 35). However, when compared with 10 long variant expressers with the highest CCND1 coding region levels (median, 110), despite similar CCND1 coding region levels, the short variant samples retained higher Ki-67 indices (median, 60% v 30%) and shorter survival times (median, 8.4 v 43 months).

Gene Expression, Blastoid Morphology, and Proliferation
Except for MCL1 (1.7-fold higher expression in blastoid samples; P = .031), there was no statistically significant association of candidate gene expression with blastoid morphology (Appendix Table A4).

High Ki-67 proliferation indices were an adverse prognostic factor (Appendix Fig A3, online only; Table 2). Expression of several genes was tightly correlated with the Ki-67 index. The strongest influence on proliferation indices (Fig 2) showed that RB1, P27, MDM2, MCL1, and ATM all had a negative influence on proliferation. A weak influence was observed for CCND1 coding region (positive), CCND1, 3'-UTR, CCND3, and BMI (all negative). KI67 mRNA expression was not reliably correlated with Ki-67 index (Spearman's = 0.06), and therefore could not substitute the immunohistochemical determination. The proliferation signature genes described by Rosenwald⁵ (CDC2, ASPM, CENPF, Tubulin) were not associated significantly with the proliferation index when tested individually (Fig 2) or as a gene group (P = .86, Goeman's global test).

View larger version (24K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig 2. Association of gene expression with the Ki-67 index. Goeman's global test. Influence of individual gene expression levels on the proliferation index. Yellow, positive; blue, negative influence. Reference line: expected height of the bar under the null hypothesis that the regression coefficient is zero indicating no influence on proliferation. Horizontal marks describe by how many standard deviations the height of the bar exceeds the reference line. UTR, untranslated region.

We performed a multivariate survival analysis including expression of all candidate genes, blastoid morphology, Ki-67 index, tumor bulk (> 10 cm), B symptoms, and IPI score (0 to 2 v 3 to 5). In addition, high-dose chemotherapy and autologous stem-cell transplantation, which appears to improve overall survival at least in younger patients,⁵⁵ was included. A combination of molecular and clinical factors was of prognostic value (Table 2). Highly significant factors were expression of MYC, MDM2, EZH2, and CCND1 (coding region and 3'-UTR), Ki-67 index, and presence of B symptoms. To illustrate the prognostic impact of this model, the patients were split into four groups based on the factors listed in Table 2, using the quartiles of the distribution of the linear predictor of the Cox model. Estimated survival curves were plotted accordingly (Fig 3). For patients assigned to quartile 1, estimated median survival was 96.5 months compared with 12.2 months for patients in quartile 4.

View larger version (11K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig 3. Survival predictor based on quartile statistics of the final Cox model. Q1 to Q4 represent the survival curve estimates corresponding to patient groups defined by the quartiles of the linear predictor of the final Cox model (Table 2).

View larger version (59K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig 4. Interrelations of gene expression levels. Shows pairwise correlations of candidate gene expression levels. Seriation was done by optimal leaf ordering for hierarchical clustering.40 Red, positive expression correlation; blue, negative correlation; white, no correlation between two candidate genes. Score: color intensity reflects the level of expression correlation [Spearman's correlation coefficients from –1 (complete negative) to 1 (complete positive correlation)].

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

Only few differentially expressed genes were identified in association with clinical factors, indicating that these might rather reflect disease progression than distinct biologic disease subsets. Similarly, overexpression of MCL1, an antiapoptotic protein, was the only characteristic feature for the MCL subgroup with blastoid morphology. This confirms an earlier report²⁶ that showed an association between MCL1 protein overexpression and high-grade morphology in MCL, indicating apoptosis-deregulation involved in the pathogenesis of blastoid MCL.

Proliferation as measured by the Ki-67 index is a powerful prognostic marker in MCL,⁹ which could be confirmed in this study. Several candidate genes showed a close correlation with the Ki-67 index, supporting a central biologic impact of proliferation in MCL, as suggested,^5,19 and pointing to underlying pathogenic events. A positive correlation with proliferation was obtained for CCND1, indicating a direct pathogenic influence of CCND1 levels on G₁/S phase transition in MCL. Remarkably, when analyzing interrelations of gene expression levels, G₁/S phase enhancers such as E2F1, CDK2, and CDK4 clustered with other proliferation-related genes, arguing for the biologic activity of the G₁/S phase pathway in proliferation deregulation of MCL. In particular, MCL with short CCND1 variant expression was characterized by high Ki-67 indices in line with previous reports,^5,56 which was proposed to be due to higher CCND1 coding region levels as a consequence of a longer mRNA half-life.^53,54,56 However, short variant expressers exhibited higher proliferation rates and shorter survival times as compared with long variant expressers with similar CCND1 coding region levels, suggesting additional biologic properties of the short CCND1 mRNA-modulating cellular proliferation in MCL, as indicated recently.^56,57

Remarkably, the genes showing the strongest influence of proliferation association were negatively correlated with the Ki-67 index, which is an important extension to the findings of Rosenwald et al.⁵ In that study, the quantitative relation between reduced survival and proliferation enhancement was related principally to gene overexpression. Our findings indicate that reduced expression of antiproliferative genes such as P27, RB1, and ATM seems to be an important factor for cell cycle deregulation in MCL, indicating a synergistic mechanism involving overexpression of cell cycle enhancers and loss of proliferation inhibitors in MCL. The gene expression interrelations observed in this study support this concept, revealing a separation of the candidate genes into two clusters: one dominated by proliferation activators, and another dominated by proliferation inhibitors and apoptosis regulators. A potential pathomechanism contributing to the dysfunction of proliferation inhibitors might be a gene–dosage-dependent transcript downregulation as a consequence of recurrent genomic aberrations, as demonstrated in this study for RB1 and TP53. In contrast to chronic lymphatic leukemia, 13q deletions in MCL are usually large and involve the RB1 locus.³⁰ Therefore, our data point to RB1 as a critical candidate gene in this region. For P16 and ATM, only a subset of samples showed evidence for a gene-dosage downregulation, indicating that additional mechanisms such as inactivation of the second allele seem to be required.^20,21

Based on cDNA microarray analysis, Rosenwald et al⁵ defined a proliferation signature that consisted of 20 genes strongly associated with proliferation and survival. Four of those genes were proposed as a surrogate for this signature. In this study, these four genes were not correlated reliably with the Ki-67 index and were correlated only partly with survival, suggesting that the information obtained by a microarray-defined signature may not be substituted easily for by a limited set of extracted genes. In addition, the study by Rosenwald et al did not allow for full correlation with the Ki-67 index (available for 14 of 101 samples), which was used as gold standard for estimating proliferation in the current study, indicating that the definition of proliferation might differ in the two studies.

Gene expression levels were the strongest prognostic factors in multivariate survival analysis in this study, supporting a direct relationship between quantitative transcript levels and disease progression. One group of prognosis-related genes was associated with proliferation, such as CCND1 or RB1, suggesting those genes influence survival times by proliferation deregulation. The association of BMI expression with reduced proliferation and prolonged survival (univariate analysis; Appendix Fig A4, online only) was unexpected due to the potential role of BMI as a transcriptional repressor of the INK4A locus.⁵⁸ However, a pathogenic role for a differential expression of BMI and EZH2, constituting two distinct polycomb complexes, was reported. Overexpression of EZH2 relative to BMI was associated with transition of resting mantle zone B-cells to rapidly dividing centroblasts and with carcinogenesis.^59-61 This might correspond to the findings of this study identifying EZH2 as negative prognostic factor with a complementary expression pattern compared with BMI.

Some of the prognosis-related genes were not associated significantly with the proliferation index, arguing for proliferation-independent mechanisms in MCL progression. This is supported by the multivariate survival analysis identifying a number of candidate genes independently of the Ki-67 index. Several of these genes are involved in apoptosis control, indicating apoptosis impairment as an additional pathomechanism in MCL. ATM and TP53 are key activators of proapoptotic responses after DNA damage, and genomic loss confers specific gene expression profiles in MCL.⁶² As discussed, a gene-dosage effect might contribute to a deregulation of these genes. In line with this concept and previous reports^25,63 is the prognostic influence of MDM2 expression in our series. MDM2 is a direct inhibitor of TP53 and thereby contributes to TP53 dysfunction.⁶⁴ Similarly, MYC was described as inhibitor of TP53-mediated apoptosis.⁶⁵ MYC expression was a prognostic factor in multivariate survival analysis, extending the findings of previous reports.⁶⁶ Interestingly, MYC levels were not associated significantly with the Ki-67 index, indicating that MYC function in MCL might rather be related to the TP53 pathway, which is supported by a negative correlation between the expression levels of the two genes (Spearman's = –0.32).

