Clinical Chemistry AACC Online Job Center
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 50: 1165-1173, 2004; 10.1373/clinchem.2003.030114
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Saussoy, P.
Right arrow Articles by Latinne, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Saussoy, P.
Right arrow Articles by Latinne, D.
Related Collections
Right arrow Molecular Diagnostics and Genetics
(Clinical Chemistry. 2004;50:1165-1173.)
© 2004 American Association for Clinical Chemistry, Inc.

Molecular Diagnostics and Genetics

Differentiation of Acute Myeloid Leukemia from B- and T-Lineage Acute Lymphoid Leukemias by Real-Time Quantitative Reverse Transcription-PCR of Lineage Marker mRNAs

Pascale Saussoy1,a, Jean-Luc Vaerman1, Nicole Straetmans4, Véronique Deneys1, Guy Cornu2, Augustin Ferrant3 and Dominique Latinne1

Cliniques Universitaires Saint Luc,1 Service de Biologie Hématologique, 2 Service de Pédiatrie, and 3 Service d’Hématologie, Brussels, Belgium. 4 Hôpital de Jolimont, Service d’Hématologie, Haine-Saint-Paul, Belgium.

aAddress correspondence to this author at: Cliniques Universitaires Saint Luc (UCL), Service de Biologie Hématologique, Clos Chapelle-aux-Champs 30-UCL 30.52, 1200 Bruxelles, Belgique. Fax 32-2-762-5855; e-mail Pascale.Saussoy{at}sang.ucl.ac.be.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: Flow cytometry of lineage markers is useful in the classification of leukemias. Our aim was to assess whether the study of lineage genes at the RNA level would enable differentiation of acute myeloid leukemias (AMLs) from B-and T-lineage acute lymphoid leukemias (ALLs).

Methods: We measured mRNA of four lineage markers [CD19, CD79a, CD3e, and myeloperoxidase (MPO)] by reverse transcription followed by real-time quantitative (RTQ)-PCR. We investigated 72 acute leukemias (40 AMLs with 23–93% blast cells plus 27 B-lineage ALLs and 5 T-lineage ALLs) defined by morphologic criteria at diagnosis. RTQ-PCR analysis was performed on bone marrow without cell sorting. The expression of each gene was calculated as the difference in the threshold cycle [{Delta}CT; CT value of target gene minus CT value of housekeeping gene (Abelson)].

Results: Three patterns of expression were detected. In the first, CD19, CD79a, and MPO mRNAs were less abundant than CD3e. In the second pattern, MPO mRNA was more abundant than the other three mRNAs. In the third, CD19 or CD79a was more highly expressed than CD3e and MPO. The three patterns corresponded to T-ALL, AML, and B-ALL, respectively. The use of cutoffs to establish qualitatively the pattern of coexpression of the four lineage markers provided the same information as the comparison among the four {Delta}CT values. Prospective use of the scoring system correctly classified each of 13 additional cases (8 AML, 4 B-lineage ALL, and 1 T-lineage ALL).

Conclusion: Study of lineage markers at diagnosis by RTQ-PCR allows differentiation of AML from B-ALL or T-ALL without cell sorting, even when the bone marrow contains few blast cells.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Golub et al. (1) showed in 1999 that it was possible to distinguish acute myeloid leukemias (AMLs)1 from acute lymphoid leukemias (ALLs) on the basis of gene expression monitored by DNA microarrays. Among the 50 genes most closely correlated with distinction between AML and ALL, some coded for proteins critical for S-phase cell cycle progression. Some of these genes were known oncogenes, whereas others had a role in transcription or in cell adhesion. In addition, it was not surprising to find genes of cell lineage markers such as CD33 or CD11c. These lineage markers belong to a family of proteins used to distinguish the cell lineage of leukemic proliferating blasts by immunophenotyping techniques (2)(3)(4). The presence of these lineage markers corresponds to a stage of cell differentiation.

Our objective was to investigate lineage markers in acute leukemias by reverse transcription followed by real-time quantitative-PCR (RTQ-PCR). We studied whether, at the mRNA level, the profiles of lineage marker expression enabled differential diagnosis among AML, B-lineage ALL, and T-lineage ALL. We selected four lineage markers already used in the immunophenotyping of acute leukemias. These markers are characterized by their persistent expression within the same lineage: CD19 and CD79a (B-lymphoid lineage), CD3e (T-lymphoid lineage), and myeloperoxidase (MPO; myeloid lineage). Using immunophenotyping, we interpreted these markers after enrichment of blast cells or in plain bone marrow, after gating. In our model, the profiles of these four lineage markers were studied without cell sorting.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
leukemic samples for rtq-pcr validation
We collected 72 bone marrow samples from adult and pediatric patients with acute leukemia at diagnosis. Diagnosis was made based on cytomorphology and cytochemistry according to the French-America-British (5)(6)(7)(8)(9)(10) and the WHO (11)(12) classifications. Immunophenotyping with immunologic classification based on the proposition of The European Group for the Immunological Classification of Leukemias (13) confirmed the morphology-based diagnosis. Immunophenotyping was performed on the plain bone marrow aspirate after lysis of the erythrocytes. Immunophenotyping analysis was performed after manual gating around blastic populations. Results are expressed as percentages of labeled blast cells. Positivity thresholds were 30% for B- and T-lineage markers and 20% for myeloid antigens (14).

