|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2007.10.8720 on June 18 2007 © 2007 American Society of Clinical Oncology. High Expression Levels of the ETS-Related Gene, ERG, Predict Adverse Outcome and Improve Molecular Risk-Based Classification of Cytogenetically Normal Acute Myeloid Leukemia: A Cancer and Leukemia Group B Study
From the Division of Hematology and Oncology, Department of Internal Medicine, and Division of Human Cancer Genetics, Department of Microbiology, Virology, Immunology and Medical Genetics, Comprehensive Cancer Center, The Ohio State University, Columbus, OH; Cancer and Leukemia Group B (CALGB) Statistical Center, Duke University Medical Center, Durham, NC; Department of Hematology and Oncology, Charité University Hospital, Berlin, Germany; Department of Genetics, University of Alabama at Birmingham, Birmingham, AL; Department of Internal Medicine, Section on Hematology and Oncology, Wake Forest University School of Medicine, Winston-Salem, NC; Department of Medicine, North Shore University Hospital, Manhasset, NY; and the Department of Medicine, University of Chicago, Chicago, IL Address reprint requests to Guido Marcucci, MD, The Ohio State University, The Comprehensive Cancer Center, A433B Starling-Loving Hall, 320 W 10th Ave, Columbus OH 43210; e-mail: guido.marcucci{at}osumc.edu
Purpose To validate ERG overexpression as an adverse predictor and assess its prognostic value in the context of other molecular markers in cytogenetically normal (CN) -acute myeloid leukemia (AML). Patients and Methods Seventy-six adult patients with primary CN-AML, younger than 60 years and treated on Cancer and Leukemia Group B (CALGB) trial 19808, were evaluated for ERG expression by quantitative reverse transcriptase polymerase chain reaction. They were then combined with 72 patients enrolled onto CALGB 9621 for analyses that included other molecular markers. Results Similar to patients enrolled onto CALGB 9621, high ERG expressers on CALGB 19808 had fewer complete remissions (CRs; P = .03) and worse event-free survival (EFS; P = .016) than low ERG expressers. In the combined set, high expressers (n = 55) had fewer CRs (P = .004) and shorter EFS (P < .001) than low expressers (n = 93). High ERG predicted failure to achieve CR (P = .004) after adjusting for BAALC expression (P = .04) and age (P = .008), and EFS (P = .004) after adjusting for FLT3 internal tandem duplication (ITD; P < .001). Among patients without FLT3-ITD (FLT3-ITD negative), only high ERG predicted shorter EFS (P = .001). Among NPM1-mutated (NPM1 positive) patients, high ERG predicted shorter EFS (P = .003), after adjusting for FLT3-ITD (P < .001). When all three markers were considered together, in the favorable FLT3-ITD–negative/NPM1-positive subset, high ERG was the only factor predicting shorter EFS (P = .002). Conclusion We validated ERG overexpression as an adverse predictor in CN-AML. Moreover, by using ERG expression levels, we improved the previously proposed molecular-risk classification of CN-AML based on the presence or absence of FLT3-ITD and NPM1 mutations, given that we identified subsets with different outcome among FLT3-ITD–negative, NPM1-positive, and FLT3-ITD–negative/NPM1-positive patients.