This study suggests a major prognostic impact of quantitative gene expression deregulation in MCL independently of the Ki-67 proliferation index and clinical factors. A limited set of factors provided a tool for the separation of risk groups differing in survival estimates, comparable to the proliferation signature described by Rosenwald et al.⁵ Gene transcripts identified as prognostic factors in multivariate survival analysis, such as MYC, MDM2, and the short and long variant of CCND1, should be evaluated in independent series and within future clinical trials. However, methodologic standardization is needed before these factors can be applied routinely.

	AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
The author(s) indicated no potential conflicts of interest.

	AUTHOR CONTRIBUTIONS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
Conception and design: Dirk Kienle, Peter Lichter, Stephan Stilgenbauer

Provision of study materials or patients: Tiemo Katzenberger, German Ott, Thomas F.E. Barth, Hans K. Müller-Hermelink, Peter Möller, Hartmut Döhner, Stephan Stilgenbauer

Collection and assembly of data: Dirk Kienle, Tiemo Katzenberger, German Ott, Doreen Saupe, Holger Kohlhammer, Thomas F.E. Barth, Sylvia Höller, Jörg Kalla, Andreas Rosenwald, Peter Möller, Hartmut Doehner, Stephan Stilgenbauer

Data analysis and interpretation: Dirk Kienle, Axel Benner, Holger Kohlhammer, Andreas Rosenwald, Stephan Stilgenbauer

Manuscript writing: Dirk Kienle, Axel Benner, Stephan Stilgenbauer

Final approval of manuscript: Dirk Kienle, Tiemo Katzenberger, German Ott, Andreas Rosenwald, Hans K. Müller-Hermelink, Peter Möller, Peter Lichter, Hartmut Döhner, Stephan Stilgenbauer

	Appendix

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
The Appendix is included in the full-text version of this article, available online at www.jco.org. It is not included in the PDF version (via Adobe® Reader®).

View larger version (20K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig A1. Definition of short and long CCND1 variant expression. Two density groups are separated at a ratio of 15 between the expression values of CCND1 coding and 3'UTR region (red line). Samples above the line (ratio < 15) correspond to long variant expressers; samples below (ratio > 15) to short variant expressers.

View larger version (10K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig A2. Overall survival in mantle-cell lymphoma with short and long CCND1 variant expression. Kaplan-Meier estimation. Median survival was 12.7 months for samples with short variant expression and 47 months for long variant expressers.

View larger version (10K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig A3. Ki-67 index and overall survival. Kaplan-Meier estimates and results of log-rank tests for samples with a Ki-67 index 0.4 (ie, more than 40% positive cells) versus the remainder. Median survival was 17.1 months for samples with indices 0.4 and 52.5 months for the remainder.

View larger version (11K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig A4. BMI expression and overall survival. Kaplan-Meier estimates separated for patients with a BMI expression above and below the median expression. Median survival was 49 months for patients with higher than median expression and 31 months for patients with lower than median expression (log-rank test P = .003).

	ACKNOWLEDGMENTS

 
We thank Annett Habermann for excellent technical assistance.

	NOTES

 
published online ahead of print at www.jco.org on June 11, 2007.

Supported by the European MCL Network (Grant No. LSHC-CT-2004-503351), Deutsche Forschungsgemeinschaft (DFG; Grant No. STI 296/1-1), and Landesstiftung Banden-Württemberg (Grant No. P-LS-Prot/54).

Presented in part at the 46th Annual Meeting of the American Society of Hematology, December 4-7, 2004, San Diego, CA, and at the Gemeinsame Jahrestagung der Deutschen, Österreichischen und Schweizerischen Gesellschaften (DGHO, ÖGHO, SGH, and SGMO) für Hamatologie und Onkologie, October 2-6, 2004, Innsbruck, Austria.

Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
1. Campo E, Raffeld M, Jaffe ES: Mantle-cell lymphoma. Semin Hematol 36:115-127, 1999[Medline]

2. Swerdlow SH, Williams ME: From centrocytic to mantle cell lymphoma: A clinicopathologic and molecular review of 3 decades. Hum Pathol 33:7-20, 2002[CrossRef][Medline]

3. Argatoff LH, Connors JM, Klasa RJ, et al: Mantle cell lymphoma: A clinicopathologic study of 80 cases. Blood 89:2067-2078, 1997[Abstract/Free Full Text]

4. Bosch F, Lopez-Guillermo A, Campo E, et al: Mantle cell lymphoma: Presenting features, response to therapy, and prognostic factors. Cancer 82:567-575, 1998[CrossRef][Medline]

5. Rosenwald A, Wright G, Wiestner A, et al: The proliferation gene expression signature is a quantitative integrator of oncogenic events that predicts survival in mantle cell lymphoma. Cancer Cell 3:185-197, 2003[CrossRef][Medline]

6. Räty R, Franssila K, Joensuu H, et al: Ki-67 expression level, histological subtype, and the International Prognostic Index as outcome predictors in mantle cell lymphoma. Eur J Haematol 69:11-20, 2002[CrossRef][Medline]

7. Ott G, Kalla J, Ott MM, et al: Blastoid variants of mantle cell lymphoma: Frequent bcl-1 rearrangements at the major translocation cluster region and tetraploid chromosome clones. Blood 89:1421-1429, 1997[Abstract/Free Full Text]

8. Beà S, Ribas M, Hernandez JM, et al: Increased number of chromosomal imbalances and high-level DNA amplifications in mantle cell lymphoma are associated with blastoid variants. Blood 93:4365-4374, 1999[Abstract/Free Full Text]

9. Tiemann M, Schrader C, Klapper W, et al: Histopathology, cell proliferation indices and clinical outcome in 304 patients with mantle cell lymphoma (MCL): A clinicopathological study from the European MCL Network. Br J Haematol 131:29-38, 2005[CrossRef][Medline]

10. Katzenberger T, Petzoldt C, Höller S, et al: The Ki67 proliferation index is a quantitative indicator of clinical risk in mantle cell lymphoma. Blood 107:3407, 2006[Free Full Text]

11. Ott MM, Bartkova J, Bartek J, et al: Cyclin D1 expression in mantle cell lymphoma is accompanied by downregulation of cyclin D3 and is not related to the proliferative activity. Blood 90:3154-3159, 1997[Abstract/Free Full Text]

12. Pinyol M, Hernandez L, Cazorla M, et al: Deletions and loss of expression of p16INK4a and p21Waf1 genes are associated with aggressive variants of mantle cell lymphomas. Blood 89:272-280, 1997[Abstract/Free Full Text]

13. Dreyling MH, Bullinger L, Ott G, et al: Alterations of the cyclin D1/p16-pRB pathway in mantle cell lymphoma. Cancer Res 57:4608-4614, 1997[Abstract/Free Full Text]

14. Chiarle R, Budel LM, Skolnik J, et al: Increased proteasome degradation of cyclin-dependent kinase inhibitor p27 is associated with a decreased overall survival in mantle cell lymphoma. Blood 95:619-626, 2000[Abstract/Free Full Text]

15. Quintanilla-Martinez L, Davies-Hill T, Fend F, et al: Sequestration of p27Kip1 protein by cyclin D1 in typical and blastic variants of mantle cell lymphoma (MCL): Implications for pathogenesis. Blood 101:3181-3187, 2003[Abstract/Free Full Text]

16. Letestu R, Ugo V, Valensi F, et al: Prognostic impact of p27KIP1 expression in cyclin D1 positive lymphoproliferative disorders. Leukemia 18:953-961, 2004[CrossRef][Medline]

17. Hernández L, Bea S, Pinyol M, et al: CDK4 and MDM2 gene alterations mainly occur in highly proliferative and aggressive mantle cell lymphomas with wild-type INK4a/ARF locus. Cancer Res 65:2199-2206, 2005[Abstract/Free Full Text]

