SEARCH

SEARCH BY CITATION

Keywords:

  • oligonucleotide microarrays;
  • immunophenotyping;
  • acute myeloid leukemia;
  • acute lymphoblastic leukemia;
  • gene expression;
  • protein expression

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. LITERATURE CITED

Background

Flow cytometry (FC) is a standard method for diagnosing and subclassifying acute myeloid (AML) and acute lymphoblastic (ALL) leukemias and allows the analysis of cell surface and intracellular proteins. In the future, diagnostic procedures may include oligonucleotide microarray analysis (MA) to detect expression patterns of large numbers of specific genes.

Methods

For comparison between methods, we performed FC and MA by using the Affymetrix GeneChip HG-U133A microarray in parallel and correlated protein expression levels and mRNA abundance of 39 relevant genes in 113 patients with newly diagnosed AML and ALL and four normal bone marrow samples.

Results

In 1,512 of 2,187 (69.1%) comparisons between methods, congruent results were obtained with regard to positivity or negativity of expression, respectively. Specifically, there was a significant correlation between protein expression and mRNA abundance for genes essential for diagnosing and subclassifying AML and ALL with regard to positivity and expression.

Conclusions

These data suggest that protein expression is highly correlated to mRNA abundance in AML and ALL. Further, expression patterns of specific genes provide important information at diagnosis for patients with AML and ALL that may be used for the discrimination from other leukemias. Cytometry Part B (Clin. Cytometry) 55B:29–36, 2003. © 2003 Wiley-Liss, Inc.

The determination of the surface and cytoplasmic expression of characteristic proteins by flow cytometry (FC) is a common method for the diagnosis and subclassification of acute myeloid (AML) and acute lymphoblastic (ALL) leukemias (1, 2). The oligonucleotide microarray analysis (MA) represents a novel technology for the simultaneous detection of the mRNA abundance of large numbers of genes (3, 4). Based on specific gene-expression patterns, distinct disease entities have been identified (5–8). Therefore, MA may gain significant importance for diagnosing acute leukemias in the near future (9, 10). However, data on the correlation between protein expression levels and mRNA abundance are limited (11–14). To analyze the relation of protein expression and mRNA abundance in AML and ALL, we performed 2,187 individual comparisons of 39 genes in 113 patients with AML and ALL at diagnosis and in four normal bone marrow samples analyzed by FC and MA in parallel (15).

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. LITERATURE CITED

Samples

Fresh bone marrow samples from thoroughly characterized patients with newly diagnosed and untreated AML and ALL as defined by the World Health Organization classification (16, 17) and from healthy volunteers were used and processed immediately. Samples were analyzed by cytomorphology, cytochemistry, immunophenotyping, cytogenetics, and molecular genetics in all cases and were characterized by specific chromosomal aberrations or molecular genetic alterations (19, 17) (Table 1). The studies were done in accordance with the rules of the local internal review board and the tenets of the revised Helsinki protocol.

Table 1. Patient Characteristics
Parametern
Acute myeloid leukemia (total)85
 t(8;21)5
 t(15;17)5
 inv(16)4
 Trisomy 82
 Normal karyotype23
 Normal karyotype, FLT3 length mutation18
 Normal karyotype, MLL internal tandem duplication13
 Complex aberrations11
 Other abnormalities4
Acute lymphoblastic leukemia (total)28
 t(9;22)14
 Other B-precursor type9
 T-precursor type5
Normal bone marrow4
% Blasts in bone marrow samples, median (range)90 (10–100)