Of the 72 samples, 40 were from AML, 27 were from B-ALL, and 5 were from T-ALL patients (see Data Supplement 1, which accompanies the online version of this article athttp://www.clinchem.org/content/vol50/issue7/). The median percentage of blast cells within the 40 AML bone marrow samples was 52% (range, 23–93%). These percentages ranged from 53% to 96% (median, 85%) and from 70% to 99% (median, 90%) in the B- and T-ALL samples, respectively.

This first cohort of 72 patients was used to develop our technique. In a second step, this technique was validated on an additional cohort of 13 patients. Of these 13 samples of bone marrow, 8 were AML, 4 were B-lineage ALL, and 1 was T-lineage ALL. The percentages of blast cells within the bone marrow samples ranged from 32% to 92% (median, 76%).

rna preparation
After the erythrocytes were lysed, the leukocytes were obtained by centrifugation and counted in a Particle Counter® Z1 (Coulter Corporation). Total RNA was extracted from 107 leukocytes with Trizol Reagent (Invitrogen) or the Tripure Isolation Reagent (Roche Diagnostics) according to the manufacturers’ recommendations. The RNA concentration was measured by ultraviolet spectroscopy.

reverse transcription reaction and conditions
The cDNA synthesis reaction was performed with 1 µg of RNA in a volume of 20 µL containing 19.95 µM random hexamers, 10 mM dithiothreitol (Invitrogen), 0.5 mM deoxynucleotide triphosphates (Roche Diagnostics), 0.025 U/µL Superscript Reverse Transcriptase (Invitrogen), 2 U/µL Ribonuclease Inhibitor Recombinant (Rnase out; Invitrogen), and 5x first-stand buffer (final dilution, 1x; Invitrogen). Reverse transcription was performed for 60 min at 37 °C. We diluted 10 µL of the total reverse transcription volume in 990 µL of nuclease-free water (Promega) and then used 5 µL of this dilution for each quantitative PCR.

pcr reaction and conditions
Primers were designed to recognize specific mRNA sequences by Primer Express Software (Applied Biosystems). These primers amplified only RNA and not human genomic DNA (which can contaminate RNA preparations). For MPO, the primers recognized all splicing variants. The primer sequences used are listed in Table 1 .


View this table:
[in this window]
[in a new window]
  		Table 1. Sequences of primers.

RTQ-PCR amplification reactions were prepared with use of the SYBR® Green PCR Core Reagent Kit (Applied Biosystems) and were analyzed on a real-time PCR instrument (GeneAmp® 5700 Sequence Detection System; Applied Biosystems). Each RTQ-PCR was performed in a final volume of 25 µL containing 5 µL of the diluted cDNA, 12.5 µL of SYBR Green PCR Core Reagent, 400 nM each amplification primer, and sterilized water (Baxter). Thermal cycling was started with a initial denaturation step at 95 °C for 10 min followed by 40 cycles at 95 °C for 15 s and 60 °C for 60 s. At the end of PCR, amplicon melting curves were obtained by increasing the temperature from 60 to 95 °C. Each PCR reaction was performed in triplicate. The same RTQ-PCR conditions were used in the second cohort except for a modification: each PCR reaction contained 25 ng of cDNA.

pcr controls
Five positive controls (cDNA) were synthesized by PCR, according to the procedure described by Morrison et al. (15), with the primers listed in Table 1Up . These amplicons were purified with Concert® Rapid PCR Purification System (Marligen) according to the manufacturer’s instructions. The amplicons were controlled by sequence analysis. Cycle sequencing was performed on a GeneAmp 2400 PCR System (Applied Biosystems) with a BigDye® Terminator Cycle Sequencing Kit (Applied Biosystems). Electrophoretic analysis was performed on an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems). The amplicons were quantified by ultraviolet spectroscopy (260 nm). The amplicon concentrations are expressed in terms of copies/5 µL.

A calibrator was prepared by mixing equimolar amounts of the five positive controls, at 106 copies/5 µL. This calibrator was diluted to 10 copies of each the five positive controls per 5 µL in Tris-EDTA buffer (10 mmol/L Tris-HCl and 1 mmol/L EDTA, pH 8) containing ribosomal RNA at 4 ng/µL (Roche Diagnostics). During dilution of the calibrator, strict precautions were taken to avoid contamination.

real-time quantification
RTQ-PCR were performed on GeneAmp® 5700 Sequence Detection System (Applied Biosystems). Real-time quantification was based on the use of the fluorochrome Sybr Green I. Quantification of the target amount in unknown samples was performed by measuring the threshold cycle (CT) (which is defined as the fractional cycle number at which the fluorescence encounters a fixed threshold). In the 72 samples, the same threshold was used for CD3e, CD19, CD79a, MPO, and Abelson (ABL). The expression of each gene was normalized to the expression of the housekeeping gene (ABL) (16)(17). The negative controls and the samples were amplified in triplicate. The calibrator was also tested for each of the five RTQ-PCR in triplicate. The reference dilution used for the calibrator was 103 copies of each five positive controls/5 µL. The mean CT was calculated. Our results were expressed in {Delta}CT (CT value of target gene minus CT value of housekeeping gene).

dna preparation
A sample of peripheral blood was taken from each of five healthy donors. The DNA was extracted according to the phenol-chloroform method (18).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
validation of rtq-pcr
The calibrator was diluted and amplified in triplicate for each of the five target sequences. For a constant threshold, the SD obtained on the CT for various dilutions of the calibrator are illustrated in Table 2 . The SD values increased with increasing CT values. The SD were similar for all positive controls.