18. Grønbaek K, Nedergaard T, Andersen MK, et al: Concurrent disruption of cell cycle associated genes in mantle cell lymphoma: A genotypic and phenotypic study of cyclin D1, p16, p15, p53 and pRb. Leukemia 12:1266-1271, 1998[CrossRef][Medline]

19. Fernàndez V, Hartmann E, Ott G, et al: Pathogenesis of mantle-cell lymphoma: All oncogenic roads lead to dysregulation of cell cycle and DNA damage response pathways. J Clin Oncol 23:6364-6369, 2005[Abstract/Free Full Text]

20. Camacho E, Hernandez L, Hernandez S, et al: ATM gene inactivation in mantle cell lymphoma mainly occurs by truncating mutations and missense mutations involving the phosphatidylinositol-3 kinase domain and is associated with increasing numbers of chromosomal imbalances. Blood 99:238-244, 2002[Abstract/Free Full Text]

21. Schaffner C, Idler I, Stilgenbauer S, et al: Mantle cell lymphoma is characterized by inactivation of the ATM gene. Proc Natl Acad Sci U S A 97:2773-2778, 2000[Abstract/Free Full Text]

22. Louie DC, Offit K, Jaslow R, et al: p53 overexpression as a marker of poor prognosis in mantle cell lymphomas with t(11;14)(q13;q32). Blood 86:2892-2899, 1995[Abstract/Free Full Text]

23. Greiner TC, Moynihan MJ, Chan WC, et al: p53 mutations in mantle cell lymphoma are associated with variant cytology and predict a poor prognosis. Blood 87:4302-4310, 1996[Abstract/Free Full Text]

24. Hernandez L, Fest T, Cazorla M, et al: p53 gene mutations and protein overexpression are associated with aggressive variants of mantle cell lymphomas. Blood 87:3351-3359, 1996[Medline]

25. Møller M, Nielsen B, Pedersen NT: Oncoprotein MDM2 overexpression is associated with poor prognosis in distinct non-Hodgkin's lymphoma entities. Mod Pathol 12:1010-1016, 1999[Medline]

26. Khoury JD, Medeiros LJ, Rassidakis GZ, et al: Expression of Mcl-1 in mantle cell lymphoma is associated with high-grade morphology, a high proliferative state, and p53 overexpression. J Pathol 199:90-97, 2003[CrossRef][Medline]

27. Monni O, Oinonen R, Elonen E, et al: Gain of 3q and deletion of 11q22 are frequent aberrations in mantle cell lymphoma. Genes Chromosomes Cancer 21:298-307, 1998[CrossRef][Medline]

28. Bentz M, Plesch A, Bullinger L, et al: T(11;14)-positive mantle cell lymphomas exhibit complex karyotypes and share similarities with B-cell chronic lymphocytic leukemia. Genes Chromosomes Cancer 27:285-294, 2000[CrossRef][Medline]

29. Stilgenbauer S, Winkler D, Ott G, et al: Molecular characterization of 11q deletions points to a pathogenic role of the ATM gene in mantle cell lymphoma. Blood 94:3262-3264, 1999[Abstract/Free Full Text]

30. Kohlhammer H, Schwaenen C, Wessendorf S, et al: Genomic DNA-chip hybridization in t(11;14)-positive mantle cell lymphomas shows a high frequency of aberrations and allows a refined characterization of consensus regions. Blood 104:795-801, 2004[Abstract/Free Full Text]

31. Schraders M, Pfundt R, Straatman HM, et al: Novel chromosomal imbalances in mantle cell lymphoma detected by genome-wide array-based comparative genomic hybridization. Blood 105:1686-1693, 2005[Abstract/Free Full Text]

32. Döhner H, Stilgenbauer S, Benner A, et al: Genomic aberrations and survival in chronic lymphocytic leukemia. N Engl J Med 343:1910-1916, 2000[Abstract/Free Full Text]

33. Korn EL: Censoring distributions as a measure of follow-up in survival analysis. Stat Med 5:255-260, 1986[Medline]

34. Kröber A, Seiler T, Benner A, et al: V(H) mutation status, CD38 expression level, genomic aberrations, and survival in chronic lymphocytic leukemia. Blood 100:1410-1416, 2002[Abstract/Free Full Text]

