ORIGINAL ARTICLE. Keywords B-CLL. ZAP-70. CD38. IgV H mutational status. Introduction

Transcription

1 Ann Hematol (2006) 85: DOI /s ORIGINAL ARTICLE Feasibility of an easily applicable method of ZAP-70 measurement in chronic lymphocytic leukemia in the routine flow cytometry setting: a methodological approach Freda Passam & Varvara Tachynopoulou & Dimitra Skoumi & Aliki Tsompanakou & Aikaterini Stavropoulos-Giokas & Chrysanthi Vadikolia & Achilles Anagnostopoulos & Georgios Paterakis Received: 20 February 2006 / Accepted: 1 June 2006 / Published online: 27 July 2006 # Springer-Verlag 2006 Abstract Zeta-associated protein 70 (ZAP-70), determined by flow cytometry, has been advocated a surrogate marker of immunoglobulin (Ig)V H unmutated status in B chronic lymphocytic leukemia (CLL). The aim of the current study was to test the applicability of an easy flow cytometry protocol for ZAP-70 measurement in CLL samples. Samples from 61 CLL patients and 44 normal subjects were analyzed using a commercial ZAP-70 monoclonal antibody (1E7.2 clone) conjugated with phycoerythrin (PE) and Alexa 488 fluorochromes. Modifications of the published methods led to the structure of a simplified inhouse method of ZAP-70 measurement. A three-color approach was used with CD19, CD3 gating comparing with the isotype control provided by the same manufacturer. The cutoff levels for ZAP-70 positivity were defined from a receiver operator characteristic curve in relation to the IgV H mutational status and from the ln normalized mean value +2 SD of normal controls. Using the 20% cutoff value for ZAP-70 positivity in CLL patients defined F. Passam : D. Skoumi : A. Stavropoulos-Giokas : G. Paterakis Immunology Department and National Histocompatibility Centre, G Gennimatas General Hospital, Athens, Greece V. Tachynopoulou : A. Tsompanakou : C. Vadikolia : A. Anagnostopoulos Hematology Department and Hematopoietic Cell Transplantation Unit, G. Papanicolaou Hospital, Thessaloniki, Greece G. Paterakis (*) 21 Leonidou Str, Athens, Greece pateraki@otenet.gr by the literature, ZAP-PE had a sensitivity of 55% and a specificity of 98% in predicting the IgV H mutational status, whereas the corresponding values for ZAP-Alexa were 55% and 84%, respectively. Using the 7% cutoff value for CD38 positivity, the sensitivity was 55%, whereas the specificity was 76%. ZAP-70-positive patients showed a shorter time to disease progression in comparison with ZAP-70-negative patients (p<0.001). In conclusion, the 100% specific prediction of mutational status is accompanied by reduced sensitivity, thus limiting ZAP-70 s applicability either as a single marker or combined with CD38 for the assessment of the mutational status of CLL. Keywords B-CLL. ZAP-70. CD38. IgV H mutational status Introduction Chronic lymphocytic leukemia (CLL) is a chronic lymphoproliferative disorder of variable clinical course. Phenotypic markers such as CD38 [1, 2] and cytogenetic abnormalities [3] have been evaluated as prognostic factors in CLL. However, one of the most robust markers associated with aggressive CLL disease is the absence of somatic mutations in rearranged immunoglobulin (Ig)V H genes [4, 5]. Zeta-associated protein 70 (ZAP-70) is a new prognostic factor for CLL and is considered a surrogate marker of the IgV H unmutated status [6 11]. ZAP-70 is a tyrosine kinase linked to the T-cell receptor, initially believed to be restricted to T and natural killer (NK) cells. B cells use the protein kinase syk for signal transduction via the B-cell receptor (BCR) [12]. The initial association of ZAP-70 mrna overexpression with IgV H unmutated status was an

2 796 Ann Hematol (2006) 85: observation derived from microarray analysis with unknown biological meaning [13]. Nevertheless, recently, various studies have ventured to investigate the physiological role of ZAP-70 in CLL B cells showing an altered response to BCR ligation in CLL cases with high ZAP-70 expression [14 21]. On clinical grounds, the aforementioned studies [7 11] have examined ZAP-70 expression in CLL samples using flow cytometry (FC) as the principal method of determination. However, to date, the implementation of a standardized FC method for the determination of ZAP-70 in the routine setting has been hampered by methodological and analytical variability [22, 23]. The methodological variability derives from several issues, including (1) sample [whole blood vs isolated peripheral blood mononuclear cells (PBMNCs) vs reconstituted cryopreserved cells], (2) time of preparation (fresh sample or after a storage interval), (3) staining protocol, (4) permeabilization and fixation reagents, (5) clone of anti-zap-70 antibody, (6) other monoclonal antibodies combined to ZAP-70 in three or four color combinations, and (7) use of an isotype control. The analytical variability has also been considerable. The result is expressed by the percentage of ZAP-70- positive B cells. However, there are differences in the assignment of the positive cells. Crespo et al. [7] used the cluster of ZAP-70-positive T cells to assign a threshold for positive B cells, whereas Orchard et al. [9] defined the positive region referring to an isotype control. The objectives of the current study were (1) to develop and evaluate an easily applicable protocol of ZAP-70 staining using the common antibody clone 1E.72 with two fluorochrome conjugates (Alexa 488 or phycoerythrin), (2) to assess reference ZAP-70 values in a group of healthy individuals, (3) to validate the specificity and sensitivity of ZAP-70 as a surrogate measurement in predicting the IgV H mutational status, and (4) to examine whether the combined use of ZAP-70 and CD38 can be superior to ZAP-70 alone. Materials and methods Patients The study population consisted of 61 CLL patients and 44 normal subjects. Patient characteristics are given in Table 1. The study was carried out in two separate FC laboratories incorporated in tertiary hospitals. CLL diagnosis was based on the finding of absolute lymphocytosis in peripheral blood (>5,000/μl) plus immunophenotypical features according to the criteria of Matutes et al. [24]. All patients and healthy controls gave informed consent, and the study was approved by the Ethical Committees of the associated Institutes. Immunoglobulin V H mutational status was known in all patients. However, the flow cytometer operators were blinded to the mutational status of the patients to obtain unbiased reporting of FC results. Using the 98% cutoff for homology to germline, 43 patients carried mutated IgV H genes, whereas 18 carried unmutated IgV H genes. The mean follow-up of the patients was 33 months. Fresh blood samples were used for 24 CLL patients and the normal subjects. Thawed isolated mononuclear cells (by Ficoll technique) were used for 37 CLL patients. For 20 patients, frozen and fresh samples were available. Materials The monoclonal antibodies used in the present study were ZAP-PE (clone 1E7.2, Caltag Laboratories, Burlingame, CA, USA), CD3 fluorescein isothiocyanate (FITC) (clone UCHT1, DakoCytomation, Glostrup, Denmark), CD19- RPECy5 (clone HD37, DakoCytomation), and ZAP-Alexa 488 (clone 1E7.2, Caltag). Matching mouse IgG1 isotype controls for ZAP were used with conjugated PE and Alexa, respectively (Caltag). Three mouse antihuman IgG1 monoclonal antibodies were used comparatively for CD38 and their corresponding isotype controls: CD38-RPE clone T16 (Beckman-Coulter, Immunotech, Marseille, France), CD38- RPE clone AT13/5 (DakoCytomation), and CD38-PE clone HB7 (Becton-Dickinson, Franklin Lakes, NJ, USA). We used three commercial fixation-permeabilization reagent kits: Fix and Perm, cell permeabilization reagents, Caltag), IntraPrep (Beckman-Coulter), and Leukoperm (Serotec Ltd., Oxford, UK). Methods Initially, the published methods of ZAP-70 determination by FC were applied in our laboratories [7 9]. Modifications were then introduced for simplification of the procedure and compared with published methods. Each set of tests was performed in triplicate. This led to the structure of an inhouse protocol, which was implemented in CLL samples with known mutational status and normal controls. The modified protocol for the measurement of ZAP-70 is summarized in Table 2. For CD38 staining, cells were stained for CD38 and CD19 for 15 min. Red blood cells (RBCs) were then lysed with ammonium chloride solution and proceeded for FC analysis. Flow cytometry analysis was performed using an Epics XL-MCL Beckman-Coulter apparatus using a three-color approach. Lymphocytes were gated according to expression of CD3 and CD19. We compared the level of expression of ZAP-70 in the CD3 and CD19 population with the corresponding isotype control. A cursor was placed in the

3 Ann Hematol (2006) 85: Table 1 Patients characteristics CD19/isotype control and CD3/isotype control quadrant plot so that the population in the positive upper right quadrant ranged between 1 and 2% (Figs. 1 and 2). The same principle was applied to the isotype control histogram gated by the CD19 subpopulation for the measurement of CD38 percentage. Statistical analysis Statistical analysis was performed using the SPSS software. Survival was plotted using the Kaplan Meier with the primary end point of disease progression. Comparisons between clinical data were performed using the χ 2 test. Differences in the rate of positivity for ZAP-70 in the T population for patients and controls were compared with the nonparametric Mann Whitney test. For the B population, receiver operator characteristic (ROC) curves were plotted to test the ZAP-70 positivity rates according to mutational status. Normal B-cell values for ZAP-70 were ln-transformed before analysis. Differences were regarded as statistically significant when p<0.05. Results IgV H mutated Standardization of ZAP-70 protocol IgV H unmutated No Age (median, min max) 60 (37 79) 64 (48 76) Male/female 27/16 11/7 Binet stage A B 7 5 C 1 2 Rai stage ZAP-70 >20% PE conjugate 1 10 Alexa conjugate 7 10 CD38 positivity (>7%) The effects of the following variables on ZAP-70 staining of the CD3-positive and CD19-positive population in normal samples were examined: 1. Various incubation times of the fixative, permeabilizer, and ZAP-70 antibody (5, 10, 30, and 60 min). It was noted that decreased incubation time with the permeabilizer (5 min) and prolonged incubation with the ZAP-70 monoclonal antibody (30 and 60 min) resulted in higher ZAP-70 staining. 2. Washing with phosphate-buffered saline (PBS) between steps 3, 5, 7, and 9 (Table 2) resulted in higher ZAP-70 staining compared with no-wash procedure between steps 3, 5, and The temperature of incubation (4 C vs room temperature) had no effect on ZAP-70 staining. 4. The washing medium (PBS vs PBS + fetal calf serum) at concentrations 2 10% had no significant effect on ZAP-70 staining. 5. For overnight storage of ZAP-70 stained samples, the percentage of CD3 staining for ZAP-70 was significantly reduced after overnight storage at 4 C. Therefore, after ZAP-70 staining, the samples should be proceeded to FC soon after preparation. 6. The three fixative and permeabilization commer cial reagents (Fix and Perm, IntraPrep, and Leukoperm) did not have a different effect on ZAP-70 staining. Table 2 Steps of proposed ZAP-70 staining protocol for CLL samples Panel FITC surface PE intracellular PeCy5 surface CD3 IgG1 isotype control CD19 CD3 ZAP-70 CD19 Alexa 488 intracellular PE surface PeCy5 surface IgG1 isotype control CD3 CD19 ZAP-70 (Alexa 488) CD3 CD19 Method 1 Whole blood, calculate volume for white blood cells 2 Addition of 10 μl surface monoclonal antibody (CD3, CD19), 15 min incubation, RT μl fixation reagent A, 10 min incubation, RT 4 Wash with PBS μl permeabilization reagent B, 5 min incubation, RT 6 Wash with PBS μg of anti-zap-70 Ab (5 μl), 30 min incubation, RT 8 Wash with PBS 1 9 Reconstitution with 500 μl PBS 10 FC analysis

4 798 Ann Hematol (2006) 85: Fig. 1 Flow cytometry results of a CLL case classified as negative with the ZAP-70 Alexa 488 fluorochrome. Gate B is on CD19-positive cells, and gate T is on CD3-positive cells. The cursor is placed for the sample with the isotype control so as to define a cutoff with 1 2% of cells lying in the positive area (W2). The cursor remains in the same position for the sample with the ZAP-70 antibody. In the current example, 7.5% of CD19 cells and 58.4% of CD3 cells stained for ZAP For cell number, the percentage of ZAP-positive B cells remained stable when samples contained cell numbers in a range of cells. 8. For separate or simultaneous incubation of permeabilizer and ZAP-70 antibody, separate incubation resulted in higher ZAP-70 staining in comparison with simultaneous incubation. 9. For Ficoll isolation of PBMNC vs whole blood, use of whole blood displayed equivalent staining to isolated PBMNC. 10. For staining of fresh vs cryopreserved samples, no difference was noted in the percentage of B ZAP-70 positivity for the samples in which both fresh and cryopreserved cells were available.

5 Ann Hematol (2006) 85: Fig. 2 Flow cytometry results of a CLL case classified as positive with ZAP-70 Alexa 488 fluorochrome. For the CD19 population, 39.1% stained in the positive area W2 (which was above the positivity threshold) 11. Dilution of ZAP-70-stained samples with PBS resulted in decreased ZAP-70 positivity. 12. The interoperator variability achieved with this simplified gating protocol according to the isotype control was less than 10%. 13. There was no significant difference in the positivity percentages of CD38 in B lymphocytes among the three clones compared (T16, AT13/5, and HB7). Implementation of ZAP-70 protocol CLL samples Compared with the isotype control, the ZAP-70 positivity rate in the CD19 population of CLL patients ranged from 1 to 80% for ZAP-PE and from 1 to 78% for ZAP-Alexa.

6 800 Ann Hematol (2006) 85: Using the ROC curve approach, ZAP-70 cutoffs for the CD19 population were calculated, which corresponded to maximum specificity and sensitivity in predicting the unmutated IgV H status. The terminology used of a ZAP- 70 as positive corresponds to a percentage of CD19 cells above a threshold, which is expected to correlate with the unmutated IgV H status. The same approach was applied for defining an optimal cutoff for CD38 positivity. We used the cutoff limit for CD38 of 7% [25 27] and 30% [11] according to the literature. Using the 7% cutoff, 20 patients were considered as CD38-positive (10/20 were IgV H unmutated). Using the 30% cutoff, ten patients were considered as CD38-positive (6/10 were IgV H unmutated). The results are shown in Table 3. According to the ROC curves, a cutoff for CD19 ZAP-PE of 26% and a cutoff for ZAP-Alexa of 33% would have to apply for 100% specificity, whereas the limit of CD38 would have to be 75%. The respective sensitivities would then be 39, 39, and 11% (Fig. 3). Using the literature reference cutoff value of 20%, there was one false-positive case (ZAP-70 positive /IgV H mutated) for ZAP-PE and seven falsepositive cases for ZAP-Alexa. On the other hand, with the same cutoff, eight cases were false-negative (ZAP-70 negative /IgV H unmutated) for ZAP-PE and eight falsenegative cases for ZAP-Alexa. As an overall percentage, discordant ZAP-70 results with the mutational status were notedin15and25%ofpatientsforzap-peandzap- Alexa, respectively. The results of the combined analysis of CD38 vs ZAP are shown in Fig. 4, taking into account the cutoff values proposed in the literature and those that were estimated in this study. It is shown that the combined assessment of the Table 3 The calculated sensitivity and specificity derived from the ROC analysis of CLL ZAP-70 and CD38 results in predicting the mutational status Method ZAP-70 PE (%) ZAP-70 Alexa 488 (%) CD38 (%) Cutoff (literature) Sensitivity Specificity Cutoff (best fit from ROC curve) Sensitivity Specificity Cutoff (for 100% specificity) Sensitivity Specificity The variable sensitivities and specificities achieved according to the cutoff levels are depicted two parameters is not superior to ZAP-70 alone due to the low specificity of CD38. Regarding the results of ZAP-70 in the CD3 population, the positivity rate in CLL patients had a median (min max) of 63% (15 98) for ZAP-PE, 74% (7 98) for ZAP-Alexa, which were not significantly different from values found in the control group [corresponding values, 65% (16 94) for ZAP-PE and 81% (37 99) for ZAP-Alexa)]. Regarding the expression of ZAP-70 in the CD3 population, comparing the mutated and unmutated group of patients, the corresponding values were not significantly different: a median of 63% in the unmutated vs 59% in the mutated group for ZAP-PE and 80 vs 69% for ZAP-Alexa. Normal controls For the normal controls, the range of ZAP-70 positivity for the CD19 population was from 1 to 25% for ZAP-PE and from 3 to 60% for ZAP-Alexa. As the distribution of the normal values was skewed, they were ln-transformed to achieve normality, and the ln average +2 SD was used to calculate the upper normal value for ZAP in the CD19 population. This value was 35% for ZAP-PE and 52% for ZAP-Alexa. Following this approach, five patients had abnormally high ZAP-PE expression (all were IgV H unmutated), and five had abnormally high ZAP-Alexa expression (all were IgV H unmutated). Survival analysis For the 61 CLL patients, 11 (26%) of 43 mutated patients showed disease progression, whereas 10 (55%) of 18 unmutated patients showed disease progression. The mean time to progression for mutated patients was 62.0 months, whereas the mean time to progression for unmutated patients was 10.4 months (p<0.001, Kaplan Meier survival analysis). According to our cutoff for ZAP- 70 positivity, patients positive for ZAP-PE and ZAP-Alexa (ZAP-PE >26% and ZAP-Alexa >33%) had a shorter time to progression compared with ZAP-70-negative patients (p<0.001) (Fig. 5a c). Discussion The attractive theory behind ZAP-70 measurement by FC is its alleged potential to be used as a surrogate marker of the IgV H mutational status in CLL patients. Initial clinical studies demonstrated that ZAP-70 correlated well with IgV H mutational status [6, 7]. Subsequent studies reported that ZAP-70 is a prognostic marker of CLL

7 Ann Hematol (2006) 85: Fig. 3 Receiver operator characteristic curve for ZAP-70 results in the CD19 population of lymphocytes for the two fluorochromes used (PE and Alexa 488) independent of IgV H status [9, 10]. Two recent large studies, including more than 200 patients each, showed the predictive value of the combined FC measurement of ZAP-70 and CD38. Combined positivity for ZAP-70 and CD38 defined a group of patients with inferior prognosis [11, 25]. The association with IgV H mutational status was not the primary end point of these studies. These promising results have led investigators to postulate the incorporation of ZAP-70 measurements in risk stratification for clinical trials [28, 29]. However, other investigators have expressed concerns over the utility of ZAP-70 assessment. The main concern is whether it is correct to regard ZAP-70 by FC determination as an independent prognostic marker regardless of IgV H mutational status. Not all studies are in line that ZAP-70 and IgV H are independent prognostic factors [25, 30, 31]. This uncertainty is reflected by the newest studies incorporating both CD38 and ZAP-70 rates for determination of prognosis instead of ZAP-70 only. Although our sample size was relatively small, we found that the coexistence of high CD38 and high ZAP-70 was found only in a subset of unmutated patients. The combined assessment did not appear to be a superior surrogate of mutational status compared with ZAP-70 alone due to the low specificity of CD38. In all clinical studies available to date, there is a consistent finding of a discordant group for ZAP-70 by FC and IgV H mutational status, which ranges between 7 and 29% of CLL cases [6, 22, 32]. Practically, this entails a risk of missing or misdiagnosing a case as unmutated based on the ZAP-70 measurement, which may have a detrimental effect if the result is used as a basis for treatment strategy. This has been depicted in the ROC analysis for our samples. The analysis showed that the established 20% cutoff value has a relatively low specificity (84%) for predicting IgV H unmutated status (especially for the Alexa-conjugated ZAP-70 antibody). Thus, the applicability of ZAP-70 as a single surrogate marker for IgV H mutational status in CLL is limited. The reason for the discordance in some CLL cases between ZAP-70 and IgV H mutational status is an intriguing scientific question. The first studies applying

8 802 Ann Hematol (2006) 85: Fig. 4 Correlation between ZAP-70 and CD38 according to mutational status in CLL patients. The ZAP-70 and CD38 cutoff values accepted in the literature (20 and 7%, respectively) are shown (dotted lines) comparably to those found in this study to have a 100% specificity (interrupted lines) for the prediction of the mutational status of CLL (ZAP-PE, 26%; ZAP-Alexa, 33%; and CD38, 75%). The combined assessment of ZAP-70 and CD38 does not appear to be superior to ZAP-70 alone due to the low specificity of CD38 ZAP-70 measurement by FC postulated that this discordant group was a separate subgroup of CLL patients with unknown clinical significance [7, 9]. A recent study has shown that the majority of ZAP-70-/IgV H -discordant cases have high rates of genetic high-risk features such as 11q deletion, 17p deletion, and VH3-21 usage, indicating a biologic mechanism for this subgroup which awaits elucidation [33]. However, the methodological difficulties in implementing ZAP-70 measurement by FC can also contribute to the reported discrepancy. In this context, it must be noted that there are very few published methodological studies. Since the original publication in 2003 of ZAP-70 overexpression in CLL, FC laboratories have faced difficulties in applying the method in routine analysis of CLL patients. One methodological difficulty relates to the use of indirect immunofluorescence in most reference articles [7 9]. However, companies have developed monoclonal ZAP-70 antibodies directly conjugated to fluorochromes, thus surpassing the need for indirect immunofluorescence. The current study extends previous observations that direct immunofluorescence can be used for ZAP-70 determination [10, 22, 23]. A study comparing various antibodies from different companies showed that there were differences in ZAP-70 percentage using the same methodology [23]. A second methodological issue concerns the effect of the specific conjugated fluorochrome on ZAP-70 measurement. There has been consideration regarding the use of fluorochromes, which have a large spill into a channel used to detect ZAP-70 (e.g., CD3 FITC with ZAP-PE). Using

9 Ann Hematol (2006) 85: Fig. 5 a Kaplan Meier plot of patients grouped according to mutational status. The end point was disease progression. The decreased time to progression noticed in the unmutated group (p<0.001) is in accordance with the established prognostic significance of the IgV H mutational status shown in other CLL studies. b Time to disease progression in ZAP-PE-positive (>26%) and ZAP-PE-negative (<26%) patients. c Time to disease progression in ZAP-Alexa-positive (>33%) and ZAP- Alexa-negative (<33%) patients current monoclonal antibodies, there is a continuum in the positivity of B cells and no defined clusters. This study is the first to compare two different fluorochromes directly conjugated to ZAP-70. We found that similar levels of sensitivity were achieved with ZAP-PE and ZAP-Alexa; however, ZAP-Alexa had relatively lower specificity. Therefore, both fluorochromes can be used in laboratory practice. A third methodological issue is the gating strategy for defining ZAP-70 positivity. The original study based the cutoff for B-cell positivity on the relative expression of ZAP-70 to the T- and NK-cell population, which may vary according to sample and patient [7]. Orchard et al. [9] used a cutoff based on an isotype control using indirect immunofluorescence for ZAP-70 staining. Gibbs et al. [22] and Bakke et al. [23] used a directly conjugated isotype control as in our study. This method presents the advantage of a simplified technique, which is also used in other staining procedures (e.g., CD38). Although the problems of cursor placement exists with isotype controls, this was an isotype control provided by the same company for the specific ZAP-70 monoclonal antibody and constructed with a similar fluorescence-to-protein (F/P) ratio. In any case, the intensity of ZAP positivity in B cells is low.

10 804 Ann Hematol (2006) 85: This makes analysis problematic with any type of quantification approach. Other alternative methods of analysis have suggested ZAP-70 measurement in the T cells of B-CLL patients because it is reported to be overexpressed in comparison with normal T cells or even T cells of IgV H mutated CLL cases [34, 35]. However, the significance of such a finding needs further evaluation. We could not demonstrate a significant difference in ZAP levels in the T-cell population of unmutated patients in comparison with mutated patients or controls. The comparison between the mean fluorescence intensity (MFI) of stained cells and a negative control has been used [34], and the molecules of equivalent soluble fuorochrome (MESF) based on a standard curve generated by FITC beads has also been proposed [36]. However, we are unsure if the determination of intensity offers any advantage in comparison with ZAP-70 percentages, and these techniques are not normally included in the routine of the average FC laboratory. Another problem being currently defined is the stability of ZAP-70 expression following activation or over time. It has been shown that ZAP-70 may change over time in the same patient and may be upregulated upon B activation [11, 14]. The effect of chemotherapy is still another issue to be looked into. The most important methodological details that we found useful were the necessity of processing samples soon after staining and the loss of ZAP-70 staining when diluting samples after preparation. Finally, alternative methods to FC for defining ZAP-70 have been investigated, such as immunohistochemistry [37 39], reverse transcription polymerase chain reaction (RT PCR) [32], or measuring ZAP-70 methylation status [40], which await further validation and still remain technically more demanding than FC. In conclusion, the use of ZAP-70 measurements in clinical trials is still problematic given the methodological difficulties of ZAP-70 determination by FC. A realistic approach for a routine laboratory would be to use a simple and reproducible protocol such as the one currently presented. However, one should take into account that when specific prediction of the mutational status of CLL is desired, a considerable reduction of sensitivity appears to limit the applicability of ZAP-70 as a surrogate marker for IgV H mutation status. References 1. Ibrahim S, Keating M, Do KA, O Brien S, Huh YO, Jilani I, Lerner S, Kantarjian HM, Albitar M (2001) CD38 expression as an important prognostic factor in B-cell chronic lymphocytic leukemia. Blood 98: Durig J, Naschar M, Schmucker U, Renzing-Kohler K, Holter T, Huttmann A, Duhrsen U (2002) CD38 expression is an important prognostic marker in chronic lymphocytic leukaemia. Leukemia 16: Dohner H, Stilgenbauer S, Dohner K, Bentz M, Lichter M (1999) Chromosome aberrations in B-cell chronic lymphocytic leukemia: reassessment based on molecular cytogenetic analysis. J Mol Med 77: Hamblin TJ, Davis Z, Gardiner A, Oscier DG, Stevenson FK (1999) Unmutated Ig V(H) genes are associated with a more aggressive form of chronic lymphocytic leukemia. Blood 94: Damle RN, Wasil T, Fais F, Ghiotto F, Valetto A, Allen SL, Buchbinder A, Budman D, Dittmar K, Kolitz J, Lichtman SM, Schulman P, Vinciguerra VP, Rai KR, Ferrarini M, Chiorazzi N (1999) IgV gene mutation status and CD38 expression as novel prognostic indicators in chronic lymphocytic leukemia. Blood 94: Wiestner A, Rosenwald A, Barry TS, Wright G, Davis RE, Henrickson SE, Zhao H, Ibbotson RE, Orchard JA, Davis Z, Stetler-Stevenson M, Raffeld M, Arthur DC, Marti GE, Wilson WH, Hamblin TJ, Oscier DG, Staudt LM (2003) ZAP-70 expression identifies a chronic lymphocytic leukemia subtype with unmutated immunoglobulin genes, inferior clinical outcome, and distinct gene expression profile. Blood 101: Crespo M, Bosch F, Villamor N, Bellosillo B, Colomer D, Rozman M, Marce S, Lopez-Guillermo A, Campo E, Montserrat E (2003) ZAP-70 expression as a surrogate marker for immunoglobulin-variable-region mutations in chronic lymphocytic leukemia. N Engl J Med 348: Durig J, Nuckel H, Cremer M, Fuhrer A, Halfmeyer K, Fandrey J, Moroy T, Klein-Hitpass L, Duhrsen U (2003) ZAP-70 is a prognostic factor in chronic lymphocytic leukemia. Leukemia 17: Orchard JA, Ibbotson RE, Davis Z, Wiestner A, Rosenwald A, Thomas PW, Hamblin TJ, Staudt LM, Oscier DG (2004) ZAP-70 expression and prognosis in chronic lymphocytic leukemia. Lancet 363: Rassenti LZ, Huynh L, Toy TL, Chen L, Keating MJ, Gribben JG, Neuberg DS, Flinn IW, Rai KR, Byrd JC, Kay NE, Greaves A, Weiss A, Kipps TJ (2004) ZAP-70 compared with immunoglobulin heavy-chain gene mutation status as a predictor of disease progression in chronic lymphocytic leukemia. N Engl J Med 351: Schroers R, Griesinger F, Trumper L, Haase D, Kulle B, Klein- Hitpass L, Sellmann L, Duhrsen U, Durig J (2005) Combined analysis of ZAP-70 and CD38 expression as a predictor of disease progression in B-cell chronic lymphocytic leukemia. Leukemia 19: Chan AC, Iwashima M, Turck CW, Weiss A (1992) ZAP-70: a 70 kd protein tyrosine-kinase that associates with the TCR ζ chain. Cell 71: Rosenwald A, Alizadeh AA, Widhopf G, Simon R, Davis RE, Yu X, Yang L, Pickeral OK, Rassenti LZ, Powell J, Botstein D, Byrd JC, Grever MR, Cheson BD, Chiorazzi N, Wilson WH, Kipps TJ, Brown PO, Staudt LM (2001) Relation of gene expression phenotype to immunoglobulin mutation genotype in B cell chronic lymphocytic leukemia. J Exp Med 194: Nolz JC, Tschumper RC, Pittner BT, Darce JR, Kay NE, Jelinek DF (2005) ZAP-70 is expressed by a subset of normal human B- lymphocytes displaying an activated phenotype. Leukemia 19: Schweighoffer E, Vanes L, Mathiot A, Nakamura T, Tybulewicz VL (2003) Unexpected requirement for ZAP-70 in pre-b cell development and allelic exclusion. Immunity 18: Chen L, Apgar J, Huynh L, Dicker F, Giago-McGahan T, Rassenti L, Weiss A, Kipps T (2005) ZAP-70 directly enhances IgM signaling in chronic lymphocytic leukemia. Blood 105:

11 Ann Hematol (2006) 85: Nedellec S, Renaudineau Y, Bordron A, Berthou C, Porakishvili N, Lydyard PM, Pers JO, Youinou P (2005) B cell response to surface IgM cross-linking identifies different prognostic groups of B-chronic lymphocytic leukemia patients. J Immunol 174: Castro J, Prada C, Loria O, Kamal A, Chen L, Burrows F, Kipps T (2005) ZAP-70 is a novel conditional heat shock protein 90 (Hsp90) client: inhibition of Hsp90 leads to ZAP-70 degradation, apoptosis and impaired signaling in chronic lymphocytic leukemia. Blood 106: Orchard J, Ibbotson R, Best G, Parker A, Oscier D (2005) ZAP-70 in B cell malignancies. Leuk Lymphoma 46: Richardson SJ, Matthews C, Catherwood MA, Alexander HD, Carey BS, Farrugia J, Gardiner A, Mould S, Oscier D, Copplestone JA, Prentice AG (2006) ZAP-70 expression is associated with enhanced ability to respond to migratory and survival signals in B cell chronic lymphocytic leukaemia (B-CLL). Blood 107: Hamblin AD, Hamblin TJ (2005) Functional and prognostic role of ZAP-70 in chronic lymphocytic leukaemia. Expert Opin Ther Targets 9: Gibbs G, Bromidge T, Howe D, Hopkins J, Johnson S (2005) Comparison of flow cytometric methods for the measurement of ZAP-70 expression in the routine diagnostic laboratory. Clin Lab Haematol 27: Bakke AC, Purtzer Z, Leis J, Huang J (2005) A robust ratio metric method for analysis of Zap-70 expression in chronic lymphocytic leukemia (CLL). Cytometry B Clin Cytom. DOI /cyto. b Matutes E, Owusu-Ankomah K, Morilla R, Garcia MJ, Houlihan A, Que TH, Catovsky D (1994) The immunological profile of B cell disorders and proposal for a scoring system for the diagnosis for CLL. Leukemia 8: Del Giudice I, Morilla A, Osuji N, Matutes E, Morilla R, Burford A, Maravelaki S, Owusu-Ankomah K, Swansbury J, A hern R, Brito-Babapulle V, Catovsky D (2005) Zeta-chain associated protein 70 and CD38 combined predict the time to first treatment in patients with chronic lymphocytic leukemia. Cancer 104: Ghia P, Stamatopoulos K, Belessi C, Moreno C, Stella S, Guida G, Michel A, Crespo M, Laoutaris N, Montserrat E, Anagnostopoulos A, Dighiero G, Fassas A, Caligaris-Cappio F, Davi F (2005) Geographic patterns and pathogenetic implications of IGHV gene usage in chronic lymphocytic leukemia: the lesson of the IGHV3-21 gene. Blood 105: Ghia P, Guida G, Stella S, Gottardi D, Geuna M, Strola G, Scielzo C, Caligaris-Cappio F (2003) The pattern of CD38 expression defines a distinct subset of chronic lymphocytic leukemia (CLL) patients at risk of disease progression. Blood 101: Binet JL, Caligaris-Cappio F, Catovsky D, Cheson B, Davis T, Dighiero G, Dohner H, Hallek M, Hillmen P, Keating M, Montserrat E, Kipps TG, Rai K (2005) Perspectives of the use of new diagnostic tools in the treatment of chronic lymphocytic leukemia. Blood 107: Del Poeta G, Del Principe MI, Consalvo MA, Maurillo L, Buccisano F, Venditti A, Mazzone C, Bruno A, Gianni L, Capelli G, Lo Coco F, Cantonetti M, Gattei V, Amadori S (2005) The addition of rituximab to fludarabine improves clinical outcome in untreated patients with ZAP-70-negative chronic lymphocytic leukemia. Cancer 104: Kim SZ, Chow KU, Kukoc-Zivojnov N, Boehrer S, Brieger A, Steimle-Grauer SA, Harder L, Hoelzer D, Mitrou PS, Weidmann E (2004) Expression of ZAP-70 protein correlates with disease stage in chronic lymphocytic leukemia and is associated with, but not generally restricted to, non-mutated Ig VH status. Leuk Lymphoma 45: Orchard J, Ibbotson R, Best G, Parker A, Oscier D (2005) ZAP-70 in B cell malignancies. Leuk Lymphoma 46: Laurenti L, Petlickovski A, Rumi C, Gobessi S, Piccioni P, Tarnani M, Puggioni P, Marietti S, Sica S, Leone G, Efremov DG (2005) Comparison of ZAP-70/Syk mrna levels with immunoglobulin heavy-chain gene mutation status and disease progression in chronic lymphocytic leukemia. Haematologica 90: Krober A, Bloehdorn J, Hafner S, Buhler A, Seiler T, Kienle D, Winkler D, Bangerter M, Schlenk RF, Benner A, Lichter P, Dohner H, Stilgenbauer S (2006) Additional genetic high-risk features such as 11q deletion, 17p deletion, and V3-21 usage characterize discordance of ZAP-70 and VH mutation status in chronic lymphocytic leukemia. J Clin Oncol 24: Herishanu Y, Kay S, Rogowski O, Pick M, Naparstek E, Deutsch VR, Polliack A (2005) T-cell ZAP-70 overexpression in chronic lymphocytic leukemia (CLL) correlates with CLL cell ZAP-70 levels, clinical stage and disease progression. Leukemia 19: Kiaii S, Choudhury A, Mozaffari F, Kimby E, Osterborg A, Mellstedt H (2005) Signaling molecules and cytokine production in T cells of patients with B-cell chronic lymphocytic leukemia (B-CLL): comparison of indolent and progressive disease. Med Oncol 22: Kay S, Pick M, Herishanu Y, Rogowsky O, Baron S, Naparstek E, Polliack A, Deutsch V (2004) A novel flow cytometry analysis method for the quantitation of zap-70 expression in B-CLL cells. Hematology J 5(S):S Admirand JH, Rassidakis GZ, Abruzzo LV, Valbuena JR, Jones D, Jeffrey Medeiros L (2004) Immunohistochemical detection of ZAP-70 in 341 cases of non-hodgkin and Hodgkin lymphoma. Mod Pathol 17: Wang J, Young L, Win W, Taylor CR (2005) Distribution and ZAP-70 expression of WHO lymphoma categories in Shanxi, China: a review of 447 cases using a tissue microarray technique. Appl Immunohistochem Mol Morphol 13: Vener C, Gianelli U, Cortelezzi A, Fracchiolla NS, Somalvico F, Savi F, Pasquini MC, Bosari S, Deliliers GL (2006) ZAP- 70 immunoreactivity is a prognostic marker of disease progression in chronic lymphocytic leukemia. Leuk Lymphoma 47: Corcoran M, Parker A, Orchard J, Davis Z, Wirtz M, Schmitz OJ, Oscier D (2005) ZAP-70 methylation status is associated with ZAP-70 expression status in chronic lymphocytic leukemia. Haematologica 90: