Journal of Surgical Oncology 49: 196-201 (1992)
Primary Pleural Mesotheliomas in South India: A 25-Year Study USHA KINI, MD, DCP, DNBE, SHAMEEM SHARIFF, MD, AND JOHNA. THOMAS, MD, DtiE From the Department of Pathology, St. john’s Medical College, Bangalore, lndia
In this report from South India, 15 patients with primary pleural mesothelioma have been diagnosed in the 25-year period 1, April 1966 through 3 1, March 1991, representing 0.02% of 76,239 biopsies received. The patients were mainly male with a mean age of 46.5 years. All except two had lived in urban Bangalore. None had been exposed to asbestos. The presentation clinically was peculiar, being continuous pricking pain, breathlessness, and cough with sputum. Physical and roentgenogram examination showed massive pleural effusion with irregular pleural thickening. Thoracotomy findings showed a distinct sessile nodularity with many slit-like spaces. Histologically, 14 were epithelial type mesotheliomas and 1 was a sarcomatous type. While the epithelial type neoplasms showed patchy squamoid differentiation, all showed mucin production. The CEA was always observed in areas of moderate differentiation. Spread occurred centrifugally to local structures on the same side as the lesion. 0 1992 Wiley-Liss, Inc. KEYWORDS: pleural neoplasms, mucin, immunohistochemistry
INTRODUCTION Mesotheliomas of the pleura have been reported from all parts of the world, with the majority of the reports being published from England, the Republic of South Africa, and the United States of America [l]. In the United States alone, the incidence of this neoplasm has been estimated to be 2.2 cases per million per year or 13 new cases per million per year [2]. Data acquired from various countries show the rate of occurrence to be between 0.02 and 0.7% [2]. Unfortunately, in India, studies on mesotheliomas have been confined chiefly to case reports alone, and even these have been few and far between [3-121. This study shows that in South India, patients commonly report relatively late in the disease process for a definitive clinical diagnosis to be made and possible treatment. This tardiness in reporting to a hospital for treatment on the part of the patient has created difficulties in early diagnosis not only for the operating surgeon but also the radiologist and pathologist where differential diagnostic confusion arises between not only primary pleural mesotheliomas and primary lung adenocarcinomas but also metastatic tumours. This study was initiated at this centre, which caters to the four Southern States of India through 73 different primary and secondary care hospitals, to essentially find 0 1992 Wiley-Liss, Inc.
out the clinicopathological characteristics of primary pleural mesothelioma (PLMT).
MATERIALS AND METHODS The archives of the Department of Pathology, St. John’s Medical College, Bangalore, India, formed the source of all data. The medical records on these patients were also examined and additional data including roentgenograms, thoracotomy findings, and follow-up were reviewed. Methods All data on primary pleural mesotheliomas were tabulated; and all specimens, tissue blocks, and primary stained paraffin sections re-examined. When considered necessary, additional tissue blocks were processed and additional stains used. The epithelial component was examined with care to determine the cytoplasmic content. The contents looked for were both keratin [13] and mucins. These mucins were examined histochemically to identify the acid and Accepted for publication November 15, 1991. Address reprint requests to Dr. John A. Thomas, MD, DHE, Department of Pathology, St. John’s Medical College, Bangalore-560 034, India.
Pleural Mesotheliomas in South India
197
neutral components using standard techniques [ 141. Mucicarmine was used as a general mucin stain; Alcian Blue 8 GX stain at pH 2.5 for acid mucins; aldehyde fuchsinAlcian Blue and Alcian Blue of different molarities (0.06M, 0.3M, and 0.9M) to identify the carboxylated, weakly sulphated, and strongly sulphated components of the acid mucins , respectively. Hyaluronidase digestion with Alcian Blue was done to detect hyaluronic acid and the presence of sialomucins was looked for with the acid hydrolysis-Alcian Blue procedure. To detect neutral mucins, besides the mucicarmine stain, the periodic acidSchiff (PAS) reaction was used with and without prior diastase digestion. The phenylhydrazine-periodic acidSchiff reaction was further done to differentiate PAS positive acid mucins from neutral mucins. The immunoperoxidase reaction to detect the expression of carcinoembryonic antigen (CEA) was sought using rabbit anti-human CEA (CRC, Cambridge, MA, USA, Product code 594015) in prediluted form, in the 4-step peroxidase-antiperoxidase method of Sternberger et al. [15]. The stroma was studied for maturity of collagen using Masson’s Trichrome stain [ 141. In addition, “ferruginous bodies” were looked for, using Perls’ reaction [ 161. “Asbestos bodies” were also looked for.
RESULTS In the 25-year period from 1, April 1966 to 3 1, March 1991, a total of 76,239 surgical biopsies were diagnosed, Fig. 1 . This 35-year-old nun presented with severe pricking pain over the left thorax and massive haemorrhagic pleural effusion. After con234 of which were from the pleura. tinuous drainage in which over 1,000 ml of fluid was drained every 24 Fifteen pleural biopsies of the 234 pleural specimens hours, this roentgenogram showed a distinct irregular patch of pleural received were diagnosed as primary PLMT, giving an thickening (arrows). No intrapulmonary lesion was observed. incidence of 0.02% of all biopsy material and 6.4% of pleural biopsies. These also include the previous series reported from this institution [ 111. The patients were in the 29-65-year age group (mean: patients required a continuous drain in which 1,000 to 46.5 years); all were male except three, the male to fe- 1,500 ml of straw-coloured to haemorrhagic turbid fluid male ratio being 4:l. All patients had lived in urban was collected every 24 hours, Roentgenograms showed all 15 patients to have had Bangalore with only two coming from rural areas. None had had any contact with either asbestos or tuberculosis. pleural effusion which was so abundant as to mask the pleural thickening and even portions of the lung. The All were referrals to this tertiary medical centre. The presenting symptoms were mainly chest pain in 14 thickening of the pleura consequently was best observed of the 15 patients. This pain was peculiar in that it was in 14 patients after drainage of the fluid (Fig. 1). The continuous and pricking in nature and of sufficient inten- thickening always appeared irregular with a tendency sity to keep the patient awake at night. Breathlessness towards occurring at the base. One patient showed no was complained of by eight patients, whose respiratory pleural thickening. Bronchoscopy did not show any lerates varied between 30 to 50 per minute at presentation. sions in any patient. Fresh fluid from pleural effusions collected for cytolCough with sputum of small quantity was present in six patients and loss of weight in two patients. Most patients ogy at regular intervals on the same patient showed irreghad combinations of these symptoms. All patients had ular and sporadic exfoliation of neoplastic cells. These been seen elsewhere by other physicians or surgeons and cells measured 50-80 pm across with altered nuclearcytoplasmic ratio, autophagocytosis, cupping, and knobs had been empirically treated with anti-tuberculin drugs. Physical examination showed all patients to have mas- on the cytoplasmic borders, satisfying the Klempman sive pleural effusion affecting the left pleural sac in ten criteria [I71 (Fig. 2). Although all such cytology preparaand the right in five. The removal of this fluid in all tions had been diagnosed as malignant, the cytopatholo-
198
Kini et al.
Fig. 2. The photomicrograph shows the exfoliate of mesothelial cells which occurred in the same patient (shown in Fig. 1) with large cells showing cannibalism (X), mitosis (Y) and a tendency towards morula ( Z ) formation and irregular cytoplasmic projections. It was also observed that these exfoliates did not occur regularly into the pleural fluid. Leishman’s stain, X400.
gist in no instance suspected a PLMT at the first presentation. Sputum when available, was non-contributory towards the diagnosis of this neoplasm. Pleural decortication was successfully done at surgical intervention on 12 patients. All specimens removed showed a definite uniform pleural thickening with the parietal layer showing prominent rib markings and numerous sessile nodules measuring 0.2 to 4 cm across (Fig. 3A,B). The sectioning of such nodules showed slit-like spaces in most instances, the slits being placed perpendicular in axis to the pleural surface (Fig. 4A). No lung involvement was observed. No evidence of a primary neoplasm was seen elsewhere in the body. In the remaining three patients, thoracotomy was not done but biopsy material from the pleura was obtained for diagnosis and showed similar appearances. Histologically, all tissue blocks examined to make the initial diagnosis on all patients showed 14 to have an epithelial type neoplasm and one a sarcomatous type. The cells lining the slit-like spaces occurring in epithelial neoplasms appeared low to tall columnar, having no basement membrane (Fig. 4B). Evidence of production of varying quantities of mucus was seen. Squamoid differentiation was observed in some but was always focal. The stroma in all 14 appeared spindle shaped and benign
looking. In the single sarcomatous neoplasm, the stroma showed marked cellular variation with mitotic activity. In no instance was an asbestos or “ferruginous body” observed. Histochemically, the acid mucin component in the 15 patient-tissues examined were found to be carboxylated in 12 of 15 patients (80%), hyaluronic acid being seen in 9 of 15 patients (60%), weakly and strongly sulphated in 8 of 15 patients (53.5%). Neutral m u c h when looked for in addition to the acid mucins were found only 3 of 15 patients (20%). Glycogen was found in 9 of 15 patients (60%). The keratin stain showed evidence of minimal quantities of keratin and prekeratin in 10 of the 15 patients. Masson’s Trichrome stain showed the stroma to be composed of principally mature collagen. In the one sarcomatous variety, a mixture of both mature and immature collagen was observed. Perls’ reaction confirmed the absence of “ferruginous bodies .” Immunohistochemically , material from 12 patients alone was examined, the material in the remaining three patients being insufficient. Nine of these showed expression for CEA which occurred intracellularly within the epithelial cells in both a predominantly focal and diffuse manner. It was also observed that this expression was seen only in areas of moderate differentiation and was
Pleural Mesotheliomas in South India
Fig. 3. Both gross photographs of decorticated specimens show the typical appearance encountered. A: Large nodules measuring 2.5 cm or greater across. B: Numerous small and moderately sized nodules measuring between 0.2 to 0.8 cm across.
seen neither in areas of good differentiation nor in areas of poor differentiation. The follow-up, though sought on all 15 patients, led to written response from relatives of three patients alone. They wrote to inform the authors that two patients had died within 6 months of the histologic diagnosis, the mode of death apparently being acute respiratory failure. One patient was still alive 6 months after diagnosis. Spread, when detected, was always on the same side of the lesion and was seen at the thoracotomy scar in one patient, axillary lymph nodes in two, along nerve trunks in one, and ribs in two cases. The remaining 12 were lost to follow-up.
DISCUSSION Mesotheliomas are neoplasms arising from mesothelial cells. These neoplasms morphologically may simulate histological appearances of carcinomas or sarcomas of various histological types [ 181. This is believed to occur
199
Fig. 4. Histologically, the nodules observed grossly showed a distinct cut surface which was better observed at low power microscopy. A: A typical nodule with slit-like spaces placed on the inner aspect of the parietal pleura. The slit like spaces were invariably at right angles to the surface of the pleura. B: A typical detail of the slit-like spaces which were lined by cuboidal epithelial-like cells, some showing evidence of mucus production. A: Haematoxylin and Eosin, X15. B: Haematoxylin and Eosin, X250.
because the mesothelial cells under the influence of various stimuli [IS] not only proliferate and become morphologically indistinguishable from epithelial cells but also resemble connective tissue cells and even endothelial cells. With pleural mesotheliomas, the pathologist is often faced with the problem of differentiating these neoplasms from primary peripheral adenocarcinomas of the lung that have infiltrated the pleura as also from metastatic neoplasms to the lung. Earlier workers considered it necessary, therefore, to autopsy every patient with PLMT [ 191 to rule out a primary neoplasm of the lung as well as a metastatic process from a primary elsewhere in the body. Wanebo et al. [20] have, however, suggested that an autopsy need not be a prerequisite towards the diagnosis of a pleural mesothelioma and a correlation
200
Kini et al.
between the clinical presentation, radiological appear- showing hyperchromatic nuclei with a surrounding colances, cytology of pleural fluid and particularly findings lagenous matrix. Such a predominant collagenous matrix at thoracotomy with characteristic gross appearances and typifies a desmoplastic response in an epithelial mesothehistological features which may enable an accurate diag- lioma [25]. A biphasic pattern with mixtures of epithelial nosis to be made. In recent years, findings at histochem- and sarcomatous patterns may also be observed. In the istry, immunohistochemistry , and electron microscopy literature, the epithelial type was the most predominant pattern followed by mixed and sarcomatous patterns in have also been added to this list [21]. PLMT is rare in the general population. This study 30% and 20%, respectively [ 161. In the present series, 14 covering a period of 25 years has reflected an incidence of the 15 were of the epithelial type and 1, the sarcomaof 0.02% and is predominantly a tumour of adult life, tous type. Mucin histochemistry has suggested that carboxylated occurring in the 50-60-year-old age group [l]. The acid mucins, particularly hyaluronic acid, were most youngest patient was 29 years old and the oldest 65 years commonly present with mixtures of weakly sulphated and with the median age being between 43 and 53 years. Most strongly sulphated groups. This has been confirmed by were males and reflected the general tendency towards [ 1 1,261. Detection of acid mucins and their several others that gender in the literature [ 11. fractions may be of help in differentiating PLMT from a Pleural mesotheliomas are found to be more common mucin producing adenocarcinoma. Eighty percent of our in coastal cities with shipyards than inland towns [22] as patients showed carboxylated acid mucins; hyaluronic this lesion is attributed to occupational exposure to asbestos [22]. Contrary to this belief, this series of cases has acid in 60%, weakly and strongly sulphated in 53.3%. shown no association with asbestos. All our patients had Neutral mucins were seen only in 20% of the patient been suspected of having tuberculosis and were being material. The keratin stain showed a positive reaction in empirically treated for this disease by their treating physi- 66.6% of the patients. Much has been written on the diagnostic potential of cians. This is understandable as tuberculosis is found to occur more frequently in the four southern states of India immunohistochemistry, particularly carcinoembryonic antigen (CEA), in differentiating PLMT from adenocar[23] with presenting features similar to PLMT. A characteristic clinical presentation of PLMT from cinomas [27,28]. The profile of strong keratin staining South India has evolved with this study. The characteris- and weak or absent CEA staining appears to characterise tic onset of PLMT appears slow and insidious, being mesotheliomas, especially the epithelial variant, and associated with continuous severe pricking chest pain and helps in distinguishing them from adenocarcinomas. In rapidly accumulating haemorrhagic pleural fluid. Addi- this study, however, a strong expression for CEA was tional irregular pleural thickening was observed on roent- seen in cells of pleural mesothelioma showing moderate genograms with a shift of the mediastinum towards the differentiation only. No expression was seen in either affected side. The typical pleural fluid cytology with well differentiated or poorly differentiated mesotheliornesothelial cell aggregates exhibiting “cupping,” canni- mas. Of the 12 cases studied, 9 showed expression for balism, knobs on the periphery of cell borders, and al- CEA. From these observations, it could be surmised that tered nuclear-cytoplasmic ratio [ 171 should characterize the expression of CEA may represent regressive foetal the lesion. The pattern was seen in 11 of the 15 cases in antigen expression with positivity neither in well differentiated nor poorly differentiated neoplasms. this study. Patients with pleural mesothelioma have a short surThe final characterisation would be at thoracotomy , with the pleura showing cobblestone nodularity encasing vival (6 months) [29]. Although metastases are often almost the entire lung but not involving the lung paren- found at autopsy in cases of PLMT, they are rarely dechyma. The appearance has been likened to a congealed tected during life. The most characteristic and usual liquid that has been poured over the lung. Decortication is method of spread appears to be contiguity, so that the difficult and results in the lung tissue being peeled off tendency of PLMT is to infiltrate along needle tracts, along with the overlying pleural neoplasm [22]. In the 12 intercostal drainage tubes, and thoracotomy scars and cases where decortication was done, the nodules were produce subcutaneous nodules [29]. Axillary lymph node seen both on the parietal and visceral layers of the pleura. involvement was seen in one patient, rib in one patient, The nodules, on further examination, showed slit-like bone marrow of rib in one patient, and infiltration into breast tissue and thoracotomy scar in one. PLMT could, spaces lying perpendicular to the pleural plane. Cytoarchitecturally PLMT shows a diversity of charac- therefore, be said to metastasize in a centrifugal manner teristics. In the epithelial type, tubules, cystic spaces, or to surrounding local structures, unlike other tumours of cystopapillary areas are seen lined by tumour epithelial- the lung. The causes of death in PLMT are pulmonary insuffilike cells exhibiting loose cohesion as well an absent basement membrane [24]. The sarcomatous pattern ciency, wasting, and metastatic disease 1251. Three of the shows spindle shaped cells of varying sizes and shapes 15 patients died of acute respiratory failure.
Pleural Mesotheliomas in South India
The authors are of the opinion that primary pleural mesotheliomas are best diagnosed thorough evaluation of all existent clinical and radiologic data as also the progression of the disease, cytologi&l appearno signifiOf pleura’ anceS On exfoliative cant bronchoscopic findings, thoracotomy obser- typical -vations at operation, gross specimen appearance, and the utilisation Of routine light microscopy aided by histochemistry and immunohistochemistry .
REFERENCES 1. Legha SS, Muggia FM: Pleural mesothelioma: clinical features
and therapeutic implications. Ann Intern Med 87:613-621, 1977. 2. Brenner J, Sordillo PP, Magill GB, Golbey RB: Malignant mesothelioma of the pleura. Review of 123 patients. Cancer 49:243 1-2435, 1982. 3. Reddy DJ, Showramma A: Malignant mesothelioma of the pleura. J Indian Med Assoc 33:139-141, 1959. 4. Lele RD, Sharma KD, Sainani GS, Rajkondawar VL: Mesothelioma of the pleura. Case report. Indian J Med Sci 14:430432, 1960. 5 . Kshirsagar VH, Patil SD, Sharma KD, Deshpande TV, Deshmane JV: Mesothelioma (Report of two autopsy cases). Indian J Cancer 8:208-214, 1971. 6. Razdan JN, Mathur KC, Bharadwaj TP: Diffuse malignant mesothelioma of the pleura. Indian J Med Sci 26:233-235, 1972. 7. Reddy DB, Raju GC, Venkatanarayana CP, Suvarnakumari G: Mesothelioma of pleura-report of a case. J Assoc Physicians India 21:453456, 1973. 8. Aikat BK, Radhakrishnan VV, Malik AK: Mesothelioma of the pleura-a study of five autopsy cases with review of literature. Indian J Cancer 11:95-101, 1974. 9. Chandurkar SN, Khandekar AL, Shivde AV, Kher AV: Primary Pleural Mesothelioma. Indian J Chest Dis 17:139-142, 1975. 10. Malik AK, Aikat BK: Primary pulmonary neoplasms-a histopathologic study. Indian J Cancer 13:149-155, 1976. 11. Mahmood S, Thomas JA: Pleural mesotheliomas. Indian J Cancer 20:247-254, 1983. 12. Bhaduri AS, Balar DB: Mesotheliomas of the pleura: A twelve year experience with fifteen cases. Indian J Cancer 25:7-15, 1988. 13. Ayoub P, Schklar G: A modification of the Mallory connective tissue stain as stain for keratin. Oral Surg Oral Med Oral Pathol 16:S8GS85, 1963. 14. Cook HC: Carbohydrates. In Bancroft JD, Stevens A (eds): “Theory and Practice of Histological Techniques.” New York: Churchill Livingstone, 1977, pp 141-167.
201
15. Sternberger LA, Hardy PH, Cuculis JJ, Meyer HG: The unla-
16.
belled antibody enzyme method by immunohistochemistry. Preparation and properties of soluble antigeniantibody complex (horseradish ueroxidase-anti-horseradish ueroxidase) and its use in ideritification of spirochaetes. J Hisiochem Cytochem 18:3 15333, 1970. Craighead JE, Abraham JL, Churg A, Green FHY, Kleinerman J, Pratt PC. Seemaver TA, Vallvathan V, Weill H: The Pathology of asbestos-associated diseases of the lungs and pleural cavities; diagnostic criteria and proposed grading schema (Report of the Pneumoconiosis Committee of the College of American Pathologists and the National Institute for Occupational Safety and Health). Arch Pathol Lab Med 106:544-596,. 1982. Klempman S: The exfoliative cytology of diffuse pleural mesothelioma. Cancer 15:691-704, 1962. McCaughey WTE: Primary tumours of the pleura. J Pathol Bacteriol76:517-529, 1958. Willis RA (ed): “Pathology of Tumours,” 4th Edition, London: Butterworths, 1967, pp 181-183. Wanebo HJ, Martini N , Melamed MR, Hilaris B, Beattie EJ: Pleural mesothelioma. Cancer 38:248 1-2488, 1976. Lucas JG, Tuttle SE: Diagnostic histochemical and immunohistochemical studies in Maligant mesothelioma. J Surg Oncol 35:30-34, 1987. Hillerdal G: Malignant mesothelioma 1982: Review of 4710 published cases. Br J Dis Chest 77:321-342, 1983. Health Information, India: By Cenral Bureau of Health Intelligence, Director General of Health Services, Ministry of Health and Family Welfare, Government of India, Nirman Bhavan, New Delhi, August 1989. Klima M, Spjit HJ, Seybold WD: Diffuse malignant mesothelioma. Am J Clin Pathol65583-600, 1976. McCaughey WTE, Kannerstein M, Churg J: Tumours and pseudotumours of the serous membranes. In McCaughey WTE, Kannerstein M, Churg J (eds): “Atlas of Tumour Pathology,” Second Series, Fascicle 20. Armed Forces Institute of Pathology, pp 1 6 7 2 , 1985. Wagner JC, Munday DE, Harington JS: Histochemical demonstration of hyaluronic acid in pleural mesotheliomas. J Pathol Bacteriol 84:73-78, 1962. Corson JM, Pinkus GS: Mesothelioma: profile of keratin proteins and carcinoembryonic antigen. An immunoperoxidase study of 20 cases and comparison with pulmonary adenocarcinomas. Am J Pathol 108:80-87, 1982. Pascal RR, Mesa-Tejada R, Bennett SJ, Garces A, Fenoglio CM: Carcinoembryonic antigen: immunohistologic identification in invasive and intraepithelial carcinomas of the lung. Arch Pathol Lab Med 101:568-571, 1977. Adams VI, et al.: Diffuse malignant mesothelioma of pleura. Diagnosis and survival in 92 cases. Cancer 58: 1540-155 1, 1986. I .
17. 18. 19.
20. 21. 22. 23.
24. 25.
26. 27.
28.
29.
Primary Pleural Mesotheliomas in South India: A 25-Year Study USHA KINI, MD, DCP, DNBE, SHAMEEM SHARIFF, MD, AND JOHNA. THOMAS, MD, DtiE From the Department of Pathology, St. john’s Medical College, Bangalore, lndia
In this report from South India, 15 patients with primary pleural mesothelioma have been diagnosed in the 25-year period 1, April 1966 through 3 1, March 1991, representing 0.02% of 76,239 biopsies received. The patients were mainly male with a mean age of 46.5 years. All except two had lived in urban Bangalore. None had been exposed to asbestos. The presentation clinically was peculiar, being continuous pricking pain, breathlessness, and cough with sputum. Physical and roentgenogram examination showed massive pleural effusion with irregular pleural thickening. Thoracotomy findings showed a distinct sessile nodularity with many slit-like spaces. Histologically, 14 were epithelial type mesotheliomas and 1 was a sarcomatous type. While the epithelial type neoplasms showed patchy squamoid differentiation, all showed mucin production. The CEA was always observed in areas of moderate differentiation. Spread occurred centrifugally to local structures on the same side as the lesion. 0 1992 Wiley-Liss, Inc. KEYWORDS: pleural neoplasms, mucin, immunohistochemistry
INTRODUCTION Mesotheliomas of the pleura have been reported from all parts of the world, with the majority of the reports being published from England, the Republic of South Africa, and the United States of America [l]. In the United States alone, the incidence of this neoplasm has been estimated to be 2.2 cases per million per year or 13 new cases per million per year [2]. Data acquired from various countries show the rate of occurrence to be between 0.02 and 0.7% [2]. Unfortunately, in India, studies on mesotheliomas have been confined chiefly to case reports alone, and even these have been few and far between [3-121. This study shows that in South India, patients commonly report relatively late in the disease process for a definitive clinical diagnosis to be made and possible treatment. This tardiness in reporting to a hospital for treatment on the part of the patient has created difficulties in early diagnosis not only for the operating surgeon but also the radiologist and pathologist where differential diagnostic confusion arises between not only primary pleural mesotheliomas and primary lung adenocarcinomas but also metastatic tumours. This study was initiated at this centre, which caters to the four Southern States of India through 73 different primary and secondary care hospitals, to essentially find 0 1992 Wiley-Liss, Inc.
out the clinicopathological characteristics of primary pleural mesothelioma (PLMT).
MATERIALS AND METHODS The archives of the Department of Pathology, St. John’s Medical College, Bangalore, India, formed the source of all data. The medical records on these patients were also examined and additional data including roentgenograms, thoracotomy findings, and follow-up were reviewed. Methods All data on primary pleural mesotheliomas were tabulated; and all specimens, tissue blocks, and primary stained paraffin sections re-examined. When considered necessary, additional tissue blocks were processed and additional stains used. The epithelial component was examined with care to determine the cytoplasmic content. The contents looked for were both keratin [13] and mucins. These mucins were examined histochemically to identify the acid and Accepted for publication November 15, 1991. Address reprint requests to Dr. John A. Thomas, MD, DHE, Department of Pathology, St. John’s Medical College, Bangalore-560 034, India.
Pleural Mesotheliomas in South India
197
neutral components using standard techniques [ 141. Mucicarmine was used as a general mucin stain; Alcian Blue 8 GX stain at pH 2.5 for acid mucins; aldehyde fuchsinAlcian Blue and Alcian Blue of different molarities (0.06M, 0.3M, and 0.9M) to identify the carboxylated, weakly sulphated, and strongly sulphated components of the acid mucins , respectively. Hyaluronidase digestion with Alcian Blue was done to detect hyaluronic acid and the presence of sialomucins was looked for with the acid hydrolysis-Alcian Blue procedure. To detect neutral mucins, besides the mucicarmine stain, the periodic acidSchiff (PAS) reaction was used with and without prior diastase digestion. The phenylhydrazine-periodic acidSchiff reaction was further done to differentiate PAS positive acid mucins from neutral mucins. The immunoperoxidase reaction to detect the expression of carcinoembryonic antigen (CEA) was sought using rabbit anti-human CEA (CRC, Cambridge, MA, USA, Product code 594015) in prediluted form, in the 4-step peroxidase-antiperoxidase method of Sternberger et al. [15]. The stroma was studied for maturity of collagen using Masson’s Trichrome stain [ 141. In addition, “ferruginous bodies” were looked for, using Perls’ reaction [ 161. “Asbestos bodies” were also looked for.
RESULTS In the 25-year period from 1, April 1966 to 3 1, March 1991, a total of 76,239 surgical biopsies were diagnosed, Fig. 1 . This 35-year-old nun presented with severe pricking pain over the left thorax and massive haemorrhagic pleural effusion. After con234 of which were from the pleura. tinuous drainage in which over 1,000 ml of fluid was drained every 24 Fifteen pleural biopsies of the 234 pleural specimens hours, this roentgenogram showed a distinct irregular patch of pleural received were diagnosed as primary PLMT, giving an thickening (arrows). No intrapulmonary lesion was observed. incidence of 0.02% of all biopsy material and 6.4% of pleural biopsies. These also include the previous series reported from this institution [ 111. The patients were in the 29-65-year age group (mean: patients required a continuous drain in which 1,000 to 46.5 years); all were male except three, the male to fe- 1,500 ml of straw-coloured to haemorrhagic turbid fluid male ratio being 4:l. All patients had lived in urban was collected every 24 hours, Roentgenograms showed all 15 patients to have had Bangalore with only two coming from rural areas. None had had any contact with either asbestos or tuberculosis. pleural effusion which was so abundant as to mask the pleural thickening and even portions of the lung. The All were referrals to this tertiary medical centre. The presenting symptoms were mainly chest pain in 14 thickening of the pleura consequently was best observed of the 15 patients. This pain was peculiar in that it was in 14 patients after drainage of the fluid (Fig. 1). The continuous and pricking in nature and of sufficient inten- thickening always appeared irregular with a tendency sity to keep the patient awake at night. Breathlessness towards occurring at the base. One patient showed no was complained of by eight patients, whose respiratory pleural thickening. Bronchoscopy did not show any lerates varied between 30 to 50 per minute at presentation. sions in any patient. Fresh fluid from pleural effusions collected for cytolCough with sputum of small quantity was present in six patients and loss of weight in two patients. Most patients ogy at regular intervals on the same patient showed irreghad combinations of these symptoms. All patients had ular and sporadic exfoliation of neoplastic cells. These been seen elsewhere by other physicians or surgeons and cells measured 50-80 pm across with altered nuclearcytoplasmic ratio, autophagocytosis, cupping, and knobs had been empirically treated with anti-tuberculin drugs. Physical examination showed all patients to have mas- on the cytoplasmic borders, satisfying the Klempman sive pleural effusion affecting the left pleural sac in ten criteria [I71 (Fig. 2). Although all such cytology preparaand the right in five. The removal of this fluid in all tions had been diagnosed as malignant, the cytopatholo-
198
Kini et al.
Fig. 2. The photomicrograph shows the exfoliate of mesothelial cells which occurred in the same patient (shown in Fig. 1) with large cells showing cannibalism (X), mitosis (Y) and a tendency towards morula ( Z ) formation and irregular cytoplasmic projections. It was also observed that these exfoliates did not occur regularly into the pleural fluid. Leishman’s stain, X400.
gist in no instance suspected a PLMT at the first presentation. Sputum when available, was non-contributory towards the diagnosis of this neoplasm. Pleural decortication was successfully done at surgical intervention on 12 patients. All specimens removed showed a definite uniform pleural thickening with the parietal layer showing prominent rib markings and numerous sessile nodules measuring 0.2 to 4 cm across (Fig. 3A,B). The sectioning of such nodules showed slit-like spaces in most instances, the slits being placed perpendicular in axis to the pleural surface (Fig. 4A). No lung involvement was observed. No evidence of a primary neoplasm was seen elsewhere in the body. In the remaining three patients, thoracotomy was not done but biopsy material from the pleura was obtained for diagnosis and showed similar appearances. Histologically, all tissue blocks examined to make the initial diagnosis on all patients showed 14 to have an epithelial type neoplasm and one a sarcomatous type. The cells lining the slit-like spaces occurring in epithelial neoplasms appeared low to tall columnar, having no basement membrane (Fig. 4B). Evidence of production of varying quantities of mucus was seen. Squamoid differentiation was observed in some but was always focal. The stroma in all 14 appeared spindle shaped and benign
looking. In the single sarcomatous neoplasm, the stroma showed marked cellular variation with mitotic activity. In no instance was an asbestos or “ferruginous body” observed. Histochemically, the acid mucin component in the 15 patient-tissues examined were found to be carboxylated in 12 of 15 patients (80%), hyaluronic acid being seen in 9 of 15 patients (60%), weakly and strongly sulphated in 8 of 15 patients (53.5%). Neutral m u c h when looked for in addition to the acid mucins were found only 3 of 15 patients (20%). Glycogen was found in 9 of 15 patients (60%). The keratin stain showed evidence of minimal quantities of keratin and prekeratin in 10 of the 15 patients. Masson’s Trichrome stain showed the stroma to be composed of principally mature collagen. In the one sarcomatous variety, a mixture of both mature and immature collagen was observed. Perls’ reaction confirmed the absence of “ferruginous bodies .” Immunohistochemically , material from 12 patients alone was examined, the material in the remaining three patients being insufficient. Nine of these showed expression for CEA which occurred intracellularly within the epithelial cells in both a predominantly focal and diffuse manner. It was also observed that this expression was seen only in areas of moderate differentiation and was
Pleural Mesotheliomas in South India
Fig. 3. Both gross photographs of decorticated specimens show the typical appearance encountered. A: Large nodules measuring 2.5 cm or greater across. B: Numerous small and moderately sized nodules measuring between 0.2 to 0.8 cm across.
seen neither in areas of good differentiation nor in areas of poor differentiation. The follow-up, though sought on all 15 patients, led to written response from relatives of three patients alone. They wrote to inform the authors that two patients had died within 6 months of the histologic diagnosis, the mode of death apparently being acute respiratory failure. One patient was still alive 6 months after diagnosis. Spread, when detected, was always on the same side of the lesion and was seen at the thoracotomy scar in one patient, axillary lymph nodes in two, along nerve trunks in one, and ribs in two cases. The remaining 12 were lost to follow-up.
DISCUSSION Mesotheliomas are neoplasms arising from mesothelial cells. These neoplasms morphologically may simulate histological appearances of carcinomas or sarcomas of various histological types [ 181. This is believed to occur
199
Fig. 4. Histologically, the nodules observed grossly showed a distinct cut surface which was better observed at low power microscopy. A: A typical nodule with slit-like spaces placed on the inner aspect of the parietal pleura. The slit like spaces were invariably at right angles to the surface of the pleura. B: A typical detail of the slit-like spaces which were lined by cuboidal epithelial-like cells, some showing evidence of mucus production. A: Haematoxylin and Eosin, X15. B: Haematoxylin and Eosin, X250.
because the mesothelial cells under the influence of various stimuli [IS] not only proliferate and become morphologically indistinguishable from epithelial cells but also resemble connective tissue cells and even endothelial cells. With pleural mesotheliomas, the pathologist is often faced with the problem of differentiating these neoplasms from primary peripheral adenocarcinomas of the lung that have infiltrated the pleura as also from metastatic neoplasms to the lung. Earlier workers considered it necessary, therefore, to autopsy every patient with PLMT [ 191 to rule out a primary neoplasm of the lung as well as a metastatic process from a primary elsewhere in the body. Wanebo et al. [20] have, however, suggested that an autopsy need not be a prerequisite towards the diagnosis of a pleural mesothelioma and a correlation
200
Kini et al.
between the clinical presentation, radiological appear- showing hyperchromatic nuclei with a surrounding colances, cytology of pleural fluid and particularly findings lagenous matrix. Such a predominant collagenous matrix at thoracotomy with characteristic gross appearances and typifies a desmoplastic response in an epithelial mesothehistological features which may enable an accurate diag- lioma [25]. A biphasic pattern with mixtures of epithelial nosis to be made. In recent years, findings at histochem- and sarcomatous patterns may also be observed. In the istry, immunohistochemistry , and electron microscopy literature, the epithelial type was the most predominant pattern followed by mixed and sarcomatous patterns in have also been added to this list [21]. PLMT is rare in the general population. This study 30% and 20%, respectively [ 161. In the present series, 14 covering a period of 25 years has reflected an incidence of the 15 were of the epithelial type and 1, the sarcomaof 0.02% and is predominantly a tumour of adult life, tous type. Mucin histochemistry has suggested that carboxylated occurring in the 50-60-year-old age group [l]. The acid mucins, particularly hyaluronic acid, were most youngest patient was 29 years old and the oldest 65 years commonly present with mixtures of weakly sulphated and with the median age being between 43 and 53 years. Most strongly sulphated groups. This has been confirmed by were males and reflected the general tendency towards [ 1 1,261. Detection of acid mucins and their several others that gender in the literature [ 11. fractions may be of help in differentiating PLMT from a Pleural mesotheliomas are found to be more common mucin producing adenocarcinoma. Eighty percent of our in coastal cities with shipyards than inland towns [22] as patients showed carboxylated acid mucins; hyaluronic this lesion is attributed to occupational exposure to asbestos [22]. Contrary to this belief, this series of cases has acid in 60%, weakly and strongly sulphated in 53.3%. shown no association with asbestos. All our patients had Neutral mucins were seen only in 20% of the patient been suspected of having tuberculosis and were being material. The keratin stain showed a positive reaction in empirically treated for this disease by their treating physi- 66.6% of the patients. Much has been written on the diagnostic potential of cians. This is understandable as tuberculosis is found to occur more frequently in the four southern states of India immunohistochemistry, particularly carcinoembryonic antigen (CEA), in differentiating PLMT from adenocar[23] with presenting features similar to PLMT. A characteristic clinical presentation of PLMT from cinomas [27,28]. The profile of strong keratin staining South India has evolved with this study. The characteris- and weak or absent CEA staining appears to characterise tic onset of PLMT appears slow and insidious, being mesotheliomas, especially the epithelial variant, and associated with continuous severe pricking chest pain and helps in distinguishing them from adenocarcinomas. In rapidly accumulating haemorrhagic pleural fluid. Addi- this study, however, a strong expression for CEA was tional irregular pleural thickening was observed on roent- seen in cells of pleural mesothelioma showing moderate genograms with a shift of the mediastinum towards the differentiation only. No expression was seen in either affected side. The typical pleural fluid cytology with well differentiated or poorly differentiated mesotheliornesothelial cell aggregates exhibiting “cupping,” canni- mas. Of the 12 cases studied, 9 showed expression for balism, knobs on the periphery of cell borders, and al- CEA. From these observations, it could be surmised that tered nuclear-cytoplasmic ratio [ 171 should characterize the expression of CEA may represent regressive foetal the lesion. The pattern was seen in 11 of the 15 cases in antigen expression with positivity neither in well differentiated nor poorly differentiated neoplasms. this study. Patients with pleural mesothelioma have a short surThe final characterisation would be at thoracotomy , with the pleura showing cobblestone nodularity encasing vival (6 months) [29]. Although metastases are often almost the entire lung but not involving the lung paren- found at autopsy in cases of PLMT, they are rarely dechyma. The appearance has been likened to a congealed tected during life. The most characteristic and usual liquid that has been poured over the lung. Decortication is method of spread appears to be contiguity, so that the difficult and results in the lung tissue being peeled off tendency of PLMT is to infiltrate along needle tracts, along with the overlying pleural neoplasm [22]. In the 12 intercostal drainage tubes, and thoracotomy scars and cases where decortication was done, the nodules were produce subcutaneous nodules [29]. Axillary lymph node seen both on the parietal and visceral layers of the pleura. involvement was seen in one patient, rib in one patient, The nodules, on further examination, showed slit-like bone marrow of rib in one patient, and infiltration into breast tissue and thoracotomy scar in one. PLMT could, spaces lying perpendicular to the pleural plane. Cytoarchitecturally PLMT shows a diversity of charac- therefore, be said to metastasize in a centrifugal manner teristics. In the epithelial type, tubules, cystic spaces, or to surrounding local structures, unlike other tumours of cystopapillary areas are seen lined by tumour epithelial- the lung. The causes of death in PLMT are pulmonary insuffilike cells exhibiting loose cohesion as well an absent basement membrane [24]. The sarcomatous pattern ciency, wasting, and metastatic disease 1251. Three of the shows spindle shaped cells of varying sizes and shapes 15 patients died of acute respiratory failure.
Pleural Mesotheliomas in South India
The authors are of the opinion that primary pleural mesotheliomas are best diagnosed thorough evaluation of all existent clinical and radiologic data as also the progression of the disease, cytologi&l appearno signifiOf pleura’ anceS On exfoliative cant bronchoscopic findings, thoracotomy obser- typical -vations at operation, gross specimen appearance, and the utilisation Of routine light microscopy aided by histochemistry and immunohistochemistry .
REFERENCES 1. Legha SS, Muggia FM: Pleural mesothelioma: clinical features
and therapeutic implications. Ann Intern Med 87:613-621, 1977. 2. Brenner J, Sordillo PP, Magill GB, Golbey RB: Malignant mesothelioma of the pleura. Review of 123 patients. Cancer 49:243 1-2435, 1982. 3. Reddy DJ, Showramma A: Malignant mesothelioma of the pleura. J Indian Med Assoc 33:139-141, 1959. 4. Lele RD, Sharma KD, Sainani GS, Rajkondawar VL: Mesothelioma of the pleura. Case report. Indian J Med Sci 14:430432, 1960. 5 . Kshirsagar VH, Patil SD, Sharma KD, Deshpande TV, Deshmane JV: Mesothelioma (Report of two autopsy cases). Indian J Cancer 8:208-214, 1971. 6. Razdan JN, Mathur KC, Bharadwaj TP: Diffuse malignant mesothelioma of the pleura. Indian J Med Sci 26:233-235, 1972. 7. Reddy DB, Raju GC, Venkatanarayana CP, Suvarnakumari G: Mesothelioma of pleura-report of a case. J Assoc Physicians India 21:453456, 1973. 8. Aikat BK, Radhakrishnan VV, Malik AK: Mesothelioma of the pleura-a study of five autopsy cases with review of literature. Indian J Cancer 11:95-101, 1974. 9. Chandurkar SN, Khandekar AL, Shivde AV, Kher AV: Primary Pleural Mesothelioma. Indian J Chest Dis 17:139-142, 1975. 10. Malik AK, Aikat BK: Primary pulmonary neoplasms-a histopathologic study. Indian J Cancer 13:149-155, 1976. 11. Mahmood S, Thomas JA: Pleural mesotheliomas. Indian J Cancer 20:247-254, 1983. 12. Bhaduri AS, Balar DB: Mesotheliomas of the pleura: A twelve year experience with fifteen cases. Indian J Cancer 25:7-15, 1988. 13. Ayoub P, Schklar G: A modification of the Mallory connective tissue stain as stain for keratin. Oral Surg Oral Med Oral Pathol 16:S8GS85, 1963. 14. Cook HC: Carbohydrates. In Bancroft JD, Stevens A (eds): “Theory and Practice of Histological Techniques.” New York: Churchill Livingstone, 1977, pp 141-167.
201
15. Sternberger LA, Hardy PH, Cuculis JJ, Meyer HG: The unla-
16.
belled antibody enzyme method by immunohistochemistry. Preparation and properties of soluble antigeniantibody complex (horseradish ueroxidase-anti-horseradish ueroxidase) and its use in ideritification of spirochaetes. J Hisiochem Cytochem 18:3 15333, 1970. Craighead JE, Abraham JL, Churg A, Green FHY, Kleinerman J, Pratt PC. Seemaver TA, Vallvathan V, Weill H: The Pathology of asbestos-associated diseases of the lungs and pleural cavities; diagnostic criteria and proposed grading schema (Report of the Pneumoconiosis Committee of the College of American Pathologists and the National Institute for Occupational Safety and Health). Arch Pathol Lab Med 106:544-596,. 1982. Klempman S: The exfoliative cytology of diffuse pleural mesothelioma. Cancer 15:691-704, 1962. McCaughey WTE: Primary tumours of the pleura. J Pathol Bacteriol76:517-529, 1958. Willis RA (ed): “Pathology of Tumours,” 4th Edition, London: Butterworths, 1967, pp 181-183. Wanebo HJ, Martini N , Melamed MR, Hilaris B, Beattie EJ: Pleural mesothelioma. Cancer 38:248 1-2488, 1976. Lucas JG, Tuttle SE: Diagnostic histochemical and immunohistochemical studies in Maligant mesothelioma. J Surg Oncol 35:30-34, 1987. Hillerdal G: Malignant mesothelioma 1982: Review of 4710 published cases. Br J Dis Chest 77:321-342, 1983. Health Information, India: By Cenral Bureau of Health Intelligence, Director General of Health Services, Ministry of Health and Family Welfare, Government of India, Nirman Bhavan, New Delhi, August 1989. Klima M, Spjit HJ, Seybold WD: Diffuse malignant mesothelioma. Am J Clin Pathol65583-600, 1976. McCaughey WTE, Kannerstein M, Churg J: Tumours and pseudotumours of the serous membranes. In McCaughey WTE, Kannerstein M, Churg J (eds): “Atlas of Tumour Pathology,” Second Series, Fascicle 20. Armed Forces Institute of Pathology, pp 1 6 7 2 , 1985. Wagner JC, Munday DE, Harington JS: Histochemical demonstration of hyaluronic acid in pleural mesotheliomas. J Pathol Bacteriol 84:73-78, 1962. Corson JM, Pinkus GS: Mesothelioma: profile of keratin proteins and carcinoembryonic antigen. An immunoperoxidase study of 20 cases and comparison with pulmonary adenocarcinomas. Am J Pathol 108:80-87, 1982. Pascal RR, Mesa-Tejada R, Bennett SJ, Garces A, Fenoglio CM: Carcinoembryonic antigen: immunohistologic identification in invasive and intraepithelial carcinomas of the lung. Arch Pathol Lab Med 101:568-571, 1977. Adams VI, et al.: Diffuse malignant mesothelioma of pleura. Diagnosis and survival in 92 cases. Cancer 58: 1540-155 1, 1986. I .
17. 18. 19.
20. 21. 22. 23.
24. 25.
26. 27.
28.
29.