Acute myeloid leukemia (AML) with normal cytogenetics has been considered an intermediate-risk group, with five-year survival rates varying between 24% and 42%.1 This difference in clinical outcome reported across studies is likely related to the genomic heterogeneity of this cytogenetic subgroup in which several molecular markers have been proven prognostic.1 Among these, FLT3 internal tandem duplication (FLT3-ITD), MLL partial tandem duplication (MLL-PTD), and high BAALC expression have been shown to influence adversely the clinical outcome, whereas mutations of the NPM1 and CEBPA genes have been associated with a favorable prognosis.1-7 Furthermore, when considered in combination, these markers may become even better predictors. Indeed, patients with mutations in the NPM1 gene (NPM1 positive) and no FLT3-ITD (FLT3-ITD negative) have been shown to have a superior outcome to FLT3-ITD–negative patients with no NPM1 mutations (NPM1 negative) and those harboring FLT3-ITD (FLT3-ITD positive).3,4,8 ETS-related gene (ERG) and other members of the ETS family are downstream effectors of signaling transduction pathways involved in the regulation of cell proliferation, differentiation, and apoptosis.9-11 Cytogenetic or molecular rearrangements involving ERG, which is located at chromosome band 21q22, have been found in AML, Ewing sarcoma, and prostate cancer.10,12,13 ERG is also overexpressed frequently in AML patients with complex karyotypes and cryptic amplification of chromosome 21,14 and cytogenetically normal (CN) -AML patients with high levels of expression of this gene have been reported previously to have a poor clinical outcome.15 More recently, it has been shown that ERG overexpression also negatively influences the outcome of patients with other types of malignancies (ie, T-cell acute lymphoblastic leukemia and prostate cancer).16,17 To date, however, only one study on the prognostic significance of ERG in CN-AML has been reported.15 Therefore, we sought to validate the prognostic significance of ERG by analyzing an independent set of uniformly treated patients with CN-AML. Consistent with the prior report, we show here that higher levels of ERG expression are associated with significantly worse outcome. Furthermore, we show that when combined with other known prognostic markers, ERG expression can improve the molecular risk-based stratification of patients with CN-AML.
Patients A total of 148 adult patients younger than 60 years with untreated primary (de novo) CN-AML were included in this study. Seventy-six patients enrolled onto the Cancer and Leukemia Group B (CALGB) treatment protocol 1980818 were newly analyzed for ERG expression by real-time reverse transcriptase polymerase chain reaction (RT-PCR). The outcome of these patients was evaluated according to ERG expression levels. To explore the impact of ERG expression in the context of other markers, these patients were then combined with a previously reported group of 72 patients enrolled onto a similar treatment protocol, CALGB 9621.15 Pretreatment cytogenetic analyses of bone marrow (BM) were performed by CALGB-approved institutional cytogenetic laboratories as part of CALGB 8461, a prospective cytogenetic companion, and were reviewed centrally, as reported previously.19 To be considered cytogenetically normal, at least 20 metaphase cells had to be analyzed and the karyotype found to be normal in each case. The presence or absence of the FLT3-ITD,20 MLL-PTD,7 and mutations in the NPM1 gene,3 and BAALC expression levels6 were also determined centrally in pretreatment samples as described previously. Written informed consent for participation in these studies was obtained from all patients.
Treatment
RNA Extraction and Real-Time RT-PCR to Measure ERG Expression Levels
Definition of Clinical End Points
Statistical Methods
To address the prognostic value of ERG expression levels in the context of the presence or absence of other molecular prognostic markers (ie, FLT3-ITD, NPM1 mutation, MLL-PTD, and BAALC overexpression), high and low ERG expressers on CALGB 19808 were combined with the previously reported high and low ERG expressers on CALGB 9621, respectively, and analyzed for outcome. Pretreatment clinical features were compared between the two groups using Fisher's two-sided exact and Wilcoxon rank sum tests for categoric and continuous variables, respectively. To analyze factors related to the probability of achieving CR, a logistic regression model was constructed using a limited backward selection procedure. Variables considered in the model were those significant at
Estimated probabilities of EFS and DFS were calculated using the Kaplan-Meier method, and the log-rank test evaluated differences between survival distributions. Proportional hazards models were constructed for survival end points, using a limited backward selection procedure. Variables considered in the model were those significant at
Validation of ERG As a Predictor of Outcome in AML Within the group of patients enrolled onto CALGB 19808, high ERG expressers differed from the low expressers in that they were more frequently FLT3-ITD positive (P = .001) and had higher WBC counts (P < .001; Table 1). The high ERG group had a significantly lower CR rate compared with the low ERG group (79% v 97%; P = .03) and, with a median follow-up of 2.9 years (range, 1.7 to 5.4 years) for patients who did not have an event, a significantly worse EFS (median, 0.8 v 3.4 years; P = .016; Fig 1A). Estimated 2-year EFS rates for the high and low expressers were 29% and 53%, respectively.
These results confirmed those obtained in 72 patients with CN-AML enrolled onto CALGB 9621. In that set of patients, we also observed significantly different CR rates for patients with high ERG expression compared with those with low expression (65% v 89%; P = .03), when levels of this marker were measured against the internal control ABL (unpublished data). Furthermore, with a median follow-up of 7.1 years (range, 5.8 to 8.9 years) for patients who did not have an event, the median EFS duration for the high ERG expressers on CALGB 9621 was 0.5 years compared with 2.5 years for the low ERG expressers (P = .001), and the estimated 2-year EFS rates were 24% and 51%, respectively (Fig 1B).
Prognostic Value of ERG Expression in the Context of Other Predictive Molecular Markers At diagnosis, high expressers had higher WBC (P = .001), a lower incidence of gum infiltration (P = .006), higher incidence of FLT3-ITD (P = .003), and a trend toward more often having high expression of BAALC (P = .08; Appendix Table A1, online only). High expressers had a lower CR rate (74% v 92%) compared with low expressers (P = .004). In a multivariable analysis, high ERG expression status independently predicted failure to achieve CR (P = .004), after adjusting for BAALC expression (P = .04) and age (P = .008). High expressers had 5.7 times the odds of not achieving CR compared with the low expressers (Table 2).
With a median follow-up of 5.8 years for patients who did not have an event (range, 1.7 to 8.9 years), high expressers also had shorter EFS (P < .001) than low expressers. In a multivariable analysis, high ERG expression independently predicted worse EFS (P = .004) after adjusting for FLT3-ITD status (P < .001; Table 2). When we analyzed patients who received APSCT after achievement of CR, there was also a significant difference in EFS duration (P < .0001) between the low and high ERG expresser groups, and the expected 2-year EFS rates were 71% and 34%, respectively. When stratifying by FLT3-ITD status, we observed no difference in EFS among FLT3-ITD–positive patients according to ERG expression levels (P = .31), whereas in the FLT3-ITD–negative subset, high ERG expressers had a worse EFS than low expressers (median, 1.3 years v not reached; P < .001), with estimated 2-year EFS rates of 35% and 65%, respectively (Fig 2A and 2B). In a multivariable analysis limited to the FLT3-ITD–negative subset, high ERG expression was the only factor that predicted for worse EFS (P = .001) and conferred a 2.8 times higher risk of an event compared with low ERG expression (Table 2).
NPM1 mutations were present in 68% of the patients and were distributed similarly among the high and low ERG groups (73% v 67%; P = .46). Within the subset of NPM1 mutated (NPM1 positive) patients, high ERG expressers had a worse EFS than low expressers (median, 0.7 years v not reached; P < .001), with estimated 2-year EFS rates of 25% and 55%, respectively (Fig 2C). In contrast, the outcome of both ERG expression groups with NPM1-negative status was similarly poor (P = .29; Fig 2D), but the number of patients in this group was too small to draw definitive conclusions (Appendix Fig A1, online only). In a multivariable analysis of NPM1-positive patients, ERG expression levels independently predicted EFS duration (P = .003), after adjusting for FLT3-ITD status (P < .001; Table 2). It has been shown previously that being both NPM1 positive and FLT3-ITD negative constitutes an independent favorable prognostic factor in CN-AML. Several groups have found that FLT3-ITD–negative/NPM1-positive patients have a significantly better outcome than the FLT3-ITD–negative/NPM1-negative and FLT3-ITD-positive patients.3,4,8 Thus, we tested whether determining ERG expression levels could improve the previously reported prognostic stratification based on the combined NPM1 mutation and FLT3-ITD status.3,4,8 We found that in the FLT3-ITD–negative/NPM1-positive subset, high ERG expressers had worse EFS (median, 1.6 years v not reached; P < .001) and DFS (median, 1.7 years v not reached; P = .003) than low expressers, with estimated 2-year EFS rates of 36% and 72% and DFS rates of 42% and 74%, respectively (Fig 3). In contrast, ERG expression had no significant prognostic impact on the FLT3-ITD–negative/NPM1-negative subset (data not shown). Among FLT3-ITD–negative/NPM1-positive patients, ERG status was the only independent predictor for EFS (P = .002; Table 2) and DFS (P = .003). In this subset, high ERG expressers had a 3.8 times higher risk of an adverse event (ie, failure to achieve CR, relapse, or death) than low expressers (Table 2).
We report here the first independent validation of ERG overexpression as a negative prognostic marker in CN-AML. We have been able to show that high ERG expression predicts failure to achieve CR and shorter EFS in independent cohorts of patients treated on two separate, albeit similar, treatment protocols. These results support ERG expression as an additional and valuable predictor for clinical outcome in patients with CN-AML, and are consistent with reports in other types of cancer (ie, T-cell acute lymphoblastic leukemia and prostate cancer), showing that high levels of ERG expression identify aggressive malignant phenotypes.16,17 To take full advantage of the predictive value of ERG expression, future larger studies in these different diseases will be necessary to establish a standardized method of quantification and define absolute cut points that would allow prospective classification of patients as high or low ERG expressers. Given that two or more genetic alterations can be present simultaneously in CN-AML patients, part of the validation process of a newly discovered predictor is to prove that it can maintain an independent prognostic value when assessed against other molecular markers. One of the limitations of this type of approach, however, often lies in the small number of patients available for subset analyses. Therefore, we combined patients with CN-AML enrolled onto CALGB 9621 and 19808 treatment protocols and investigated the prognostic value of ERG expression once FLT3-ITD, MLL-PTD, and NPM1 mutations and high BAALC expression were taken into account (Fig 4). Using this approach, our data support the existence of a complex prognostic network constituted by these markers. First, we observed that levels of ERG expression affected the outcome of only FLT3-ITD–negative patients, but not that of FLT3-ITD–positive patients. This is consistent with the notion that patients harboring FLT3-ITD have a dismal prognosis and are relatively unaffected by the presence or absence of other prognostic markers. In contrast, FLT3-ITD–negative patients seem to constitute a more heterogeneous prognostic group, consisting of subsets with different outcomes that can be identified on additional molecular characterization. Second, we showed that expression levels of ERG also influenced the outcome of the NPM1-positive subset only, whereas the prognosis of the NPM1-negative subset seemed not to be affected, although too few patients were analyzed to draw definitive conclusions (Fig A1). Finally, when we tested whether ERG expression could improve the previously proposed molecular risk classification of CN-AML based on a combined FLT3-ITD and NPM1 mutation status,3,4,8 we identified a subset of patients (ie, FLT3-ITD negative/NPM1 positive/low ERG) with a favorable prognosis, with an estimated 2-year EFS rate exceeding 70% (Fig A1).
Our data reinforce the view that from a molecular standpoint, CN-AML is a heterogeneous disease. We have proposed previously a prioritized schema that stratifies patients with CN-AML to risk-adapted therapies based on molecular markers.1 In light of the information derived from the current work, we can now integrate ERG expression levels into this schema. Our data indicate that FLT3-ITD–negative/NPM1-positive patients, considered to have a favorable prognosis,3,4,8 can be separated into prognostic subsets based on expression levels of ERG. FLT3-ITD–negative/NPM1-positive/low ERG patients are likely to benefit from high-dose cytarabine and APSCT-based therapies as administered in CALGB 9621 and 19808. In contrast, FLT3-ITD–negative/NPM1-negative/high ERG patients perhaps should be considered for alternative and/or more aggressive treatments, given that their estimated 2-year EFS is less than 40% (Fig A1). Likewise, CN-AML patients with FLT3-ITD who have a poor prognosis should be considered for clinical trials investigating compounds that directly inhibit FLT3 tyrosine kinase activity or for allogeneic SCT, although a definitive role for the latter in first CR remains to be established.23,24
The author(s) indicated no potential conflicts of interest.
Conception and design: Guido Marcucci, Clara D. Bloomfield Financial support: Clara D. Bloomfield Administrative support: Guido Marcucci, Richard A. Larson, Clara D. Bloomfield Provision of study materials or patients: Bayard L. Powell, Jonathan E. Kolitz, Richard A. Larson, Clara D. Bloomfield Collection and assembly of data: Kati Maharry, Susan P. Whitman, Tamara Vukosavljevic, Peter Paschka, Christian Langer, Claudia D. Baldus, Andrew J. Carroll Data analysis and interpretation: Guido Marcucci, Kati Maharry, Susan P. Whitman, Krzysztof Mrózek, Clara D. Bloomfield Manuscript writing: Guido Marcucci, Krzysztof Mrózek, Clara D. Bloomfield Final approval of manuscript: Guido Marcucci, Kati Maharry, Susan P. Whitman, Tamara Vukosavljevic, Peter Paschka, Christian Langer, Krzysztof Mrózek, Claudia D. Baldus, Andrew J. Carroll, Bayard L. Powell, Jonathan E. Kolitz, Richard A. Larson, Clara D. Bloomfield
published online ahead of print at www.jco.org on June 18, 2007. Supported by National Cancer Institute (Bethesda, MD) Grants No. CA101140, CA77658, CA102031, CA31946, CA09512, CA16058, CA98933, CA90469, CA96887, and CA089341, and the Coleman Leukemia Research Foundation. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Mrózek K, Marcucci G, Paschka P, et al: Clinical relevance of mutations and gene-expression changes in adult acute myeloid leukemia with normal cytogenetics: Are we ready for a prognostically prioritized molecular classification? Blood 109:431-448, 2007 2. Fröhling S, Schlenk RF, Breitruck J, et al: Prognostic significance of activating FLT3 mutations in younger adults (16 to 60 years) with acute myeloid leukemia and normal cytogenetics: A study of the AML Study Group Ulm. Blood 100:4372-4380, 2002 3. Döhner K, Schlenk RF, Habdank M, et al: Mutant nucleophosmin (NPM1) predicts favorable prognosis in younger adults with acute myeloid leukemia and normal cytogenetics: Interaction with other gene mutations. Blood 106:3740-3746, 2005 4. Thiede C, Koch S, Creutzig E, et al: Prevalence and prognostic impact of NPM1 mutations in 1485 adult patients with acute myeloid leukemia (AML). Blood 107:4011-4020, 2006 5. Fröhling S, Schlenk RF, Stolze I, et al: CEBPA mutations in younger adults with acute myeloid leukemia and normal cytogenetics: Prognostic relevance and analysis of cooperating mutations. J Clin Oncol 22:624-633, 2004 6. Baldus CD, Tanner SM, Ruppert AS, et al: BAALC expression predicts clinical outcome of de novo acute myeloid leukemia patients with normal cytogenetics: A Cancer and Leukemia Group B study. Blood 102:1613-1618, 2003 7. Whitman SP, Liu S, Vukosavljevic T, et al: The MLL partial tandem duplication: Evidence for recessive gain-of-function in acute myeloid leukemia identifies a novel patient subgroup for molecular-targeted therapy. Blood 106:345-352, 2005 8. Schnittger S, Schoch C, Kern W, et al: Nucleophosmin gene mutations are predictors of favorable prognosis in acute myelogenous leukemia with a normal karyotype. Blood 106:3733-3739, 2005 9. Oikawa T: ETS transcription factors: Possible targets for cancer therapy. Cancer Sci 95:626-633, 2004[CrossRef][Medline] 10. Ichikawa H, Shimizu K, Hayashi Y, et al: An RNA-binding protein gene, TLS/FUS, is fused to ERG in human myeloid leukemia with t(16;21) chromosomal translocation. Cancer Res 54:2865-2868, 1994 11. Hart AH, Corrick CM, Tymms MJ, et al: Human ERG is a proto-oncogene with mitogenic and transforming activity. Oncogene 10:1423-1430, 1995[Medline] 12. Sorensen PHB, Lessnick SL, Lopez-Terrada D, et al: A second Ewing's sarcoma translocation, t(21;22), fuses the EWS gene to another ETS-family transcription factor, ERG. Nat Genet 6:146-151, 1994[CrossRef][Medline] 13. Tomlins SA, Rhodes DR, Perner S, et al: Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science 310:644-648, 2005 14. Baldus CD, Liyanarachchi S, Mrózek K, et al: Acute myeloid leukemia with complex karyotypes and abnormal chromosome 21: Amplification discloses overexpression of APP, ETS2, and ERG genes. Proc Natl Acad Sci U S A 101:3915-3920, 2004 15. Marcucci G, Baldus CD, Ruppert AS, et al: Overexpression of the ETS-related gene, ERG, predicts a worse outcome in acute myeloid leukemia with normal karyotype: A Cancer and Leukemia Group B study. J Clin Oncol 23:9234-9242, 2005 16. Baldus CD, Burmeister T, Martus P, et al: High expression of the ETS transcription factor ERG predicts adverse outcome in acute T-lymphoblastic leukemia in adults. J Clin Oncol 24:4714-4720, 2006 17. Wang J, Cai Y, Ren C, et al: Expression of variant TMPRSS2/ERG fusion messenger RNAs is associated with aggressive prostate cancer. Cancer Res 66:8347-8351, 2006 18. Kolitz JE, George SL, Marcucci G, et al: A randomized comparison of induction therapy for untreated acute myeloid leukemia (AML) in patients < 60 years using P-glycoprotein (Pgp) modulation with Valspodar (PSC833): Preliminary results of Cancer and Leukemia Group B study 19808. Blood 106:122a-123a, 2005 (abstr 407) 19. Byrd JC, Mrózek K, Dodge RK, et al: Pretreatment cytogenetic abnormalities are predictive of induction success, cumulative incidence of relapse, and overall survival in adult patients with de novo acute myeloid leukemia: Results from Cancer and Leukemia Group B (CALGB 8461). Blood 100:4325-4336, 2002 20. Thiede C, Steudel C, Mohr B, et al: Analysis of FLT3-activating mutations in 979 patients with acute myelogenous leukemia: Association with FAB subtypes and identification of subgroups with poor prognosis. Blood 99:4326-4335, 2002 21. Kolitz JE, George SL, Dodge RK, et al: Dose escalation studies of cytarabine, daunorubicin, and etoposide with and without multidrug resistance modulation with PSC-833 in untreated adults with acute myeloid leukemia younger than 60 years: Final induction results of Cancer and Leukemia Group B study 9621. J Clin Oncol 22:4290-4301, 2004 22. Klein JP, Moeschberger ML: Survival Analysis: Techniques for Censored and Truncated Data. New York, NY, Springer-Verlag, 1997 23. Tallman MS: New agents for the treatment of acute myeloid leukemia. Best Pract Res Clin Haematol 19:311-320, 2006[Medline] 24. Gale RE, Hills R, Kottaridis PD, et al: No evidence that FLT3 status should be considered as an indicator for transplantation in acute myeloid leukemia (AML): An analysis of 1135 patients, excluding acute promyelocytic leukemia, from the UK MRC AML10 and 12 trials. Blood 106:3658-3665, 2005 Submitted January 23, 2007; accepted May 1, 2007.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
1,500/µL and platelets 
= .20 from the univariable models. Variables remaining in the final model for CR were significant at 