View this table:
[in this window]
[in a new window]
  		Table 2. SD of CTABL for the calibrator.1

Real-time PCR data were analyzed according to the procedure described by Livak and Schmittgen (19)(20). The variation in {Delta}CT according to concentration over the range 10–104 copies/5 µL is shown in Table 3 . We plotted the logarithm of the input amount vs {Delta}CT for each PCR amplification. A linear regression was performed for each plot. The linear regression slope for each PCR amplification is shown in Table 3 . The slopes were all <0.1.


View this table:
[in this window]
[in a new window]
  		Table 3. Validation of the {Delta}CT calculation: Variations of {Delta}CT at different input amounts and slope of logarithm input amount vs {Delta}CT.1

We sometimes observed amplification curves for the negative controls for each five-positive control RTQ-PCR (with CT values of 33.0–39.9), but in all cases, melting curves showed melting temperature (Tm) values different from the Tm values of the PCR amplicons.

For the five-positive control RTQ-PCR, no amplification signal was observed when we used human genomic DNA as template.

quantification by rtq-pcr
All quantitative data for the two cohorts are detailed in online Data Supplement 2. In the AML subgroup, all subtypes of AML were present except for AML-M7. In the lymphoid leukemias, all stages of differentiation described by The European Group for the Immunological Classification of Leukemias (13) were analyzed. Our samples also included secondary acute leukemias (5 AML with preceding myelodysplastic syndrome). No undifferentiated or biphenotypic acute leukemias were investigated in this study.

The percentage of blasts in the analyzed samples was always >20% (12) (online Data Supplement 2). For each sample, the morphology-based diagnosis and the expression ({Delta}CT) of the four lineage markers were evaluated. Three samples (39, 43, and 50) were uninterpretable according to the CT value of the housekeeping gene. The 69 other samples were classified into two subsets according to the CT values obtained for the target genes. The first subset included samples characterized by a lineage marker CT value <30 (which corresponded to concentrations >100 copies/PCR). These samples were referred to as positive and quantifiable, and their results are expressed in terms of {Delta}CT. In the second subset, the target gene CT values were >30. The results for these samples, referred to as positive and nonquantifiable, are indicated as "+". To have fewer positive and nonquantifiable samples, each PCR of the second cohort of patient contained 10-fold more cDNA (25 ng/PCR), and all samples were then positive and quantifiable. For all samples, analysis of the amplicon melting curves confirmed the specificity of the amplification.

As shown in Figs. 1 and 2 , which illustrate the dispersion of {Delta}CT for each lineage marker, the number of samples varied among markers because positive and nonquantifiable samples were not included.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 1. Range of {Delta}CT values for each lineage marker.

The expression of each target gene was normalized to the expression of the housekeeping gene (ABL). The y axis is arranged with –10 at the top and +6 at the bottom. {Delta}CT values were negatively correlated with expression; i.e., the higher the expression of a lineage marker, the lower the CT value for this marker, leading to a {Delta}CT (CT value of lineage marker minus CT value of housekeeping gene) that is more negative. Cutoffs (horizontal lines) were located at {Delta}CT values of –1, –1.5, 0.5, and –2.5 for CD19, CD79a, CD3e, and MPO, respectively.



View larger version (25K):
[in this window]
[in a new window]
  		Figure 2. Patterns of coexpression of the four lineage markers for positive and quantifiable samples.

(A), patterns of T-lineage ALL samples. (B), patterns of AML samples. Sample 61 ({blacksquare}; B-lineage ALL) expressed more MPO than the other three lineage markers; sample 10 (•) is AML CD19+. (C), patterns of B-lineage ALL samples. Sample 1 ({blacksquare}; AML) expresses more CD79a than CD3e and MPO.

Depending on the lineage marker, the difference in expression between the highest and the lowest {Delta}CT values varied from 8.7 (CD3e) to 11.2 (MPO; Fig. 1Up ). The dispersion of {Delta}CT values was irregular except for MPO. For CD19 and CD79a, one-half of the {Delta}CT values were located in a small range (–3.6 to –1.2 and –4.7 to –1.6, respectively). For CD3e, most samples were observed in the 0.6–3.3 range.

To isolate these samples, cutoff criteria were chosen for CD19, CD79a, and CD3e at {Delta}CT values of –1, –1.5, and 0.5, respectively. For MPO, the cutoff was fixed at the median {Delta}CT (–2.5). On the basis of these four cutoffs, each sample was interpreted as pass or fail (P or F in online Data Supplement 2a). Samples that were positive but nonquantifiable were listed as failed. For CD19, all AML and T-lineage ALL samples failed, and all B-lineage ALL (except for samples 41, 48, 55, 56, and 57) passed. The cutoff used for CD79a separated B-lineage ALL from other acute leukemias. For CD3e, all T-lineage ALL samples passed, but B-lineage ALL and AML samples were observed on each side of the cutoff. For MPO, all T-lineage ALL samples failed, but B-lineage ALL and AML samples were observed on each side of the median {Delta}CT. Thus, with these cutoffs, none of the four lineage markers allowed acute leukemias to be differentiated correctly except for B-lineage ALL with the expression of CD79a.

We next compared the expression of the four lineage markers. Three patterns of coexpression were distinguished (Fig. 2Up ). In the first pattern (Fig. 2AUp ), CD3e was expressed more than the other three markers ({Delta}CTCD3e < {Delta}CTCD19, {Delta}CTCD79a, and {Delta}CTMPO). The second pattern, illustrated in Fig. 2BUp , expressed MPO more than the other three markers. In the third pattern (Fig. 2CUp ), expression of CD19 or CD79a was stronger than expression of MPO and CD3e. Each of these profiles corresponded to only one morphology-based diagnosis. The three profiles corresponded to T-ALL, AML, and B-ALL, respectively. There were, however, two exceptions. Sample 61 was associated with the second profile but corresponded to a B-lineage ALL, and sample 1 was associated with the third profile but corresponded to an AML.