35. Kienle D, Kröber A, Katzenberger T, et al: VH mutation status and VDJ rearrangement structure in mantle cell lymphoma: Correlation with genomic aberrations, clinical characteristics, and outcome. Blood 102:3003-3009, 2003[Abstract/Free Full Text]

36. Korz C, Pscherer A, Benner A, et al: Evidence for distinct pathomechanisms in B-cell chronic lymphocytic leukemia and mantle cell lymphoma by quantitative expression analysis of cell cycle and apoptosis-associated genes. Blood 99:4554-4561, 2002[Medline]

37. Kienle D, Korz C, Hosch B, et al: Evidence for distinct pathomechanisms in genetic subgroups of chronic lymphocytic leukemia revealed by quantitative expression analysis of cell cycle, activation, and apoptosis-associated genes. J Clin Oncol 23:3780-3792, 2005[Abstract/Free Full Text]

38. Bijwaard KE, Aguilera NS, Monczak Y, et al: Quantitative real-time reverse transcription-PCR assay for cyclin D1 expression: Utility in the diagnosis of mantle cell lymphoma. Clin Chem 47:195-201, 2001[Abstract/Free Full Text]

39. Pollard KS, van der Laan MJ: Resampling-based Multiple Testing: Asymptotic Control of Type I Error and Applications to Gene Expression Data. Division of Biostatistics, University of California, Berkeley, Technical Report 121, 2003. http://www.bepress.com/ucbbiostat/paper121

40. Bar-Joseph Z, Demaine ED, Gifford DK, et al: Fast optimal leaf ordering for hierarchical clustering. Bioinformatics 17:22-29, 2001 (suppl 1)

41. Wand MP, Jones MC: Kernel Smoothing. London, United Kingdom, Chapman and Hall, 1995

42. Goeman JJ, van de Geer SA, de Kort F, et al: A global test for groups of genes: Testing association with a clinical outcome, Bioinformatics 20:93-99, 2004[Abstract/Free Full Text]

43. Goeman JJ, Oosting J, Cleton-Jansen A, et al: Testing association of a pathway with survival using gene expression data. Bioinformatics 21:1950-1957, 2005[Abstract/Free Full Text]

44. Kaplan E, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958[CrossRef]

45. Harrell FE: Hmisc S function library. http://biostat.mc.vanderbilt.edu/twiki/bin/view/Main/Hmisc

46. R Development Core Team: R: A language and environment for statistical computing. Vienna, Austria, R Foundation for Statistical Computing, 2005

47. Gentleman R, Carey VJ, Bates DM, et al: Bioconductor: Open software development for computational biology and bioinformatics. Genome Biol 5:R80, 2004[CrossRef][Medline]

48. Beà S, Tort F, Pinyol M, et al: BMI-1 gene amplification and overexpression in hematological malignancies occur mainly in mantle cell lymphomas. Cancer Res 61:2409-2412, 2001[Abstract/Free Full Text]

49. Camacho FI, Algara P, Rodriguez A, et al: Molecular heterogeneity in MCL defined by the use of specific VH genes and the frequency of somatic mutations. Blood 101:4042-4046, 2003[Abstract/Free Full Text]

50. Walsh SH, Thorselius M, Johnson A, et al: Mutated VH genes and preferential VH3-21 use define new subsets of mantle cell lymphoma. Blood 101:4047-4054, 2003[Abstract/Free Full Text]

51. Orchard J, Garand R, Davis Z, et al: A subset of t(11;14) lymphoma with mantle cell features displays mutated IgVH genes and includes patients with good prognosis, nonnodal disease. Blood 101:4975-4981, 2003[Abstract/Free Full Text]

52. Nodit L, Bahler DW, Jacobs SA, et al: Indolent mantle cell lymphoma with nodal involvement and mutated immunoglobulin heavy chain genes. Hum Pathol 34:1030-1034, 2003[CrossRef][Medline]

53. Rimokh R, Berger F, Bastard C, et al: Rearrangement of CCND1 (BCL1/PRAD1) 3' untranslated region in mantle-cell lymphomas and t(11q13)-associated leukemias. Blood 83:3689-3696, 1994[Abstract/Free Full Text]

54. Lebwohl DE, Muise-Helmericks R, Sepp-Lorenzino L, et al: A truncated cyclin D1 gene encodes a stable mRNA in a human breast cancer cell line. Oncogene 9:1925-1929, 1994[Medline]

55. Dreyling M, Lenz G, Hoster E, et al: Early consolidation by myeloablative radiochemotherapy followed by autologous stem cell transplantation in first remission significantly prolongs progression-free survival in mantle-cell lymphoma: Results of a prospective randomized trial of the European MCL Network. Blood 105:2677-2684, 2005[Abstract/Free Full Text]

56. Sander B, Flygare J, Porwit-Macdonald A, et al: Mantle cell lymphomas with low levels of cyclin D1 long mRNA transcripts are highly proliferative and can be discriminated by elevated cyclin A2 and cyclin B1. Int J Cancer 117:418-430, 2005[CrossRef][Medline]

57. Culjkovic B, Topisirovic I, Skrabanek L, et al: EIF4E promotes nuclear export of cyclin D1 mRNAs via an element in the 3'UTR. J Cell Biol 169:245-256, 2005[Abstract/Free Full Text]

58. Itahana K, Zou Y, Itahana Y, et al: Control of the replicative life span of human fibroblasts by p16 and the polycomb protein Bmi-1. Mol Cell Biol 23:389-401, 2003[Abstract/Free Full Text]

59. Raaphorst FM, van Kemenade FJ, Fieret E, et al: Cutting edge: Polycomb gene expression patterns reflect distinct B cell differentiation stages in human germinal centers. J Immunol 164:1-4, 2000[Abstract/Free Full Text]

60. Breuer RH, Snijders PJ, Smit EF, et al: Increased expression of the EZH2 polycomb group gene in BMI-1-positive neoplastic cells during bronchial carcinogenesis. Neoplasia 6:736-743, 2004[CrossRef][Medline]

61. Raaphorst FM, Meijer CJ, Fieret E, et al: Poorly differentiated breast carcinoma is associated with increased expression of the human polycomb group EZH2 gene. Neoplasia 5:481-488, 2003[Medline]

62. Greiner TC, Dasgupta C, Ho VV, et al: Mutation and genomic deletion status of ataxia telangiectasia mutated (ATM) and p53 confer specific gene expression profiles in mantle cell lymphoma. Proc Natl Acad Sci U S A 103:2352-2357, 2006[Abstract/Free Full Text]

63. Solenthaler M, Matutes E, Brito-Babapulle V, et al: p53 and mdm2 in mantle cell lymphoma in leukemic phase. Haematologica 87:1141-1150, 2002[Abstract/Free Full Text]

64. de Oca Luna RM, Tabor AD, Eberspaecher H, et al: The organization and expression of the mdm2 gene. Genomics 33:352-357, 1996[CrossRef][Medline]

65. Ceballos E, Delgado MD, Gutierrez P, et al: C-Myc antagonizes the effect of p53 on apoptosis and p21WAF1 transactivation in K562 leukemia cells. Oncogene 19:2194-2204, 2000[CrossRef][Medline]

66. Nagy B, Lundan T, Larramendy ML, et al: Abnormal expression of apoptosis-related genes in haematological malignancies: Overexpression of MYC is poor prognostic sign in mantle cell lymphoma. Br J Haematol 120:434-441, 2003[CrossRef][Medline]

Submitted August 21, 2006; accepted April 4, 2007.

This article has been cited by other articles:

				I. Wlodarska, D. Dierickx, V. Vanhentenrijk, K. Van Roosbroeck, H. Pospisilova, F. Minnei, G. Verhoef, J. Thomas, P. Vandenberghe, and C. De Wolf-Peeters Translocations targeting CCND2, CCND3, and MYCN do occur in t(11;14)-negative mantle cell lymphomas Blood, June 15, 2008; 111(12): 5683 - 5690. [Abstract] [Full Text] [PDF]

				S. Sander, L. Bullinger, E. Leupolt, A. Benner, D. Kienle, T. Katzenberger, J. Kalla, G. Ott, H. K. Muller-Hermelink, T. F.E. Barth, et al. Genomic aberrations in mantle cell lymphoma detected by interphase fluorescence in situ hybridization. Incidence and clinicopathological correlations Haematologica, May 1, 2008; 93(5): 680 - 687. [Abstract] [Full Text] [PDF]