Analyses of the prognostic significance of the Epstein-Barr virus transactivator ZEBRA protein and diagnostic value of its two synthetic peptides in nasopharyngeal carcinoma

(1)

Analyses of the prognostic significance of the Epstein–Barr virus transactivator ZEBRA protein and diagnostic value of its two synthetic peptides in nasopharyngeal carcinoma

Rkia Dardari

^a,b

, Jos´e Menezes

^b

, Emmanuel Drouet

^c

, Irene Joab

^d

, Abdellatif Benider

^e

, Hanae Bakkali

^f

, Lamya Kanouni

^f

, Hassan Jouhadi

^e

, Nourredine Benjaafar

^f

,

Brahim El Gueddari

^f

, Mohammad Hassar

^a

, Meriem Khyatti

^a,∗

aPasteur Institute of Morocco, 1, Rue Abou Kacem Ez-Zahraoui, 20 100 Casablanca, Morocco

bLaboratory of Immunovirology, Viral and Immune Diseases Program, Paediatric Research Center, Sainte-Justine Hospital, University of Montreal, Montreal, Qc, Canada H3T 1C5

cUnit of Virus Host Cell Interactions (UVHCI) UMR 5233 UJF-EMBL-CNRS B.P. 181. 6, rue Jules Horowitz. F-38042 Grenoble Cedex 9, France

dMolecular genetic Institute, 27, Rue Juliette Dodu, 75 010 Paris, France

eOncology Center, CHU Averroes, Casablanca, Morocco

fNational Institute of Oncology, CHU Avicenne Rabat, Morocco

Received 18 December 2006; received in revised form 13 September 2007; accepted 28 September 2007

Abstract

Background: Although numerous serological studies have determined the diagnostic and prognostic values of Epstein–Barr virus (EBV) antibodies in adult patients with nasopharyngeal carcinoma (NPC), little data about the anti-EBV immune response in children with NPC is available.

Objectives: To examine the diagnostic value of IgG antibodies against BamHI Z Epstein–Barr replication activator (ZEBRA) protein and two related synthetic peptides (Zp125 and Zp130). To compare the prognostic value of IgA antibodies against early antigens (EA) and viral capsid antigen (VCA), and IgG antibodies against ZEBRA protein, of Moroccan children treated for NPC with their prognostic value for young and adult NPC patients.

Study design: Sera were collected from 255 newly diagnosed Moroccan NPC patients and 226 healthy donors. IgA antibody against VCA and EA was measured by immunofluorescence assays. IgG antibody against ZEBRA, Zp125, and Zp130 was measured by ELISA.

Results: No significant difference in the detection of IgG-Zp125 and Zp130 antibodies was observed in children with NPC. IgG-Zp130 were detected less frequently than IgG-Zp125 in young and adult patients, as compared to children. High specificity of IgG-Zp125 and -Zp130 antibodies was found in the three age groups. A decrease in IgG-ZEBRA was observed in patients with NPC in clinical remission, whereas patients with NPC who died or developed metastases maintained or had an increase in these titers.

Conclusion: IgG-ZEBRA is a better diagnostic and post-therapeutic prognostic marker in children with NPC, who showed very low titers of IgA -VCA and -EA.

Keywords: Children; NPC; ZEBRA; Diagnosis; Prognosis

∗Corresponding author at: Laboratory of Virology, Pasteur Institute of Morocco, 1, Rue Abou Kacem Ez-Zahraoui, 20 100 Casablanca, Morocco.

Tel.: +212 22 43 44 67; fax: +212 22 26 09 57.

E-mail address:meriem.khyatti@pasteur.ma(M. Khyatti).

1. Introduction

Nasopharyngeal carcinoma (NPC) is a tumour that arises from the epithelium of the retro-nasal cavity. NPC is the most frequent neoplasia of the nasopharynx and respiratory tract in children from North and East Africa, accounting for 5%–20%

doi:10.1016/j.jcv.2007.09.010

(2)

of childhood malignancies (Benider et al., 1995; Ellouz et al., 1978), although even in NPC endemic areas, the incidence among children is low (about 0.10%) (Chow et al., 1997; Huang, 1990). NPC in this setting has a poor prognosis, since only a small proportion of patients with NPC are diagnosed at an early stage, and distant metastases are frequent at late stages (Sahraoui et al., 1999; EL-Husseiny et al., 2000).

NPC is strongly associated with Epstein–Barr virus (EBV). Studies indicate that antibodies against EBV are suitable markers for early diagnosis of NPC, particularly in high risk areas, and can be used to monitor the recur- rence and progression of this malignant disease. However, several case-control studies differ in which of the EBV serological markers are most specific and sensitive. The use of recombinant proteins or synthetic peptides of viral capsid antigen (VCA), early antigen (EA) and EBV nuclear antigen (EBNA) complexes, has improved the sensitivity and the specificity of tests for these markers. Attention has also focused on the diagnostic and prognostic value of IgG antibody against BamHI Z Epstein–Barr replication activator (ZEBRA) antigen (IgG-ZEBRA) in patients with NPC.

A high prevalence of IgG-ZEBRA was observed in NPC patients, including those negative for IgG and IgA antibodies against VCA and EA (IgA-VCA and IgA-EA) (Dardari et al., 2000; Mathew et al., 1994). Moreover,Yip et al. (1994) showed that IgG-ZEBRA correlated positively with patients’

survival and could be detected in NPC patients 1–6 months

metastases to the lung or liver were clinically evident. In addi- tion, a positive correlation of IgG-ZEBRA with lymph-node involvement was reported in young Moroccan patients with NPC (Dardari et al., 2000).

Although numerous serological studies investigated the diagnostic and prognostic value of EBV antibodies in adult patients with NPC, there is little information about the anti- EBV immune response in children with NPC. Therefore, we evaluated the diagnostic value of IgG antibodies directed against ZEBRA synthetic peptides, Zp125 and Zp130 (IgG- Zp125 and IgG-Zp130) and the post-therapeutic prognostic value of IgG-ZEBRA in Moroccan patients aged 10–70 years with NPC. The pediatric cohort was analyzed separately.

2. Materials and methods 2.1. Patients

Two hundred fifty-five newly diagnosed Moroccan patients with NPC were studied at the National Institute of Oncology in Rabat or at the Oncology Center in Casablanca.

The clinical features of these patients are summarized in Table 1. The age of patients varied between 10 and 70 years old. Ten percent were children (≤15 years old), 23% were young adults (>15≤30 years old), and 67% were older adults (older than 30 years). Histopathological examination showed that 100% of children, 97% of young adult patients, and 87%

Table 1

Characteristics of Moroccan patients with NPC

Parameters Children≤15 years (%) Young > 15≤30 years (%) Adults > 30 years (%)

WHO classification

I: Squamous cell carcinoma 0 0 3

II: Non keratinizing carcinoma 0 3 7

III: Undifferentiated carcinoma 100 97 87

TNM classification Tumour (T)

T1 0 0 4

T2 0 18 25

T3 29 36 25

T4 71 41 44

Lymph node

N0 0 0 12

N1 29 9 13

N2 43 59 49

N3 14 32 26

Distant metastasis (M)

M0 70 42 39

M1 11 5 2

Mx 22 53 59

Symptoms

Nasal obstruction 8 35 39

Epistaxis 58 65 35

Otalgia 33 25 37

Hypoacousia 25 20 27

Headaches 8 25 21

Stage I (T1N0), stage II (T2N0, T3N1, T3N2), stage III (T3N0, T3N1, T3N2, T1N2, T2N2), stage IV (N3T1, N3T2, N3T3, N3T4).

(3)

of older adult patients had undifferentiated NPC (WHO III).

All of these patients were at stage III or IV, according to UICC-TNM classification (1979). Regarding the symptoms of NPC, less frequent nasal obstruction and headaches in children was observed as compared to the other groups of age. All patients were treated with chemotherapy followed by radiotherapy. Two hundred twenty-six unrelated healthy subjects were included in this study. Controls were frequency matched for sex and age. Blood samples were collected after informed consent and approval by the institutional ethics committee. Sera collected from patients and controls were stored in aliquots at−20^◦C freezer.

2.2. EBV-related serology

Indirect Immunofluorescence assays (IFA), according to Henle and Henle (1976), were performed to measure IgA- VCA and IgA-EA. P3HR-1 cell line and Raji cells were used as the source of EBV antigens. Acetone fixed slides of P3HR- 1 cells were used to detect IgA-VCA and slides prepared from Raji cells treated with 12-O-tetradecanoylphorbol-13- acetate (40 ng/ml) and n-butyrate (17 mM/ml) were used to detect IgA-EA. IgA titers represent the last serum dilution at which clear IFA staining of positive target cells was observed. Sera from NPC patients were tested for IgG- ZEBRA, using the ELISA test described by Marechal et al. (1993). The recombinant ZEBRA protein used was puri- fied from Escherichia coli transformed by a recombinant pET3c plasmid containing theBZLF1gene under the tran- scriptional control of the T7 RNA polymerase promoter. The titer of IgG-ZEBRA in NPC sera was defined using the end point dilution method described byVaur et al. (1986). IgG antibodies against the two synthetic peptides, Zp125 and Zp130, were determined using the ELISA test described by Brousset et al. (1994). Zp125 corresponds to the entire amino acid sequence of the ZEBRA activation domain, located between amino-acid 59 and 93, and Zp130 corresponds to the entire DNA binding protein, a basic region located between amino acids 145 and 200. Zp125 and Zp130 ELISA tests were done in Emmanuel Drouet’s Laboratory (Unit of Virus Host Cell Interactions (UVHCI) UMR 5233 UJF-EMBL- CNRS B.P. 181. 6 rue Jules Horowitz. F-38042 Grenoble Cedex 9).

2.3. Serological follow-up

A serological follow-up study was performed on 6 children and 12 adults with NPC from whom serial serum samples were available. No other selection procedure was applied. These patients were followed-up for 12–48 months after their first admission. Sera samples were collected prior to therapy and usually at 6 months after initiation of treatment, but were sometimes collected at shorter or longer intervals, depending on the clinical course or the time of check-up visits, and examined for EBV IgA-EA, IgA-VCA and IgG- ZEBRA.

2.4. Statistical methods

Analysis of variance (ANOVA) was used to determine the statistical significance of the differences in the mean titer of antibodies between the different age groups. The Mantel–Haenszel chi-square test was used to test the statistical significance of the trend in seropos- itivity rates of various antibodies between the three NPC age groups. Pearson’s correlation analysis was performed to determine correlation between the IgG-ZEBRA titers and the OD values of IgG-Zp125. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were defined as follows:

sensitivity =a/(a+b), specificity =d/(c+d), PPV =a/(a+c), NPV =d/(b+d), (a) = number of NPC patients showing posi- tive reaction, (b) = number of NPC patients showing negative reaction, (c) = number of healthy donors showing positive reaction, (d) = number of healthy donors showing negative reaction.

3. Results

3.1. Serological analysis

IgA-VCA was less often detected in children (45%) compared to young (70%, p1< 0.02) and older adult patients (85%,p1< 0.01) (Table 2). A significant difference in the mean titer of IgA-VCA between the three age groups (66vs.

281,p1< 0.01; 66vs. 898,p2< 0.01) was also observed. Only

Table 2

Prevalence of IgA-EA, -VCA and IgG-ZEBRA antibodies in children, young and adults patients with NPC

NPC groups p1 p2 p3 Control groups

Antibodies Children≤15 years (N= 25)

Young≥15 > 30 years (N= 59)

Adults >30 years (N= 171)

(N= 226)

% Mean titer % Mean titer % Mean titer % Mean titer

IgA-VCA 45 66 70 281 85 898 <0.02/<0.01 <0.001/<0.01 <0.001/<0.01 2 10

IgA-EA 16 30 30 161 58 346 NS/<0.01 <0.001/<0.01 <0.001/<0.01 0 –

IgG-ZEBRA 100 17 800 93 20 950 96 21 600 NS/NS NS/NS NS/NS 2 500

p1values for positivity rates (percent)/mean titers between children and young patients with NPC are based on analysis of variance (ANOVA) test and Mantel–Haenszel chi-square test for linear trend, respectively.p2values for positivity rates (percent)/mean titers between children and adult patients with NPC.

p3values for positivity rates (percent)/mean titers between young and adult patients with NPC. The bold values represent the significantpvalues.

(4)

Table 3a

Sensitivity and specificity of IgG-ZEBRA, -Zp125 and -Zp130 antibodies in Moroccan patients with NPC

Sensitivity (%) p1 p2 p3 Specificity (%) p1 p2 p3

ZEBRA Zp125 Zp130 ZEBRA Zp125 Zp130

Children≤15 years 100 80 67 <0.01 <0.01 NS 100 88 100 NS NS NS

Young≥15 > 30 years 93 69 41 <0.01 <0.01 <0.01 100 94 100 NS NS NS

Adults >30 years 96 73 48 <0.01 <0.01 <0.01 96 83 100 NS NS NS

p4 NS NS <0.01 NS NS NS

p5 NS NS <0.01 NS NS NS

p6 NS NS NS NS NS NS

The bold values represent the significantpvalues.

16% of children had detectable IgA-EA, and they had a low mean titer compared to adult patients (30vs. 346,p2< 0.01).

Although, no difference was observed for the presence of IgA-EA between children and young adult patients with NPC (16%vs. 30%,p1< 0.10), the mean titer of IgA-EA was lower in children than in young adult patients (30vs. 161,p1< 0.01).

The three age groups had comparable IgG-ZEBRA.

3.2. Immunoreactivity to Zp125 and Zp130 synthetic peptides

As shown in Table 3a, there was no significant difference in IgG-Zp125, as compared to IgG-Zp130 (80% vs.

67%,p3< 0.10) in children. In contrast, IgG-Zp130 was less often present than IgG-Zp125 in young adults (69%vs.41%, p3< 0.01) and adults (73% vs. 48%, p3< 0.01). The presence of IgG-Zp130 was higher in children, as compared to young adults (67%vs.41%,p4< 0.01) and adults (76%

vs. 48%, p5< 0.01). IgG-ZEBRA was more often present than IgG-Zp125 and IgG-Zp130 in children (100%vs.80%, p1< 0.01, 100% vs. 67%; p2< 0.01, respectively); young adults (93% vs. 69%, p1< 0.01; 93% vs. 41%, p2< 0.01, respectively); and in adults (96%vs.73%,p1< 0.01; 96%vs.

48%,p2< 0.01, respectively). Of note, IgG-Zp125 and IgG- Zp130 had higher specificity than IgG-ZEBRA. The results presented inTable 3bshowed a significant difference in the NPV of IgG-ZEBRA as compared to IgG-Zp125 (93%vs.

65%,p1< 0.01 in young; 93%vs.67%,p1< 0.01, in adults) and to IgG-Zp130 (93% vs.54%,p2< 0.01 in young; 93%

vs. 55%,p2< 0.01 in adults), whereas there was no significant difference in the PPV of IgG-Zp125 and IgG-Zp130 compared to IgG-ZEBRA, in any age group. Eighty-five percent of sera from children with NPC who were negative for IgA-VCA reacted with Zp125, as compared to 42%

(p3< 0.05) and 50% (p4< 0.05) in young and older adult patients with NPC, respectively (Table 4). In contrast, no significant difference was found in reactivity against Zp125, of IgA-EA negative sera, between children, young adult and older adult patients with NPC (72%, 60%, 61%, respectively).

The immunoreactivity of IgA-EA negative sera with ZEBRA protein was lower than with Zp125 in children (100% vs.

72%,p2< 0.01), young adult (88%vs.60%,p2< 0.05), and older adult patients with NPC (88%vs.61%,p2< 0.05). The immunoreactivity of IgA-VCA negative sera with ZEBRA protein was higher than with Zp125 in young adults (82%vs.

42%, p1< 0.05) and older adults (81%vs.50%, p1< 0.05) than in children. No correlation was observed in NPC patients between sera IgG-Zp125 titers and IgG-ZEBRA titers (data not shown).

3.3. Serological follow-up

Fig. 1 illustrates IgG-ZEBRA, IgA-VCA, and IgA-EA antibody patterns observed in 6 children and 12 adults with

Table 3b

Positive and negative predictive values of IgG-ZEBRA, -Zp125 and -Zp130 in Moroccan patients with NPC

Positive predictive value (%) p1 p2 p3 Negative predictive value (%) p1 p2 p3

ZEBRA Zp125 Zp130 ZEBRA Zp125 Zp130

Children≤15 years 100 89 100 NS NS NS 100 79 71 NS NS NS

Young≥15 > 30 years 100 94 100 NS NS NS 93 65 54 <0.01 <0.01 NS

Adults >30 years 98 97 100 NS NS NS 93 67 55 <0.01 <0.01 NS

P6 NS NS NS NS NS NS

p1 value, to compare sensitivity, specificity, PPV and NPV of IgG-ZEBRA to IgG-Zp125 antibodies in children, young and adults with NPC.p2 value, to compare sensitivity, specificity, PPV and NPV of IgG-ZEBRA to IgG-Zp130 antibodies in children, young and adults with NPC.p3 value, to compare sensitivity, specificity, PPV and NPV of IgG-Zp125 to Zp130 antibodies in children, young and adults with NPC.p4 value, to compare sensitivity, specificity, PPV and NPV of IgG-ZEBRA, IgG-Z125 and Zp130 antibodies in children and young with NPC.p5 value, to compare sensitivity, specificity, PPV and NPV of IgG-ZEBRA, IgG-Z125 and Zp130 antibodies in children and adults with NPC.p6 value, to compare sensitivity, specificity, PPV and NPV of IgG-ZEBRA, IgG-Zp125 and IgG-Zp130 antibodies in young and adults with NPC. The bold values represent the significantpvalues.

(5)

Table 4

Evaluation of IgG antibody responses against ZEBRA protein and ZEBRA-p125 synthetic peptide in IgA-VCA and -EA negative patients with NPC Children≤15 years Young≥15 > 30 years Adults >30 years p3/p4values Total Number positive/

number tested (%)

Number positive/

number tested (%)

Number positive/

number tested (%)

Number positive/

number tested (%)

IgA-VCA negative/IgG-ZEBRA positive 13/13 (100%) 14/17 (82%) 21/26 (81%) NS 48/56 (86%)

IgA-VCA negative/IgG-ZEBRA-p125 positive 11/13 (85%) 5/12 (42%) 13/26 (50%) <0.05/<0.05 29/51 (57%)

p1value NS <0.05 <0.05 <0.05

IgA-EA negative/IgG-ZEBRA positive 21/21 (100%) 36/41 (88%) 58/66 (88%) NS 115/128 (90%)

IgA-EA negative/IgG-ZEBRA-p125 positive 13/18 (72%) 18/30 (60%) 40/66 (61%) NS 71/114 (62%)

p2value <0.01 <0.05 <0.05 <0.01

p1value for positivity rates (percent) of IgG-ZEBRA antibodies and IgG-ZEBRA-p125 between IgA-VCA negative and IgA-VCA positive patients.p2value for positivity rates (percent) of IgG-ZEBRA antibodies and IgG-ZEBRA-p125 between IgA-EA negative and IgA-EA positive patients.p3value to compare antibody responses between children and young patients,p4value to compare antibody responses between children and adult patients. The bold values represent the significantpvalues.

NPC. Each curve represents a patient, longitudinally monitored for one anti-EBV antibody. IgG-ZEBRA, IgA-VCA and IgA-EA antibody patterns in children are presented in Fig. 1A1–A3, respectively. According to their IgG-ZEBRA titers, two patients (NPC-5, -6) who responded well to treatment, showed moderate and stable titers of their IgG-ZEBRA from the beginning; while NPC-1, -3 and -4, who had high titers of IgG-ZEBRA at the time of diagnosis, developed a striking decline in their IgG-ZEBRA titers. Patient NPC-2, who died of cranial extension without nodal metastases, had the highest IgA-EA and IgA-VCA titers and an eight-fold increase in IgG-ZEBRA titers within the first year of follow- up. Serological follow-up of IgG-ZEBRA, IgA-VCA and IgA-EA in adult patients (NPC-7, -8 and -9) who were in clinical remission, is illustrated inFig. 1B1–B3. These showed a significant decrease in IgG-ZEBRA, IgA-VCA and IgA- EA titers, except for NPC-8 in whom IgA-EA titer remained stable. In two patients who died during treatment, a decrease in IgG-ZEBRA titer occurred (NPC-10 and -12), whereas a little increase in IgG-ZEBRA was observed in patient NPC- 11 (Fig. 1C1). IgA-VCA titer increased significantly in all of them (Fig. 1C2). In contrast, IgA-EA was very low at the time of diagnosis, and remained unchanged in two of them. The antibody patterns in patients (NPC-13, -14, -15, -16, -17, and -18) who have developed metastases are illustrated inFig. 1D1–D3. Patients NPC-15 and -17 maintained high stable titers of IgG-ZEBRA, IgA-VCA and IgA-EA, whereas, an increase in IgG-ZEBRA was found in patient NPC-13, who had high titer of IgA-VCA and low titer of IgA-EA. No significant change in the IgG-ZEBRA and IgA- EA titers was observed in patients NPC-16 and -18, in whom IgA-VCA increased significantly. In contrast, NPC-14 maintained low IgA-VCA, but showed a decrease in IgG-ZEBRA and IgA-EA.

4. Discussion

IgG-ZEBRA antibodies represent a more sensitive marker for the diagnosis of NPC in children than IgA-VCA and IgA-EA antibodies, which have been recognized as specific

markers for this tumour. The present study also indicates that Zp125, identified as the most immunogenic epitope of the activation domain of ZEBRA protein (Brousset et al., 1994), showed a high degree of immunoreactivity with sera from children, young adult and older adult patients with NPC. The IgG-Zp130 were less frequently detected than IgG-Zp125 in young adult and older adult patients. Taking into account the high molecular diversity of IgG and IgA immune responses against different EBV proteins in young adult and older adult patients with NPC (Fachiroh et al., 2004; Karray et al., 2005), the difference observed between Zp125 and Zp130 reactivity is not surprising. These results support those ofTedeschi et al. (1995), who identified three major specific epitopes of ZEBRA protein (i.e.ZEBRA-1, ZEBRA-19 and ZEBRA- 22), which had variable sensitivity and specificity for NPC diagnosis. NPC patients’ sera were reactive with ZEBRA- 22 epitope in 63% of cases, whereas IgG antibodies against ZEBRA-1 were less specific for NPC since elevated titres were detected in sera collected from infectious mononu- cleosis and lymphomas. Interestingly, both IgG-Zp125 and IgG-Zp130 were highly specific for NPC in the three age groups. The absence of correlation between the titer of IgG-ZEBRA and the immunoreactivity of Zp125 suggest that other epitopes of the ZEBRA protein might be impli- cated in the IgG immune response of NPC patients, and the combination of several peptides of ZEBRA protein is required to improve their use for diagnosis of this malignant disease.

The analysis of antibody patterns in patients with NPC indicates that IgG-ZEBRA had better prognostic value than IgA-EA and IgA-VCA. Stable low IgG-ZEBRA antibody titer, or a striking decline in IgG-ZEBRA antibodies, was observed in children during treatment. Of note, children showing low IgG-ZEBRA titers were also negative for IgA- VCA and IgA-EA; the latter have been identified as being produced following frequent reactivation of latent EBV, repeated EBV infection, or both (Gutierrez et al., 2001; Yip et al., 1996). Predominance of the IgG response against EA- R has been reported in children, as compared to IgG-EA-D which is most often used to detect reactivated EBV infection (Schaade et al., 2001). Taken together, these results suggest

(6)

Fig. 1. Serological follow-up of EBV antibodies in patients with NPC. Each curve represents a patient longitudinally monitored for one anti-EBV antibody.

A1, A2 and A3 represent, respectively, IgG-ZEBRA, IgA-VCA and -EA antibody patterns in six children. B1, B2 and B3 represent, respectively, IgG-ZEBRA, IgA-VCA and -EA in three adults in clinical remission. C1, C2 and C3 represent, respectively, IgG-ZEBRA, IgA-VCA and -EA in three patients who died during treatment. C1’, C2’ and C3’ represent, respectively, IgG-ZEBRA, IgA-VCA and -EA in six patients who had metastases.

that low viral reactivation, as demonstrated by low IgA specific antibody titers and low IgG-ZEBRA, indicate a good prognosis in children with NPC; children with this combination of EBV specific responses remained clinically in remission 4 years post-treatment. Moreover, all children who remained disease free after treatment, but were not in the long-term follow-up group, had a low titer of IgG-ZEBRA at the time of diagnosis (data not shown). Similarly, decline

in IgG-ZEBRA, IgA-VCA and -EA antibodies was observed in adult patients who were clinically in remission, but these IgG or IgA responses remained higher than those observed in children with NPC. The difference in the immune responses between children and adults might be explained by genetic recombination diversity, that may occur in adult patients co-infected with different strains of EBV (Srivastava et al., 2000), leading thus to the creation of new epitopes which

(7)

could be presented to the patient’s immune system during disease progression. A decrease in the ability of the immune system with age to clear EBV infected cells might also serve as a booster for the humoral response, and consequently in the persistence of antibody production in adult patients (Glaser et al., 1985). It is noteworthy that, comparatively to adult patients in clinical remission, those with metastasis showed high titers of IgG-ZEBRA, IgA-VCA and -EA, without significant fluctuations during treatment.Sigel et al.

(1994) suggest that the strong humoral responses in NPC patients reflects impaired T cells function, resulting in an increase in EBV-infected cells, which produce high levels of antibodies. Studies have underlined the role of the cytokines level such tumour growth factor (TGF-␤) and IL-10 on the immune responses in EBV infection. In this regard, positive correlation between high TGF-␤and high IgA antibody titers in patients with NPC was found (Xu et al., 1999, 2000).

This correlation is not surprising since TGF-␤ alone, or in combination with IL-10 or IL-5, induces switching to IgA in antibody B-producing cells (Stavnezer, 1995). One may note here that EBV is able to induce the production of IL-10 by B lymphocytes; this could suppress the antiviral immu- nity and enhance survival and the growth of B cells (Burdin et al., 1993). However,Tedeschi et al. (2007)reported that humoral immune response decreases in very advanced stage of NPC disease. This is clearly supported by our data showing a decrease in IgG-ZEBRA titer and the persistence of a stable and very low titer of IgA-EA in adult NPC patients who died. Whether the low IgA and IgG immune responses in children are correlated with a difference in cytokines production deserves more investigation.

Although, only a small number of patients were included in a follow-up longitudinal study, the results clearly showed the potential significance of IgG-ZEBRA antibodies for the post-therapeutic surveillance of NPC in children. Another consideration is that the use of IgG-ZEBRA is less expensive than the DNA test, and might be preferable in resource-poor settings. IgG-ZEBRA could be especially helpful for those negative for IgA-VCA and IgA-EA.

Disclosure

The authors have no financial conflict of interest.

Acknowledgements

We thank the medical staff at the National Institute of Oncology (INO), Rabat, and the Oncology Center, Casablanca, for the invaluable assistance in recruiting patients for the collection of sera. We are grateful to the Cana- dian Institutes of Health Research for support. Rkia Dardari was supported by fellowship from the International Agency for Research on Cancer (Lyon, France) and the Sainte-Justine Hospital Foundation.

References

Benider A, Sahraoui S, Acharki A, Samlali R, Kahlain A. Carcinomas of the nasopharynx in children. Analysis of 50 cases. Bull Cancer 1995;82:155–61.

Brousset P, Drouet E, Schlaifer D, Icart J, Payen C, Meggeto F, Mar- chou B, Massip P, Delsol G. Epstein–Barr virus (EBV) replicative gene expression in tumour cells of AIDS-related non-Hodgkin’s lym- phoma in relation to CD4 cell number and antibody titres to EBV. AIDS 1994;8:583–90.

Burdin N, P´eronne C, Banchereau J, Rousset F. Epstein–Barr virus transfor- mation induces B lymphocytes to produce human interleukin 10. J Exp Med 1993;177:295–304.

Chow KC, Ma J, Lin LS, Chi KH, Yen SH, Liu SM, Liu WT, Chen WK, Chang TH, Chen KY. Serum responses to the combination of Epstein–Barr virus antigens from both latent and acute phases in nasopharyngeal carcinoma: complementary test of EBNA-1 with EA-D.

Cancer Epidemiol Biomarkers Prev 1997;6:363–8.

Dardari R, Khyatti M, Benider A, Jouhadi H, Kahlain A, Cochet C, Mansouri A, El Gueddari B, Benslimane A, Joab I. Antibodies to the Epstein–Barr virus transactivator protein (ZEBRA) as a valuable biomarker in young patients with nasopharyngeal carcinoma. Int J Cancer 2000;86:

71–5.

EL-Husseiny G, Allam A, Khafaga Y, Kandil A, Belal A, Shalaby L, Mohamad G, Al-Shabanah M, Al-Rajhi N, Al-Amro A, Derek Jenkin R. Nasopharyngeal carcinoma in children and adolescents. J Egy Nat Cancer Inst 2000;12:151–5.

Ellouz R, Cammoun M, Attia RB, Bahi J. Nasopharyngeal carcinoma in children and adolescents in Tunisia: clinical aspects and the paraneoplastic syndrome. IARC Sci Publ 1978:115–29.

Fachiroh J, Schouten T, Hariwiyanto B, Paramita DK, Harijadi A, Haryana SM, Ng NH, Middeldorp JM. Molecular diversity of Epstein–Barr virus IgG and IgA antibody responses in nasopharyngeal carcinoma: a comparison of Indonesian, Chinese, and European subjects. J Infect Dis 2004;190:53–62.

Glaser R, Strain EC, Tarr KL, Holliday JE, Donnerberg RL, Kiecolt-Glaser JK. Changes in Epstein–Barr virus antibody titers associated with aging.

Proc Soc Exp Biol Med 1985;179:352–5.

Gutierrez J, Rodriguez M, Soto MJ, Suarez S, Morales P, Piedrola G, Maroto MC. An evaluation of a polyantigenic ELISA to detectEpstein–Barr virus reactivation. Microbios 2001;106:49–54.

Henle W, Henle G. The sero-epidemiology of Epstein–Barr virus. Adv Pathobiol 1976;5:5–17.

Huang TB. Cancer of the nasopharynx in childhood. Cancer 1990;1:

968–71.

Karray H, Ayadi W, Fki L, Hammami A, Daoud J, Drira MM, Frikha M, Jlidi R, Middeldorp JM. Comparison of three different serological techniques for primary diagnosis and monitoring of nasopharyngeal carcinoma in two age groups from Tunisia. J Med Virol 2005;75:593–602.

Marechal V, Meyohas MC, Joab I, Gaha S, Giot JF, Sergeant A, Nicolas JC. Enzymelinked immunosorbent assay for antibodies to ZEBRA, an Epstein–Barr transactivator. Res Virol 1993;144:397–404.

Mathew A, Cheng HM, Sam CK, Joab I, Prasad U, Cochet C. A high incidence of serum IgG antibodies to the Epstein–Barr virus replication activator protein in nasopharyngeal carcinoma. Cancer Immunol Immunother 1994;38:68–70.

Sahraoui S, Acharki A, Benider A, Bouras N, Kahlain A. Nasopharyngeal carcinoma in children under 15 years of age: a retrospective review of 65 patients. Ann Oncol 1999;10:1499–502.

Schaade L, Kleines M, Hausler M. Application of virus-specific immunoglobulin M (IgM), IgG, and IgA antibody detection with a polyantigenic enzyme-linked immunosorbent assay for diagnosis of Epstein–Barr virus infections in childhood. J Clin Microbiol 2001;39:3902–5.

Sigel G, Schillinger M, Henninger K, Bauer G. IgA directed against early antigen of Epstein-barr virus is no specific marker for the diagnosis of nasopharyngeal carcinoma. J Med Virology 1994;43:222–27.

(8)

Srivastava G, Wong KY, Chiang AK, Lam KY, Tao Q. Coinfection of multiple strains of Epstein–Barr virus in immunocompetent normal individuals:

reassessment of the viral carrier state. Blood 2000;95:2443–5.

Stavnezer J. Regulation of antibody production and class switching by TGF- beta. J Immunol 1995;155:1647–51.

Tedeschi R, Foong YT, Cheng HM, dePaoli P, Lehtinen T, Elfborg T, Dill- ner J. The disease associations of the antibody response against the Epstein–Barr virus transactivator protein ZEBRA can be separated into different epitopes. J Gen Virol 1995;76:1393–400.

Tedeschi R, Pin E, Martorelli D, Bidoli E, Marus A, Pratesi C, Bortolin MT, Zanussi S, Vaccher E, Dolcetti R, De Paoli P. Serum antibody response to lytic and latent Epstein–Barr virus antigens in undifferentiated nasopharyngeal carcinoma patients from an area of nonendemicity. Clin Vaccine Immunol 2007;14:435–41.

Vaur L, Agut H, Garbarg-Chenon A, Prud’Homme de Saint-Maur G, Nicolas JC, Bricout F. Simplified enzyme-linked immunosorbent assay

for specific antibodies to respiratory syncytial virus. J Clin Microbiol 1986;24:596–9.

Xu JW, Menezes J, Prasad U, Ahmad A. Elevated serum levels of transforming growth factor beta1 in Epstein–Barr virus-associated nasopharyngeal carcinoma patients. Int J Cancer 1999;84:396–9.

Xu JW, Ahmad A, Jones JF, Dolcetti R, Vaccher E, Prasad U, Menezes J.

Elevated serum transforming growth factor beta1 levels in Epstein–Barr virus-associated diseases and their correlation with virus-specific immunoglobulin A (IgA) and IgM. J Virol 2000;74:2443–6.

Yip TT, Ngan RK, Lau WH, Poon YF, Joab I, Cochet C, Cheng AK. A possi- ble prognostic role of immunoglobulin-G antibody against recombinant Epstein–Barr virus BZLF-1 transactivator protein ZEBRA in patients with nasopharyngeal carcinoma. Cancer 1994;74:2414–24.

Yip TT, Lau WH, Ngam RKC, Poon YF, Cochet C, Ho JHC, Lo TY. Role of Epstein–Barr virus serology in the prognosis of nasopharyngeal carcinoma: The present and the future. Epstein–Barr virus Rep 1996;3:25–33.