In a third and final analysis of data, cutoff criteria were chosen according to morphology-based diagnosis. For CD3e, the cutoff was placed at the lower limit of the highest {Delta}CT observed in T-lineage ALL. For CD19 and CD79a, cutoffs were placed at the lower limit of the highest {Delta}CT observed in B-lineage ALL. The cutoff for MPO was placed at the upper limit of the lowest {Delta}CT observed in B- and T-lineage ALL. Cutoffs for CD19, CD79a, CD3e, and MPO were at {Delta}CT values of 1, –1.5, –2.5, and –3.5, respectively. These cutoffs, different from those established previously, allowed a second qualitative interpretation of the samples (P or F in online Data Supplement 2). Again, samples that were positive but nonquantifiable appeared as failed. According to these cutoffs, T-lineage ALLs were characterized by failed expression of CD19, CD79a, and MPO. The expression of CD3e passed. In AML, MPO passed and CD19, CD79a, and CD3e failed, or the four lineage markers failed. B-Lineage ALLs were characterized by passing expression of CD19 and CD79a and a failing expression of MPO and CD3e. Table 4 summarizes the four coexpression patterns of these three morphology-based diagnoses. On the 69 analyzed samples, only 1 (sample 10) could not be incorporated into one of these patterns. In this sample, CD19a and MPO passed and CD79a and CD3e failed.


View this table:
[in this window]
[in a new window]
  		Table 4. Coexpression patterns of the four lineage markers at the mRNA level.1

A scoring system was established to classify acute leukemias. For CD19, CD79a, and CD3e, a value of 2 was assigned when a passed qualitative expression was observed. For passing qualitative expression of MPO, a value of 1 was assigned. Five scoring levels were applied as shown in Table 4Up . This system identified three types of acute leukemias: B-lineage ALL (score = 4); T-lineage ALL (score = 2); and AML (score = 0 or 1). A score of 3 corresponded to AML CD19+.

This scoring system was applied in a second cohort of patients to differentiate acute leukemias. For the 13 samples analyzed, we observed 100% of correlation with the morphology-based diagnosis.

We next compared RTQ-PCR data with immunophenotyping results for the four lineage markers (Table 5 ). The {Delta}CT values (in terms of failing or passing) and the percentages of labeled blast cells (in terms of positive or negative) were correlated according to positivity cutoffs for both techniques. The correlation for all four markers was good, but it was better for CD19 and CD3e than for MPO and CD79a: 100% vs 89–96%.


View this table:
[in this window]
[in a new window]
  		Table 5. Concordance of RTQ-PCR and immunophenotyping.1


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
One advantage—and disadvantage—of SYBR Green is that it links to any double-stranded DNA. Specific PCR products and nonspecific products and/or primer-dimers are detected equally well (21). In preliminary studies, we checked the specificity of our PCR amplifications by agarose gel electrophoresis, sequence analysis, and amplicon melting curve analysis. With the primers described in Table 1Up , primer-dimers were reduced to concentrations (CT values, 33.0–39.9) that was not important for target detection and quantification. Furthermore, primer-dimer Tms were different from Tms of specific amplicons. For CD19, CD79a, CD3e, MPO, and ABL amplicons, we observed Tms of 81.0, 87.0, 82.0, 87.0, and 81.0 °C, respectively (Table 1Up ).

In view of the SD on the CT values, we estimated that experimental precision would be insufficient when the CT value was >30. Target mRNAs with such CT values were not quantified because sampling error contributed significantly to the variance in the experimental data. In the second cohort, the increased quantity of cDNA in each RTQ-PCR allowed greater sensitivity for the detection of targets genes. The formation of primer-dimers and the samples profile were not modified by this increase in cDNA quantity.

The analysis of {Delta}CT at various concentrations showed no differences between the PCR efficiencies of the target genes and the housekeeping gene. This indicated that all PCRs had the same efficiencies. In agreement with the procedure of Livak and Schmittgen (19)(20), the {Delta}CT method can be used for the relative quantification of target gene without the need to run calibration curves on the same plate.

ABL was chosen as a housekeeping gene on the basis of two published reports (16)(17).

All of the leukemic samples (except for samples 39, 43, and 50) jointly expressed CD3e, CD19, CD79a, and MPO. However, the expression of the different lineage markers differed among samples and within the same sample. This coexpression, which was observed even in samples containing many blasts, suggests the persistence of a normal cell population from bone marrow or peripheral blood (contamination during cell sampling). CD19, CD79a, CD3e, and MPO are not specific markers of leukemic cells. Moreover, RTQ-PCR is performed on the whole population of leukocytes, which is more complex than a population containing only blast cells. In addition, our technique does not allow cells to be analyzed individually, as in immunophenotyping using a gating system. The mRNA of residual normal cells and of blast cells is thus quantified together.

The CT value for the housekeeping gene in samples 39, 43, and 50 was >30, corresponding to a low concentration of RNA (or cDNA) or poor RNA (or cDNA) quality. CT values from these samples could therefore not be interpreted.

The 69 leukemic samples were classified according to the expression of each lineage marker, not simply according to the presence or absence of that marker. For CD19, the range of expression was the same for B-lineage ALL and for AML and T-lineage ALL (Fig. 1Up ). For CD79a, the range of expression was wider for AML and T-lineage ALL than for B-lineage ALL. The expression range for CD3e was narrower for T-lineage ALL than for AML and B-lineage ALL. In Table 1Up , cutoffs were placed on the basis of visual criteria. These cutoffs would have been different if our sampling had contained more T-lineage ALL and if all of the samples had been positive and quantifiable. Because the cutoffs for CD79a, CD3e, and MPO corresponded to negative {Delta}CT values, samples that were positive but nonquantifiable were interpreted as failed in online Data Supplement 2a. Because these samples had one target gene CT value >30 and a CT value for the housekeeping gene <30, {Delta}CT values were positive and were located under these cutoffs in the samples with failed expression. For CD19, the {Delta}CT of positive and nonquantifiable samples were calculated without a quantification limit. All of theses samples had a {Delta}CTCD19 >2 and were located in the group with failed expression.

Because the use of these cutoffs did not allow differentiation of acute leukemias, the coexpression profile of the four lineage markers was studied. In this second approach we also worked without knowing the morphology-based diagnosis of the samples. As shown in Fig. 2Up , three profiles were observed. Each profile corresponded to only one pathology, except for samples 1 and 61. In sample 1, CD79a was expressed more than CD3e and MPO, but the expression of CD19 and CD79a was less in this sample than in all B-lineage ALLs (Fig. 2CUp ). The same observation was made for sample 61. Sample 61 had an AML profile, but the expression of CD19 and CD79a was higher in this sample than in all AML samples (Fig. 2BUp ). Expression of MPO enzyme activity in leukemic blasts is commonly used to distinguish myeloid from lymphoid leukemias (9). However, several studies have shown that a small minority of cases of MPO enzyme-negative acute leukemia in adults and children exhibited MPO protein and/or MPO RNA in the leukemic blast cells (22)(23)(24). Austin et al.(25) reported that in patients with infant B-precursor ALL (younger than 366 days of age), the leukemic lymphoblasts frequently expressed MPO at the RNA or protein level. Our sample 61 was a patient who was 396 days of age. This young patient had a BIII-ALL that expressed CD33 at the protein level as the lone myeloid marker. However, we observed that although sample 61 had an AML profile, the difference between the {Delta}CTMPO (–2.84) and the {Delta}CTCD79a (–2.725) was not statistically different.

Sample 1 was an AML-M0. Immunophenotyping analysis confirmed the myeloid lineage assignment by positive expression of CD13, CD33, and CD117. Moreover, immunophenotyping analysis showed no lymphoid-associated antigen expression. Negative results for CD79a, CD3, and CD22 are major criteria for myeloid lineage assignment (26)(27).

In the third data analysis, our approach was different. The morphology-based diagnosis for each sample and the corresponding data were known, and then cutoffs were established. With these cutoffs, four coexpression patterns were distinguished that allowed acute leukemias to be differentiated correctly. Each pattern corresponded to only one pathology. There was, however, one exception. Sample 10 could not be assigned to any of these patterns (Fig. 2BUp ). This sample was an AML with a t(8;21) translocation. These AMLs have a particular morphology (28)(29)(30)(31)(32) and more frequently coexpress CD19 and CD56. Controversial findings have been described regarding the frequency and the intensity of CD19 and/or the expression of CD56 in AML with t(8;21) (33)(34), and have led to questions as to whether these aberrant phenotypic features occur frequently enough to allow selection of cases for molecular screening on the basis of immunophenotyping (31). With these four cutoffs, samples 1 and 61 were correctly interpreted.

None of the four selected lineage markers alone provided the necessary information to differentiate the acute leukemias by RTQ-PCR. However, the combined study of CD79a and CD3e enabled us to develop an algorithm (Fig. 3 ) in which the diagnosis of AML was obtained after exclusion of the diagnosis of B- and T-lineage ALL. However, such an algorithm is not applicable in routine practice because of the risk of aberrant expression of one of these two markers. A second algorithm based on the four lineage markers was thus defined to decrease the risk of aberrant expression (Fig. 4 ); this algorithm presents only one of several combinations. When the scoring system was established, we attributed a lower value to MPO than the other three lineage markers to distinguish the scoring of T-lineage ALL and the scoring of AML with strong expression of MPO. This also allowed separation of the scorings of AML CD19+ and B-lineage ALL. The presence of a normal cell population that also expresses these lineage markers most probably influences the results. Semiqualitative interpretation using an algorithm enables correct conclusions. Even without cell sorting, two lineage markers (CD3e and CD79a) are sufficient for this differential diagnosis.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 3. Algorithm based on two lineage markers.

The diagnosis was a B-lineage ALL if the {Delta}CT value of CD79a passed (less than –1.5). The diagnosis was a T-lineage ALL if the {Delta}CT value of CD79a failed (greater than –1.5) and the {Delta}CT value of CD3e passed (less than –2.5). The diagnosis was AML if the {Delta}CT values of CD79a and CD3e failed (greater than –1.5 and greater than –2.5, respectively). P, pass; F, fail.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 4. Algorithm based on four lineage markers.

