|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Pyothorax-Associated Lymphoma: A Review of 106 CasesByFrom the Department of Pathology, Osaka University Graduate School of Medicine, Suita; and Pathology and Surgery Sections, National Kinki-Chuo Hospital, Sakai, Osaka, Japan. Address reprint requests to Katsuyuki Aozasa, MD, Department of Pathology (C3), Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan; email: aozasa{at}molpath.med.osaka-u.ac.jp
PURPOSE: Pyothorax-associated lymphoma (PAL) is a non-Hodgkin’s lymphoma developing in the pleural cavity after a long-standing history of pyothorax. Full details of PAL are provided here. PATIENTS AND METHODS: Clinical and pathologic findings were reviewed in 106 patients with PAL collected through a nationwide survey in Japan. RESULTS: Age of the patients with PAL was 46 to 82 years (median, 64 years), with a male/female ratio of 12.3:1. All patients had a 20- to 64-year (median, 37-year) history of pyothorax resulting from artificial pneumothorax for treatment of pulmonary tuberculosis (80%) or tuberculous pleuritis (17%). The most common symptoms on admission were chest and/or back pain (57%) and fever (43%). Laboratory data showed that the serum neuron-specific enolase level was occasionally elevated (3.55 to 168.7 ng/mL; median, 18.65 ng/mL), suggesting a possible diagnosis of small-cell lung cancer. Histologically, PAL usually showed a diffuse proliferation of large cells of B-cell type (88%). In situ hybridization study showed that PAL in 70% of the patients was Epstein-Barr virus (EBV)-positive. PAL was responsive to chemotherapy, but the overall prognosis was poor, with a 5-year survival of 21.6%. CONCLUSION: This study established the distinct nature of PAL as a disease entity. PAL is a non-Hodgkin’s lymphoma of exclusively B-cell phenotype in the pleural cavity of patients with long-standing history of pyothorax, and is strongly associated with EBV infection. Development of PAL is closely related to antecedent chronic inflammatory condition; therefore, PAL should be defined as malignant lymphoma developing in chronic inflammation.
IN 1987, WE REPORTED the development of pleural lymphoma in three (2.2%) of 134 patients with chronic pyothorax (CP) at one of the hospitals specializing in chest diseases in Osaka, Japan, during the period from 1971 to 1985.1 We regarded CP to be etiologically important in the development of pleural lymphoma because (1) pleural lymphomas were never found in 984 cases of malignant lymphomas and related disorders in general hospitals in Osaka during the same period2; (2) a review of the Annual of Pathological Autopsy Cases in Japan (1974 to 1985) revealed six cases of pleural lymphoma, all of which were associated with CP; and (3) a review of pleural lymphoma cases reported in Japanese journals on chest diseases revealed that all were associated with CP. From the subsequent clinicopathologic examination in 37 cases of pleural lymphoma,3 Aozasa et al2 concluded that this is a distinct type of lymphoma and proposed the term pyothorax-associated lymphoma (PAL), defined as non-Hodgkin’s lymphoma (NHL) of exclusively B-cell phenotype developing in the pleural cavity of patients with CP. In 1993, Japanese investigators suggested the association of Epstein-Barr virus (EBV) with PAL development on the basis of a study of a total of nine cases of PAL.4 They reported (1) elevated serum values of anti-EBV antibodies in all patients; (2) the presence of EBV genomes in the nuclei of tumor cells; and (3) the expression of latent infection genes such as EB nuclear antigen 2 (EBNA-2) and latent membrane protein 1 (LMP-1) in the tumor cells. Our subsequent studies on EBV in histologic samples5 and cell lines established from PAL cases6 also showed the etiologic importance of EBV in the development of PAL. Here, we summarize the clinical and pathologic findings in 106 patients with PAL collected through a nationwide survey in Japan during the period 1985 to 2002.
Case Collection After the first nationwide study of PAL in Japan, we collected data from a total of 126 cases of PAL. Clinical and pathologic findings could be reviewed in 106 cases. Twenty-five of these cases were collected from our affiliated hospitals and the remaining 81 cases were from a review of the Annual of Pathological Autopsy Cases in Japan or from the records of meetings on thoracic diseases held in Japan. These patients were admitted to hospitals during the period 1972 to 2000. The age of the patients at first admission for PAL was 46 to 82 years (median, 64 years), with a male/female ratio of 12.3:1, thus showing male predominance. Diagnosis of PAL was made by histologic examination of biopsy or surgically resected samples in 96 (91%) of the 106 patients and by autopsy in 10 (9%) patients (Fig 1). These 10 patients had been treated for nonneoplastic diseases or malignancy of an unknown nature. Histologic specimens were fixed in 10% formalin and routinely processed for paraffin embedding. Histologic sections were cut at 4 µm and stained with hematoxylin and eosin and by immunoperoxidase procedure. Tumors were histologically classified according to the World Health Organization classification.7
Immunohistochemistry Immunohistochemical study on the paraffin-embedded sections was carried out using the avidin-biotin-peroxidase complex method. The primary antibodies used in the study, their suppliers, and dilutions were as follows: CD3 (Dakopatts, Glostrup, Denmark; 1:100), CD43 (Bioscience, Emmenbrucke, Switzerland; 1:50), CD20 (Kyowa Medex, Tokyo, Japan; 1:200), CD45RO (Dakopatts; 1:100), MB-1 (Bioscience; 1:50), CS1-4 (LMP-1) (Dakopatts; 1:20), and PE-2 (EBNA-2) (Dakopatts; 1:10). The sections were treated with 0.1% trypsin solution (Sigma, St Louis, MO) at 37°C for 30 minutes before reaction with anti-CD3 for antigen retrieval.
In Situ Hybridization for EBV
Clinical Findings History of pyothorax. All patients with PAL were admitted to hospital after a 20- to 64-year (median, 37-year) history of pyothorax resulting from artificial pneumothorax for treatment of pulmonary tuberculosis (80% of 98 patients) or tuberculous pleuritis (17%). Three patients had suffered from nontuberculous pyothorax. Fistula formation from the empyema space was found in 11 (31%) of 35 patients (internal fistula in seven patients, external in one, and unknown in three). Records regarding examination for the causal organisms of pyothorax were available in 45 patients. No bacteria were identified in 20 patients (44%), Mycobacterium tuberculosis in five (11%), and nontuberculous bacilli in 20 (44%). Surgical procedures for the treatment of pyothorax, such as decortication, obliteration of the empyema space, and thoracoplasty, were performed in 11 patients after admission for PAL. Presenting symptoms. The common presenting symptoms of PAL were pain in the chest and/or back (57% of 104 patients) and fever (43%). A few patients complained of pain in the abdomen (7%), shoulder (7%), and upper limbs (2%). Forty percent of the patients presented with a tumor or swelling in the chest wall. Respiratory symptoms such as productive cough, often with hemoptysis and dyspnea, were also observed in 24% of patients. One patient was admitted with cardiac tamponade as a result of invasion of the tumor into the pericardium. Two patients showed paralysis of the lower limbs on admission caused by tumor invasion into the spinal cord. Radiographic and physical findings. Diagnoses at admission were nonneoplastic diseases in 53% of 87 patients: CP (35 patients), chest wall tuberculosis (five), and pleuritis (four) (Table 1). Forty percent of the patients were diagnosed as having CP complicated with malignancies. The remaining patients were diagnosed as having lung cancer (three patients), metastatic tumor (two patients), or primary liver tumor (one patient). Chest radiography and computed tomography (CT) revealed mass shadows in 47% of 93 patients and 85% of 81 patients, respectively. The main masses detected by radiographic investigations were situated in the pleura (80% of 99 patients), pleura and lung (10%), and lung near the pleura (7%).
Laboratory data. Thirty-one (39%) of 79 patients showed leukocytosis (more than 10 x 103/mm3) without leukemic cells. The serum levels of CRP and LDH were often highly elevated; the values ranged from 0 to 26.1 mg/dL (n = 44; median, 9.25 mg/dL) and 142 to 4,723 IU/L (n = 76; median, 575 IU/L), respectively. As for tumor markers, the serum levels of neuron-specific enolase (NSE) were measured in some PAL patients, showing mildly elevated values: 3.55 to 168.7 ng/mL (n = 22; median, 18.65 ng/mL), suggesting a possible diagnosis of small-cell lung cancer. Treatment. Nineteen patients received surgical resection of the tumor, because involvement of the tumor was confined to the primary tumor site. Eighty-four (79%) of 106 patients received chemotherapy and/or radiotherapy. The chemotherapeutic protocols are listed in Table 2: the most common agents used were cyclophosphamide, doxorubicin, vincristine, and prednisolone (CHOP, VEPA). The radiation dosage ranged from 0.4 to 108 Gy (median, 48 Gy). The clinical outcome was evaluated according to the guideline of the International Workshop to standardize response criteria for NHL.9 After chemotherapy and/or radiotherapy, 57% of 84 patients showed complete or partial response, indicating a similar response of PAL to adjuvant therapy found in large B-cell lymphomas in other organs. Twenty-six patients did not respond to adjuvant therapy; 11 had progressive disease and 15 showed no response. Fifteen patients did not receive any therapy for PAL.
Complications. Infection was the most common complication throughout the clinical course of PAL (23 of 52 patients [44%]): sepsis in six, pneumonia in 10, relapse of active inflammation in the empyema space in two, urinary tract infection in two, and herpes zoster in one. Other complications were respiratory failure in eight cases, bone marrow suppression in seven, and bleeding tendency including disseminated intravascular coagulation syndrome in four. Malignancies other than PAL were found in nine patients: gastric cancer in four; colon cancer in two; and one case each of cervical cancer, prostate cancer, and breast cancer. All but two malignancies were detected more than 4 years before the diagnosis of PAL. Prognosis. Follow-up data were available in 102 patients. Until now, seventy-eight patients have died. Among these, 58 patients died from PAL and 17 patients died from intercalated diseases. Information regarding the cause of death was not available in three patients. The follow-up period was calculated from the date of first admission for PAL. The median survival was 9 months (range, 1 to 178 months). The estimated 1-, 3-, and 5-year survival rates were 48.6%, 27.0%, and 21.6%, respectively. Those of patients who showed complete response to chemotherapy and/or radiotherapy were better (ie, 88.2%, 63.0%, and 49.0%, respectively).
Pathologic Findings
Autopsy findings. Autopsy was performed in 36 patients. The tumors showed contiguous invasion to adjacent structures, such as the lung (58%), liver (31%), diaphragm (28%), pericardium (22%), mediastinum (22%), and spleen (11%). Noncontiguous metastatic lesions were found in the adrenal glands (17%) and intrathoracic (25%) and extrathoracic (28%) lymph nodes. CNS involvement was found in five patients (14%).
Detection of EBV Infection by Immunohistochemistry and In Situ Hybridization
Approximately 80% of 98 patients received artificial pneumothorax as treatment for lung tuberculosis. At the artificial pneumothorax, air must be injected into the pleural cavity every 2 to 4 weeks to collapse the lung affected by tuberculosis. This procedure must be continued for 2 to 4 years, and in the course of the procedure, pyothorax, usually nontuberculous in nature, might develop. The artificial pneumothorax, originally established in Western countries as a form of surgical therapy for lung tuberculosis, had been more widely performed in Japan than in Western countries, especially in the 1930s to 1950s. In the literature from Western countries, malignant lymphoma has been relatively rarely described as a complication of CP,10,11 but reports of malignant mesothelioma12 or squamous cell carcinoma13 in CP patients were common. The higher frequency of PAL among CP patients in Japan might be related to the frequent use of artificial pneumothorax or to genetic factors. A previous case-control study revealed that the artificial pneumothorax was the only significant risk factor for the development of PAL among CP patients.14 In addition, there have been no reports describing the development of pleural neoplasias in CP patients receiving surgical resection of the pleural pyogenic membrane. These findings suggest that an artificial pneumothorax leaves chronic nonhealing inflammation in the pleural cavity, which might subsequently result in the development of PAL (Fig 4). Repetitive chest radiographic examination at the artificial pneumothorax might be one of the causes of PAL development.15
A previous study showed the male/female ratio in CP to be almost equal; thus, the remarkably increased male/female ratio (12.3:1) in PAL patients suggests that males are extremely susceptible to this form of cancer. The median age of PAL patients (64 years) was close to the age of those with diffuse large B-cell lymphoma at other sites in our district (mean, 61.9 years; data not published). Chest and/or back pain were the most common presenting symptoms, which are not features of CP. CT is recommended for follow-up of patients with CP and PAL, because mass shadows could be detected in 85% of patients by CT. The usefulness of CT for detecting lymphomatous masses was also reported in other extranodal site. Physical and radiographic examination of PAL patients often suggested a diagnosis of lung cancer, especially small-cell lung carcinoma, after histologic evaluation of biopsy specimens usually showing a proliferation of round cells. Thus, the serum levels of NSE are occasionally measured in PAL patients, showing mildly elevated values. The NSE levels in PAL decreased after chemotherapy in most cases; thus, the serum NSE levels could be a useful marker for monitoring the disease activity of PAL. PAL usually showed a responsiveness to chemotherapy and radiotherapy. However, the overall prognosis was poor, with a 5-year survival of 21.6%. We previously reported that poor performance status and high serum levels of ALT and blood urea nitrogen were significantly associated with a shortened survival among PAL patients as determined using the Cox proportional hazards model.16 The performance status of PAL was 0 or 1 in 48% of 29 patients and 2 or more in the remaining cases, suggesting that the short survival in PAL might be because of the large proportion of patients with poor performance status. In this series, 70% of PAL cases were positive for EBV, indicating a close relationship between EBV infection and PAL. One previous study with combined PCR, in situ hybridization method, and immunohistochemistry showed that 85% of PALs were EBV-positive.5 EBV infection might occur at an early stage of PAL development because the fusion pattern of EBV terminal repeats in the tumor cells is monoclonal.4,6 The expression patterns of EBV latent genes among EBV-associated neoplasias are categorized as latency I, II, or III according to the expression of EBNA-1, EBNA-2, and LMP-1.17 PAL showed the same expression pattern as malignant lymphomas developing in immunocompromised patients, ie, EBNA-1+, EBNA-2+, and LMP-1+, thus categorized as latency III.4-6 PAL is a distinctive type of malignant lymphoma of mostly B-cell nature, strongly associated with EBV infection. PAL develops in chronically inflamed tissue, and is thus defined as malignant lymphoma developing in chronic inflammation. Malignant lymphoma of mucosa-associated lymphoid tissue could also be included in this category because Helicobacter pylori–induced gastritis and Hashimoto’s thyroiditis are known to have roles in predisposition to gastric lymphoma18 and thyroid lymphoma,19 respectively. These lymphomas might share common underlying mechanisms for disease processes; among them, inflammatory cytokines20 and genetic alterations induced by reactive oxygen species21 might be involved. In conclusion, the present study largely confirmed the findings of PAL found in 37 previously reported cases.3 The new findings described in the present study included causal organisms for pyothorax, the adjuvant therapies used and responsiveness to them, complications during the course, and the usefulness of the estimation of serum NSE levels for monitoring the disease activity. A more marked male/female predominance (12.3:1) and a more frequent presentation of masses or swelling in the chest (40%) compared with the previous study, and a rare example of T-cell PAL were shown. As a result, this study established the distinct nature of PAL as a disease entity.
Supported by grant nos. 12576004, 12670159, 12770087, 1270159, and 14031213 from the Ministry of Science, Culture and Sports, Japan. We thank Z. Dong, MD, A. Liu, MD, and Y. Kabutomori for their technical assistance.
1. Iuchi K, Ichimiya A, Akashi A, et al: Non-Hodgkin’s lymphoma of the pleural cavity developing from long-standing pyothorax. Cancer 60: 1771-1775, 1987[CrossRef][Medline] 2. Aozasa K, Tsujimoto M, Sakurai M, et al: Non-Hodgkin’s lymphomas in Osaka, Japan. Eur J Cancer Clin Oncol 21: 487-492, 1985[CrossRef][Medline]
3. Iuchi K, Aozasa K, Yamamoto S, et al: Non-Hodgkin’s lymphoma of the pleural cavity developing from long-standing pyothorax: Summary of clinical and pathological findings in thirty-seven cases. Jpn J Clin Oncol 19: 249-257, 1989 4. Fukayama M, Ibuka T, Hayashi Y, et al: Epstein-Barr virus in pyothorax-associated pleural lymphoma. Am J Pathol 143: 1044-1049, 1993[Abstract] 5. Ohsawa M, Tomita Y, Kanno H, et al: Role of Epstein-Barr virus in pleural lymphomagenesis. Mod Pathol 8: 848-853, 1995[Medline] 6. Kanno H, Yasunaga Y, Ohsawa M, et al: Expression of Epstein-Barr virus latent infection genes and oncogenes in lymphoma cell lines derived from pyothorax-associated lymphoma. Int J Cancer 67: 86-94, 1996[CrossRef][Medline] 7. Jaffe ES, Harris NL, Stein H, Vardiman JW: World Health Organization Classification of Tumors, Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon France, IARC Press, 2001 8. Weiss LM, Chen YY, Liu XF, et al: Epstein-Barr virus and Hodgkin’s disease: A correlation in situ hybridization and polymerase chain reaction study. Am J Pathol 139: 1259-1265, 1991[Abstract]
9. Cheson BD, Hornings SJ, Coiffier B, et al: Report of an international workshop to standardize response criteria for non-Hodgkin’s lymphomas: NCI Sponsored International Working Group. J Clin Oncol 17: 1244-1253, 1999 10. Martin A, Capron F, Lignory-Brunaud MD, et al: Epstein-Barr virus-associated primary lymphomas of the pleural cavity occurring in long-standing pleural chronic inflammation. Hum Pathol 25: 1314-1318, 1994[CrossRef][Medline] 11. Molinie V, Pouchot J, Navratil E, et al: Primary Epstein-Barr virus-related non-Hodgkin’s lymphoma of the pleural cavity following long-standing tuberculous empyema. Arch Pathol Lab Med 120: 288-291, 1996[Medline] 12. Roviano GC, Sartori F, Calabro F, et al: The association of pleural mesothelioma and tuberculosis. Am Rev Respir Dis 126: 569-571, 1982[Medline] 13. Willen R, Bruce T, Dahlstrom G, et al: Squamous epithelial cancer in metaplastic pleura following extrapleural pneumothorax for pulmonary tuberculosis. Virchows Arch A Pathol Anat Histopathol 370: 225-231, 1976[CrossRef] 14. Aozasa K, Ohsawa M, Iuchi K, et al: Artificial pneumothorax as a risk factor for development of pleural lymphoma. Jpn J Cancer Res 84: 55-57, 1993[CrossRef][Medline]
15. Hongyo T, Kurooka M, Taniguchi E, et al: Frequent p53 mutations at dipyrimidine sites in patients with pyothorax-associated lymphoma. Cancer Res 58: 1105-1107, 1998
16. Aozasa K, Ohsawa M, Iuchi K, et al: Prognostic factors for pleural lymphoma patients. Jpn J Clin Oncol 21: 417-421, 1991
17. Rowe M, Lear AL, Croom-Carter D, et al: Three pathways of Epstein-Barr virus gene activation from EBNA-1 positive latency in B lymphocytes. J Virol 66: 122-1231, 1992
18. Parsonnet J, Hausen S, Rodriguez L, et al: Helicobacter pylori infection and gastric lymphoma. N Engl J Med 329: 1835-1841, 1994 19. Aozasa K, Inoue A, Tajima K, et al: Malignant lymphomas of the thyroid gland: Analysis of 79 patients with emphasis on histologic prognostic factors. Cancer 58: 100-104, 1896 20. Kanno H, Naka N, Yasunaga Y, et al: Production of the immunosuppressive cytokine interleukin-10 by Epstein-Barr-virus-expressing pyothorax-associated lymphoma: Possible role in the development of overt lymphoma in immunocompetent hosts. Am J Pathol 150: 349-357, 1997[Abstract] 21. Miwa H, Kanno H, Munakata S, et al: Induction of chromosomal aberrations and growth-transformation of lymphoblastoid cell lines by inhibition of reactive oxygen species-induced apoptosis with interleukin-6. Lab Invest 80: 725-734, 2000[Medline] Submitted September 5, 2001; accepted June 24, 2002.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
