Clinical Chemistry Link to Randox Laboratories Web Site
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 53: 1757-1766, 2007. First published August 16, 2007; 10.1373/clinchem.2007.089326
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow 089326.Supplemental Data
Right arrow All Versions of this Article:
clinchem.2007.089326v1
53/10/1757    most recent
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 (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stamatopoulos, B.
Right arrow Articles by Lagneaux, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stamatopoulos, B.
Right arrow Articles by Lagneaux, L.
Related Collections
Right arrow Molecular Diagnostics and Genetics
(Clinical Chemistry. 2007;53:1757-1766.)
© 2007 American Association for Clinical Chemistry, Inc.

Molecular Diagnostics and Genetics

Quantification of ZAP70 mRNA in B Cells by Real-Time PCR Is a Powerful Prognostic Factor in Chronic Lymphocytic Leukemia

Basile Stamatopoulos1,a, Nathalie Meuleman1, Benjamin Haibe-Kains2,3, Hughes Duvillier1, Martine Massy1, Philippe Martiat1, Dominique Bron1 and Laurence Lagneaux1

1 Laboratory of Experimental Hematology, 2 Functional Genomics and Translational Research Unit, Faculty of Medicine, Institut Jules Bordet, Université Libre de Bruxelles (ULB), Brussels, Belgium.
3 Machine Learning Group, Faculty of Sciences, Université Libre de Bruxelles (ULB), Brussels, Belgium.

aAddress correspondence to this author at: Université Libre de Bruxelles, Institut Jules Bordet, Laboratoire d’Hématologie Expérimentale, Boulevard de Waterloo no. 121-1000 Bruxelles, Belgium. Fax 32-0-2-541-3453; e-mail bstamato{at}ulb.ac.be.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: Chronic lymphocytic leukemia (CLL) is heterogeneous with respect to prognosis and clinical outcome. The mutational status of the immunoglobulin variable heavy chain region (IGHV) has been used to classify patients into 2 groups in terms of overall survival (OS) and clinical characteristics, but the labor-intensive nature and the cost of this time-consuming analysis has prompted investigations of surrogate markers.

Methods: We developed a standardized quantitative real-time reverse transcription-PCR (qPCR) method to measure zeta-chain (TCR)-associated protein kinase (ZAP70) mRNA in purified CD19+ cells. We evaluated this and other methods (flow cytometry analyses of ZAP70 and CD38 proteins and qPCR analysis of lipoprotein lipase mRNA) in a cohort of 108 patients (median follow-up, 82 months) to evaluate any associations with IGHV mutational status, OS, and treatment-free survival (TFS).

Results: The association between qPCR-measured ZAP70 and IGHV mutational status was statistically significant [{chi}2 (1) = 50.95; P <0.0001], and the value of Cramer’s V statistic (0.72) indicated a very strong relation. This method also demonstrated sensitivity, specificity, and positive and negative predictive values of 87.8%, 85.7%, 87.5%, and 86%, respectively. ZAP70 expression was significantly associated with OS (P = 0.0021) and TFS (P <0.0001). ZAP70+ patients had significantly shorter median TFS (24 months) than ZAP70– patients (157 months) (P <0.0001). Moreover, qPCR-measured ZAP70 expression has greater prognostic power than IGHV mutational status and the other prognostic markers tested.

Conclusions: ZAP70 mRNA quantification via qPCR is a strong surrogate marker of IGHV mutational status and a powerful prognostic factor.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
B-cell chronic lymphocytic leukemia (CLL) 1 is the most common leukemia in western countries. CLL mainly affects elderly individuals, but about one-third of patients are younger than 60 years at diagnosis (1). CLL has an extremely variable clinical course, with overall survival (OS) times ranging from months to decades. The disease runs an indolent clinical course in some patients, and life expectancy is not shortened. For others, the disease is aggressive and progresses rapidly, with survival times after diagnosis of 2–3 years (2). Therefore, it is very important to identify factors that can predict a poor prognosis and the patients who will benefit from intensive therapy at an early disease stage.