The diagnosis was a B-lineage ALL if the {Delta}CT values of CD19 and CD79a passed (<1 and less than –1.5, respectively). The diagnosis was a T-lineage ALL if the {Delta}CT values of CD19 and MPO failed (>1 and greater than –3.5, respectively) and the {Delta}CT value of CD3e passed (less than –2.5). The diagnosis was AML if the {Delta}CT value of CD19 failed (>1) and the {Delta}CT value of MPO passed (less than –3.5) or if the {Delta}CT values of CD19, MPO, and CD3e failed (>1, greater than –3.5, and greater than –2.5, respectively). P, pass; F, fail.

To improve the diagnosis beyond the differentiation of the three lineages, it would be necessary to increase the number of genes studied. Although the study of the specific translocations of acute leukemias may seem obvious, other genes are most probably important for the prognosis, for residual disease follow-up, or for treatment of acute leukemias (11)(35)(36)(37)(38)(39)(40)(41). Like low-density microarrays, RTQ-PCR allows study of several hundred genes but confers more precise quantification (42). Furthermore, RTQ-PCR uses very little cDNA in each reaction. Our approach would perhaps be of help in classifying cases in which cytomorphology and immunophenotyping are inconclusive in establishing the cell lineage. We could imagine the inclusion of a "lineage expression panel" in a RTQ-PCR screening plate already used for detection of fusion transcripts.

In conclusion, the study of four lineage markers by RTQ-PCR enables the differential diagnosis of AML, B-lineage ALL, and T-lineage ALL to be made without cell sorting, even when the sample contains few blasts. The lowest percentages of blasts necessary to adequately differentiate the various acute leukemias are 23%, 53%, and 66% for AML, B-lineage ALL, and T-lineage ALL, respectively. These limits are based on the percentages of blasts observed in our two cohorts at the time of cytomorphologic analysis (online Data Supplement 2). Dilutions of leukemic samples by normal bone marrow would be necessary to define the lowest limit for each type of acute leukemia. The dispersion observed in each profile is most probably the result of the absence of cell sorting. In spite of this dispersion and with a few exceptions, the coexpression profile established with cutoffs or by comparison of the four {Delta}CT values provides the same information. Many questions remain to be solved. Currently, we are studying the expression profiles of these four lineage markers in peripheral blood and in bone marrow of healthy donors. We would like to know how the persistence of a normal cell population influences the results. Despite the absence of cell sorting in RTQ-PCR analysis and without considering the CV of the immunophenotyping technique, the correlation between the semiquantitative expression at the mRNA and protein levels was 89–100%, according to lineage marker. Currently we are comparing the quantitative expression profiles of these four lineage markers at the mRNA and protein levels (with and without gating), and we are especially interested in a comparison between the {Delta}CT and the median fluorescence intensity.


   Acknowledgments
 
We thank Prof. D. Van Bockstaele and Dr. O. Cornu for critically reading the manuscript. We acknowledge the assistance of all those who took part in the collection of samples (in alphabetic order): Bernard Chatelain (Mont-Godinne UCL, Yvoir), Olivier Ketelslegers (CHR Citadelle, Liège), Estelle Lacher (Centre Hospitalier du Luxembourg, Luxembourg), and Frédéric Lambert (CHU Sart-Tilman, Liège). This work was supported by grants from Fonds National de la Reserche Scientifique (Grant Télévie No. 7.4519.00) and by Salus Sanguinis.


