1197 Malignant Lymphoma in Thailand Changes in the Frequency of Malignant Lymphoma Determined from a Histopathologic and Immunophenotypic Analysis of 425 Cases at Siriraj Hospital Sanya Sukpanichnant, M.D.1 Dhira Sonakul, M.B., B.Ch., DMT&H1 Anong Piankijagum, M.D., Ph.D.2 Wanchai Wanachiwanawin, M.D., B.Sc.2 Gavivann Veerakul, M.D.3 Chularatana Mahasandana, M.D.3 Voravarn S. Tanphaichitr, M.D., M.S.3 Vinai Suvatte, M.D., Ph.D.3 1 Department of Pathology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand. 2 Department of Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand. 3 Department of Pediatrics, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand. Presented in part at the 8th Meeting of the European Association for Haematopathology, La Villette-Paris, France, April 22–25, 1996. Supported by Siriraj-China Medical Board Fund Grant No. 75-348-213. The authors thank Miss Kanittar Srisook for her expertise in paraffin section immunoperoxidase. Address for reprints: Sanya Sukpanichnant, M.D., Department of Pathology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, Thailand. Received November 17, 1997; revision received March 20, 1998; accepted March 20, 1998. © 1998 American Cancer Society BACKGROUND. Analysis of malignant lymphoma in a single institution at different periods of time can determine the changing status of the disease in the region. METHODS. To compare with the large series of 1095 lymphoma cases reported between 1957–1971 at Siriraj Hospital, the largest hospital in Thailand, a similar study was performed through histopathologic evaluation of 425 lymphoma cases diagnosed consecutively at the same institution between August 1993 and October 1995. Phenotypic analysis was performed by paraffin section-immunoperoxidase studies. RESULTS. A striking increase in lymphoma cases was noted from 73 cases/year in the first series to 189 cases/year in the second series (an increase of 158.9%). Lymphoma occurred in all age groups, with a peak incidence at the seventh decade of life. The male to female ratio decreased from 2:1 in 1957–1971 to 1.3:1 in the more recent series. The incidence of Hodgkin’s disease (HD) was found to have decreased from 28.9% to 8.5%. There were 36 cases (8.5%) of HD and 389 cases (91.5%) of non-Hodgkin’s lymphoma (NHL) reported in the second series. The subtypes of HD included 16 cases of mixed cellularity, 13 cases of nodular sclerosis, 6 cases of lymphocyte depletion, and 1 case of lymphocyte predominance. According to the Working Formulation, the 389 NHL cases included low grade (14.1%), intermediate grade (57.3%), high grade (11.3%), and miscellaneous groups (17.2%). They were classified as small lymphocytic (9.5%), follicular (11.1%), diffuse (50.9%), immunoblastic (4.1%), small noncleaved (4.4%), lymphoblastic (2.8%), anaplastic large cell (9.0%), mycosis fungoides (1.8%), hairy cell leukemia (0.3%), true histiocytic (0.5%), and extramedullary plasmacytoma (1.0%). The immunophenotypes of the 359 NHL cases available for paraffin section-immunoperoxidase studies were B-cell (71.0%), T-cell (24.5%), histiocyte (0.6%), and undetermined phenotypes (3.9%). CONCLUSIONS. The incidence of malignant lymphoma is increasing in Thailand, with a high frequency of intermediate to high grade NHL of B-cell phenotype reported. Cancer 1998;83:1197–204. © 1998 American Cancer Society. KEYWORDS: lymphoma, Hodgkin’s disease, non-Hodgkin’s lymphoma, Thailand, classification, immunophenotype, extranodal lymphoma, childhood lymphoma. I n 1980, Piankijagum et al. from Siriraj Hospital, Bangkok, Thailand, reported a series of 1,095 cases of malignant lymphoma.1 This series has been regarded as the largest series of lymphoma cases ever published from a single institution in Thailand. The data were based on biopsied materials received in the Department of Pathology at Siriraj Hospital during a 15-year period (1957–1971). The patients originated from all four regions of the country: central, north, northeast, and south. Of the total 140,323 specimens acquired during the study period, 1095 cases of malignant lymphoma were identified. The 1198 CANCER September 15, 1998 / Volume 83 / Number 6 histologic types of malignant lymphoma in that series were classified as Hodgkin’s disease (HD) (316 cases, 28.9%); reticulum cell sarcoma (472 cases; 43.1%); lymphosarcoma, including lymphocytic and lymphoblastic lymphoma (229 cases; 20.9%); giant follicular lymphoma (20 cases; 1.8%); Burkitt’s lymphoma (1 case; 0.1%); and unclassified lymphoma (57 cases; 5.2%) according to the criteria used by Thai pathologists during that period,2 which was prior to the publication regarding tumors of the hemopoietic system by Rappaport in 1966.3 The relative incidence of malignant lymphoma in different regions of Thailand followed the same pattern as the overall incidence. Male preponderance was shown with male to female ratio for non-Hodgkin’s lymphoma (NHL) ranging from 2.1:1 to 2.7:1 and was very striking for HD occurring in childhood (5.7:1). The mean age for HD was 35.4 years, whereas the mean ages for reticulum cell sarcoma and lymphosarcoma were 41.8 years and 40.5 years, respectively. Malignant lymphoma between 1957–1971 was considered uncommon at Siriraj Hospital (approximately 73 cases reported per year). However, hematologic malignancies, including malignant lymphoma and leukemia, currently are ranked as the second most common cancer at Siriraj Hospital,4 which is the largest hospital in Thailand with 2400 inpatient beds. The hospital also is regarded as the largest referral center for hematologic diseases in Thailand. Patients from all regions of the country have access to the hospital, which is a government-based hospital and one of the best known medical schools in Thailand. Due to the better understanding and more precise diagnosis of malignant lymphoma by histologic approach in conjunction with immunophenotypic studies, we believed it would be interesting to determine whether the status of malignant lymphoma has been changing since the first series of malignant lymphomas was reported from our institution by Piankijagum et al.1 The current study was conducted to analyze cases of malignant lymphoma by histopathologic and immunophenotypic means. Comparison with the recent study of NHL in Thailand reported by Intragumtornchai et al.5 also was performed. MATERIALS AND METHODS Consecutive cases of lymphoma were collected at Siriraj Hospital during the 27-month period between August 1993 and October 1995 based on biopsied materials received at the Department of Pathology, similar to the approach used in the first series of malignant lymphoma reported from our institution.1 Malignant lymphoma was established in all cases by histology in conjunction with immunophenotypic studies by paraffin section-immunoperoxidase and classified according to the Working Formulation (National Cancer Institute)6 for NHL and the Rye classification7 for HD. In all cases, the diagnosis was made in consensus by two hematopathologists (S.S. and D.S.). The Working Formulation was used in this study because it has been recommended among Thai pathologists since 1987 and is well accepted by Thai hematologists. Paraffin section-immunoperoxidase was the only technique available in our institution for immunophenotyping malignant lymphoma; therefore, no attempt was made to classify malignant lymphoma according to the revised European–American classification of lymphoid neoplasms (Revised EuropeanAmerican classification of Lymphoid neoplasms [R.E.A.L.] classification)8 due to incomplete data regarding the immunophenotypes, many of which required frozen-section-immunoperoxidase or flow cytometry. Paraffin section-immunoperoxidase was performed using the labeled streptavidin-biotin-peroxidase method to characterize B-cell by CD20 (L-26), T-cell by CD45RO (UCHL-1), CD43 (Leu22) or polyclonal CD3, histiocyte by CD68 (KP-1), Reed–Sternberg cells by CD15 (LeuM-1) and CD30 or Ki-1 (BerH2). Plasma cells were evaluated for immunoglobulins G, A, and M, as well as k, and l light chains. Leukocyte common antigen (LCA) (CD45), epithelial membrane antigen (EMA), and cytokeratin also were included in the study. In conjunction with the histologic evaluation of malignant lymphoma, the immunoperoxidase stainings were evaluated to establish the phenotype in each case as described in the standard hematopathologic textbooks.9 –13 Briefly stated, the neoplastic cells in all subtypes of HD, with the exception of nodular lymphocyte predominant subtype, showed immunoreactivity with CD15 and CD30 in the absence of immunoreactivity with LCA and EMA. The neoplastic cells in nodular, lymphocyte-predominant HD showed immunoreactivity of B-cell phenotype (CD201, CD451) and were EMA positive (1) with negative CD15 and CD30 stainings. The B-cell phenotype of NHL was established when the neoplastic cells showed immunoreactivity with CD20 and CD45. They may bear monoclonal immunoglobulin or have light chain restriction. They were negative for T-cell and histiocytic markers. The T-cell phenotype of NHL was established when the neoplastic cells showed immunoreactivity with at least one of the three T-cell markers (CD45RO, CD43, and polyclonal CD3) and CD45 with negative CD20, immunoglobulins, CD68, and CD15 stainings. The histiocytic phenotype of NHL was established when the neoplastic cells showed immunoreactivity with CD68 in the absence of pan-B, pan- Malignant Lymphoma in Thailand/Sukpanichnant et al. 1199 TABLE 1 Histologic Grades and Immunophenotypes of 389 Cases of Non-Hodgkin’s Lymphoma at Siriraj Hospital According to the Working Formulation for Clinical Usage in NHL3 Histologic grade No. of cases B-cell T-cell Histiocytic Undetermined NA Low 55 (14.1%) 223 (57.3%) 44 (11.3%) 67 (17.2%) 389 (100%) 36 (78.3%) 163 (77.6%) 27 (69.2%) 29 (45.3%) 255 (71.0%) 8 (17.4%) 41 (19.5%) 11 (28.2%) 28 (43.8%) 88 (24.5%) — 2 (4.3%) 6 (2.9%) 1 (2.6%) 5 (7.8%) 14 (3.9%) 9 Intermediate High Miscellaneous Total cases — — 2 (3.1%) 2 (0.6%) 13 5 3 30 NHL: non-Hodgkin’s lymphoma; NA: not available for paraffin section-immunoperoxidase. T-cell markers, CD15, and CD30. Plasmacytoma showed monoclonal immunoglobulin staining or light chain restriction. CD30 could be positive in NHL cases, especially NHL with anaplastic features; in such cases the neoplastic cells showed a definite B- or T-cell phenotype as described earlier. When a B- or T-cell phenotype could not be demonstrated in some CD301 NHL cases the histology in such cases was not that of HD and immunoreactivity with CD45 in the absence of cytokeratin was necessary. RESULTS Of the total 32,847 biopsied specimens acquired during the 27-month period, 425 consecutive cases of malignant lymphoma were diagnosed (average, 189 cases per year): 36 cases of HD (8.5%) and 389 cases of NHL (91.5%). The histologic subtypes of HD included 16 mixed cellularity (MCHD) (44.4%), 13 nodular sclerosis (NSHD) (36.1%), 6 lymphocyte depletion (16.7%), and 1 lymphocyte predominance (2.8%). According to the Working Formulation for clinical usage in NHL, intermediate and high grade lymphomas constituted the majority of NHL (68.6%) whereas low grade lymphoma comprised only 14.1% of all NHL (Table 1). Both HD and NHL occurred in all age groups (Fig. 1). The mean age of HD patients was 27.1 years (range, 4 –70 years) whereas the mean age of NHL patients was 47.1 years (range, 1–98 years). The age peak in HD was in the third decade. Nine of 36 cases of HD (2 NSHD and 7 MCHD) were in children (age , 15 years). The majority of NHL occurred between the third and eighth decades with an age peak in the seventh decade. Intermediate grade NHL was the most common in all age groups except for the first decade of life in which high grade NHL was most common. Low grade NHL occurred in the fourth decade and later. The male to female ratio was 1.3:1 overall. We found no FIGURE 1. Age distribution of the patients with malignant lymphoma at Siriraj Hospital (1993–1995) demonstrated by histogram. Low, intermediate, and high grades of non-Hodgkin’s lymphoma were classified according to the Working Formulation of the National Cancer Institute.6 HD: Hodgkin’s disease; NHL: non-Hodgkin’s lymphoma. gender differences in adult HD (1.3:1) or low and intermediate grade NHL (1.2:1 each), but there was male preponderance in childhood HD (3.5:1), high grade NHL (1.9:1), and anaplastic large cell lymphoma (2.5:1). There were 359 cases (92.3%) available for paraffin section-immunoperoxidase studies. B-cell phenotype was the most common type of NHL (71.0%); T-cell phenotype comprised approximately one-fourth of all NHL (24.5%). We found only 2 cases of true histiocytic lymphoma (0.6%) and failed to demonstrate a phenotype in 3.9% of NHL (Table 1). Both B-cell and T-cell NHL occurred in all age groups. The mean age in B-cell NHL was 51.1 years (range, 2–98 years) whereas that in T-cell NHL was 36.5 years (range, 1–91 years). 1200 CANCER September 15, 1998 / Volume 83 / Number 6 TABLE 2 Histologic Subtypes and Immunophenotypes of NHL According to the Working Formulation Histologic type-no. of cases (%) No. tested B-cell (%) T-cell (%) Histiocyte (%) U (%) Small lymphocytic-37 (9.5%) Follicular, small cell-11 (2.8%) Follicular, mixed-7 (1.8%) Follicular, large cell-25 (6.4%) Diffuse, small cell-4 (1.0%) Diffuse, mixed-88 (22.6%) Diffuse, large cell-106 (27.2%) Immunoblastic-16 (4.1%) Small noncleaved-17 (4.4%) Lymphoblastic-11 (2.8%) True histiocytic-2 (0.5%) Anaplastic large cell-35 (9.0%) Hairy cell leukemia-1 (0.3%) Plasmacytoma-4 (1.0%) Mycosis fungoides-7 (1.8%) Unclassifiable-18 (4.6%) 34 7 5 23 4 85 98 16 13 10 2 35 1 4 6 16 24 (70.6%) 7 (100%) 5 (100%) 23 (100%) 4 (100%) 52 (61.2%) 84 (85.7%) 15 (93.8%) 12 (92.3%) — — 15 (42.9%) 1 (100%) 4 (100%) — 9 (56.3%) 8 (23.5%) — — — — 33 (38.8%) 8 (8.2%) 1 (6.3%) — 10 (100%) — 15 (42.9%) — — 6 (100%) 7 (43.8%) — — — — — — — — — — 2 (100%) — — — — — 2 (5.9%) — — — — — 6 (6.1%) — 1 (7.7%) — — 5 (14.3%) — — — — NHL: non-Hodgkin’s Lymphoma; U: undetermined phenotype. The age peak in B-cell NHL was in the sixth decade. The age distribution of T-cell NHL was bimodal with one peak in the second decade and the other in the seventh decade. No gender predilection was noted in either B- or T-cell NHL (male:female 5 1.3:1 and 1:1, respectively). According to the subtypes of NHL in the Working Formulation, diffuse, large cell type and diffuse, mixed cell type were the two most common types (106 cases [27.2%] and 88 cases [22.6%], respectively) (Table 2). B-cell was the most common phenotype among these two types of lymphomas (85.7% and 61.2%, respectively). In terms of histologic correlation with particular phenotypes of lymphomas in this series, follicular lymphomas of all subtypes, when tested, were of B-cell phenotype. Plasmacytoma was another B-cell neoplasm. Hairy cell leukemia, a rare condition in Thailand, was found to be of B-cell phenotype. However, we encountered a case of T-cell small lymphocytic lymphoma simulating hairy cell leukemia morphologically in the bone marrow section, but no circulating cells with cytoplasmic projections were found. Nearly all small noncleaved cell lymphomas in our series were of B-cell phenotype (92.3%); in only 1 case (7.7%) could the phenotype not be determined. The majority of immunoblastic lymphomas were B-cell neoplasms (93.8%). It is interesting to note that T-cell immunoblastic lymphoma does exist; one case was found in our series. When tested in our series all lymphoblastic lymphoma and mycosis fungoides (cutaneous T-cell lymphoma) were of T-cell phenotype. We also identified a case of diffuse large B-cell lymphoma of the paranasal sinus found to have epidermotropism by lymphoma cells in the skin nodule, simulating mycosis fungoides, at the time of recurrence. Based on tissue samples received at the Department of Pathology, Siriraj Hospital, we found that 214 of 389 cases of NHL (55.0%) had extranodal involvement by lymphoma (Table 3). The upper aerodigestive tract and gastrointestinal (GI) tract were the two most common extranodal sites of NHL (59 cases [15.2%] and 33 cases [8.5%], respectively). Among the lymphomas involving the upper aerodigestive tract were low grade (4 cases), intermediate grade (48 cases), high grade (6 cases), and plasmacytoma (1 case). T-cell lymphoma was predominant in the nasal passages (11 of 15 cases) and B-cell lymphoma was predominant in the tonsil, oral cavity and palate, and nasopharynx (Table 4). The mean age of the patients with lymphoma involving the upper aerodigestive tract was 51 years (range, 14 – 85 years). The male to female ratio was 1.1:1. Among the lymphomas involving the GI tract, 30 cases (91%) were primary GI lymphoma and the other 3 cases were comprised of GI involvement in advanced nodal lymphoma. The histologic subtypes of GI lymphomas were low grade (9 cases), intermediate grade (16 cases), and high grade (8 cases). Nearly all were B-cell lymphoma when tested for immunophenotype (B-cell: 26 cases; T-cell: 1 case; and undetermined phenotype: 3 cases). Four of nine small lymphocytic lymphomas were characteristic of low grade lymphoma of mucosa-associated lymphoid tissue, a recently recognized lymphoma14; two additional cases were suggestive of this specific lymphoma. The only case of low grade T-cell lymphoma of the ileum was Malignant Lymphoma in Thailand/Sukpanichnant et al. 1201 TABLE 3 The Immunophenotypes of NHL Involving Extranodal Sites Based on Tissue Samples Submitted for Pathologic Examination Site of involvement No. of cases B-cell T-cell U NA Upper aerodigestive Gastrointestinal Skin Soft tissue Liver Bone Spleen Brain Lacrimal gland, eyelid, and conjunctiva Pleura Retroperitoneum Mediastinum Salivary gland Subcutaneous tissue Ovary Testis Orbit Pelvic mass Abdominal mass Thyroid gland Pancreas Trachea Breast Extradural Total 59 33 19 17 13 12 9 8 6 6 5 4 4 4 2 2 2 2 2 1 1 1 1 1 214 34 27 4 12 6 6 2 7 6 — 5 2 4 — 1 2 2 2 1 1 1 — — — 125 14 1 14 2 7 1 7 — — 5 — 2 — 3 — — — — — — — 1 — — 57 3 3 — — — 2 — — — 1 — — — — — — — — — — — — — — 9 8 2 1 3 — 3 — 1 — — — — — 1 1 — — — 1 — — — 1 1 23 NHL: non-Hodgkin’s lymphoma; U: undetermined phenotype; NA: not available for paraffin section-immunoperoxidase. TABLE 4 The Immunophenotypes of NHL Involving the Upper Aerodigestive Tract TABLE 5 Histologic and Immunophenotypic Findings in 40 Cases of Childhood NHL Site of involvement (no.) B-cell T-cell U NA Histologic type (no. of cases) B-cell T-cell U NA Nasal (15) Nasopharynx (9) Paranasal sinuses (5) Tonsil (18) Oral cavity and palate (8) Tongue (4) 2 6 3 13 6 4 11 1 1 — 1 — 1 — 1 1 — — 1 2 — 4 1 — Small lymphocytic (4) Large cell (15) Small noncleaved cell (11) Lymphoblastic (5) Unclassifiable (5) Total cases (40) — 4 7 — 2 13 (36.1%) 3 10 — 4 3 20 (55.6%) 1 1 1 — — 3 (8.3%) — — 3 1 — 4 NHL: non-Hodgkin;s lymphoma; U: undetermined phenotype; NA: not available for paraffin sectionimmunoperoxidase. enteropathy-associated, T-cell lymphoma, another recently described entity.14 This patient had chronic diarrhea and was misdiagnosed by other hospitals as having immunoproliferative small intestinal disease for years. The mean age of the patients with lymphoma involving the GI tract was 45.3 years (range, 2– 81 years). The male to female ratio was 2.3:1. Childhood NHL constituted 10.3% of all NHL in our series because only 40 patients were age , 15 years. The histologic types according to the Working Formulation and the phenotypes of childhood NHL NHL: non-Hodgkin’s lymphoma; U: undetermined phenotype; NA: not available for paraffin sectionimmunoperoxidase. are demonstrated in Table 5. Practically, childhood NHL mainly includes lymphoblastic, small noncleaved cell, and large cell lymphomas.15 Therefore, large cell lymphoma constituted 37.5% of all childhood NHL when diffuse lymphoma and anaplastic large cell lymphoma were included. Small noncleaved cell lymphoma was the second most common (27.5%) and lymphoblastic lymphoma constituted 12.5% of all childhood NHLs. T-cell lymphomas were more prevalent in childhood (55.6%) when compared with the 1202 CANCER September 15, 1998 / Volume 83 / Number 6 TABLE 6 The Changing Status of Malignant Lymphoma at Siriraj Hospital: Comparison between the Series Reported by Piankijagum et al.1 and the Current Study Period of study No. of patients Average no. of cases/yr Age peak incidence Male:female NHL/HD (%) Piankijagum et al. The current study 15 years (1957–1971) 1095 73 6th decade 2:1 71.1/28.9 27 months (1993–1995) 425 189 7th decade 1.3:1 91.5/8.5 NHL: non-Hodgkin’s lymphoma; HD: Hodgkin’s disease. frequency of T-cell lymphoma in the overall cases of NHL (24.5%). Only 5 cases of acquired immunodeficiency syndrome (AIDS)-associated lymphoma could be documented. All these patients were male, ranging in age from 21–27 years. Three of these patients had lymph node enlargement that was proven to be small noncleaved cell lymphoma; two of these patients had a B-cell phenotype when tested. Another patient had lymph node enlargement that was proven to be a high grade lymphoma of unclassifiable type due to extensive necrosis; however, B-cell phenotype was established in this case. The fifth patient had advanced disease presenting with a tibial lesion proven to be anaplastic large cell lymphoma of T-cell phenotype. DISCUSSION To our knowledge, the current series of malignant lymphoma is the largest one in Thailand to study both histology and immunophenotypes of NHL. When compared with the first series from our institution conducted between 1957–1971,1 the rising number of cases of malignant lymphoma was striking (Table 6), from 73 cases per year to 189 cases per year (an increase of 158.9%), whereas the number of biopsied specimens increased from 9355 specimens per year to 14,599 specimens per year (56.1%). We believe that there are several reasons for this changing status in malignant lymphoma. First, histologic evaluation in conjunction with immunophenotypic studies plays an important role in the diagnosis of malignant lymphoma in several cases that might have been diagnosed as undifferentiated neoplasm, carcinoma, or even sarcoma in the past based on histology alone. Second, there are more hematologists to recognize more cases of malignant lymphoma. At the time of the first series from our institution, there were only 2 centers for hematologic diseases and approximately 15 hematologists throughout the country. To date, there are 14 centers (8 of which are in medical schools: 5 in the central region and 1 each in the north, northeast, and south) and 105 registered hematologists (data provided by the Society of Hematology of Thailand in 1996). Third, the improved standard of the health care system in Thailand provides easier patient access to the hospitals. Nevertheless, we believe that the incidence of malignant lymphoma in Thailand is increasing when we combine the number of adult NHL from the series recently reported by Intragumtornchai et al.5 (278 cases per year) with the number of malignant lymphomas in the current series (189 cases per year) (at least 467 cases of malignant lymphoma that are newly diagnosed per year in Thailand). A tumor registry of malignant lymphoma and a nationwide epidemiologic study are needed to confirm the increasing incidence of malignant lymphoma. Moreover, the incidence of HD has decreased significantly. We believe the main reason for such a change is the better characterization of lymphoma with the help of immunohistochemistry. Intragumtornchai et al.5 recently reported a large series of 1391 cases of adult NHL (age $ 15 years) in Thailand from a compilation of cases diagnosed in 6 medical centers during a 5-year period (1988 –1992) from 4 different regions of the country (central [3 centers; 554 cases], north [1 center; 391 cases], northeast [1 center; 223 cases], and south [1 center; 223 cases]) varying from 20 –78 cases per year in these 6 medical centers. The histologic types of NHL in each center were not reviewed by the panel of pathologists for a consensus diagnosis; however, the authors concluded that the frequencies of the pathologic subtypes reported from each center followed a similar pattern. Only 130 of the total 1391 cases (9.3%) were immunophenotyped in their series. Nevertheless, our results agree with their series and other series in Asia that show that low grade lymphoma is less common in Asia than in the U.S. whereas intermediate to high grade lymphomas are more common in Asia than in the U.S.16 This remains true when we examine adult NHL in our series. Of the 349 cases, there were 51 cases of low grade (14.6%), 212 cases of intermediate grade (60.7%), 28 cases of high grade (8.0%), and 58 cases of miscellaneous group including anaplastic large cell lymphoma (ALCL) (16.6%). The immunophenotypes of NHL from our series confirms those from the series reported by Intragumtornchai et al.5 (Table 7). This relative frequency is similar to that published in the standard textbooks from the U.S. and Europe.9 –13 When the relative frequency of B-cell and T-cell lymphoma is compared with that reported in Taiwan, China, Hong Kong, and Japan, there is less T-cell lymphoma in Thailand.16 Human T-cell lymphoma virus- Malignant Lymphoma in Thailand/Sukpanichnant et al. TABLE 7 Comparison of the Immunophenotype in NHL with Histologic Correlation between the Current Study at Siriraj Hospital and the Series Reported by Intragumtornchai et al.5 Immunophenotype and histologic grade The current study (359 cases) Intragumtornchai et al. (130 cases) B-cell Low grade Intermediate grade High grade Miscellaneous T-cell Low grade Intermediate grade High grade Miscellaneous Histiocytic Undetermined phenotype 255 (71.0%) 36 (10.0%) 163 (45.4%) 27 (7.5%) 29 (8.1%) 88 (24.5%) 8 (2.2%) 41 (11.4%) 11 (3.1%) 28 (7.8%) 2 (0.6%) 100 (76.9%) 14 (10.8%) 78 (60.0%) 6 (4.6%) 2 (1.5%) 21 (16.2%) — 11 (8.5%) 3 (2.3%) 7 (5.4%) — 14 (3.9%) 9 (6.9%) type 1 may be associated with the higher frequency of T-cell lymphoma in Japan and Taiwan.17 To the best of our knowledge, adult T-cell leukemia/lymphoma has not been reported in Thailand. The incidence of follicular lymphoma in our series was low (43 cases; 11.1%) when compared with that encountered in the U.S.9 However, the frequency of follicular lymphoma in our series was higher than that reported by Intragumtornchai et al. (3.8%). Interpathologist agreement and intrapathologist reproducibility in making a histologic diagnosis of follicular lymphoma are well recognized.18 In Thailand, to our knowledge no study has addressed such reproducibility in diagnosing follicular lymphoma. In our experience, we have encountered a number of cases of follicular lymphoma with a vaguely follicular or nodular appearance that were misdiagnosed as diffuse lymphoma by outside pathologists. However, these cases demonstrated a more distinctly follicular appearance at the time of recut for better histologic sections or subsequent biopsies. The frequency of small lymphocytic lymphoma in our series (9.5%) was comparable to that of Intragumtornchai et al. (10%). We have previously confirmed the usefulness of histologic recognition for certain types of NHL to predict the immunophenotypes of lymphoma, including follicular lymphoma (B-cell), plasmacytoma (B-cell), small noncleaved cell lymphoma (B-cell), lymphoblastic lymphoma (T-cell), and mycosis fungoides or cutaneous T-cell lymphoma (T-cell).9 –13 However, we should emphasize that at times distinguishing between various types of NHL is not possible even in cases with appropriate fixation of the specimens and high quality histologic sections. In this situation, mul- 1203 tiparameter analysis to establish a definite diagnosis includes immunophenotyping by paraffin section- or frozen section-immunoperoxidase or by flow cytometry, electron microscopy, cytogenetic study, and molecular genetic study.9 –13 ALCL in our series was diagnosed by the anaplastic features of the lymphoma cells, the sinus distribution, and the immunoreactivity with CD30.19 We found that ALCL was of either B-cell or T-cell phenotype equally (42.9%) and 14.3% were of undetermined phenotype. No attempt was made to follow the R.E.A.L. classification that placed ALCL of B-cell phenotype into the diffuse large B-cell lymphoma category.8 The common extranodal involvement in NHL is well documented in the literature9 –13 and confirmed in our series (55.0%). Lymphoma can occur in any organ or tissue due to the ubiquity of hematopoietic lymphoid system; therefore, lymphoma often is included in the differential diagnosis of neoplasm. Awareness of the possible existence of extranodal lymphoma correctly should lead to a rapid diagnosis. This fact should be recognized by physicians who are not hematologists or oncologists. It is interesting to examine why certain types of lymphomas tend to have one particular immunophenotype. The higher frequency of T-cell lymphoma in childhood is another interesting issue. Because malignant lymphoma is comprised of many different types with regard to biologic and clinical behavior, it will be very useful to determine which particular groups of lymphomas that can be cured, controlled, or even prevented. For instance, the Epstein–Barr virus (EBV) is among a number of etiologic agents proposed in lymphomagenesis.20 Recently, EBV genomes were demonstrated in most nasal T-cell/natural killer cell lymphomas.21 One case of nasal T-cell lymphoma in our series was included in that article via a T-cell lymphoma workshop organized in Hong Kong in 1995; in this case EBV genomes were demonstrated by in situ hybridization.21 EBV genomes have been found in many types of malignant lymphoma of either B-cell or T-cell phenotype.20 Despite the high rate of human immunodeficiency virus infection in Thailand, only 5 cases of documented AIDS-associated lymphoma were noted in this series (5 of 389 NHL; 1.3%). Similarly, Kaposi’s sarcoma is not common in AIDS patients in Thailand.22 The underlying reason for this unique feature of AIDS-related neoplasms in Thailand is not known and further investigation is needed. The current study emphasizes the significance of lymphoma especially of NHL. The significant increase in the number of cases of NHL encountered at Siriraj 1204 CANCER September 15, 1998 / Volume 83 / Number 6 Hospital requires further study regarding epidemiology and lymphomagenesis. REFERENCES 1. Piankijagum A, Pacharee P, Wasi P. Malignant lymphomas in Thailand: an analysis of 1,095 cases. J Med Assoc Thai 1980;63:181–91. 2. Gall EA, Mallory TB. Malignant lymphoma. A clinicopathologic survey of 618 cases. Am J Pathol 1942;18:381– 429. 3. Rappaport H. Tumors of the hematopoietic system. Washington, DC: Armed Forces Institute of Pathology, 1966. 4. Siriraj Cancer Center. Tumour registry. Statistical report 1996. Bangkok: Faculty of Medicine Siriraj Hospital, Mahidol University, 1996. 5. Intragumtornchai T, Wannakrairoj P, Chaimongkol B, Bhoopat L, Lekhakula A, Thamprasit T, et al. Non-Hodgkin’s lymphomas in Thailand. A retrospective pathologic and clinical analysis of 1391 cases. Cancer 1996;78:1813–9. 6. The Non-Hodgkin’s Lymphoma Pathologic Classification Project. National Cancer Institute sponsored study of classifications of non-Hodgkin’s lymphomas: summary and description of a working formulation for clinical usage. Cancer 1982;49:2112–35. 7. Lukes RJ, Craver LF, Hall TC, Rappaport H, Rubin P. Report of the nomenclature committee. Cancer Res 1966;26:1311. 8. Harris NL, Jaffe ES, Stein H, Banks PM, Chan JKC, Cleary ML, et al. A revised European–American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood 1994;84:1361–92. 9. Lukes RJ, Collins RD. Tumor of the hematopoietic system. Washington, DC: Armed Forces Institute of Pathology, 1992. 10. Lennert K, Feller AC. Histopathology of non-Hodgkin’s lymphomas (based on the updated Kiel classification). Berlin: Springer-Verlag, 1990. 11. Warnke RA, Weiss LM, Chan JKC, Cleary ML, Dorfman RF. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. Tumors of the lymph nodes and spleen. Washington, DC: Armed Forces Institute of Pathology, 1995. Jaffe ES. Surgical pathology of the lymph nodes and related organs. 2nd edition Philadelphia: W.B. Saunders, 1995. Knowles DM, editor. Neoplastic hematopathology. Baltimore: Williams & Wilkins, 1992. Isaacson PG. Gastrointestinal lymphomas and lymphoid hyperplasias. In: Knowles DM, editor. Neoplastic hematopathology. Baltimore: Williams & Wilkins, 1992:953–78. Kjeldsberg CR, Wilson JF, Berard CW. Non-Hodgkin’s lymphoma in children. Hum Pathol 1983;14:612–27. Shih L-Y, Liang D-C. Non-Hodgkin’s lymphomas in Asia. Hematol Oncol Clin North Am 1991;5:983–1001. Watanabe S, Mukai K, Shimomaya M. Adult T cell leukemia/ lymphoma. In: Knowles DM, editor. Neoplastic hematopathology. Baltimore: Williams & Wilkins, 1992:1281–94. NCI Non-Hodgkin’s Classification Project Writing Committee. Classification of non-Hodgkin’s lymphomas. Reproducibility of major classification systems. Cancer 1985;55:91–5. Kinney MC, Greer JP, Glick AD, Salhany KE, Collins RD. Anaplastic large cell Ki-1 malignant lymphomas: recognition, biological and clinical implications. Pathol Annu 1991: 26:1–24. Weiss LM, Chang KL. Association of the Epstein-Barr virus with hematolymphoid neoplasia. Adv Anat Pathol 1996;3:1– 15. Jaffe ES, Chan JK, Su IL, Frizzera G, Mori S, Feller AC, et al. Report of the workshop on nasal and related extranodal angiocentric T/natural killer cell lymphomas. Definitions, differential diagnosis, and epidemiology. Am J Surg Pathol 1996;20:103–11. Bhoopat L, Thamprasert K, Chaiwun B, Attasiri C, Vithayasai P, Chaimongkol B, et al. Histopathologic spectrum of AIDSassociated lesions in Maharaj Nakorn Chaing Mai Hospital. Asian Pac J Allergy Immunol 1994;12:95–104.