Several prognostic markers have been identified in the last 3 decades. Initially, Rai et al. (3) and Binet et al. (4) developed clinical staging systems based on patient clinical characteristics, but these systems were unable to prospectively distinguish early-stage CLL that progresses rapidly to aggressive disease from disease destined to remain in an early stage for a long time (5). This difficulty prompted the increasing use of other genetic and biological markers for predicting the prognosis of CLL (6). Many prognostic factors have recently been found to predict clinical outcome, with one of the most important molecular genetic variable markers being the mutational status of the immunoglobulin variable heavy chain region (IGHV) (7). This new marker can be used to separate patients into 2 groups: a group with unmutated IGHV and a worse outcome, and a group with mutated IGHV associated with a good prognosis. However, this analysis is laborious and costly and is inaccessible for most clinical laboratories. Identifying a surrogate marker for IGHV mutational status is therefore an important goal. CD38 was the first marker to be correlated with IGHV mutational status (8), but the relationship is not absolute. In 2001, a comparison of gene expression profiles for the 2 patient groups revealed a small number of differentially expressed genes (9)(10), of which the lipoprotein lipase gene (LPL) 2 was closely related to IGHV mutational status. Further studies confirmed the predictive value of this gene (11)(12). In 2003, Wiestner et al. (13) found that zeta-chain (TCR)-associated protein kinase (ZAP70) expression, which is usually found in T lymphocytes and natural killer cells, was correlated with IGHV mutational status in 93% of cases. Patients with <20% ZAP70+ B cells as measured by flow cytometry (FC) generally had a mutated IGHV status, and patients with ≥20% ZAP70+ cells had an unmutated IGHV status. Further clinical studies confirmed the prognostic value of the ZAP70 protein (14)(15)(16)(17)(18)(19)(20); however, FC measurement of ZAP70 status is often inaccurate at the positivity limit (21) because of low resolution of the positive and negative populations, and can also be influenced by the gating procedure and antibody choice (22). To offset these drawbacks, investigators have proposed absolute quantification of ZAP70 mRNA (23)(24).

We describe the validation of a new standardized quantitative real-time reverse transcription-PCR (qPCR) analysis for measuring ZAP70 mRNA in purified B lymphocytes and its power as both a surrogate for IGHV mutational status and a prognostic marker for survival and treatment-free time in CLL.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
patients, sample collection, and rna extraction
This study was approved by the Bordet Institute Ethics Committee and was based on peripheral blood samples obtained with informed consent from 108 CLL patients. Patients either were untreated or had received no treatment for at least 6 months before the study. The 63 male and 45 female patients had a typical CD19+CD5+CD23+ phenotype and blood leukocyte counts between 1 x 1010 and 25 x 1010 cells/L. The median age at diagnosis was 65 years (range, 46–90 years). The median follow-up duration was 82 months (range, 8–299 months). Table 1 summarizes other patient features. Some information is unavailable because of limited biological material or the absence of clinical data. Treatment-free survival (TFS) and disease-related mortality were calculated from the time of diagnosis. Peripheral blood mononuclear cells were isolated by density-gradient centrifugation over Linfosep (Biomedics). B cells were purified with a CD19+ magnetic-bead system (MidiMACS, Miltenyi Biotec) according to the manufacturer’s instructions. Mean B-cell purity as measured by FC was >99%. Total RNA was extracted from purified CD19+ cells in a single step with TriPure Isolation Reagent (Roche Applied Science) (25).


View this table:
[in this window]
[in a new window]

  		Table 1. Cross-tabulations of prognostic markers vs IGHV mutational status and ZAP70 mRNA measured by qPCR.

fc measurement of zap70 and cd38 expression
We measured the expression of cytoplasmic ZAP70 protein by FC with the Fix and Perm Permeabilization Kit (ImTec Diagnostics), a ZAP70 phycoerythrin-conjugated antibody (clone 1E7.2, eBioscience), fluorescein-isothiocyanate–conjugated CD3, and phycoerythrin-Cy5–conjugated CD19 (Immunotech) (16). Because the choice of the threshold for ZAP70 positivity can critically affect the decision regarding ZAP70 status (see Text 1 and Figs. 1 and 2 in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol53/issue10), we defined it on the basis of the lower limit of the region which included 99% of ZAP70+CD3+ cells. This threshold maximized the concordance between ZAP70 status and mutational status. After establishing the appropriate gating on CD3+ cells, we fixed the cutoff for ZAP70 positivity and measured ZAP70 in CD19+ cells. We evaluated the expression of CD38 on cell surfaces by FC in a CD19+ gate with a panel of fluorochrome-labeled monoclonal antibodies (phycoerythrin-conjugated CD38, Immunotech). CD38 expression was deemed positive if 7% of the cells stained positive in a standard 3-color FC analysis (8).

qpcr analysis of zap70 and lpl expression
We used 25 ng cDNA (produced by a standard reverse transcription) in a qPCR reaction with SYBR® Green PCR Master Mix (Applied Biosystems) and 0.32 µmol/L of gene-specific forward and reverse primers (Invitrogen). The sequences of the ZAP70, PPIA [peptidylprolyl isomerase A (cyclophilin A)], and LPL primers have been published [ (26), (27), and (11), respectively]. We also tested 5 housekeeping genes [LMNB1, lamin B1; EIF1AX, eukaryotic translation initiation factor 1A, X-linked; CASC3 (also known as MLN51), cancer susceptibility candidate 3; PPIA; and PGK1, phosphoglycerate kinase 1] as endogenous controls (data not shown). Finally, we standardized all results using PPIA gene expression, which was the most stable. Standard real-time PCR was performed on an ABI Prism 7900 HT (Applied Biosystems). A calibrator sample (cDNA from the Namalwa cell line, a human B-lymphoid leukemia cell line that expresses ZAP70 at a low level; ATCC) was included as a control in each experiment. In all cases, we created dissociation curves to confirm PCR specificity. We analyzed the data with the comparative {Delta}{Delta}Ct method (for details, see Text 2 in the online Data Supplement).