   Footnotes
 
1 Nonstandard abbreviations: AML, acute myeloid leukemia; ALL, acute lymphoid leukemia; RTQ-PCR, real-time quantitative-PCR; MPO, myeloperoxidase; CT, threshold cycle; and Tm, melting temperature.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M, Mesirov JP, et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999;286:531-537.[Abstract/Free Full Text]
  2. Dinndorf PA, Buckley JD, Nesbit ME, Lampkin BC, Piomelli S, Feig SA, et al. Expression of myeloid differentiation antigens in acute nonlymphocytic leukemia: increased concentration of CD33 antigen predicts poor outcome—a report from the Childrens Cancer Study Group. Med Pediatr Oncol 1992;20:192-200.[ISI][Medline] [Order article via Infotrieve]
  3. Master PS, Richards SJ, Kendall J, Roberts BE, Scott CS. Diagnostic application of monoclonal antibody KB90 (CD11c) in acute myeloid leukaemia. Blut 1989;59:221-225.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  4. Buccheri V, Mihaljevic B, Matutes E, Dyer MJ, Mason DY, Catovsky D. mb-1: a new marker for B-lineage lymphoblastic leukemia. Blood 1993;82:853-857.[Abstract/Free Full Text]
  5. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, et al. Proposals for the classification of the acute leukaemias. French-American-British (FAB) Co-operative Group. Br J Haematol 1976;33:451-458.[ISI][Medline] [Order article via Infotrieve]
  6. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, et al. A variant form of hypergranular promyelocytic leukemia (M3). Ann Intern Med 1980;92(2 Pt 1):261.
  7. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol 1982;51:189-199.[ISI][Medline] [Order article via Infotrieve]
  8. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, et al. Criteria for the diagnosis of acute leukemia of megakaryocyte lineage (M7). A report of the French-American-British Cooperative Group. Ann Intern Med 1985;103:460-462.
  9. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, et al. Proposed revised criteria for the classification of acute myeloid leukemia. A report of the French-American-British Cooperative Group. Ann Intern Med 1985;103:620-625.
  10. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, et al. Proposal for the recognition of minimally differentiated acute myeloid leukaemia (AML-MO). Br J Haematol 1991;78:325-329.[ISI][Medline] [Order article via Infotrieve]
  11. Harris NL, Jaffe ES, Diebold J, Flandrin G, Muller-Hermelink HJK, Vardiman J, et al. World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues: report of the ClinicalAdvisory Committee Meeting-Airlie House, Virginia, November 1997. J Clin Oncol 1999;17:3835-3849.[Abstract/Free Full Text]
  12. Harris NL, Jaffe ES, Diebold J, Flandrin G, Muller-Hermelink HK, Vardiman J, et al. The World Health Organization classification of hematological malignancies report of the Clinical Advisory Committee Meeting, Airlie House, Virginia, November 1997. Mod Pathol 2000;13:193-207.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  13. Bene MC, Bernier M, Casasnovas RO, Castoldi G, Knapp W, Lanza F, et al. The reliability and specificity of c-kit for the diagnosis of acute myeloid leukemias and undifferentiated leukemias. The European Group for the Immunological Classification of Leukemias (EGIL). Blood 1998;92:596-599.[Abstract/Free Full Text]
  14. Garand R, Bene MC. A new approach of acute lymphoblastic leukemia immunophenotypic classification: 1984–1994 the GEIL experience. Groupe d’Etude Immunologique des Leucemies. Leuk Lymphoma 1994;13(Suppl 1):1-5.
  15. Morrison TB, Weis JJ, Wittwer CT. Quantification of low-copy transcripts by continuous SYBR Green I monitoring during amplification. Biotechniques 1998;24:954-958960, 962.[ISI][Medline] [Order article via Infotrieve]
  16. Schnittger S, Weisser M, Schoch C, Hiddemann W, Haferlach T, Kern W. New score predicting for prognosis in PML-RARA+, AML1-ETO+, or CBFBMYH11+ acute myeloid leukemia based on quantification of fusion transcripts. Blood 2003;102:2746-2755.[Abstract/Free Full Text]
  17. Beillard E, Pallisgaard N, Van Der Velden VH, Bi W, Dee R, Van Der Schoot E, et al. Evaluation of candidate control genes for diagnosis and residual disease detection in leukemic patients using ‘real-time’ quantitative reverse-transcriptase polymerase chain reaction (RQ-PCR)—a Europe Against Cancer program. Leukemia 2003;17:2474-2486.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  18. Moore DD. Phenol extraction and ethanol precipitation of DNA. Ausubel FM Brent R Kingston RE Moore DD Seidman JG Smith JAet al eds. Current protocols in molecular biology 1996:2.1.1 John Wiley Boston. .
  19. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2–{Delta}{Delta}C(T) method. Methods 2001;25:402-408.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  20. Livak KJ. ABI Prism 7700 Sequence Detection System. User bulletin 2 1997 PE Applied Biosystems Foster City, CA. .
  21. Rasmussen R, Morrison T, Herrmann M, Wittwer C. Quantitative PCR by continuous fluorescence monitoring of a double strand DNA-specific binding dye. Biochemica 1998;2:8-11.
  22. Ferrari S, Ceccherelli G, Tagliafico E, Zucchini P, Manfredini R, Torelli G, et al. Detection of low abundance mRNA of myeloid specific genes in cells of acute and chronic lymphoid leukemias by cRNA hybridization. Leukemia 1990;4:688-693.[ISI][Medline] [Order article via Infotrieve]
  23. Urbano-Ispizua A, Matutes E, Villamor N, Ribera JM, Feliu E, Montserrat E, et al. Clinical significance of the presence of myeloid associated antigens in acute lymphoblastic leukaemia. Br J Haematol 1990;75:202-207.[ISI][Medline] [Order article via Infotrieve]
  24. Zhou M, Findley HW, Zaki SR, Little F, Coffield LM, Ragab AH. Expression of myeloperoxidase mRNA by leukemic cells from childhood acute lymphoblastic leukemia. Leukemia 1993;7:1180-1183.[ISI][Medline] [Order article via Infotrieve]
  25. Austin GE, Alvarado CS, Austin ED, Hakami N, Zhao WG, Chauvenet A, et al. Prevalence of myeloperoxidase gene expression in infant acute lymphocytic leukemia. Am J Clin Pathol 1998;110:575-581.[ISI][Medline] [Order article via Infotrieve]
  26. Stasi R, Amadori S. AML-M0: a review of laboratory features and proposal of new diagnostic criteria. Blood Cells Mol Dis 1999;25:120-129.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  27. Bene MC, Castoldi G, Knapp W, Ludwig WD, Matutes E, Orfao A, et al. Proposals for the immunological classification of acute leukemias. European Group for the Immunological Characterization of Leukemias (EGIL). Leukemia 1995;9:1783-1786.[ISI][Medline] [Order article via Infotrieve]
  28. Porwit-MacDonald A, Janossy G, Ivory K, Swirsky D, Peters R, Wheatley K, et al. Leukemia-associated changes identified by quantitative flow cytometry. IV. CD34 overexpression in acute myelogenous leukemia M2 with t(8;21). Blood 1996;87:1162-1169.[Abstract/Free Full Text]
  29. Kita K, Nakase K, Miwa H, Masuya M, Nishii K, Morita N, et al. Phenotypical characteristics of acute myelocytic leukemia associated with the t(8;21)(q22;q22) chromosomal abnormality: frequent expression of immature B-cell antigen CD19 together with stem cell antigen CD34. Blood 1992;80:470-477.[Abstract/Free Full Text]
  30. Hurwitz CA, Raimondi SC, Head D, Krance R, Mirro J, Jr, Kalwinsky DK, et al. Distinctive immunophenotypic features of t(8;21)(q22;q22) acute myeloblastic leukemia in children. Blood 1992;80:3182-3188.[Abstract/Free Full Text]
  31. Andrieu V, Radford-Weiss I, Troussard X, Chane C, Valensi F, Guesnu M, et al. Molecular detection of t(8;21)/AML1-ETO in AML M1/M2: correlation with cytogenetics, morphology and immunophenotype. Br J Haematol 1996;92:855-865.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  32. Baer MR, Stewart CC, Lawrence D, Arthur DC, Byrd JC, Davey FR, et al. Expression of the neural cell adhesion molecule CD56 is associated with short remission duration and survival in acute myeloid leukemia with t(8;21)(q22;q22). Blood 1997;90:1643-1648.[Abstract/Free Full Text]
  33. Creutzig U, Harbott J, Sperling C, Ritter J, Zimmermann M, Loffler H, et al. Clinical significance of surface antigen expression in children with acute myeloid leukemia: results of study AML-BFM-87. Blood 1995;86:3097-3108.[Abstract/Free Full Text]
  34. Barnard DR, Kalousek DK, Wiersma SR, Lange BJ, Benjamin DR, Arthur DC, et al. Morphologic, immunologic, and cytogenetic classification of acute myeloid leukemia and myelodysplastic syndrome in childhood: a report from the Childrens Cancer Group. Leukemia 1996;10:5-12.[ISI][Medline] [Order article via Infotrieve]
  35. Grimwade D, Walker H, Oliver F, Wheatley K, Harrison C, Harrison G, et al. The importance of diagnostic cytogenetics on outcome in AML: analysis of 1,612 patients entered into the MRC AML 10 trial. The Medical Research Council Adult and Children’s Leukaemia Working Parties. Blood 1998;92:2322-2333.[Abstract/Free Full Text]
  36. Van Dongen JJ, Macintyre EA, Gabert JA, Delabesse E, Rossi V, Saglio G, et al. Standardized RT-PCR analysis of fusion gene transcripts from chromosome aberrations in acute leukemia for detection of minimal residual disease. Report of the BIOMED-1 Concerted Action: investigation of minimal residual disease in acute leukemia. Leukemia 1999;13:1901-1928.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  37. Leith CP, Kopecky KJ, Godwin J, McConnell T, Slovak ML, Chen IM, et al. Acute myeloid leukemia in the elderly: assessment of multidrug resistance (MDR1) and cytogenetics distinguishes biologic subgroups with remarkably distinct responses to standard chemotherapy. A Southwest Oncology Group study. Blood 1997;89:3323-3329.[Abstract/Free Full Text]
  38. Barjesteh van Waalwijk van Doorn-Khosrovani S, Erpelinck C, van Putten WL, Valk PJ, van der Poel-van de Luytgaarde S, Hack R, et al. High EVI1 expression predicts poor survival in acute myeloid leukemia: a study of 319 de novo AML patients. Blood 2003;101:837-845.[Abstract/Free Full Text]
  39. Baldus CD, Tanner SM, Kusewitt DF, Liyanarachchi S, Choi C, Caligiuri MA, et al. BAALC, a novel marker of human hematopoietic progenitor cells. Exp Hematol 2003;31:1051-1056.[ISI][Medline] [Order article via Infotrieve]
  40. Zwaan CM, Meshinchi S, Radich JP, Veerman AJ, Huismans DR, Munske L, et al. FLT3 internal tandem duplication in 234 children with acute myeloid leukemia: prognostic significance and relation to cellular drug resistance. Blood 2003;102:2387-2394.[Abstract/Free Full Text]
  41. Whitman SP, Archer KJ, Feng L, Baldus C, Becknell B, Carlson BD, et al. Absence of the wild-type allele predicts poor prognosis in adult de novo acute myeloid leukemia with normal cytogenetics and the internal tandem duplication of FLT3: a cancer and leukemia group B study. Cancer Res 2001;61:7233-7239.[Abstract/Free Full Text]
  42. Rajeevan MS, Ranamukhaarachchi DG, Vernon SD, Unger ER. Use of real-time quantitative PCR to validate the results of cDNA array and differential display PCR technologies. Methods 2001;25:443-451.[CrossRef][ISI][Medline] [Order article via Infotrieve]



The following articles in journals at HighWire Press have cited this article:


Home page
 Am. J. Pathol.Home page
Q. H. Meng, G. Yang, W. Yang, B. Jiang, L. Wu, and R. Wang
Protective Effect of Hydrogen Sulfide on Balloon Injury-Induced Neointima Hyperplasia in Rat Carotid Arteries
Am. J. Pathol., April 1, 2007; 170(4): 1406 - 1414.
[Abstract] [Full Text] [PDF]

Home page
 Appl. Environ. Microbiol.Home page
B. Martin, A. Jofre, M. Garriga, M. Pla, and T. Aymerich
Rapid Quantitative Detection of Lactobacillus sakei in Meat and Fermented Sausages by Real-Time PCR
Appl. Envir. Microbiol., September 1, 2006; 72(9): 6040 - 6048.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
G. Yang, K. Cao, L. Wu, and R. Wang
Cystathionine {gamma}-Lyase Overexpression Inhibits Cell Proliferation via a H2S-dependent Modulation of ERK1/2 Phosphorylation and p21Cip/WAK-1
J. Biol. Chem., November 19, 2004; 279(47): 49199 - 49205.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Saussoy, P.
Right arrow Articles by Latinne, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Saussoy, P.
Right arrow Articles by Latinne, D.
Related Collections
Right arrow Molecular Diagnostics and Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS