1069 Nasopharyngeal Carcinoma In Situ Two Cases of an Emerging Diagnostic Entity Florence Cheung, M.B.B.S.1 Siu Wah Pang, M.B.B.S.1 Fei Hioe, M.B.B.S.1 Kin-Nam Cheung, M.Phil.1 Anne Lee, M.B.B.S.2 Tsz-kok Yau, M.B.B.S.2 1 Department of Clinical Pathology, Pamela Youde Nethersole Eastern Hospital, Chai Wan, Hong Kong, People’s Republic of China. 2 Department of Clinical Oncology, Pamela Youde Nethersole Eastern Hospital, Chai Wan, Hong Kong, People’s Republic of China. BACKGROUND. The association of Epstein–Barr virus (EBV) with the oncogenesis of nasopharyngeal carcinoma (NPC) is well established. Latent infection by EBV with clonal proliferation has also been demonstrated in preinvasive lesions of NPC. In situ hybridization for EBV-encoded RNA (ISH EBER) now serves as an ancillary test in the definitive diagnosis of these lesions. METHODS. Two cases of nasopharyngeal carcinoma in situ (NPCIS) are presented in this study. Their biopsies were studied by ordinary light microscopy, the ISH EBER technique, and immunostaining for bcl-2. Tissue samples from 100 high risk subjects negative for NPC and NPCIS, who served as controls, were also studied using the ISH EBER technique. RESULTS. NPCIS was characterized by abnormal light microscopic appearance as well as positive staining by the ISH EBER technique; these features were not observed in samples from the 100 high risk subjects. Immunostaining for bcl-2 protein was positive but less specific. Postradiotherapy biopsies of the two patients were negative for NPCIS. CONCLUSIONS. With the help of the ISH EBER technique, the diagnosis of NPCIS is now possible in routine surgical pathology. As this entity is rare, it is necessary to have a high degree of suspicion when evaluating biopsies from high risk individuals. Radiotherapy for patients with NPCIS is justified in view of the risk of cancer progression and the possibility of a coexisting invasive carcinoma. Cancer 1998;83:1069 –73. © 1998 American Cancer Society. KEYWORDS: nasopharyngeal carcinoma in situ, preinvasive lesions, in situ hybridization Epstein–Barr virus– encoded RNA. N The authors thank Ms. Teresa Chan for clerical support. Address for reprints: Florence Cheung, M.B.B.S., Department of Clinical Pathology, Pamela Youde Nethersole Eastern Hospital, 3 Lok Man Road, Chai Wan, Hong Kong, People’s Republic of China. Received December 15, 1997; revision received March 11, 1998; accepted March 11, 1998. © 1998 American Cancer Society asopharyngeal carcinoma (NPC), an epithelial carcinoma that arises from the nasopharyngeal mucosa, is one of the most common cancers in southern China and parts of Southeast Asia. In Hong Kong, in the year 1992, it accounted for 6.3% of all newly diagnosed cancers, ranking third among the most common cancers in males and eighth among those of females.1 That Epstein–Barr virus (EBV) is associated with the oncogenesis of NPC and lymphoepithelioma-like carcinomas of other organs is well established.2,3 EBV DNA, RNA, and proteins have been found in the tumor cells of almost all NPCs, implying a latent infection.4,5 There is evidence that tumor cells are clonal proliferations derived from a single EBV-infected cell.4 On the other hand, latent infection by EBV does not occur in normal nasopharyngeal epithelial cells.6 NPC patients and healthy people who later develop NPC have high levels of antibodies against EBV proteins.7,8 Immunoglobulin (Ig)A titer to EBV viral capsid antigen (VCA) is now widely used as a screening test for high risk populations or relatives of NPC patients.9 If this test is used in conjunction with nasopharyngeal endoscopy and prompt biopsy, it can be anticipated 1070 CANCER September 15, 1998 / Volume 83 / Number 6 TABLE 1 Clinical Data on Two Patients with NPCIS Gender/age (yrs) Symptoms Case 1 F/45 Case 2 M/55 Nasopharyngeal IgA EBV endoscopy VCA titer findings Tinnitus 1:160 Blood-stained postnasal drip 1:40 Bulge at left roof Bulge at right roof NPCIS: nasopharyngeal carcinoma in situ; IgA: immunoglobulin A; EBV: Epstein–Barr virus; VCA: viral capsid antigen. that more and more early stage NPCs and preinvasive lesions will be encountered in clinical practice. In the past, diagnostic criteria for preinvasive lesions of NPC based on morphology alone have been vague, subjective, or impracticable.10 –13 Now, with the help of more specific and nonisotopic molecular techniques, these lesions can be diagnosed with confidence.14,15 Precedented by the success of cervical carcinoma screening, recognition of preinvasive NPC lesions holds the promise of early treatment and hence better prognosis. However, purely preinvasive lesions have rarely been documented, and it is possible that the time interval for their progression to invasion is short.16 Previous reports of lesions classified as preinvasive came from retrospective studies of archival material only.14,15 It was at this state of the art that we encountered two cases of pure nasopharyngeal carcinoma in situ (NPCIS) unassociated with an invasive component. Diagnosis was based on histologic features and confirmed by nonisotopic in situ hybridization for EBV-encoded RNA (ISH EBER). The patients received radiotherapy, and postradiotherapy biopsy revealed no lesions. These two cases are presented herein, with data from control studies that were conducted on nasopharyngeal tissue from 100 high risk patients who presented to our hospital during the same period. These 100 patients had raised IgA EBV VCA titers but negative biopsies for invasive and in situ lesions histologically. MATERIALS AND METHODS Patients and Specimens Over a 2-year period (November 1995 to October 1997), a total of 1220 nasopharyngeal biopsies were performed at our hospital. They were performed in response to screening of asymptomatic individuals with raised IgA EBV VCA titer, patients with signs and symptoms suggestive of NPC, or NPC patients after completion of radiotherapy. Of these, 94 new cases of NPC and 2 cases of NPCIS were diagnosed. Clinical data for the latter two patients are shown in Table 1. All tissue samples were fixed in 10% buffered formalin and embedded in paraffin, and sections were prepared for light microscopy. To confirm the neoplastic nature of the 2 NPCIS cases, we performed ISH EBER studies on them, as well as on negative biopsies from 100 high risk patients who served as controls. Their IgA EBV VCA titers varied from 1:10 to 1:640 (according to an immunofluorescence assay). Immunohistochemical studies to detect bcl-2 protein were also performed on the two NPCIS cases to assess their degree of staining. Nonisotopic ISH EBER Technique Tissue sections 4 mm thick were cut and mounted onto slides coated with 3-aminopropyltriethoxysilane. The sections were deparaffinized, rehydrated, and airdried. The dewaxed slides were digested with 2.5 mg/mL pepsin in 0.1N HCl for 30 minutes at 37oC. After the digestion, sections were dehydrated in 95% ethanol and air-dried. The hybridization was done by incubating the sections with fluorescein-conjugated DNA probe to EBV EBER mRNA (BioGenex, San Ramon, CA) for 1 hour at 37oC. After incubation, the sections were washed with 0.05 M Tris buffer saline (pH 7.6), and fluorescein-conjugated probes were detected with alkaline phosphatase conjugated mouse antifluorescein monoclonal antibody (Dako, Glostrup, Denmark). The color reaction was performed with nitroblue tetrazolium salt and 5-bromo-4-chloro-3-indolyl phosphate solution in the dark for 20 minutes. Appropriate positive controls were included in each assay. The sections were counterstained with 0.1% methyl green. Immunostaining for bcl-2 protein Standard immunoperoxidase technique was performed on paraffin sections for the detection of bcl-2 oncoprotein by incubation with bcl-2 monoclonal antibody (Dako, Glostrup, Denmark) at a dilution of 1:30. The sections were counterstained with hematoxylin. RESULTS Case 1 had 2 biopsies, and NPCIS was detected only in the first. Case 2 had 4 biopsies within a period of 2 months, with near-stripping of the mucosal bulge during the fourth biopsy. NPCIS was present in all four biopsies. There was no accompanying invasive component in any of the above specimens. Microscopically, NPCIS changes in these two patients were similar, with patchy involvement of both covering and cryptal epithelium (Figs. 1 and 2). These areas were characterized by epithelial cells with loss of polarity, enlarged vesicular nuclei, and large eosinophilic nucleoli. The basement membrane was disrupted by infiltrates of lymphocytes and plasma cells dispersed Nasopharyngeal Carcinoma In Situ/Cheung et al. 1071 FIGURE 1. Case 1: (A) The nasopharyngeal mucosa shows changes characteristic of nasopharyngeal carcinoma in situ, involving the whole thickness of the epithelium. (B) A high-power view shows carcinoma in situ cells with large, vesicular nuclei and prominent eosinophilic nucleoli (H & E). FIGURE 2. Case 2: (A) Nasopharyngeal carcinoma in situ involves the mucosal crypt. (B) A high-power view highlights the loss of polarity and disruption of the basement membrane by lymphocytic infiltrates (H & E). among the epithelial cells. Junction with adjacent respiratory epithelium was either abrupt or indistinct. ISH EBER technique demonstrated moderate nuclear staining of these carcinoma in situ (CIS) cells, accentuating their intraepithelial distribution (Fig. 3). Positive staining of the nuclei was concentrated along the nuclear membranes. The nucleoli were not stained, but instead highlighted by a ringlike outline. The staining intensity was slightly less than that in classical NPC cells used as controls. As expected, studies of the 100 negative biopsies that served as controls showed absolutely no uptake by the nonneoplastic epithelium. Immunohistochemical staining for bcl-2 protein, similar to a recent study,17 showed positive staining of the basal cells in normal epithelium and mild-to-mod- erate cytoplasmic staining of the CIS cells (Fig. 4). Distinction of neoplastic from normal epithelial cells was much inferior to that offered by ISH EBER technique. Both patients were treated by radiotherapy to a total dose of 66 gray (Gy) delivered in 33 daily fractions. Nasopharyngeal endoscopy was performed 8 weeks after completion of radiotherapy and revealed no visible tumor. Biopsy at the site where a bulge was previously present revealed no residual CIS changes in either of the two patients, and the ISH EBER studies were negative. DISCUSSION The demonstration of EBV DNA and RNA in preinvasive lesions of the nasopharynx, and evidence confirming 1072 CANCER September 15, 1998 / Volume 83 / Number 6 FIGURE 3. (A) An in situ hybridization Epstein–Barr virus– encoded RNA study of nasopharyngeal carcinoma in situ (NPCIS) from Case 2 shows ribbonlike staining of an intraepithelial lesion. (B) A high-power view shows positively stained nuclei with ringlike nucleoli (methyl green counterstaining). FIGURE 4. Immunohistochemistry for the detection of bcl-2 protein demonstrates positive staining of basal cells and, to a lesser degree, nasopharyngeal carcinoma in situ cells (hematoxylin counterstaining). that they are premalignant clones of EBV-infected cells, mark an important step in understanding NPC oncogenesis.15,18 Although they are not long-lasting, these lesions do occupy a place in the multistep process that culminates in invasive cancer. With the provision of molecular techniques, especially the ISH EBER technique, more and more NPCIS lesions will be diagnosed in routine surgical pathology, requiring appropriate patient management. However, it is necessary to have a high degree of suspicion when looking for foci of abnormal epithelial cells, especially in high risk subjects with raised IgA EBV VCA, to detect these subtle lesions. In the two cases that we present herein, NPCIS is characterized by epithelial cells with loss of polarity, large vesicular nuclei, promi- nent eosinophilic nucleoli, disrupted basement membrane, and lymphocytic infiltrates. The low-power configuration of these foci recaptures that of the normal reticular epithelium overlying lymphoid follicles described by Ali in his study of nasopharyngeal mucosa.19 The high-power morphology of the NPCIS cells resembles that of undifferentiated classical NPC, except that NPCIS has a lesser degree of nuclear pleomorphism. Whether different histologic types of NPC are preceded by different morphologic forms of NPCIS is still unknown. It is the recognition of these abnormal cells that alerts the surgical pathologist to the need for ISH EBER studies. The NPCIS cells show moderate nuclear staining in a pattern that is slightly different from the strong, uniform nuclear staining usually seen in NPC tumor cells during ISH EBER. Whether this reflects a lesser degree of EBER expression in preinvasive tumor cells remains to be verified. According to studies by Sam et al.,6 normal nasopharyngeal biopsies from 23 high risk subjects were negative for latent markers of EBV. Within the context of nasopharyngeal epithelium, latent infection has only been demonstrated in neoplastic NPC cells and preinvasive lesions. This was borne out by the negative results of our study of specimens from the 100 high risk control subjects. It is also our experience that squamous metaplasia, with or without atypia, and postirradiation dysplasia that we occasionally encounter are ISH EBER negative (data not shown). Hence, the ISH EBER technique is highly specific in picking up NPCIS lesions, with its all-or-none staining characteristic, as distinct from the immunostaining by bcl-2 antibodies. Awareness of the existence of preinvasive NPC Nasopharyngeal Carcinoma In Situ/Cheung et al. lesions and the availability of confirmatory ancillary studies will certainly increase the number of these lesions that are diagnosed. This poses a management dilemma: to treat or not to treat. There is currently little clinical data to guide treatment decisions. We recommend definitive treatment, the justification being that coexisting invasive NPC can never be completely ruled out, even with repeated biopsies. The risk of progression to invasive NPC is too real to ignore. According to the studies of Pathmanathan et al.,15 5 of 8 of their patients with preinvasive lesions developed cancer within 12 months of biopsy. The nasopharynx is not a site that is easily accessible to observation, especially in view of the frequent submucosal infiltrations that occur in NPC. Besides the inherent risk of delay in detecting invasive carcinoma, the psychologic distress and anxiety caused is likely to be substantial if an observation policy is adopted. Hence, we decided to treat our patients to a total dose of 66 Gy in 33 daily fractions. Histologic confirmation of complete remission was achieved in both, but longer follow-up is required to assess the ultimate control rate. Although highly responsive to radiotherapy, NPC is associated with mortality and its treatment with unpleasant morbidity. Hopefully, endoscopic biopsy of high risk subjects coupled with molecular studies of suspicious intraepithelial lesions will screen out the early lesions and alleviate the suffering of some patients. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. REFERENCES 1. 2. 3. 4. Cancer incidence and mortality in Hong Kong, 1992. Hong Kong Cancer Registry, Hospital Authority, 1996. Zur Hausen H, Schulte-Holthausen H, Klein G, Henle W, Henle G, Clifford P, et al. EBV DNA in biopsies of Burkitt tumors and anaplastic carcinoma of the nasopharynx. Nature 1970;228:1056 – 8. Iezzoni JC, Gaffey MJ, Weiss LM. The role of Epstein–Barr virus in lymphoepithelioma-like carcinomas. Am J Clin Pathol 1995;103:308 –15. Raab-Traub N, Flynn K. The structure of the termini of the Epstein–Barr virus as a marker of clonal cellular proliferation. Cell 1986;47:883–9. 16. 17. 18. 19. 1073 Liebowitz D, Kieff E. Epstein–Barr virus. In: Roizman B, Whitley RJ, Lopez C. The human herpes viruses. New York: Raven Press, 1993:107–72. Sam CK, Brooks LA, Niedobitek G, Young LS, Prasad U, Rickinson AB. Analysis of Epstein–Barr virus infection in nasopharyngeal biopsies from a group at high risk of nasopharyngeal carcinoma. Int J Cancer 1993;53:957– 62. Henle G, Henle W. Epstein–Barr virus specific IgA serum antibodies as an outstanding feature of nasopharyngeal carcinoma. Int J Cancer 1976;17:1–17. Ho HC, Kwan HC, Ng MH, de The G. Serum IgA antibody to Epstein–Barr virus capsid antigen preceding symptoms of nasopharyngeal carcinoma. Lancet 1978;1:436 –7. Zong YS, Sham JST, Ng MH, Ou XT, Guo YQ, Zheng SA, et al. Immunoglobulin A against viral capsid antigen of Epstein– Barr virus and indirect mirror examination of the nasopharynx in the detection of asymptomatic nasopharyngeal carcinoma. Cancer 1992;69:3–7. Teoh TB. Epidermoid carcinoma of the nasopharynx among Chinese: a study of 31 necropsies. J Pathol Bacterial 1957; 73:451– 65. Zong YS, Li QX. Histopathology of paracancerous nasopharyngeal carcinoma in situ. Chinese Med J 1986,99:763–71. Lee JCK. Precancerous change. In: van Hasselt CA, Gibb AG. Nasopharyngeal carcinoma. Hong Kong: Chinese University Press, 1991:31– 45. Chan CW, Nicholls JM, Sham JST, Dickens P, Choy D. Nasopharyngeal carcinoma in situ in nasopharyngeal carcinoma. J Clin Pathol 1992;45:898 –901. Yeung WM, Zong YS, Chin CT, Chan KH, Sham JST, Choy DTK, Ng MH. Epstein–Barr virus carriage by nasopharyngeal carcinoma in situ. Int J Cancer 1993;53:746 –50. Pathmanathan R, Prasad U, Sadler R, Flynn K, Raab-Traub N. Clonal proliferations of cells infected with Epstein–Barr virus in preinvasive lesions related to nasopharyngeal carcinoma. N Engl J Med 1995;333:693– 8. Nicholls JM. Nasopharyngeal carcinoma: classification and histological appearances. Adv Anat Pathol 1997;4:71– 84. Sheu LF, Chen A, Meng CL, Ho KC, Lin FG, Lee WH. Analysis of bcl-2 expression in normal, inflamed, dysplastic nasopharyngeal epithelia, and nasopharyngeal carcinoma: association with p53 expression. Hum Pathol 1997;28:556 – 62. Kieff E. Epstein–Barr virus: increasing evidence of a link to carcinoma. N Engl J Med 1995;333:724 – 6. Ali MY. Histology of the human nasopharyngeal mucosa. J Anat 1965;99:657–72.