ighv gene mutational analysis
We conducted IGHV gene mutational analysis as previously described (28) and aligned sequences with those in the international ImMunoGeneTics information system database (http://imgt.cines.fr). Sequences with ≤2% deviation from any germ line IGHV sequence were considered unmutated (7).

statistical analysis
We analyzed ROC curves with GraphPad Prism 5.0 (GraphPad Software) to determine the ZAP70, LPL, and CD38 expression cutoff values that best distinguished mutated and unmutated cases. We generated time-dependent ROC curves with the R package ROCsurvival (29). We used {chi}2 Pearson statistics (with the Yates continuity correction for 2 x 2 tables) to describe associations between clinical markers and used Cramer’s V statistic to quantify the strength of association between 2 variables (information unobtainable from the P value). Values of 0.20–0.35 indicate a moderate relation, 0.36–0.49 indicate a substantial relation, and values ≥0.50 a strong relation. We plotted OS and TFS distributions with the Kaplan–Meier method and used the log-rank test with GraphPad Prism 5.0 to compare the distributions. Univariate and multivariate Cox regression analyses evaluated the effects of the different prognostic variables on TFS and/or OS. We included clinical stage [early (Binet A) vs advanced (Binet B and C)], CD38 expression (≥7% vs <7%), FC-measured ZAP70 expression (≥20% vs <20%; we obtained a 21% cutoff with ROC curve analysis but kept the cutoff commonly described in the literature), qPCR measurement of ZAP70 mRNA (cutoff of 115-fold that of ZAP70 mRNA expression in the Namalwa calibrator line), and LPL mRNA expression (cutoff of 6-fold that of LPL mRNA expression in the Namalwa line) as potential prognostic factors in our analysis. All tests were 2-sided. A P value <0.05 was considered statistically significant. All analyses were performed with SPSS 13.0 software.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
association between binet stage and ighv mutational status
Cross-tabulations showed that mutational status was significantly associated with Binet stage (P = 0.004), treatment status (P <0.0001), and being alive or dead (P = 0.004) (Table 1Up ). Cramer’s V values for association strength were 0.33 (moderate), 0.40 (substantial relation), and 0.31 (moderate), respectively. We also evaluated TFS with respect to Binet stage (Kaplan–Meier data not shown). Patients presented median TFS times of 88 months, 29 months, and 19 months for stages A, B, and C, respectively (P <0.0001).

strong association of QPCR-measured zap70 expression with mutational status
Table 1Up and Fig. 1 summarize cross-tabulations for mutational status and other prognostic factors. Mutational status was significantly associated with all markers tested. The value for Cramer’s V statistic (0.72) indicated a very strong relation between qPCR-measured ZAP70 expression and mutational status; the other markers showed a less good relation (0.33 to 0.56) (Table 1Up ). Table 1 in the online Data Supplement presents rates of concordance with mutational status and association by cross-tabulation. To estimate the powers of these markers and their ability to correctly predict mutational status, we evaluated sensitivity, specificity, and positive and negative predictive values (see Table 2 in the online Data Supplement). These values for qPCR-measured ZAP70 expression were 87.8%, 85.7%, 87.5%, and 86%, respectively, and were better than for the other markers. Moreover, we compared the areas under the ROC curve (AUCs) with a nonparametric statistical test (30) (Fig. 2A ). The AUC reflects the probability of correctly discriminating between true-positive and true-negative findings. The AUC for qPCR-measured ZAP70 expression is significantly different from the AUCs for LPL expression (P = 0.014) and CD38 expression (P = 0.007) but not significantly different from FC-measured ZAP70 expression (P = 0.424), indicating that ZAP70 expression is a better marker for predicting mutational status. Moreover, the AUCs for FC-measured ZAP70 expression and LPL expression are not significantly different (P = 0.165), indicating that these 2 methods are globally good predictors of mutational status, independently of the chosen cutoff value. With respect to the optimal cutoff, however, qPCR-measured ZAP70 expression gives better results.


Figure 1
View larger version (20K):
[in this window]
[in a new window]

  		Figure 1. Correlations between IGHV mutational status and other prognostic factors.

(A and B), ZAP70 expression was measured by qPCR and FC; (C), LPL expression was measured by qPCR; and (D), CD38 expression was measured by FC. Optimal cutoff values were based on ROC curve analyses and are indicated by a horizontal line. Closed circles represent CLL-related deaths.


Figure 2
View larger version (20K):
[in this window]
[in a new window]

  		Figure 2. ROC curve analysis of the different prognostic factors vs IGHV mutational status and time-dependent ROC curves.

(A), AUC values are in Table 2Up . AUC values calculated for all markers were compared with a nonparametric test. The qPCR-measured ZAP70 AUC was significantly different from the LPL AUC (P = 0.014) and the CD38 AUC (P = 0.007). The AUC for FC-measured ZAP70 was not significantly different from the AUCs for qPCR-measured ZAP70 and LPL (P = 0.424 and 0.165, respectively). (B), time-dependent ROC curves at 1 year after diagnosis. (C) time-dependent ROC curves at 2 years after diagnosis.

prognostic value of qpcr-measured zap70 expression

tfs and os analyses
Need for treatment and patient death are clearly associated with both IGHV mutational status and qPCR-measured ZAP70 expression (Table 1Up ). The median TFS times for qPCR-measured ZAP70+ and ZAP70– patients were 24 months and 157 months, respectively (P <0.0001). Moreover, OS was significantly associated in log-rank tests with IGHV mutational status (P = 0.0034) and qPCR- and FC-measured analyses of ZAP70 expression (P = 0.0021 and 0.0006, respectively), but not with LPL or CD38 expression (P = 0.1972 and 0.2267). Table 3 in the online Data Supplement and Fig. 3 summarize the effects of other prognostic factors on TFS and OS.


Figure 3
View larger version (23K):
[in this window]
[in a new window]

  		Figure 3. Kaplan–Meier survival curves for TFS and OS.

(A and F), indicated are TFS and OS curves for IGHV mutational status; (B and G), ZAP70 by qPCR; (C and H), ZAP70 by FC; (D and I), LPL by qPCR; and (E and J), CD38 by FC. Assessment of mutational status was based on a 98% cutoff value (n = 98), ROC curve analyses were used to determine cutoffs for ZAP70 (n = 101) and LPL (n = 101) in qPCR measurements of mRNA in purified B cells, and 20% and 7% cutoffs for positive cells were used in FC analyses of ZAP70 expression (n = 87) and CD38 expression (n = 82), respectively.

patients exhibiting discordance between ighv mutational status and other prognostic factors
We also investigated TFS for patients who presented discordance between mutational status and other prognostic factors (see Table 4 in the online Data Supplement). We evaluated TFS for each marker and compared the 2 groups with a log-rank test. We observed a significant difference in TFS only for qPCR-measured ZAP70 expression. Unmutated IGHV/ZAP70– and mutated IGHV/ZAP70+ patients have median TFS times of 178 months and 67 months, respectively (P = 0.0395). To evaluate whether qPCR-measured ZAP70 expression is the predominant prognostic factor for TFS, we also compared patients with discordance between qPCR-measured ZAP70 expression and other prognostic markers. Although there were no significant differences because of the small number of patients, patients evaluated as ZAP70+ by qPCR had an apparently shorter median TFS time (see Table 4 in the online Data Supplement).

univariate and multivariate cox regression
We used univariate Cox regression to evaluate the impact of the binarized data (using Table 1Up cutoffs). All of the tested markers were significant univariate predictors of TFS, but IGHV mutational status and qPCR- and FC-measured ZAP70 expression were the only significant predictors of OS (P = 0.011, 0.008, and 0.004, respectively; see Table 5 in the online Data Supplement). A multivariate Cox regression analysis that included ZAP70 measurement (either by qPCR or by FC; because the 2 variables are highly correlated, fitting of the Cox model may become unstable if both are used), qPCR-measured LPL expression, mutational status, and CD38 expression also indicated that ZAP70 measurement (by qPCR, P = 0.0209; by FC, P = 0.0068) better predicts TFS than mutational status and the other markers (see Table 5 in the online Data Supplement).

time-dependent roc curves
We generated time-dependent ROC curves to evaluate the power of the tested markers at 1 and 2 years after diagnosis. The AUC for ZAP70 expression (by either method) was higher than for any of the other prognostic factors, including mutational status (Table 2 and Fig. 2Up , B and C).


View this table:
[in this window]
[in a new window]

  		Table 2. Summary of all analyses.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Clinical staging systems fail to identify early-stage patients most likely to experience disease progression, and this limitation has prompted a search for new prognostic markers. IGHV mutational status was recently identified as one of the most predictive factors for time from diagnosis to initial treatment. CD38, ZAP70, and LPL have been shown to have similar prognostic values and have been considered surrogate markers of mutational status (8)(11)(12)(13)(31). The aim of this study was to develop a qPCR assay for ZAP70 expression in B cells isolated from CLL patients. We also evaluated IGHV mutational status, LPL mRNA expression, CD38 expression, and FC-measured ZAP70 expression in our cohort of 108 patients (median follow-up, 82 months). Because the initial method for measuring ZAP70 expression by FC was promising but imprecise, we also investigated the use of a qPCR assay. Furthermore, FC measurement of ZAP70 expression requires cytoplasmic staining and an internal control, such as T cells or natural killer cells, to evaluate the positivity limit. These 2 variables plus antibody choice, the conjugated fluorochrome, and the subjective choice of the positivity limit can influence the final result and may lead to incorrect patient classification. The requirement of a standardized methodology therefore led us to develop a standardized qPCR assay with an absolute positivity limit, which was normalized with a stable endogenous control tested among 5 housekeeping genes and expressed as the fold expression of ZAP70 over that in the calibrator Namalwa cell line.

We also evaluated LPL and CD38 expression because these markers have been associated with more aggressive disease and have been correlated with IGHV status (11)(32)(33). qPCR measurement of ZAP70 mRNA shows sensitivity, specificity, and positive and negative predictive values that better predict mutational status than the other markers tested. Similarly with other markers, qPCR-measured ZAP70 expression was associated with IGHV mutational status but with a higher concordance rate (86%). AUC analysis indicated that measurement of ZAP70 expression by either method is an appreciably better predictor of IGHV status than LPL or CD38 expression. Evaluation of Cramer’s V statistic, a measure of the strength of these associations, indicated a very strong relation between mutational status and qPCR-measured ZAP70 expression and indicated substantial to strong relationships for FC-measured ZAP70 expression, LPL expression, and CD38 expression. We conclude that qPCR measurement of ZAP70 mRNA is a strong surrogate marker for IGHV mutational status and is more powerful than the other tested markers. Other researchers have also reported a clear correlation between mutational profile and ZAP70 expression, with concordance rates of 83% and 81% for qPCR and 77% for FC analysis (24)(34). The degree of concordance between ZAP70 expression and mutational status thus varied, depending on the method. We therefore calculated all cutoff values for each method independently to maximize the concordance with mutational status, the most robust biological prognostic factor. After this operation we considered all correlations optimal.

Several clinical studies have shown mutational status to be a good predictor of TFS and OS (7)(35). We confirmed these findings with our patient cohort. All markers tested were significant predictors of TFS in log-rank tests [ZAP70 expression measured by either method (P <0.0001), LPL expression (P = 0.0063), and CD38 expression (P = 0.0017)]. ZAP70+ patients by qPCR had a significantly shorter median TFS time (24 months) than ZAP70– patients (157 months). Measurement of ZAP70 expression by FC has been identified as a significant predictor of disease progression and OS in CLL (15)(19). The Kaplan–Meier estimates of the survival function for ZAP70+ and ZAP70– patients were significantly different (by qPCR, P = 0.0021; by FC, P = 0.0006), but no significant differences were apparent for LPL or CD38 expression despite the numbers of patients included in the analyses. Our results for the predictive value of CD38 expression with the 7% cutoff are in accord with those of Domingo-Domenech et al. (36). The 7% threshold is apparently not the best cutoff for identifying patients with a poor outcome, given that the use of 20% and 30% cutoff thresholds have shown higher survival rates for CD38– patients than for CD38+ patients (8). In our study, however, even CD38 expression with a 20% or 30% cutoff was unable to predict OS. Furthermore, CD38 expression may vary over time (37). Recent studies have suggested that IGHV mutations and CD38 expression are independent prognostic factors (38), with CD38 expression probably reflecting the disease’s proliferative potential (39).

A recent study demonstrated that LPL expression was a predictor of CLL survival and also more reliable than ZAP70 expression for predicting mutational status (11); however, the median follow-up in this study was 17 months (range, 0–57 months). Our results are in line with those of Heintel et al. (12), who reported high LPL expression to be significantly associated with a shorter TFS time; however, the median OS times of the 2 groups were not significantly different. The median follow-up time in the Heintel et al. (12) study was 48 months, compared with 82 months (range, 8–299 months) for our cohort. In contrast, we found the difference in OS survival times for ZAP70 expression to be highly significant (P = 0.0021). The median OS time for ZAP70+ patients was 12.7 years, which is similar to previously reported findings (14)(16)(18). Only 3 studies have demonstrated a correlation between ZAP70 expression and OS; the other studies evaluated ZAP70’s prognostic value only in terms of TFS.

Univariate Cox regression analysis indicated that only IGHV mutational status and ZAP70 expression evaluated by either method were good predictors of OS. In the multivariate analysis, only ZAP70 expression was a significant independent factor for predicting TFS. These data suggest that ZAP70 expression is the best of the prognostic markers tested and that the qPCR method can offset the limitations of FC.

Regarding the patients who displayed discordance (i.e., unmutated IGHV and negative for qPCR- or FC-measured ZAP70, LPL expression, or CD38 expression), only qPCR-measured ZAP70 status discordant for mutational status was able to predict TFS (P = 0.0395). We thus conclude that ZAP70 expression is the strongest predictor of the need for treatment and that ZAP70 expression is a better predictor than IGHV mutational status. When we plotted discordant cases for qPCR- and FC-measured ZAP70 expression by the Kaplan–Meier method, the qPCR+/FC– TFS curve showed a clear trend toward a shorter TFS time (median TFS, 57 months) compared with the qPCR–/FC+ TFS curves (median TFS, 80 months), but the difference was not significant (P = 0.23) because of the small number of patients (n = 11) with discordance.