Flow Cytometry

The studies were performed on cells isolated from bone marrow by Ficoll-Hypaque density gradient centrifugation, as described previously (18). Triple stainings, isotype controls, and monoclonal antibodies against 39 antigens were used in the following combinations as designed for diagnostic purposes (conjugated with the fluorochromes fluorescein isothiocyanate, phycoerythrin, and phycoerythrin cyanine 5 (PC-5), respectively): CD34/CD2/CD33, CD7/CD33/CD34, CD34/CD56/CD33, CD11b/CD117/CD34, CD64/CD4/CD45, CD15/CD13/CD33, HLA-DR/CD33/CD34, CD65/CD87/CD34, CD34/CD135/CD33, CD34/CD116/CD33, CD34/NG2/CD33, CD38/CD133/CD34, CD90/CD117/CD34, CD61/CD14/CD41, CD36/CD235a/CD45, CD9/CD33/CD34, CD97/CD33/CD34, CD34/CD10/CD19, CD5/CD19/CD20, CD2/CD1a/CD3, CD3/CD4/CD8, MPO/LF/cyCD15, TdT/cyCD22/cyCD3, and TdT/cyCD79a/cyCD3. All antibodies were purchased from Immunotech (Marseilles, France), except for CD64 and CD15 (Medarex, Annandale, NJ); CD133 (Milteny Biotech, Bergisch Gladbach, Germany), and MPO and LF (Caltag, Burlingame, CA). The respective combinations of antibodies were added to 1 × 106 cells (volume, 100 μl) and incubated for 10 min at room temperature. The samples were then washed twice in phosphate buffered saline and resuspended in 0.5 ml of phosphate buffered saline. For analysis of cytoplasmic antigens, cells were fixed and permeabilized before staining with Fix & Perm (Caltag). FC analysis was performed on a FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA). Analysis of list-mode files was performed by means of the CellQuest Pro software (Becton Dickinson). To obtain analyses on the same cells for FC and MA, i.e., all nucleated cells of each sample, the analysis gate was set in a forward versus side scatter plot and included lymphocyte, blast, monocyte, and granulocyte populations. Antigen expression was rated positive at a cutoff level of 20% of the cells within the mononuclear gate for membrane proteins and at a cutoff level of 10% for cytoplasmic antigens as compared with isotype controls (in patients with analysis of isotype controls). Mean fluorescence intensity values were calculated for all events, with fluorescence values higher than those of isotype controls.

Microarray Experiments

For MA the GeneChip System (Affymetrix, Santa Clara, CA) was used. The targets for GeneChip analyses were prepared according to the current Expression Analysis Technical Manual. Briefly, mononuclear cells were purified from bone marrow aspirates by Ficoll-Hypaque density gradient centrifugation (18). Subsequently, 1 × 107 cells were lysed in 300 μl of RLT lysis buffer (RNeasy Mini Kit, Qiagen, Hilden, Germany). The lysates were stored from 1 week to 37 months at −80°C. Before extraction of total RNA (RNeasy Mini Kit), the lysates were homogenized (QIAshredder columns, Qiagen). Approximately 5 μg of total RNA was used as starting material in the subsequent cDNA synthesis using oligo[(dT)24T7promotor]65 primer (cDNA Synthesis System, Roche Applied Science, Mannheim, Germany). The cDNA was purified by phenol:chlorophorm:isoamylalcohol extraction (Ambion, Austin, TX) and acetate/ethanol precipitated overnight. For detection of the hybridized target nucleic acid, biotin-labeled ribonucleotides were incorporated during the in vitro transcription (Enzo BioArray HighYield RNA Transcript Labeling Kit, Enzo, Farmingdale, NY, USA). After quantification of the purified cRNA (RNeasy Mini Kit), 15 μg was fragmented by alkaline treatment (200 mM Tris-acetate, pH 8.2; 500 mM potassium acetate, 150 mM magnesium acetate) and added to the hybridization cocktail sufficient for five hybridizations on standard-format GeneChip microarrays. Before hybridization to U133A, Test3 microarrays (Affymetrix) were chosen, in some cases, for monitoring of labeling efficiency and the integrity of the cRNA. Washing and staining of the probe arrays were performed according to current protocols (Micro_1v1, EukGE-WS2v4). The Affymetrix software (Microarray Suite 5.0) extracted fluorescence intensities from each element on the microarrays as detected by confocal laser scanning according to the manufacturer's recommendations. This software qualitatively rates the mRNA abundance of genes as present, marginal, and absent calls. For comparisons with FC results, present and marginal calls were rated positive and absent calls were rated negative. To compare different experiments, the global microarray intensities were scaled to a common target intensity. Each new human GeneChip expression array features 100 human maintenance genes that serve as a tool to normalize and scale the data before performing data comparisons. These 100 probe sets were used for normalization, as recommended by the manufacturer (U133A mask file, http://www.affymetrix.com/support/technical/mask_files.affx). Scaling was performed to a target intensity of 5,000. This resulted in the following parameters: background of approximately 70, scaling factor of approximately 1.000, average signal intensity of a present called gene of approximately 1,000, average signal intensity of an absent called gene of approximately 45, and noise of approximately 6. All probe sets representing genes of interest for this study were functionally annotated by the NetAffx database (Affymetrix), and HGNC (HUGO Gene Nomenclature Committee) approved gene symbols were proven. In some cases, genes were represented by more than one probe set. If the mean average fluorescence intensity differed between different probe sets for identical genes, the probe set revealing the highest average fluorescence intensity was used for comparative analyses. In general, the results obtained by redundant probe sets were consistent with regard to positivity and negativity.

Statistics

A total of 39 genes were analyzed in 117 samples. The congruence of positivity and negativity of the expression of the respective genes as determined by FC and MA was analyzed for each gene in each patient. Comparisons of microarray intensities were performed by the Mann-Whitney U test. Analyses for bivariate correlations of mRNA and protein expression levels were performed by Pearson's correlation coefficient using SPSS 11.0.1.

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. LITERATURE CITED

One hundred seventeen samples were analyzed in parallel by FC and MA for the expression of 39 genes (Table 1). Eighty-five were AML, 28 were ALL, and four were normal bone marrow samples. A total of 2,187 comparisons of individual expression data obtained by both methods was analyzed with regard to positivity. Of these, 1,512 (69.1%) revealed congruent results for positivity of protein expression and mRNA abundance (881 cases, 40.3% with positive expression and 631 cases, 28.9% with negative expression, respectively; Table 2.) In 509 comparisons (23.3%), MA detected positivity for mRNA expression (call: present), whereas the results of FC indicated negativity. In 166 cases (7.6%), protein expression was demonstrated by FC, whereas no mRNA expression was detected by MA (call: absent). The incongruent cases were not due to differences of cells analyzed because both methods were applied to the same samples of cells obtained by Ficoll-Hypaque density gradient centrifugation.

Table 2. Comparisons of Protein Expression and mRNA Abundance in Acute Leukemia as Assessed by Flow Cytometry and Microarray Analysisa
AntigenNo. of comparisonsFC and MA positiveFC and MA negativeMA positive and FC negativeFC positive and MA negativeCorrelation of mean fluorescence intensity by flow cytometry and average fluorescence intensity by microarray analysis
Coefficient of correlation (Spearman)Significance of correlation (Spearman)
  • a

    Protein expression level and mRNA abundance of 39 markers were compared in 117 patients. “No. of comparisons” indicates the number of patients analyzed for the respective antigens (maximum number, 117 patients). Mean fluorescence intensity values obtained by flow cytometry were calculated for all events with fluorescence values higher than those of isotypic controls using the CellQuest Pro software (Becton Dickinson). Average fluorescence intensity values obtained by microarray analyses were calculated by Affymetrix software (Microarray Suite 5.0). FC, flow cytometry; MA, microarray.

CD10331712040.6080.000
CD116652903510.1590.166
CD1177328232110.3780.000
CD11b613322600.6070.000
CD13755851110.5900.000
CD133452113101−0.1350.272
CD135654222100.1020.367
CD147217272800.5850.000
CD1568305312−0.0140.888
CD19742743310.1060.253
CD1a23016700.2080.319
CD2731395100.1090.248
CD2030201−0.5000.391
CD22632026017−0.3790.007
CD235a614243030.0920.358
CD337543142160.4940.000
CD34724591710.5450.000
CD36562791820.7030.000
CD3860521160.6120.000
CD3e69192011190.1190.296
CD4671037812−0.1960.040
CD41500500.2980.050
CD4571710000.2760.004
CD520101−0.1000.873
CD56730601120.1260.180
CD6163056070.0410.679
CD646817163320.5240.000
CD7721254060.1440.132
CD79a2261015−0.1960.094
CD87661515000.4220.028
CD8a503200.1820.572
CD940265630.3690.002
CD9060342780.0290.842
CD973835021−0.2530.041
HLA-DR73631900.3220.059
Laktoferrin6315143130.2090.027
MPO744322900.0570.558
NG270163150.3590.044
TdT6939132150.0260.790
Total2187 (100.0%)881 (40.3%)631 (28.9%)509 (23.3%)166 (7.6%)  
  1512 (69.1%) congruent675 (30.9%) not congruent  

Although the congruence observed in 69.1% of the comparisons might have been anticipated, the incongruent cases need specific considerations. In general, positive results in MA and negative results in FC most likely indicate that the abundance of the respective mRNA is not sufficient to result in positivity for protein expression as defined in these analyses, i.e., detection of protein by FC in at least 20% (≥10% for cytoplasmic antigens) of gated cells with limit for positivity set by 99% of cells analyzed as isotype controls. It must be taken into consideration that the analytic strategy applied for FC included gating on all cells, and a better sensitivity is anticipated for gating on blasts or specific subpopulations only, which is not possible for MA. However, this is not within the scope of the present analyses. Conversely, positivity by FC and negativity by MA may be due to nonspecific binding of antibodies in FC or to lack of sensitivity of MA, which could not be further substantiated in the present setting. In addition, it is obvious that the results that were positive in FC and negative in MA mainly involved lymphoid-associated markers, whereas the results that were positive in MA and negative in FC typically involved myeloid-associated antigens. Thus, in the latter cases, MA might be detecting mRNA from residual normal cells, which account for fewer than 10% to 20% of all cellularity in the sample, and might be related to the lower sensitivity of FC, which is, at least in part, limited in the present analyses by the placement of arbitrary cutoff levels for positivity. This is not in accordance with the consensus for diagnostic purposes (19), but this strategy provided the only way to directly compare FC and MA results with regard to positivity and negativity. These shortcomings must be taken into account when speculating on a complementary role for MA in addition to FC for diagnosing acute leukemias. However, MA might detect mRNA of disease-specific genes when FC produces negative results and thus may optimize diagnostic procedures. To substantiate the results in this regard and to determine the degree of correlation between protein expression and mRNA abundance within leukemia subtypes, i.e., AML, B-precursor ALL, and T-precursor ALL, these analyses also were performed separately (Tables 3–8).

Table 3. Comparisons of Protein Expression and mRNA Abundance in Acute Myeloid Leukemia as Assessed by Flow Cytometry and Microarray Analysisa
AntigenNo. of comparisonsFC and MA positiveFC and MA negativeMA positive and FC negativeFC positive and MA negative
  • a

    Protein expression level and mRNA abundance of 39 markers were compared in 85 patients. “No. of comparisons” indicates the number of patients analyzed for the respective antigens (maximum number, 85 patients). Mean fluorescence intensity values obtained by flow cytometry were calculated for all events with fluorescence values higher than those of isotypic controls using the CellQuest Pro software (Becton Dickinson). Average fluorescence intensity values obtained by microarray analyses were calculated by Affymetrix software (Microarray Suite 5.0). FC, flow cytometry; MA, microarray.

CD10123801
CD11646250210
CD11748288111
CD11b41231170
CD135044051
CD1333015951
CD13545291150
CD14491517170
CD1548261201
CD195064031
CD1a80620
CD24996340
CD2000000
CD2243524014
CD235a39414201
CD3350362111
CD3447249131
CD3639224130
CD384135015
CD3e471114715
CD447102269
CD4150050
CD454848000
CD500000
CD5649038110
CD614203804
CD6449157252
CD748103503
CD79a1301012
CD8747130340
CD8a30300
CD92714562
CD904122964
CD972523020
HLA-DR4839180
Laktoferrin441212173
MPO50380120
NG24604114
TdT451813113
Total1459 (100.0%)602 (41.3%)409 (28.0%)329 (22.5%)119 (8.2%)
  1011 (69.3%) congruent448 (30.7%) not congruent
Table 4. Congruence of Results of Flow Cytometry and Microarray Analysis for Genes Essential for Diagnosing Acute Myeloid Leukemias
AntigenCongruent assignment
MPO76%
CD1388%
CD3376%
Table 5. Comparisons of Protein Expression and mRNA Abundance in B-Precursor Acute Lymphoblastic Leukemia as Assessed by Flow Cytometry and Microarray Analysisa
AntigenNo. of comparisonsFC and MA positiveFC and MA negativeMA positive and FC negativeFC positive and MA negative
  • a

    Protein expression level and mRNA abundance of 39 markers were compared in 23 patients. “No. of comparisons” indicates the number of patients analyzed for the respective antigens (maximum number, 23 patients). Mean fluorescence intensity values obtained by flow cytometry were calculated for all events with fluorescence values higher than those of isotypic controls using the CellQuest Pro software (Becton Dickinson). Average fluorescence intensity values obtained by microarray analyses were calculated by Affymetrix software (Microarray Suite 5.0). FC, flow cytometry; MA, microarray.

CD101813203
CD1161740121
CD1172101560
CD11b188190
CD132111550
CD133134450
CD1351811160
CD141918100
CD151742101
CD192020000
CD1a130850
CD22123160
CD2020101
CD221915202
CD235a180882
CD332161104
CD342118030
CD36155541
CD381715101
CD3e195644
CD41701421
CD4100000
CD451919000
CD510100
CD562002000
CD611701601
CD64172870
CD72001901
CD79a86002
CD871721140
CD8a00000
CD91110001
CD901711213
CD971110001
HLA-DR2121000
Laktoferrin1732120
MPO2041150
NG22011801
TdT2118012
Total622 (100%)239 (38.4%)195 (31.4%)155 (24.9%)33 (5.3%)
  434 (69.8%) congruent188 (30.2%) not congruent
Table 6. Congruence of Results of Flow Cytometry and Microarray Analysis for Genes Essential for Diagnosing B-Precursor Acute Lymphoblastic Leukemias
AntigenCongruent assignment
CD2289%
CD79a75%
CD19100%
CD1083%
TdT86%
Table 7. Comparisons of Protein Expression and mRNA Abundance in T-Precursor Acute Lymphoblastic Leukemia as Assessed by Flow Cytometry and Microarray Analysisa
AntigenNo. of comparisonsFC and MA positiveFC and MA negativeMA positive and FC negativeFC positive and MA negative
  • a

    Protein expression level and mRNA abundance of 39 markers were compared in 5 patients. “No. of comparisons” indicates the number of patients analyzed for the respective antigens (maximum number, 5 patients). Mean fluorescence intensity values obtained by flow cytometry were calculated for all events with fluorescence values higher than those of isotypic controls using the CellQuest Pro software (Becton Dickinson). Average fluorescence intensity values obtained by microarray analyses were calculated by Affymetrix software (Microarray Suite 5.0). FC, flow cytometry; MA, microarray.

CD1021100
CD11620020
CD11730030
CD11b22000
CD1333000
CD13322000
CD13522000
CD1430210
CD1530210
CD1930300
CD1a20200
CD232010
CD2010100
CD2210001
CD235a30210
CD3331101
CD3433000
CD3620011
CD3822000
CD3e33000
CD430102
CD4100000
CD4533000
CD510001
CD5630201
CD6130102
CD6420110
CD731002
CD79a10001
CD8720020
CD8a20020
CD922000
CD9020101
CD9722000
HLA-DR32010
Laktoferrin20020
MPO30120
NG230300
TdT33000
Total91 (100%)34 (37.4%)24 (26.4%)20 (22.0%)13 (14.3%)
  58 (63.8%) congruent33 (36.3%) not congruent
Table 8. Congruence of Results of Flow Cytometry and Microarray Analysis for Genes Essential for Diagnosing T-Precursor Acute Lymphoblastic Leukemias
AntigenCongruent assignment
CD3100%
TdT100%

Thus, when focusing on the genes most specific for the diagnosis of AML, i.e., myeloperoxidase, CD13, and CD33, a strong correlation between protein expression and mRNA abundance was observed (Tables 3 and 4; congruence in 76%, 88%, and 76%, respectively). However, these three genes were rated positive by MA and negative by FC in 24%, 10%, and 2%, respectively. This result is in accordance with published data on the lack of expression of disease-specific antigens as analyzed by FC in 25% to 30% of AML cases (20). These cases in fact suggest that MA may add to diagnostic certainty. In only 1%, 2%, and 12%, respectively, myeloperoxidase, CD13, and CD33 were rated positive by FC and negative by MA. Further, the data were similar for most other AML-specific antigens and for antigens necessary to subclassify AML. Thus, the percentages of congruent cases, MA-positive and FC-negative cases, and FC-positive and MA-negative cases were 75%, 23%, and 2% for CD117; 59%, 41%, and 0% for CD11b; 80%, 17%, and 3% for CD133; 65%, 35%, and 0% for CD14; 56%, 42%, and 2% for CD15; 46%, 51%, and 3% for CD235a; and 67%, 33%, and 0% for CD36.

Turning to B-precursor ALL, the overall percentage of congruent cases was 69.8% (Table 5). It was even higher for the antigens most relevant for establishing the diagnosis and for subclassification, i.e., CD22, CD79a, CD19, CD10, and TdT (congruence in 75% to 100%; Table 6). Similar data were obtained for T-precursor ALL, although the total number of patients analyzed was relatively small (Tables 7 and 8).

Most importantly, the strong correlations between protein expression and mRNA abundance were not limited to congruence in positivity but were demonstrated also quantitatively. To confirm this observation, the protein expression levels and mRNA abundance were compared by Pearson's correlation coefficient (Table 2). These comparisons showed significant, although in many cases rather low, correlations for the fluorescence intensities as assessed by FC and MA for many of the analyzed genes and emphasizes the high coherence of expression patterns for protein and mRNA (Fig. 1). The relatively low correlations in some genes may have been due mainly to an overall low abundance of mRNA and low protein expression in the respective cases. Further, the quantitative analysis of antigen expression only in cells rated positive based on isotype controls may have added some variability to the results. However, even when considering these restrictions, the present results support the overall correlation between methods.

thumbnail image

Figure 1. Correlations between protein expression levels and mRNA abundance. Expression levels were compared by Pearson's correlation coefficient (Table 2). Mean fluorescence intensity values obtained by flow cytometry were calculated for all events, with fluorescence values higher than those of isotype controls, by using the CellQuest Pro software (Becton Dickinson). Average fluorescence intensity values obtained by micorarray analyses were calculated by Affymetrix software, Microarray Suite 5.0.

Download figure to PowerPoint

Overall, these results demonstrate for the first time that there is a significant correlation between protein expression and gene expression in AML and ALL, and that the antigens identified thus far as essential for the diagnosis and subclassification of AML and ALL by FC may represent additional candidate genes when using MA as a diagnostic tool for molecular cancer class prediction (21, 22). Further, it is anticipated that the present analyses represent a prime example and will be reproduced for a variety of other entities such as lymphoid malignancies. Due to their high potential to assess the expression patterns of large numbers of genes and due to their excellent reproducibility features, microarrays may be a promising future diagnostic tool (8) in combination with current standard methods that, which must be emphasized, have the advantage of separate evaluations of different subpopulations by cytomorphology and multiparameter immunophenotyping. Further studies are needed to comprehensively prove the utility of MA in the routine clinical setting.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. LITERATURE CITED
  • 1
    Czuczman MS, Dodge RK, Stewart CC, Frankel SR, Davey FR, Powell BL, Szatrowski TP, Schiffer CA, Larson RA, Bloomfield CD. Value of immunophenotype in intensively treated adult acute lymphoblastic leukemia: cancer and leukemia Group B study 8364. Blood 1999; 93: 39313939.
  • 2
    Weir EG, Borowitz MJ. Flow cytometry in the diagnosis of acute leukemia. Semin Hematol 2001; 38: 124138.
  • 3
    Kohlmann A, Dugas M, Schoch C, Schnittger S, Mergenthaler S, Kern W, Haferlach T, Hiddemann W. Gene expression profiles of distinct AML subtypes in comparison to normal bone marrow. Blood 2001; 98: 91a.
  • 4
    Liotta L, Petricoin E. Molecular profiling of human cancer. Nat Rev Genet 2000; 1: 4856.
  • 5
    Alizadeh AA, Eisen MB, Davis RE, Ma C, Lossos IS, Rosenwald A, Boldrick JC, Sabet H, Tran T, Yu X, Powell JI, Yang L, Marti GE, Moore T, Hudson J Jr, Lu L, Lewis DB, Tibshirani R, Sherlock G, Chan WC, Greiner TC, Weisenburger DD, Armitage JO, Warnke R, Staudt LM. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000; 403: 503511.
  • 6
    Hedenfalk I, Duggan D, Chen Y, Radmacher M, Bittner M, Simon R, Meltzer P, Gusterson B, Esteller M, Kallioniemi OP, Wilfond B, Borg A, Trent J. Gene-expression profiles in hereditary breast cancer. N Engl J Med 2001; 344: 539548.
  • 7
    Khan J, Wei JS, Ringner M, Saal LH, Ladanyi M, Westermann F, Berthold F, Schwab M, Antonescu CR, Peterson C, Meltzer PS. Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks. Nat Med 2001; 7: 673679.
  • 8
    Yeoh EJ, Ross ME, Shurtleff SA, Williams WK, Patel D, Mahfouz R, Behm FG, Raimondi SC, Relling MV, Patel A, Cheng C, Campana D, Wilkins D, Zhou X, Li J, Liu H, Pui CH, Evans WE, Naeve C, Wong L, Downing JR. Classification, subtype discovery, and prediction of outcome in pediatric acute lymphoblastic leukemia by gene expression profiling. Cancer Cell 2002; 1: 133143.
  • 9
    Armstrong SA, Staunton JE, Silverman LB, Pieters R, den Boer ML, Minden MD, Sallan SE, Lander ES, Golub TR, Korsmeyer SJ. MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia. Nat Genet 2002; 30: 4147.
  • 10
    Schoch C, Kohlmann A, Schnittger S, Bros R, Dugas M, Mergenthaler S, Kern W, Hiddemann W, Eils R, Haferlach T. Acute myeloid leukemias with reciprocal rearrangements can be distinguished by specific gene expression profiles. Proc Natl Acad Sci USA 2002; 99: 1000810013.
  • 11
    Ahram M, Best CJ, Flaig MJ, Gillespie JW, Leiva IM, Chuaqui RF, Zhou G, Shu H, Duray PH, Linehan WM, Raffeld M, Ornstein DK, Zhao Y, Petricoin EF, III, Emmert-Buck MR. Proteomic analysis of human prostate cancer. Mol Carcinog 2002; 33: 915.
  • 12
    Falini B, Mason DY. Proteins encoded by genes involved in chromosomal alterations in lymphoma and leukemia: clinical value of their detection by immunocytochemistry. Blood 2002; 99: 409426.
  • 13
    Serrano J, Roman J, Jimenez A, Castillejo JA, Navarro JA, Sanchez J, Garcia-Castellanos JM, Martin C, Maldonado J, Torres A. Genetic, phenotypic and clinical features of acute lymphoblastic leukemias expressing myeloperoxidase mRNA detected by RT-PCR. Leukemia 1999; 13: 175180.
  • 14
    Zhan F, Hardin J, Kordsmeier B, Bumm K, Zheng M, Tian E, Sanderson R, Yang Y, Wilson C, Zangari M, Anaissie E, Morris C, Muwalla F, van Rhee F, Fassas A, Crowley J, Tricot G, Barlogie B, Shaughnessy J Jr. Global gene expression profiling of multiple myeloma, monoclonal gammopathy of undetermined significance, and normal bone marrow plasma cells. Blood 2002; 99: 17451757.
  • 15
    Kern W, Kohlmann A, Wuchter C, Schnittger S, Schoch C, Mergenthaler S, Ratei R, Ludwig WD, Hiddemann W, Haferlach T. Correlation of gene expression and protein expression in acute myeloid leukemia as assessed by microarray analysis and flow cytometry. Blood 2001; 98: 92a.
  • 16
    Harris NL, Jaffe ES, Diebold J, Flandrin G, Muller Hermelink HK, Vardiman J, Lister TA, Bloomfield CD. World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues: report of the Clinical Advisory Committee meeting, Airlie House, Virginia, November 1997. J Clin Oncol 1999; 17: 38353849.
  • 17
    Jaffe ES, Harris NL, Stein H, Vardiman JW. World Health Organization classification of tumours. Pathology and genetics of tumours of haematopoietic and lymphoid tissues. Lyon: IARC Press; 2001.
  • 18
    Wuchter C, Harbott J, Schoch C, Schnittger S, Borkhardt A, Karawajew L, Ratei R, Ruppert V, Haferlach T, Creutzig U, Dorken B, Ludwig WD. Detection of acute leukemia cells with mixed lineage leukemia (MLL) gene rearrangements by flow cytometry using monoclonal antibody 7.1. Leukemia 2000; 14: 12321238.
  • 19
    Borowitz MJ, Bray R, Gascoyne R, Melnick S, Parker JW, Picker L, Stetler-Stevenson M. U.S.–Canadian consensus recommendations on the immunophenotypic analysis of hematologic neoplasia by flow cytometry: data analysis and interpretation. Cytometry 1997; 30: 236244.
  • 20
    Campana D, Behm FG. Immunophenotyping of leukemia. J Immunol Methods 2000; 243: 5975.
  • 21
    Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M, Mesirov JP, Coller H, Loh ML, Downing JR, Caligiuri MA, Bloomfield CD, Lander ES. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999; 286: 531537.
  • 22
    Ramaswamy S, Tamayo P, Rifkin R, Mukherjee S, Yeang CH, Angelo M, Ladd C, Reich M, Latulippe E, Mesirov JP, Poggio T, Gerald W, Loda M, Lander ES, Golub TR. Multiclass cancer diagnosis using tumor gene expression signatures. Proc Natl Acad Sci USA 2001; 98: 1514915154.