| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Oncology, Markers, Clinical Correlates |
Departments of Anatomical and Cellular Pathology [J. H. M. T., K-W. L., J. K., M. W. Y. C., A. R. C., K-F. T.] and Surgery [R. K. Y. T., J. K. S. W., C. A. v. H.] and Institute of Molecular Oncology [D. P. H.], Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong Special Administrative Region, People’s Republic of China
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Experimental Design: Twenty-eight NPC cases and 26 noncancerous subjects were prospectively recruited. NP brushing samples were subjected to quantitative real-time PCR analysis of EBV DNA and methylation-specific PCR analysis of the DAP-kinase, RASSF1A, and p16 genes.
Results: EBV DNA quantity in NP brushing samples from NPC patients (median, 8.94 copies/actin) was significantly higher than that of controls (median, 0 copies/actin; P < 0.0001). Twenty-seven of 28 NPC patients had detectable EBV DNA in NP brushes, whereas 25 of 26 controls had undetectable or very low levels of EBV DNA. Elevated EBV DNA level in brushing samples as a tumor marker had a sensitivity of 96.4% and a specificity of 96.2% for NPC detection. Moreover, T1 disease had a significantly lower EBV DNA level as compared with locally more advanced disease (P = 0.037). In brushing samples of NPC patients, the frequencies of DAP-kinase, RASSF1A, and p16 promoter hypermethylation were 50.0%, 39.3%, and 46.4%, respectively. Seventy-eight percent of cases showed methylation of at least one gene. No aberrant hypermethylation was detected in control samples.
Conclusions: Our study demonstrated the feasibility of detecting multiple molecular tumor markers in NP brushing samples with a high sensitivity and specificity for NPC detection. It offers a powerful yet noninvasive approach for the diagnosis of NPC in high-risk populations.
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We adopted a simple, noninvasive, and effective procedure, NP brushing, to obtain samples directly from the nasopharynx to detect EBV DNA quantity and tumor-specific epigenetic changes. We aimed to assess whether quantitative EBV DNA analysis and multiple gene hypermethylation might serve as useful diagnostic tumor markers in our high-risk population.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and 2
).
|
|
After the brush with covering sheath engaged the nasopharynx, the sheath was retracted, and the brush was turned for at least three complete rotations while brushing across the nasopharynx. The sheath was then pushed forward to cover the brush, followed by the withdrawal of the brush from the nasopharynx. To avoid contamination, the sheath and the terminal plastic knob at the end of the brush were wiped with 70% alcohol pads. The terminal plastic knob was then cut away with a wire cutter. The brush was then cut and soaked in PBS and promptly sent to the laboratory. Bilateral brushing was performed for each case, and the samples were processed separately. After centrifugation, the supernatant of the brushing samples was discarded. The cell pellet was collected for standard phenol-chloroform extraction.
DNA Extraction from Formalin-fixed Paraffin-embedded Tissue Samples and Plasma Samples.
Paraffin-embedded tumor tissues from 16 NPC patients were available for molecular analysis. The primary biopsies of the remaining 12 NPC cases were not available because they were referral cases. Ten 10-µm sections were cut and subjected to standard phenol-chloroform extraction. For tumor content < 80%, manual microdissection was performed to enrich the tumor content as described previously (17)
. For plasma, 400 µl were purified for DNA extraction using High Pure PCR Template Preparation Kit (Boehringer Mannheim) according to the manufacturer’s protocol.
Detection of EBV DNA by Quantitative Real-time PCR.
EBV DNA quantities in NP brushing samples were measured by real-time quantitative PCR analysis. All corresponding plasma samples, except those from one NPC patient (case 62) and two control subjects (cases 35 and 36), were available for EBV DNA quantitative analysis. The Taqman assay was based on DNA amplification of a 75-bp fragment of EBV BamHI-W region as described previously (5)
. The PCR primers were custom-synthesized by Invitrogen Life Technologies, Inc. (Hong Kong). The nonextendible DNA probe with a fluorescent dye (FAM) attached to the 5'-end and a second fluorescent dye (TAMRA) linked to the 3'-end was synthesized by Applied Biosystems (Foster City, CA). Commercial reagents and PLATINUM Quantitative PCR SUPERMIX-UDG kit (Life Technologies, Inc.) were used according to the manufacturer’s protocol. In brief, the PCR was set up in a reaction volume of 50 µl, including 5 µl of DNA template, 25 µl of 2x PLATINUM Quantitative PCR SUPERMIX-UDG, 300 nM of each of the amplification primers, and 50 nM of the fluorescent probe. The reactions were performed and analyzed using the iCycler iQ Real-Time PCR Detection System (Bio-Rad, Hercules, CA). A standard curve was run in parallel with each analysis using DNA extracted from an EBV-positive cell line, Namalwa (CRL-1432, American Type Culture Collection). Namalwa is a diploid cell line that has been previously reported to contain two integrated viral genomes/cell. Serial dilutions of Namalwa DNA were made. To express the DNA quantity in genome equivalents for the unknown samples, a conversion factor of 6.6 pg DNA/diploid cell was used. To adjust for the variation in the number of cells in different samples, we performed quantification of a fragment of the ß-actin nuclear gene using a commercially available kit from Applied Biosystems. Each sample was analyzed in duplicate. Water blanks were included as controls in every analysis.
The EBV DNA level in brushing samples was expressed as the ratio of the copy number of the EBV genome:the copy number of the ß-actin gene, whereas in the plasma sample, it was expressed as copies of EBV genome/ml plasma.
Bisulfite Modification and COBRA.
The C promoter of EBV DNA within the virion is unmethylated. However, in latent EBV from NPC cells, the C promoter is methylated (18)
. To investigate the nature of EBV DNA in brushing samples, we examined EBV C promoter methylation status in some of our samples. Five NP brushing samples and five primary NPC biopsy samples were investigated by COBRA. Blood and NP brushing samples from a patient with infectious mononucleosis were used as unmethylated controls. Genomic DNA from primary tumor and brushing samples was modified by bisulfite treatment and purified using the CpGenome DNA Modification Kit (Intergen, Purchase, NY) according to the manufacturer’s instructions.
The COBRA protocol has been described previously (19) . Bisulfite-modified DNA was amplified with specific primers for EBV C promoter. Primer sequences for bisulfite-treated DNA were 5'-CGTAGGATCCCTTAAACTCTCTTATTAACTATA-3' (upstream) and 5'-TCGAAGATCTAATGTGTTTTAATTAGAAATTT-3' (downstream; Ref. 20 ). PCR reactions were performed in a volume of 50 µl containing 1x PCR buffer, 4 mM MgCl2, 0.25 mM deoxynucleotide triphosphate, 10 pmol of each primer, and 1 unit of AmpliTaq Gold polymerase (Applied Biosystems). The PCR involved an initial denaturation at 95°C for 12 min; followed by 40 cycles of denaturation at 95°C for 1 min, annealing at 58°C for 1 min, and polymerization at 72°C for 1 min; followed by a final extension at 72°C for 5 min. PCR products were then digested by restriction enzyme TaqI at 60°C overnight. Digested samples were analyzed on 10% polyacrylamide gels and stained by ethidium bromide.
MSP.
NP brushing samples of 28 NPC cases and 16 corresponding primary tumor biopsies were studied. NP brushing samples from 12 noncancerous control cases were also included. Three cancer-related genes with a high frequency of methylation in NPC (DAP-kinase, RASSF1A, and p16) were included (14)
. The methylation statuses of these genes were detected by MSP analysis. The modified DNA was used as a template for PCR amplification using primers specific for either methylated or modified unmethylated DNA. The primer sequences and PCR conditions for DAP-kinase (21)
, RASSF1A (22)
, and p16 (23)
were described previously. Fifteen µl of PCR products were loaded onto a 10% nondenaturing polyacrylamide gel, stained with ethidium bromide, and visualized under UV illumination.
Statistical Analysis.
Mann-Whitney test (two-tailed) was used to compare the nonparametric variables between groups. P <0.05 was considered as statistical significance. The statistical analysis was performed using the statistical software SPSS Version 10.0.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and 2
. Raised EBV IgA VCA or EA titer (
5) was seen in 12 of 26 (46.7%) control subjects and 25 of 28 (89.3%) NPC patients. The sensitivity for cancer detection was 89.3% and the specificity was 53.8% when anti-VCA IgA 5 and/or EA 5 were set as cutoff points. The positive predictive value was 67.5%, and the negative predictive value was 82.3%.
EBV DNA Quantity in NP Brushing and Plasma samples.
Quantitative analysis of EBV DNA was performed on all brushing samples and corresponding plasma samples when available from 28 NPC patients and 26 control noncancerous subjects. Bilateral NP brushing was performed for each case, and the samples were processed separately. The one with higher EBV DNA quantity was used for data analysis. All NP brushing samples were positive for PCR analysis of the ß-actin gene, suggesting sufficient yields of genomic DNA. EBV DNA was detectable in 27 of 28 (96.4%) brushing samples of NPC patients, and the median EBV DNA quantity was 8.94 copies/actin (range, 0–12,953 copies/actin; Table 1
). For noncancerous subjects, EBV was not detectable or was detectable at an extremely low level in 25 of 26 cases (96%), and the median EBV DNA quantity was 0 copies/actin (range, 0–0.8 copies/actin; Table 2
). A statistically significant difference in EBV DNA quantities in brushing samples between NPC and noncancerous subjects was demonstrated (P < 0.0001; Fig. 1
). Although we set the threshold of a positive test as 0.05 EBV copy/actin, it resulted in one false positive (case 60) and one false negative case (case 58) for NPC detection. The test thus predicted NPC with a sensitivity of 96.4% and had a specificity of 96.2%. The positive predictive value was 96.4%, and the negative predictive value was 96.2%.
|
). T1 disease had a significantly lower level of EBV DNA quantity in NP brushing samples as compared with locally more advanced disease (T2-T4; P = 0.037; Table 4
). No statistically significant difference was demonstrated between node-negative and node-positive cases and between lower stage and higher stage diseases.
|
|
and 2
; Fig. 2
). Three of 24 (12.5%) noncancerous control subjects showed elevated EBV DNA in plasma (case 32, 100 copies/ml; case 39, 38 copies/ml; and case 67, 88 copies/ml). One of these subjects had a history of sinonasal inverted papilloma (case 32). All of these patients had almost undetectable EBV DNA in brushing samples, and the NP biopsies were negative for malignancy. The one NPC patient with no detectable EBV DNA in plasma (case 58) also showed negative EBV DNA in NP brushing samples and has been mentioned earlier. Detectable plasma EBV DNA has a sensitivity of 96.3%, a specificity of 87.5%, a positive predictive value of 89.7%, and a negative predictive value of 95.5% for the detection of NPC.
|
). The plasma level of EBV DNA was significantly higher in node-positive cases than in node-negative cases (P = 0.003) and higher stage cases (stage III and IV) than in lower stage cases (stage I and II; P = 0.006; Table 4
), but no statistically significant difference was seen in T1 versus T2, T3, and T4 cases.
Methylation Status of EBV C Promoter as Determined by COBRA.
To investigate the nature of EBV DNA, we examined the methylation status of the EBV C promoter from five NP brushing samples and five primary NPC biopsies by COBRA. All of these samples were completely methylated at the selected TaqI recognition site (Fig. 3)
. The absence of the unmethylated sequence of the EBV C promoter implied that only latent EBV was detected in the NP brushing samples.
|
and 2
, and representative MSP results are shown in Fig. 4
. In brushing samples of NPC, 78.6% (22 of 28) exhibited aberrant promoter hypermethylation in at least one of the three genes studied. Thirty-six percent (10 of 28) of the brushing samples from NPC patients had only one gene methylated, 29% (8 of 28) had two genes methylated, and 14% (4 of 28) had all three genes methylated. In NPC primary tumor tissues of the 16 available cases, identical methylation status was detected in both NP brushing samples and biopsy samples in 100% of cases (16 of 16) for DAP-kinase, 69% of cases (11 of 16) for RASSF1A, and 75% of cases (12 of 16) for p16. In 62.5% (10 of 16) of the cases, the MSP result of NP brushing exactly matched the corresponding tumor for all three genes. Notably, gene hypermethylation was not found in brushing samples when such alteration was not seen in the corresponding primary tumor. Moreover, promoter hypermethylation of any of the three genes was not detected in NP brushing samples of 12 noncancerous control cases. The methylation status of these genes did not correlate with the T, N, or overall staging of the diseases.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our results indicated that NP brushing was a simple and noninvasive procedure to collect NP tissue samples for molecular analysis. Quantitative EBV DNA analysis represented a sensitive (sensitivity, 96.4%) and specific (specificity, 96.2%) marker for NPC in this endemic region. One of the advantages of quantitative analysis is the allowance of a good cutoff between cases and controls. The only false negative case (case 58) had undetectable EBV DNA in both NP brushing and plasma samples. This patient was a 47-year-old female with a family history of NPC and was found to have elevated EBV IgA titers. NP biopsy was performed, and she proved to have early-stage (T1N0M0, stage I) undifferentiated NPC. In situ hybridization for EBV-encoded mRNAs (EBER) was positive. The reason(s) accounting for the absence of detectable EBV DNA in brushing and plasma samples remained uncertain. The false positive case in a noncancerous control subject (case 60) showed slightly elevated EBV DNA (0.8 copy/actin) in brushing sample. The NP biopsy of this patient was negative for malignancy, and in situ hybridization for EBER was also negative. Contamination during sample collection might be one of the explanations. However, we cannot exclude the possibility of obscured malignancy. This patient has been followed-up regularly to monitor any clinical evidence of NPC.
In our study, EBV DNA quantity was correlated with T stages of disease (Table 4)
. It was logical to assume that more tumor cells (hence, a larger amount of EBV DNA) could be brushed out from a more locally advanced tumor. On the other hand, the lack of association between EBV DNA quantity in NP brushes and nodal status and staging was not unexpected.
Our results also implied that EBV DNA originated from latently infected NPC cells. The status of latent infection could also be demonstrated in the NP brushing samples in our study. In lytic replication with the production of new virions, the EBV C promoter is not methylated, whereas in latent EBV infection, the C promoter is inactivated and hypermethylated (18 , 28) . Five NP brushing samples and five primary tumor biopsy samples were studied using COBRA. All of these samples harbored the methylated sequence of the EBV C promoter only, indicating the latent status of EBV, and no lytic component was detected.
For comparison, corresponding plasma EBV DNA quantitative analysis was performed. Plasma EBV DNA quantity has been shown to be a good molecular marker for NPC (5)
. It was associated with disease staging and shown to be useful for monitoring of disease progression (29
, 30)
. Our current data also demonstrated that plasma EBV DNA quantity was a sensitive and specific maker for NPC detection. In addition, node-positive versus node-negative cases and higher stage versus lower stage cases contained a significantly larger quantity of plasma EBV DNA (Table 4)
. In contrast to NP brushing, no significant difference was demonstrated among T stages. It appeared that the EBV DNA quantity detected in the NP brushing samples reflected the local tumor load in the nasopharynx, whereas the plasma EBV DNA quantity reflected the tumor load in the whole body. In our series, three noncancerous control subjects had elevated plasma EBV DNA. One of the control cases had a history of sinonasal inverted papilloma. The possible association between sinonasal inverted papilloma and EBV may deserve further investigation (8)
. Nevertheless, various EBV-associated conditions may result in detectable plasma EBV DNA. These included infectious mononucleosis and EBV-associated malignancies, such as certain subsets of lymphoma and gastric carcinoma (31
, 32)
. The causes of elevated plasma EBV DNA in the two other control subjects remained unclear. However, for these three subjects, significant elevation of EBV DNA in corresponding NP brushing samples was not detected. Detection of an elevated level of EBV DNA from NP brushing would help to localize the disease in the nasopharynx. The potential diagnostic value of quantitative EBV DNA analysis may be illustrated by one of our NPC cases. The patient was a 57-year-old man (case 51) who presented with minor ENT symptoms with a IgA VCA titer of 20. No obvious NP tumor mass was detected in endoscopic examination, and the initial NP biopsy was negative for malignancy. However, both NP brushing and plasma EBV DNA were elevated (Table 1)
. Repeated multiple random NP biopsies were then performed under general anesthesia. Only one of the biopsy specimens revealed undifferentiated carcinoma. An early-stage NPC (T1N0M0, stage I) was diagnosed. Detection of NPC at an early stage is critical to improve patient prognosis. However, the nasopharynx remained a difficult area for routine examination. It is not unusual to have repeated biopsies before one can confirm the diagnosis in suspicious cases. Our data indicated that the EBV quantities in both NP brushing and plasma of control subjects were significantly lower than those of NPC cases. More importantly, the differences could be demonstrated even when comparing the controls with individuals with early-stage disease (T1, N0, or stage I/II; Table 3
). Quantitative EBV DNA analysis in NP samples would help to localize the disease process in the nasopharynx and might contribute to early recognition of the disease.
We also explored the utility of other molecular markers for NPC detection in NP brushing samples. Gene promoter hypermethylation is an increasingly recognized mechanism for inactivation of cancer-related genes (10)
. Such epigenetic changes could also be detected in various types of body fluid, for example, serum, urine, saliva, and bronchial brushing specimens (11, 12, 13
, 33, 34, 35)
. These observations suggested that detection of promoter hypermethylation of various tumor-specific genes might serve as a useful tumor marker (12)
. The multiple gene methylation profile in NPC has been demonstrated in our previous study (14)
. Simultaneous inactivation of several pathways by aberrant methylation in NPC has been documented with high frequencies of DAP-kinase, RASSF1A, and p16 gene methylation (14)
. The current study demonstrated for the first time that aberrant promoter hypermethylation of DAP-kinase, RASSF1A, and p16 could be detected in NP brushing samples from NPC patients. The frequencies of gene promoter hypermethylation of DAP-kinase, RASSF1A, and p16 were 50.0% (14 of 28), 39.3% (11 of 28), and 46.4% (13 of 28), respectively. In 10 of 16 NPC cases for whom the primary tumor specimens were available, identical methylation status of three genes was identified in both brushing specimens and biopsy specimens (Table 1)
. Gene hypermethylation was not found in brushing samples when such alteration was not seen in the corresponding primary tumor. Moreover, only unmethylated alleles of the three genes were detected in 12 noncancerous control subjects, including the one with elevated EBV DNA quantity in NP brushing (case 60). These results demonstrated that the test was highly specific for cancer detection. Taking the presence of at least one gene showing methylation in NP brushing samples as a positive test, the sensitivity of detecting NPC was 78.6% with 100% specificity. Notably, hypermethylation of p16 was detected in brushing samples as well as the primary tumor sample from NPC case 58 (Table 1)
. However, the brushing samples were not positive for RASSF1A as in the primary tumor tissue. As described earlier, this case was negative for EBV DNA in both brushing and plasma samples. Combining methylation and quantitative EBV DNA analysis in NP brushes would increase the sensitivity for NPC detection to 100%. These results further demonstrated the potential diagnostic value of multiple tumor marker detection in brushing samples. Recently, a real-time PCR-based assay (quantitative MSP) for the detection and quantification of CpG island methylation has been introduced. The potential advantages of quantitative MSP analysis of multiple tumor markers may deserve additional studies.
In conclusion, NP brushing provided sufficient materials for multiple tumor marker analysis. It offered a powerful yet noninvasive approach for the detection of NPC. Our data provide justification for a larger prospective study for this approach in detecting NPC in high-risk populations.
|
FOOTNOTES |
|---|
1 Supported in part by the Research Grant Council of Hong Kong Special Administrative Region (Earmarked Grant CU4301/99M). 
2 To whom requests for reprints should be addressed, at Department of Anatomical and Cellular Pathology, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong. Phone: 852-26322352; Fax: 852-26497286; E-mail: kfto{at}cuhk.edu.hk
3 The abbreviations used are: NPC, nasopharyngeal carcinoma; NP, nasopharyngeal; ENT, ear, nose, and throat; COBRA, combined bisulfite restriction analysis; MSP, methylation-specific PCR; AJCC, American Joint Committee on Cancer; VCA, viral capsid antigen; EA, early antigen.
Received 1/28/02; revised 4/29/02; accepted 5/ 6/02.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. L. Gulley and W. Tang Laboratory Assays for Epstein-Barr Virus-Related Disease J. Mol. Diagn., July 1, 2008; 10(4): 279 - 292. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. C. Stevens, S. A. W. M. Verkuijlen, B. Hariwiyanto, Harijadi, J. Fachiroh, D. K. Paramita, I. B. Tan, S. M. Haryana, and J. M. Middeldorp Diagnostic Value of Measuring Epstein-Barr Virus (EBV) DNA Load and Carcinoma-Specific Viral mRNA in Relation to Anti-EBV Immunoglobulin A (IgA) and IgG Antibody Levels in Blood of Nasopharyngeal Carcinoma Patients from Indonesia J. Clin. Microbiol., July 1, 2005; 43(7): 3066 - 3073. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Cerimele, T. Battle, R. Lynch, D. A. Frank, E. Murad, C. Cohen, N. Macaron, J. Sixbey, K. Smith, R. S. Watnick, et al. Reactive oxygen signaling and MAPK activation distinguish Epstein-Barr Virus (EBV)-positive versus EBV-negative Burkitt's lymphoma PNAS, January 4, 2005; 102(1): 175 - 179. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
X174 DNA marker (MBI Fermentas); U, undigested DNA; D, DNA digested with TaqI. A fully unmethylated pattern was seen in DNA samples from blood (IMB) and NP brushing (IMBr) of a patient with infectious mononucleosis. Fully methylated pattern was seen in NPC biopsies (26Bx and 65Bx) and NP brushing (14Br, 15Br, and 23Br) samples.