Furthermore, time-dependent ROC curves and values for Cramer’s V statistic confirmed the superior clinical impact of ZAP70 compared with mutational status and other tested markers. These results agree with Del Principe et al. (15), who reported that FC measurement of ZAP70 protein better predicts outcome than mutational status or CD38 expression. On the contrary, others have found LPL to be more reliable than ZAP70 for predicting mutational status (11) and survival or to be at least as powerful as ZAP70 (32). However, when we analyzed discordant cases for ZAP70 and LPL with the Kaplan–Meier method, the TFS curve showed a clear trend to shorter TFS in the ZAP70+/LPL– group but the difference was not statistically significant (126 months vs 157 months, P = 0.056), probably because of the small sample size (n = 25).

We conclude that ZAP70 is the most powerful prognostic marker among those tested (Table 2Up ). The choice of ZAP70 method is more complicated, but qPCR-measured ZAP70 is strongly associated with mutational status, prevails over mutational status in discordant cases, and clearly trends to prevailing in cases of discordance with FC-measured ZAP70 expression or LPL expression (see Table 4 in the online Data Supplement). Moreover, the qPCR method is more accurate than the FC method.

In conclusion, we have demonstrated that quantifying ZAP70 mRNA in B cells by real-time PCR is a strong surrogate marker of IGHV mutational status and that this marker is highly associated with TFS and OS. This straightforward and standardized assay can be routinely used in laboratories to better evaluate the outcomes and therapeutic needs of CLL patients.


   Acknowledgments
 
Grant/funding support: This work was financed by F.R.I.A. grant (Fond de Recherche pour l’Industrie et l’Agriculture) and the Télévie fund, both of which are affiliated with the F.R.S-F.N.R.S. (Fonds de la Recherche Scientifique – FNRS).

Financial disclosures: None declared.

Acknowledgments: We thank Hakim El Housni, Eric Van Den Neste, Pascale Saussoy, and Bassam Badran for their help.


   Footnotes
 
1 Nonstandard abbreviations: CLL, chronic lymphocytic leukemia; OS, overall survival; IGHV, immunoglobulin variable heavy chain region; ZAP70, zeta-chain (TCR)-associated protein kinase; FC, flow cytometry; qPCR, quantitative real-time reverse transcription-PCR; TFS, treatment-free survival; AUC, area under the ROC curve.

2 Human genes: LPL, lipoprotein lipase; ZAP70, zeta-chain (TCR) associated protein kinase 70 kDa; PPIA, peptidylprolyl isomerase A (cyclophilin A); LMNB1, lamin B1; EIF1AX, eukaryotic translation initiation factor 1A, X-linked; CASC3, cancer susceptibility candidate 3; IGHV, immunoglobulin variable heavy chain region; PGK1, phosphoglycerate kinase 1; CD38, CD38 molecule.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Zwiebel JA, Cheson BD. Chronic lymphocytic leukemia: staging and prognostic factors. Semin Oncol 1998;25:42-59.[ISI][Medline] [Order article via Infotrieve]
  2. Hamblin T. Chronic lymphocytic leukaemia: one disease or two?. Ann Hematol 2002;81:299-303.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  3. Rai KR, Sawitsky A, Cronkite EP, Chanana AD, Levy RN, Pasternack BS. Clinical staging of chronic lymphocytic leukemia. Blood 1975;46:219-234.[Abstract/Free Full Text]
  4. Binet JL, Auquier A, Dighiero G, Chastang C, Piguet H, Goasguen J, et al. A new prognostic classification of chronic lymphocytic leukemia derived from a multivariate survival analysis. Cancer 1981;48:198-206.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  5. Keating MJ, Chiorazzi N, Messmer B, Damle RN, Allen SL, Rai KR, et al. Biology and treatment of chronic lymphocytic leukemia. Hematology Am Soc Hematol Educ Program 2003;:153-175.
  6. Shanafelt TD, Geyer SM, Kay NE. Prognosis at diagnosis: integrating molecular biologic insights into clinical practice for patients with CLL. Blood 2004;103:1202-1210.[Abstract/Free Full Text]
  7. Hamblin TJ, Davis Z, Gardiner A, Oscier DG, Stevenson FK. Unmutated Ig V(H) genes are associated with a more aggressive form of chronic lymphocytic leukemia. Blood 1999;94:1848-1854.[Abstract/Free Full Text]
  8. Damle RN, Wasil T, Fais F, Ghiotto F, Valetto A, Allen SL, et al. Ig V gene mutation status and CD38 expression as novel prognostic indicators in chronic lymphocytic leukemia. Blood 1999;94:1840-1847.[Abstract/Free Full Text]
  9. Rosenwald A, Alizadeh AA, Widhopf G, Simon R, Davis RE, Yu X, et al. Relation of gene expression phenotype to immunoglobulin mutation genotype in B cell chronic lymphocytic leukemia. J Exp Med 2001;194:1639-1647.[Abstract/Free Full Text]
  10. Klein U, Tu Y, Stolovitzky GA, Mattioli M, Cattoretti G, Husson H, et al. Gene expression profiling of B cell chronic lymphocytic leukemia reveals a homogeneous phenotype related to memory B cells. J Exp Med 2001;194:1625-1638.[Abstract/Free Full Text]
  11. van’t Veer MB, Brooijmans AM, Langerak AW, Verhaaf B, Goudswaard CS, Graveland WJ, et al. The predictive value of lipoprotein lipase for survival in chronic lymphocytic leukemia. Haematologica 2006;91:56-63.[ISI][Medline] [Order article via Infotrieve]
  12. Heintel D, Kienle D, Shehata M, Krober A, Kroemer E, Schwarzinger I, et al. High expression of lipoprotein lipase in poor risk B-cell chronic lymphocytic leukemia. Leukemia 2005;19:1216-1223.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  13. Wiestner A, Rosenwald A, Barry TS, Wright G, Davis RE, Henrickson SE, et al. ZAP-70 expression identifies a chronic lymphocytic leukemia subtype with unmutated immunoglobulin genes, inferior clinical outcome, and distinct gene expression profile. Blood 2003;101:4944-4951.[Abstract/Free Full Text]
  14. Zanotti R, Ambrosetti A, Lestani M, Ghia P, Pattaro C, Remo A, et al. ZAP-70 expression, as detected by immunohistochemistry on bone marrow biopsies from early-phase CLL patients, is a strong adverse prognostic factor. Leukemia 2007;21:102-109.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  15. Del Principe MI, Del Poeta G, Buccisano F, Maurillo L, Venditti A, Zucchetto A, et al. Clinical significance of ZAP-70 protein expression in B-cell chronic lymphocytic leukemia. Blood 2006;108:853-861.[Abstract/Free Full Text]
  16. Crespo M, Bosch F, Villamor N, Bellosillo B, Colomer D, Rozman M, et al. ZAP-70 expression as a surrogate for immunoglobulin-variable-region mutations in chronic lymphocytic leukemia. N Engl J Med 2003;348:1764-1775.[Abstract/Free Full Text]
  17. Durig J, Nuckel H, Cremer M, Fuhrer A, Halfmeyer K, Fandrey J, et al. ZAP-70 expression is a prognostic factor in chronic lymphocytic leukemia. Leukemia 2003;17:2426-2434.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  18. Orchard JA, Ibbotson RE, Davis Z, Wiestner A, Rosenwald A, Thomas PW, et al. ZAP-70 expression and prognosis in chronic lymphocytic leukaemia. Lancet 2004;363:105-111.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  19. Rassenti LZ, Huynh L, Toy TL, Chen L, Keating MJ, Gribben JG, et al. ZAP-70 compared with immunoglobulin heavy-chain gene mutation status as a predictor of disease progression in chronic lymphocytic leukemia. N Engl J Med 2004;351:893-901.[Abstract/Free Full Text]
  20. Vener C, Gianelli U, Cortelezzi A, Fracchiolla NS, Somalvico F, Savi F, et al. ZAP-70 immunoreactivity is a prognostic marker of disease progression in chronic lymphocytic leukemia. Leuk Lymphoma 2006;47:245-251.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  21. Chen YH, Peterson LC, Dittmann D, Evens A, Rosen S, Khoong A, et al. Comparative analysis of flow cytometric techniques in assessment of ZAP-70 expression in relation to lgVH mutational status in chronic lymphocytic leukemia. Am J Clin Pathol 2007;127:182-191.[ISI][Medline] [Order article via Infotrieve]
  22. Gibbs G, Bromidge T, Howe D, Hopkins J, Johnson S. Comparison of flow cytometric methods for the measurement of ZAP-70 expression in a routine diagnostic laboratory. Clin Lab Haematol 2005;27:258-266.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  23. Catherwood MA, Matthews C, Niblock R, Dobbin E, Morris TC, Alexander HD. ZAP-70 mRNA quantification in B-cell chronic lymphocytic leukaemia. Eur J Haematol 2006;76:294-298.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  24. Van Bockstaele F, Janssens A, Piette A, Callewaert F, Pede V, Offner F, et al. Kolmogorov-Smirnov statistical test for analysis of ZAP-70 expression in B-CLL, compared with quantitative PCR and IgV(H) mutation status. Cytometry B Clin Cytom 2006;70:302-308.[Medline] [Order article via Infotrieve]
  25. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156-159.[ISI][Medline] [Order article via Infotrieve]
  26. Kienle DL, Korz C, Hosch B, Benner A, Mertens D, Habermann A, et al. Evidence for distinct pathomechanisms in genetic subgroups of chronic lymphocytic leukemia revealed by quantitative expression analysis of cell cycle, activation, and apoptosis-associated genes. J Clin Oncol 2005;23:3780-3792.[Abstract/Free Full Text]
  27. Mertens D, Wolf S, Schroeter P, Schaffner C, Dohner H, Stilgenbauer S, et al. Down-regulation of candidate tumor suppressor genes within chromosome band 13q14.3 is independent of the DNA methylation pattern in B-cell chronic lymphocytic leukemia. Blood 2002;99:4116-4121.[Abstract/Free Full Text]
  28. Li AH, Rosenquist R, Forestier E, Holmberg D, Lindh J, Lofvenberg E, et al. Clonal rearrangements in childhood and adult precursor B acute lymphoblastic leukemia: a comparative polymerase chain reaction study using multiple sets of primers. Eur J Haematol 1999;63:211-218.[ISI][Medline] [Order article via Infotrieve]
  29. Heagerty PJ, Lumley T, Pepe MS. Time-dependent ROC curves for censored survival data and a diagnostic marker. Biometrics 2000;56:337-344.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  30. Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology 1983;148:839-843.[Abstract/Free Full Text]
  31. Zucchetto A, Bomben R, Bo MD, Nanni P, Bulian P, Rossi FM, et al. ZAP-70 expression in B-cell chronic lymphocytic leukemia: evaluation by external (isotypic) or internal (T/NK cells) controls and correlation with IgV(H) mutations. Cytometry B Clin Cytom 2006;70:284-292.[Medline] [Order article via Infotrieve]
  32. Van Bockstaele F, Pede V, Janssens A, Callewaert F, Offner F, Verhasselt B, et al. Lipoprotein lipase mRNA expression in whole blood is a prognostic marker in B cell chronic lymphocytic leukemia. Clin Chem 2007;53:204-212.[Abstract/Free Full Text]
  33. Boonstra JG, van Lom K, Langerak AW, Graveland WJ, Valk PJ, Kraan J, et al. CD38 as a prognostic factor in B cell chronic lymphocytic leukaemia (B-CLL): comparison of three approaches to analyze its expression. Cytometry B Clin Cytom 2006;70:136-141.[Medline] [Order article via Infotrieve]
  34. Sheikholeslami MR, Jilani I, Keating M, Uyeji J, Chen K, Kantarjian H, et al. Variations in the detection of ZAP-70 in chronic lymphocytic leukemia: comparison with IgV(H) mutation analysis. Cytometry B Clin Cytom 2006;70:270-275.[Medline] [Order article via Infotrieve]
  35. Krober A, Seiler T, Benner A, Bullinger L, Bruckle E, Lichter P, et al. V(H) mutation status, CD38 expression level, genomic aberrations, and survival in chronic lymphocytic leukemia. Blood 2002;100:1410-1416.[Abstract/Free Full Text]
  36. Domingo-Domenech E, Domingo-Claros A, Gonzalez-Barca E, Beneitez D, Alonso E, Romagosa V, et al. CD38 expression in B-chronic lymphocytic leukemia: association with clinical presentation and outcome in 155 patients. Haematologica 2002;87:1021-1027.[ISI][Medline] [Order article via Infotrieve]
  37. Montillo M, Hamblin T, Hallek M, Montserrat E, Morra E. Chronic lymphocytic leukemia: novel prognostic factors and their relevance for risk-adapted therapeutic strategies. Haematologica 2005;90:391-399.[Abstract/Free Full Text]
  38. Hamblin TJ, Orchard JA, Ibbotson RE, Davis Z, Thomas PW, Stevenson FK, et al. CD38 expression and immunoglobulin variable region mutations are independent prognostic variables in chronic lymphocytic leukemia, but CD38 expression may vary during the course of the disease. Blood 2002;99:1023-1029.[Abstract/Free Full Text]
  39. Patten P, Buggins A, Richards J, Wotherspoon A, Hamblin TJ, Mufti G, et al. CD38 expression in B-CLL is dynamic and changes due to contact with activating stimuli found within the leukaemic microenvironment [Abstract]. Annual Meeting of the American Society of Hematology, 2006 Orlando, FL. http://health.groups.yahoo.com/group/CLLResearch/message/8776 (accessed May 2007)..



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


Home page
 haematolHome page
A. W. Hauswirth and U. Jager
Impact of cytogenetic and molecular prognostic markers on the clinical management of chronic lymphocytic leukemia
Haematologica, January 1, 2008; 93(1): 14 - 19.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow 089326.Supplemental Data
Right arrow All Versions of this Article:
clinchem.2007.089326v1
53/10/1757    most recent
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 (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stamatopoulos, B.
Right arrow Articles by Lagneaux, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stamatopoulos, B.
Right arrow Articles by Lagneaux, L.
Related Collections
Right arrow Molecular Diagnostics and Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS