Primary sinonasal non-Hodgkin's lymphomas (NHL) are relatively common in Asia, comprising up to 7[percnt] of all NHL with a marked preponderance demonstrating a NK/T-cell phenotype. 6,24 Other areas with a relatively high frequency of sinonasal NK/T-cell lymphomas have been reported from Peru 1,36 and Mexico. 2 In contrast, sinonasal NHL are uncommonly seen in the United States and Europe, accounting for just 1.5[percnt] of NHL in one study, 18 with several studies reporting a predominance of B-cell neoplasms. 10,15,17 Nasal or nasal-type NK/T-cell lymphomas are characterized by their strong association with EBV, cytoplasmic CD3 positivity, frequent CD56 positivity, expression of cytotoxic markers (TIA-1, granzyme B, perforin), and lack of T-cell receptor gene rearrangements. 5,9,23,25,27,30,33,39 Other NK cell markers such as CD16 have been variably reported as usually negative 9,39 or positive. 33,34 Largely because of the rarity of these lesions in the West, only a few studies have attempted to characterize these lesions in the U.S. 10,11,16,37,42 or European population. 12,27,35,41 Several of these studies have confirmed a strong association with EBV; however, molecular studies for TCR gene rearrangement or thorough immunophenotypic analyses of Western cases are limited. In particular, no thorough studies of U.S. populations, including immunophenotyping, EBV status, and TCR gene rearrangement studies, have been reported. In this study, we fully characterized 15 non-B-cell primary lymphomas of the sinonasal or oropharyngeal area presenting in the United States with regard to histopathologic features, immunophenotype, EBV status, and T-cell receptor gene rearrangement.
MATERIALS AND METHODS
Twelve cases of primary sinonasal and three primary oropharyngeal non-B-cell lymphoproliferative disorders presenting in the United States and diagnosed at the City of Hope National Medical Center (Duarte, CA), Harbor-UCLA Medical Center (Torrance, CA), or Kaiser Permanente Medical Center (Los Angeles, CA) from 1988 to 1999 were studied. Paraffin-embedded, formalin-fixed tissue was available from 14 cases and B-5-fixed tissue from one case. Most of the cases were seen in consultation from throughout the United States, and clinical information and follow up were limited. Cases originating from outside the United States were excluded.
Immunohistochemistry was performed on formalin-fixed (or in one case B-5-fixed), paraffin-embedded tissue sections using heat-induced epitope retrieval 3 by steaming for 20 minutes, followed by the avidin-biotin complex method on an automated immunostainer (Techmate 1000, Ventana, Tucson, AZ, USA), as previously described. 38 The following antibodies were used[colon] CD2 (Novocastra, Newcastle upon Tyne, UK), CD3 (Dako, Carpinteria, CA, USA), CD4 (Novocastra), CD5 (Novocastra), CD8 (Dako), CD16 (Novocastra), CD20 (Dako), CD30 (Dako), CD43 (Becton-Dickinson, Mountain View, CA, USA), CD56 (Zymed, South San Francisco, CA, USA), CD57 (Immunotech, Marseille, France), ALK1 (Dako), Granzyme B (Monosan, The Netherlands), TIA-1 (Immunotech), perforin (Endogen, Woburn, MA, USA), and LMP-1 (Dako). All staining was performed with appropriate positive and negative controls.
EBV RNA in-situ hybridization studies were performed using a 30-base oligonucleotide complementary to a portion of the EBER1 gene, as previously described. 8 Briefly, 10-[mgr]m sections cut from paraffin blocks were deparaffinized, rehydrated, predigested with pronase, prehybridized, and then hybridized overnight at a concentration of 0.25 ng/[mgr]L of probe. After washing, detection was accomplished using avidin-alkaline phosphatase conjugate followed by development with McGadey's substrate. A blue or brown[ndash]blue color within the nucleus was considered a positive reaction. A known EBV-positive tumor was used as a positive control in each run. Viability of total RNA was tested using a poly d(T) probe.
Double-label immunohistochemical and in-situ hybridization studies were performed on several cases on 10-[mgr]m sections from formalin-fixed, paraffin-embedded tissue. In-situ hybridization was performed first as described, followed by immunohistochemistry.
Rearrangement of TCR-[ggr] were analyzed by PCR using primer sets directed against all TCR-[ggr] variable regions (V[ggr]1[ndash]8, V[ggr]9, V[ggr]10, and V[ggr]11) and all TCR-[ggr] joining regions, as previously described. 20 Amplifiability of the DNA was confirmed by concurrent PCR amplification of the [bgr]-globin sequence.
The clinical and morphologic features of the 15 cases are summarized in Table 1. There were 12 men and 3 women, ranging from 29 to 75 years of age (median, 49 yrs). All resided in the United States. Three were Latino, and one each were of known Caucasian, Filipino, Korean, and Vietnamese background. Whether these represented first-generation immigrants or otherwise was not known. The ethnicities of the remaining patients were not known. Eleven patients presented with nasal masses, with some having ulceration, septal deviation or destruction, and/or involvement of contralateral nasal cavity or adjacent sinuses. The oropharynx (tonsils, uvula, soft palate, base of tongue) was primarily involved in three cases without clinically recognized sinonasal lesions. Lymphadenopathy was not noted on presentation in any of the cases. Only limited follow-up information could be obtained. One patient had persistence of the nasopharyngeal mass despite six cycles of CHOP. He also developed multiple skin and subcutaneous nodules on his back and upper chest 6 months after initial diagnosis, which were biopsied and shown to be involved by NK/T-cell lymphoma. Another patient showed a partial clinical response of his nasopharyngeal lesion to four cycles of CHOP but developed a large 8 cm right cervical neck mass in addition to persistence of the primary nasopharyngeal lesion.
The original diagnostic classification of the tumors varied somewhat, depending on the scheme used when the diagnosis was made. Seven cases were diagnosed as nasal or nasal-type NK/T-cell lymphoma, three cases as diffuse mixed small and large cell lymphoma, two as angiocentric T-cell lymphoma, and one case each as peripheral T-cell lymphoma (mixed cell type), angioimmunoproliferative lesion (grade 3), and polymorphic reticulosis. All cases were reviewed and reclassified according to the Working Formulation 31 solely on morphologic features (Table 1). Nine cases were thus classified as diffuse mixed small and large cell lymphoma, five as diffuse large cell lymphoma, and one case as a diffuse small cleaved cell lymphoma based on sheets of irregular small-to medium-sized cells. Some representative histologic patterns are shown in Figure 1. All cases would be classified as angiocentric lymphoma in the REAL classification 22 or as extranodal NK/T-cell lymphoma, nasal type in the provisional WHO Classification. 21 Case no. 10 could alternatively be considered a peripheral T-cell lymphoma because of negative CD56 and the presence of TCR-[ggr] chain gene rearrangement, although it was positive for EBV EBER-ISH and cytotoxic markers. Twelve of the 15 cases showed both necrosis and an angiocentric/angiodestructive pattern. The remaining three cases lacked either feature. Most of the cases had large numbers of admixed histiocytes with more variable numbers of plasma cells. One case had numerous eosinophils.
The results of the immunophenotypic studies are summarized in Table 2. All cases were positive for cytoplasmic CD3 and CD43, and negative for CD20. Twelve of 13 were CD2-positive. The natural killer cell-associated marker CD56 was positive in 10 of 15. CD8 was positive in 3 of 15, whereas CD4 was negative in all cases. CD5 was positive in 5 of 12 cases. Cytotoxic markers TIA-1 and granzyme B were positive in all 14 cases studied, whereas perforin was positive in only 5 of 14. The natural killer cell-associated markers CD16 and CD57 were negative in all cases studied (0 of 14), although numerous CD16-positive histiocytes were seen in many of the cases. Epstein-Barr virus LMP-1 was positive in 3 of 14 cases. A typical immunoprofile is shown in Figure 2. CD30 was positive in 3 of 15 cases, each of which had a large pleomorphic cell morphology. CD30 positivity ranged from only a subpopulation of cells to strong, uniform staining in most tumor cells (Fig. 3). Each of these three cases was negative for ALK-1 and two were CD56-positive.
EBV RNA-positive cells were detected by EBER-ISH in all 14 cases, with the majority of tumor cells being positive in every case. One case was rejected as a result of a lack of poly d(T) staining. Double-label studies on three cases showed that the EBER-ISH-positive tumor cells were not CD16-positive, although numerous admixed CD16-positive histiocytes were present (Fig. 2D).
Only one of 12 cases tested demonstrated a TCR-[ggr] gene rearrangement by the PCR technique. This case was CD2[plus], CD8[plus], CD4-, CD56-, TIA-1[plus], granzyme B [plus], perforin [plus], and EBER-ISH-positive. The remaining 11 cases tested were negative for TCR-[ggr] gene rearrangement consistent with a true natural killer cell derivation.
Primary sinonasal non-Hodgkin's lymphomas are more common in Asian and certain South and Central American populations, but also occur rarely in the United States and Europe. The Western cases have been shown to have a similar morphology and strong association with EBV, 10[ndash]12,27,41,42 but thorough immunophenotypic and genotypic analyses have been limited to date. In fact, the one study with an extensive immunophenotypic analysis on Western cases concluded that most cases were true T-cell neoplasms, 41 contrary to the Asian and Hispanic experience. However, others 26 have questioned this conclusion because of a lack of supporting molecular studies in that report. We attempted to further characterize our cases of primary sinonasal non-B-cell lymphomas arising in U.S. patients with regard to histology, immunophenotype, EBV EBER-ISH positivity, and TCR-[ggr] status to compare them with their more frequent Asian and Hispanic counterparts.
The most common patterns were either a diffuse mixed small and large cell population (9 of 14) or a diffuse large cell infiltrate (5 of 14). Necrosis and an angiodestructive pattern were commonly seen (12 of 15) and were helpful to identify a malignant process when present. Notably, however, both necrosis and an angiodestructive pattern were absent in three cases. This may reflect sampling bias resulting from the small specimen size. Many of the cases had a prominent reactive component with many histiocytes and plasma cells obscuring the atypical lymphoid infiltrate. Immunohistochemistry and/or EBV EBER-ISH were valuable in these cases to confidently identify the malignant infiltrate. The primary sinonasal location is a clue to the possibility of a NK/T-cell lymphoma; however, the diagnosis should also be considered at other sites. Three of our cases primarily involved the oropharynx but were otherwise identical to the remaining cases. Nasal-type NK/T-cell lymphomas may also occur at primary sites in the skin, subcutis, testis, or gastrointestinal tract. 7,25
Similar to the Asian cases, all of our cases were positive for cytoplasmic CD3 and were strongly EBV EBER-ISH-positive. Cytoplasmic CD3 can be detected with polyvalent antiserum in natural killer cells as well as T-lymphocytes because it detects the epsilon fragment of CD3 that is present in the cytoplasm of NK cells. 32 Cytoplasmic CD3 is therefore not an entirely specific marker for T lineage differentiation. Surface CD3 is absent in natural killer cells using monoclonal antibodies; however, this requires fresh or frozen tissue which is not generally available. Normal NK cells are also negative for CD5, although in our series 5 of 12 cases were positive. This may represent another example of the failure of neoplastic cells to strictly adhere to the phenotype of their proposed normal counterparts. Clonality of EBV has been demonstrated in nasal NK/T-cell lymphomas, 33 implicating EBV in the pathogenesis of these neoplasms. EBV EBER-ISH is useful diagnostically in these lesions, as strong positivity has not been shown in nonneoplastic lymphoid proliferations at this site. 42 Nasopharyngeal carcinomas, which occur in a similar location and may occasionally simulate lymphoma morphologically, are also EBV-positive (albeit in the epithelial cells), but should be readily distinguished by immunohistochemistry for keratin and lymphoid markers. It is important to recognize that immunohistochemistry for EBV LMP-1, which was positive in only a minority of cases (3 of 14), is not a useful surrogate for EBER-ISH to document EBV in these tumors.
The natural killer cell-associated marker CD56 (N-CAM) was positive in 10 of 15 cases, similar to other studies of nasal NK/T-cell lymphomas. CD56 appears to be a useful marker for NK/T-cell lymphomas and is often cited as evidence of NK cell differentiation, along with the absence of TCR gene rearrangements. However, CD56 expression has also been reported in other hematolymphoid neoplasms, including aggressive NK-cell leukemia/lymphoma, blastoid NK-cell lymphoma, a subset of unspecified peripheral T-cell lymphomas, occasional intestinal T-cell lymphomas, T-cell CLL/PL, hepatosplenic T-cell lymphoma, lymphoblastic lymphoma, and anaplastic large cell lymphomas. 5,7,14 CD56 expression is also not uncommonly seen in acute myeloid leukemia. 28 This lack of specificity emphasizes the need to perform a panel of immunohistochemical markers and to incorporate other data (EBER-ISH and TCR status) to arrive at the correct diagnosis.
The other natural killer cell-associated markers, CD16 and CD57, were universally negative in our cases. Some studies show similar negative results with CD16, 9,39 whereas others paradoxically report most of their cases to be CD16-positive. 33,34 The reason for the discrepancy is unclear. In most of our cases, there were numerous reactive histiocytes throughout the lesion which were CD16-positive. Distinction between tumor cells and reactive infiltrate would be extremely difficult by frozen section immunohistochemistry. Not all of the studies reporting CD16 positivity used frozen tissue, however. In our cases, lack of tumor cell expression of CD16 was confirmed by double-label with EBV EBER-ISH. CD57 has been more consistently reported negative.
Cytotoxic markers TIA-1 and granzyme B were positive in all our cases, and perforin was positive in 5 of 14. Both normal cytotoxic T- and NK cells may express the pore-forming protein perforin and cytotoxic granule-associated proteins TIA-1 and granzyme B. Cytotoxic cell lysis results when granzymes and other proteins are delivered through pores formed by perforin in the target cell membrane. These markers are not specific for nasal NK/T-cell lymphomas and are frequently positive in a variety of other T or NK neoplasms, including large granular lymphocytic leukemia, hepatosplenic T-cell lymphoma, intestinal T-cell lymphomas, subcutaneous T-cell lymphomas, anaplastic T-cell lymphomas, and NK-like T-cell lymphomas. 5,13,14
Three cases with large pleomorphic morphology were CD30-positive. CD30 positivity ranged from a subpopulation of cells to strong and diffuse positivity. The differential diagnosis in these cases included anaplastic large cell lymphoma (ALCL); however, all were ALK-1-negative, strongly EBV EBER-ISH-positive, and presented as nasal lesions. CD30 positivity in nasal or nasal-type NK/T-cell lymphoma has been reported by others, 27,35 with both studies indicating only a subset of tumor cells as positive. The potential for confusion with ALCL is highlighted by the previously mentioned positivity of some cases of ALCL for CD56 and cytotoxic markers. There appears then to be some immunophenotypic and morphologic overlap with nasal-type NK/T-cell lymphomas, which may present in the skin as primary or metastasis. The most helpful distinguishing features appear to be EBV EBER-ISH positivity in nasal-type NK/T-cell lymphoma and ALK-1-positive immunohistochemistry in ALCL. Prior studies have confirmed a low frequency of EBV infection in ALCL, 29 whereas EBER-ISH is one of the defining features of NK/T-cell lymphoma. ALK-1 protein expression not only supports a diagnosis of ALCL, but also has been shown to be a favorable prognostic indicator. 4,19 Separation of ALK-negative ALCL from NK/T-cell lymphoma would seem to be rely more on EBV studies, although the prognostic difference between these two groups may not be significant. Aggressive clinical courses were noted by Tsang et al. in their cases of non-nasal CD56-and EBER-positive lymphomas, including several cases involving the skin. 40 Finally, diffuse strong CD30 positivity should be expected for ALCL, with most cases of CD30-positive NK/T-cell lymphoma showing only a subpopulation of tumor cells as positive. However, one of our cases of nasal NK/T-cell lymphoma showed strong, diffuse staining for CD30. Because the prognostic difference between a ALK[plus] ALCL and the more aggressive nasal-type NK/T-cell lymphomas is marked, EBV EBER-ISH studies and ALK-1[plus] immunohistochemistry may be of great value in distinguishing these two entities in difficult cases in which there is overlap of morphologic and other immunohistochemical features.
Primary sinonasal NK/T-cell lymphomas are encountered less frequently in Western populations than in Asia or parts of South or Central America, but are otherwise similar. In fact, the limited ethnic information from our patients suggests that most U.S. cases have a similar ethnic background to those non-U.S. populations with a high frequency of nasal NK/T-cell lymphoma. Although typically presenting as a destructive sinonasal lesion, identical tumors may also be seen primarily involving the oropharynx. Necrosis and an angiodestructive pattern are frequently found, although not universally. The predominant immunophenotype is cytoplasmic CD3[plus], CD56[plus], and cytotoxic marker [plus]. Other NK markers CD16 and CD57 are negative. EBV EBER-ISH is almost always positive and is useful diagnostically. TCR gene rearrangements are usually negative, suggestive of a true natural killer cell origin in most cases. Sinonasal NK/T-cell lymphomas show some morphologic and immunophenotypic overlap with other non-B-cell non-Hodgkin's lymphomas and a combined approach using a panel of immunohistochemical stains, EBV studies, and even TCR gene rearrangement studies is helpful to avoid misdiagnosis. 4
1. Arber DA, Weiss LM, Albujar PF, et al. Nasal lymphomas in Peru. Am J Surg Pathol 1993; 17[colon]392[ndash]9.
2. Aviles A, Rodriguez L, Guzman R, et al. Angiocentric T-cell lymphoma of the nose, paranasal sinuses and hard palate. Hematol Oncol 1992; 10[colon]141[ndash]7.
3. Battifora H, Alsabeh R, Jenkins KA, et al. Epitope retrieval (unmasking) in immunohistochemistry. Adv Pathol Lab Med 1995; 8[colon]101[ndash]18.
4. Cataldo KA, Jalal SM, Law ME, et al. Detection of t(2;5) in anaplastic large cell lymphoma[colon] comparison of immunohistochemical studies, FISH, and RT-PCR in paraffin-embedded tissue. Am J Surg Pathol 1999; 23[colon]1386[ndash]92.
5. Chan ACL, Ho JWY, Chiang AKS, et al. Phenotypic and cytotoxic characteristics of peripheral T-cell and NK-cell lymphomas in relation to Epstein-Barr virus association. Histopathology 1999; 34[colon]16[ndash]24.
6. Chan JKC, Ng CS, Lau WH, et al. Most nasal/nasopharyngeal lymphomas are peripheral T-cell neoplasms. Am J Surg Pathol 1987; 11[colon]418[ndash]29.
7. Chan JKC, Sin VC, Wong KF, et al. Nonnasal lymphoma expressing the natural killer cell marker CD56[colon] a clinicopathologic study of 49 cases of an uncommon aggressive neoplasm. Blood 1997; 89[colon]4501[ndash]13.
8. Chang KL, Chen Y, Shibata D, et al. Description of an in situ hybridization methodology for detection of Epstein-Barr virus RNA in paraffin-embedded tissues, with a survey of normal and neoplastic tissues. Diagn Mol Pathol 1992; 1[colon]246[ndash]55.
9. Chiang AKS, Chan ACL, Srivastava G, et al. Nasal T/natural killer (NK)-cell lymphomas are derived from Epstein-Barr virus-infected cytotoxic lymphocytes of both NK-and T-cell lineage. Int J Cancer 1997; 73[colon]332[ndash]8.
10. Cuadra-Garcia I, Proulx GM, Wu CL, et al. Sinonasal lymphoma[colon] a clinicopathlogic analysis of 58 cases from the Massachusetts General Hospital. Am J Surg Pathol 1999; 23[colon]1356[ndash]69.
11. Davison SP, Habermann TM, Strickler JG, et al. Nasal and nasopharyngeal angiocentric T-cell lymphomas. Laryngoscope 1996; 106[colon]139[ndash]43.
12. Emile JF, Boulland ML, Haioun C, et al. CD5-CD56[plus] T-cell receptor silent peripheral T-cell lymphomas are natural killer cell lymphomas. Blood 1996; 87[colon]1466[ndash]73.
13. Felgar RE, Macon WR, Kinney MC, et al. TIA-1 expression in lymphoid neoplasms. Identification of subsets with cytotoxic T lymphocyte or natural killer cell differentiation. Am J Pathol 1997; 150[colon]1893[ndash]900.
14. Felgar RE, Salhany KE, Macon WR, et al. The expression of TIA-1[plus] cytolytic-type granules and other cytolytic lymphocyte-associated markers in CD30[plus] anaplastic large cell lymphomas (ALCL)[colon] correlation with morphology, immunophenotype, ultrastructure, and clinical features. Hum Pathol 1999; 30[colon]228[ndash]36.
15. Fellbaum C, Hansmann ML, Lennert K. Malignant lymphomas of the nasal cavity and paranasal sinuses. Virchows Arch 1989; 414[colon]399[ndash]405.
16. Ferry JA, Sklar J, Zukerberg LR, et al. Nasal lymphoma[colon] a clinicopathologic study with immunophenotypic and genotypic analysis. Am J Surg Pathol 1991; 15[colon]268[ndash]79.
17. Frierson HF, Innes DJ, Mills SE, et al. Immunophenotypic analysis of sinonasal non-Hodgkin's lymphomas. Hum Pathol 1989; 20[colon]636[ndash]42.
18. Frierson HF, Mills SE, Innes DJ. Non-Hodgkin's lymphomas of the sinonasal region[colon] histologic subtypes and their clinicopathologic features. Am J Clin Pathol 1984; 81[colon]721[ndash]7.
19. Gascoyne RD, Aoun P, Wu D, et al. Prognostic significance of anaplastic lymphoma kinase (ALK) protein expression in adults with anaplastic large cell lymphoma. Blood 1999; 93[colon]3913[ndash]21.
20. Greiner TC, Raffeld M, Lutz C, et al. Analysis of T cell receptor-[ggr] gene rearrangements by denaturing gradient gel electrophoresis of GC-clamped polymerase chain reaction products. Correlation with tumor-specific sequences. Am J Pathol 1995; 146[colon]46[ndash]55.
21. Harris NL, Jaffe ES, Diebold J, et al. The World Health Organization Classification of neoplastic diseases of the hematopoietic and lymphoid tissues[colon] report of the clinical advisory committee meeting[ndash]Airlie House, Virginia, November 1997. J Clin Oncol 1999; 17[colon]3835[ndash]49.
22. Harris NL, Jaffe ES, Stein H, et al. A revised European-American classification of lymphoid neoplasms[colon] a proposal from the International Lymphoma Study Group. Blood 1994; 84[colon]1361[ndash]92.
23. Ho FCS, Choy D, Loke SL, et al. Polymorphic reticulosis and conventional lymphomas of the nose and upper aerodigestive tract. Hum Pathol 1990; 21[colon]1041[ndash]50.
24. Ho FCS, Todd D, Loke SL, et al. Clinico-pathologic features of malignant lymphomas in 294 Hong Kong Chinese patients, retrospective study covering an eight-year period. Int J Cancer 1984; 34[colon]143[ndash]8.
25. Jaffe ES, Chan JKC, Su IJ, et al. Report of the workshop on nasal and related extranodal angiocentric T/natural killer cell lymphomas. Am J Surg Pathol 1996; 20[colon]103[ndash]11.
26. Jaffe ES. Nasal and nasal-type T/NK cell lymphoma[colon] a unique form of lymphoma associated with the Epstein-Barr virus [lsqb]Commentary[rsqb]. Histopathology 1995; 27[colon]581[ndash]3.
27. Kanavaros P, Lescs MC, Briere J, et al. Nasal T-cell lymphomas[colon] a clinicopathologic entity associated with peculiar phenotype and with Epstein-Barr virus. Blood 1993; 81[colon]2688[ndash]95.
28. Khalidi HS, Medeiros LJ, Chang KL, et al. The immunophenotype of adult acute myelogenous leukemia[colon] high frequency of lymphoid antigen expression and comparison of immunophenotype, French-American[ndash]British classification, and karyotypic abnormalities. Am J Clin Pathol 1998; 109[colon]211[ndash]20.
29. Lopategui JR, Gaffey MJ, Chan JKC, et al. Infrequent association of Epstein-Barr virus with CD30-positive anaplastic large cell lymphomas from American and Asian patients. Am J Surg Pathol 1995; 19[colon]42[ndash]9.
30. Ng CS, Lo STH, Chan JKC, et al. CD56[plus] putative natural killer cell lymphomas[colon] production of cytolytic effectors and related proteins mediating tumor cell apoptosis? Hum Pathol 1997; 28[colon]1276[ndash]82.
31. Non-Hodgkin's Lymphoma Pathologic Classification Project. National Cancer Institute sponsored study of classifications of non-Hodgkin's lymphomas[colon] summary and description of a working formulation for clinical usage. Cancer
32. Ohno R, Yamaguchi M, Oka K, et al. Frequent expression of CD3[amp]epsiv; in CD3 (Leu 4)-negative T-cell lymphomas. Leukemia 1995; 9[colon]44[ndash]52.
33. Ohsawa M, Nakatsuka S, Kanno H, et al. Immunophenotypic and genotypic characterization of nasal lymphoma with polymorphic reticulosis morphology. Int J Cancer 1999; 81[colon]865[ndash]70.
34. Ohshima K, Suzumiya J, Shimazaki K, et al. Nasal T/NK cell lymphomas commonly express perforin and Fas ligand[colon] important mediators of tissue damage. Histopathology 1997; 31[colon]444[ndash]50.
35. Petrella T, Delfau-Larue MH, Caillot D, et al. Nasopharyngeal lymphomas[colon] further evidence for a natural killer cell origin. Hum Pathol 1996; 27[colon]827[ndash]33.
36. Quintanilla-Martinez L, Franklin J, Guerrero I, et al. Histological and immunophenotypic profile of nasal NK/T cell lymphomas from Peru[colon] high prevalence of p53 overexpression. Hum Pathol 1999; 30[colon]849[ndash]55.
37. Ratech H, Burke JS, Blayney DW, et al. A clinicopathologic study of malignant lymphomas of the nose, paranasal sinuses, and hard palate, including cases of lethal midline granuloma. Cancer 1989; 64[colon]2525[ndash]31.
38. Sheibani K, Tubbs RR. Enzyme immunohistochemistry[colon] technical aspects. Semin Diagn Pathol 1984; 1[colon]235[ndash]50.
39. Tao Q, Ho FCS, Loke SL, et al. Epstein-Barr virus is localized in the tumor cells of nasal lymphomas of NK, T or B cell type. Int J Cancer 1995; 60[colon]315[ndash]20.
40. Tsang WYW, Chan JKC, Yip TTC, et al. In situ localization of Epstein-Barr virus encoded RNA in non-nasal/nasopharyngeal CD56-positive and CD56-negative T-cell lymphomas. Hum Pathol 1994; 25[colon]758[ndash]65.
41. Van Gorp J, DeBruin PC, Sie-Go DMDS, et al. Nasal T-cell lymphoma[colon] a clinicopathologic and immunophenotypic analysis of 13 cases. Histopathology 1995; 2[colon]139[ndash]48.
42. Weiss LW, Gaffey MJ, Chen YY, et al. Frequency of Epstein-Barr viral DNA in `Western' sinonasal and Waldeyer's ring non-Hodgkin's lymphomas. Am J Surg Pathol 1992; 16[colon]156[ndash]62.