Pleomorphic carcinoma is a rare epithelial malignant. Personal Experience in Surgical Management of Pulmonary Pleomorphic Carcinoma

Transcription

1 Personal Experience in Surgical Management of Pulmonary Pleomorphic Carcinoma Federico Raveglia, MD, Maurizio Mezzetti, MD, Tiziana Panigalli, MD, Simone Furia, MD, Luigi Giuliani, MD, Serena Conforti, MD, and Stefano Meda, MD University of Milan, Milan, Italy Background. Pleomorphic carcinoma is a rare epithelial malignant tumor. Pulmonary pleomorphic carcinoma was introduced by the 1999 World Health Organization classification as a new peculiar type of lung carcinoma showing concurrent malignant epithelial and sarcomatoid spindle cell elements. Few reports describe its clinical behavior. My colleagues and I report a series of patients surgically treated for pulmonary pleomorphic carcinoma to describe our experience with this malignant neoplasm. Methods. Twenty cases of pleomorphic pulmonary carcinoma were collected and studied clinicopathologically. All patients underwent surgical resection. The cases were as follows: 6 stage I, 12 stage II, and 2 stage IIIA. Histologic diagnosis was established by using light microscopic examination and immunohistochemistry. rates were calculated with the Kaplan-Meier method. Results. We postoperatively diagnosed 20 cases of pleomorphic carcinoma: 14 cases were exclusively spindle and giant-cell carcinomas, 2 cases were spindle and giant-cell carcinoma combined with adenocarcinoma, 2 were combined with squamous cell carcinoma, and 2 were combined with large cell carcinoma. At last followup, 4 patients were still alive; they were postoperative T1 N0 and T2 N0. The remaining 16 patients died from early distant metastases. The median duration of disease-free survival was 5 months. The median duration of overall survival was 8 months. Conclusions. The prognosis of patients with pleomorphic carcinoma was poor, despite surgery and adjuvant chemotherapy, because of early relapse of disease. Nodal involvement was a determinant prognostic variable, because advanced stages were related to worse prognosis. In case of preoperatively proven pulmonary pleomorphic carcinoma, surgery should be recommended to N0 patients. (Ann Thorac Surg 2004;78:1742 7) 2004 by The Society of Thoracic Surgeons Pleomorphic carcinoma is a rare epithelial malignant tumor [1]. It has been described in several human organs, including the lung parenchyma [2, 3]. The 1999 World Health Organization (WHO) classification identified pulmonary pleomorphic carcinoma as a specific type of lung cancer among neoplasms with pleomorphic, sarcomatoid, or sarcomatous elements [4]. This is a group of poorly differentiated non small-cell carcinomas that contain a component of sarcomatoid elements. According to the 1999 WHO definition, pleomorphic carcinoma is a pulmonary carcinoma consisting of spindle or giants cells (or both) combined with squamous cell carcinoma, adenocarcinoma, or large-cell carcinoma. Pure spindle and giant-cell carcinomas are rare. The spindle or giant-cell component should comprise at least 10% of the neoplasm for diagnosis. The adenocarcinoma, squamous cell carcinoma, or large-cell carcinoma components of the tumor should be always documented when present. The WHO considers pleomorphic carcinoma as an entity separate from squamous cell carcinoma, adenocarcinoma, or large-cell carcinoma on the basis of the mutational spectrum seen in these tumor types [5, 6]. Accepted for publication April 27, Address reprint requests to Dr Raveglia, Piazza Leonardo da Vinci 7, Milan, Italy; ravegliafederico@tiscali.it. Before 1999 pleomorphic carcinoma was considered as a variant of other well-known lung carcinomas because of its biphasic appearance and its frequent association with the other histologic types [7]. Diagnosis of this histotype was problematic and confusing because of the lack of uniform diagnostic criteria. The lack of widely accepted criteria and the rarity of this pulmonary carcinoma made its behavior unclear and its appropriate treatment controversial. In 1999 the WHO reclassified spindle cell carcinoma combined with squamous cell carcinoma, adenocarcinoma, giant-cell carcinoma, or large-cell carcinoma as a new class of lung tumor named pleomorphic carcinoma [8]. Classification of these tumors is based on light microscopic criteria, sometimes supported by immunohistochemistry [4]. Few large series of patients with pulmonary pleomorphic carcinoma according to the 1999 WHO classification have been described in the international literature, and most of these articles focused on the histologic features. The pleomorphic carcinoma histopathologic classification is universally adopted, but its clinical relevance and the behavior of the tumor are still uncertain. My colleagues and I diagnosed pulmonary pleomorphic carcinoma in a small number of patients surgically treated for lung carcinoma. We collected all clinical data 2004 by The Society of Thoracic Surgeons /04/$30.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg RAVEGLIA ET AL 2004;78: PLEOMORPHIC CARCINOMA SURGICAL OUTCOMES 1743 about these patients from 1999 to obtain more information about this low-frequency tumor. This article describes our clinical experience with pulmonary pleomorphic carcinoma and reports histologic records, preoperative and postoperative staging, and the correlated outcomes. The purpose of the study was to better understand the clinical relevance in terms of the survival of this histotype by using information from the literature and our personal experience. With this article, focused on the outcomes in patients surgically treated for pulmonary pleomorphic carcinoma, we broached the role of surgery in this subgroup of non small-cell lung cancer. and Methods From the beginning of 1999 to the end of 2002, 513 patients underwent surgical resection for primary lung carcinoma at the Department of Thoracic Surgery, University of Milan. Of these, 20 cases were primary pleomorphic carcinoma. The 20 patients with pleomorphic carcinoma were registered for age, sex, presenting symptoms, smoking habits, and stage. There were 14 men and 6 women. The mean age at diagnosis was 60.3 years (range, 43 to 76 years); 80% of patients smoked. Eighteen patients presented with symptoms such as coughing, dyspnea, fever, hemoptysis, or obstructive pneumonia in case of endobronchial involvement. Two patients had an incidental roentgenogram finding of a pulmonary lesion. All patients were studied with contrast-enhanced chest computed tomography (CT) that demonstrated 8 leftlung and 12 right-lung single lesions with radiologic signs of malignancy. Overall, 12 lesions were in the right upper lobe, 6 in the left upper lobe, and 2 in the left lower lobe. We used contrast-enhanced CT for noninvasive intrathoracic staging. In addition, cervical mediastinoscopy was performed before operation to analyze superior mediastinal lymph nodes in 7 patients. One patient presented with a lesion in the right lung and dubious mediastinal lymphadenopathy at CT scan. The other 6 cases had a left upper lobe single lesion, and mediastinoscopy was performed to exclude right mediastinal nodal involvement because contralateral lymphatic spread is more likely for left upper lobe tumors. None of these patients presented with N3 disease. Two patients were positive for N2 nodal malignancy, which is compatible with non small-cell lung cancer. Metastatic disease was routinely detected by brain CT, abdominal CT, and bone scan. None of these patients presented with metastatic disease. All of the lesions were described as solitary masses. Twelve were centrally situated, and 8 were peripherally situated. Bronchoscopy was always performed to define the airway involvement and the extent of the pathologic tissue in the bronchus. Eight cases were described as endobronchial neoplasms, and endoscopic biopsy resulted in the diagnosis of 4 squamous cell carcinomas and 4 non small-cell carcinomas. In case of a preoperative established diagnosis, we elected surgery because of the malignancy. In 12 patients no proven pathologic preoperative diagnosis was achieved; we elected surgery on the basis of radiologic findings. In these 12 patients intraoperative histologic examination of the lung tumor was performed and was positive for malignant disease. On the basis of the frozen section examination results, we proceeded with surgical resection. The 2 patients preoperatively staged as N2 with mediastinoscopy underwent surgical resection because they were classified as having non small-cell lung cancer stage IIIA. All patients had pulmonary function tests and cardiac evaluation. Two patients with compromised pulmonary function underwent parenchyma-sparing resection (typical segmentectomy). None had preoperative chemotherapy. Anesthesiologic procedures were typical for major thoracic surgery. Ventilation was always by double-lumen tube. The incision was always a standard posterolateral thoracotomy. This allowed thorough evaluation of the anterior and posterior hilar structures and permitted complete ipsilateral mediastinal node dissection (nodal stations 2, 4, 5, 6, 7, 8, and 9). Operations included 2 left pneumonectomies, 2 right pneumonectomies, 8 right upper lobectomies, 4 left upper lobectomies, 2 left lower lobectomies, and 2 right upper lobe segmentectomies. Expert pathologists at the University of Milan examined all the pathologic material. Diagnosis was obtained by light microscopic findings and was completed with immunohistochemical examinations. Each case met the WHO criteria. Cases were classified as spindle cell carcinoma if at least 10% of the tumor was composed of fusiform malignant cells. Histologic components were classified as follows: spindle cell carcinoma, giant-cell carcinoma, squamous cell carcinoma, adenocarcinoma, or large-cell carcinoma. Immunohistochemical procedures were performed by using antibodies against cytokeratin and vimentin [9 11]. A positive reaction to 1 epithelial marker was useful to confirm epithelial differentiation in the sarcomatoid component in case of poor carcinomatous differentiation at the light microscopic examination. All patients received adjuvant chemotherapy except in cases with N0 disease. They received a combination of chemotherapy agents. Chemotherapy cycles were repeated every 28 days according to the following pattern: cisplatin on the 1st day and gemcitabine on the 2nd, 8th, and 15th days. Follow-up was obtained for all patients with periodic clinical and radiographic controls (including standard chest radiographs and CT scanning) at our outpatient clinic until the patient s death. Flexible bronchoscopy was performed in patients who presented with endobronchial carcinoma or in case bronchial recurrence was suspected. and disease-free survival rates were calculated on the basis of the Kaplan-Meier method. The log-rank test was used to test for significant differences in survival and disease-free survival [12].

3 1744 RAVEGLIA ET AL Ann Thorac Surg PLEOMORPHIC CARCINOMA SURGICAL OUTCOMES 2004;78: Table 1. Summary of Clinical Data Patient No. Sex Smoker ptnm Histology Disease- Free (mo) 1 F Y T2 N1 Spindle/giant-cell large-cell carcinoma M Y T2 N1 Spindle/giant-cell squamous cell carcinoma M Y T1 N0 Spindle/giant-cell 32 a 32 b 4 F N T2 N1 Spindle/giant-cell M Y T2 N0 Spindle/giant-cell adenocarcinoma 22 a 22 b 6 M Y T2 N0 Spindle/giant-cell M Y T3 N1 Spindle/giant-cell M Y T2 N2 Spindle/giant-cell F N T2 N1 Spindle/giant-cell M Y T2 N1 Spindle/giant-cell squamous cell carcinoma F Y T2 N1 Spindle/giant-cell large-cell carcinoma M Y T2 N1 Spindle/giant-cell F N T2 N1 Spindle/giant-cell M Y T1 N0 Spindle/giant-cell 31 a 31 b 15 M Y T2 N0 Spindle/giant-cell M Y T2 N0 Spindle/giant-cell adenocarcinoma 21 a 21 b 17 M Y T3 N1 Spindle/giant-cell M Y T2 N2 Spindle/giant-cell F N T2 N1 Spindle/giant-cell M Y T2 N1 Spindle/giant-cell 6 4 a Patient still alive; b Absence of metastasis. ptnm pathologic tumor-node-metastasis stage. Results Clinical data were tabulated according to histologic findings, pathologic tumor-node-metastasis staging, and clinical disease behavior (Table 1). The tumors ranged from 2.1 to 9 cm. Nineteen lesions were intraparenchymal, and 8 showed intrabronchial components in the tumor. Twenty cases of primary pulmonary pleomorphic carcinoma were identified. Six cases had identifiable epithelial components. Fourteen cases were classified as pure pleomorphic carcinomas consisting only of spindle and giant cells (Table 1). All tumors expressed both cytokeratin and vimentin. Most cases expressed diffuse vimentin, as opposed to focal stain for cytokeratin [13]. We compared postoperative with preoperative diagnosis (8 cases), and 2 cases of squamous cell carcinoma were reclassified as pleomorphic carcinoma combined with squamous cell carcinoma. The other cases of squamous cell carcinoma were reclassified as spindle and giant-cell carcinoma combined with large-cell carcinoma. Four preoperative non small-cell lung cancers were reclassified as 2 pleomorphic carcinomas combined with adenocarcinoma and 2 pure pleomorphic carcinomas. According to the tumor-node-metastasis classification, cases were postoperatively classified as follows: 6 stage I (2 T1 N0 and 4 T2 N0), 12 stage II (10 T2 N1 and 2 T3 N1), and 2 stage IIIA (2 T2 N2). These findings matched the preoperative staging. The in-hospital and 30-day mortality rates were 0%. There were no severe postoperative operative complications. No re-resections were performed for neoplastic involvement of surgical edges. The overall survival was 20% (4/20) at the time of last follow-up (December 2003). The remaining 80% of these patients died because of recurrence of disease. The median duration of survival was 8 months (Fig 1; Table 2). The disease-free survival time was extremely short; in our series the median disease-free survival was 5 months (Fig 2; Table 3). Metastases involved different areas, Fig 1. Overall survival function estimates for pleomorphic carcinoma. Median survival was 8 months.

4 Ann Thorac Surg RAVEGLIA ET AL 2004;78: PLEOMORPHIC CARCINOMA SURGICAL OUTCOMES 1745 Table 2. Data Related to Figure 1 time (mo) Dead Rate including the mediastinal lymph nodes, brain, bone, adrenal glands, liver, and lung. At detection of metastatic disease, all patients presented more than 1 involved area. When we considered the nodal status, 4 patients of the 6 N0 cases were still alive and free from metastatic disease at last follow-up. The remaining 14 cases who had nodal involvement (12 N1 and 2 N2) at postoperative staging died of recurrence of disease. We used the Kaplan-Meier method to generate 1 set of survival curves that compared patients on the basis of nodal disease status (Fig 3; Table 4). We classified patients into 2 categories (stage I and stage II to III) to obtain survival curves for nodal involvement. We found a significant difference in survival for patients with nodal involvement when we compared those with N1 and N2 disease with those with N0 disease (log-rank test; p 0.001). Table 3. Data Related to Figure 2 Disease-Free (mo) Comment Relapse of Disease Disease-Free Rate Pulmonary pleomorphic cell carcinoma is a rare malignant lung tumor. This rarity has made its classification confusing and its diagnosis difficult. Before 1999 this tumor was classified as a variant of squamous cell carcinoma. In 1999 the WHO reclassified the pleomorphic carcinoma and set widely adopted criteria. According to these criteria, pleomorphic carcinoma is now considered a peculiar type of lung carcinoma. We reported our clinical experience with pulmonary pleomorphic carcinoma to add new information regarding the behavior and prognosis of this tumor and the role of surgery in its management. In accordance with other studies, we noted a male preponderance [14], a mean age of 60.3 years [15], and a high correlation with cigarette smoking. Most lesions were found in the upper lobes. Contrary to other series, most of our cases were pure spindle and giant-cell Fig 2. Overall disease-free survival function estimates for pleomorphic carcinoma. Median disease-free survival was 5 months. Fig 3. functions for postoperative stage I versus stage II or III (p 0.001). Data 1 patients with nodal involvement (stage II or III); data 2 patients without nodal involvement (stage I).

5 1746 RAVEGLIA ET AL Ann Thorac Surg PLEOMORPHIC CARCINOMA SURGICAL OUTCOMES 2004;78: Table 4. Data Related to Figure 3 (mo) Data 2 Data 1 Dead Rate Dead Rate carcinomas. Clinically most of the patients presented with nonspecific symptoms such as coughing and dyspnea, ranging from 2 to 4 months before the diagnosis of pulmonary disease. Our data showed that establishing a pathologic preoperative diagnosis was complicated. We obtained an adequate specimen only in case of proximal endobronchial disease involvement. The similarity to other tumors and the intimate mixture in foci of different elements made differential diagnosis of pleomorphic carcinoma difficult when based on endoscopic sampling. Preoperative diagnosis, when available, was significant in detecting malignancy; a correct differential diagnosis of pleomorphic carcinoma has always been established on the basis of definitive postoperative specimen examination. Our experience indicated that the prognosis of patients with pleomorphic carcinoma was poor despite surgery and adjuvant chemotherapy, even in case of local disease (stage I). Pleomorphic carcinoma presented an aggressive clinical behavior, and most patients experienced early multiple disease spreading (Fig 2). The overall median postoperative survival was 8 months (Fig 1). Some investigators who have described the malignant behavior of sarcomatoid carcinomas did not find significant differences with patients with nonsarcomatoid non small-cell carcinoma of the lung [16]. However, our data, in accordance with those of Fishback and colleagues [10], suggest that the prognosis of this pulmonary tumor in terms of median survival is poorer than in conventional non small-cell lung carcinoma (according to literature [17]: 20 months for adenocarcinoma, 18.5 months for squamous cell carcinoma, and 12.6 months for large cell carcinoma). Our figures showed that survival was even more unfavorable than in small-cell lung cancer when compared with many series of patients surgically treated for small-cell lung cancer before the chemotherapy era [18]. We noted that multiple distant metastases occurred earlier and worsened prognosis in patients with nodal involvement. No patient with lymph node involvement survived. Comparison between the survival curve in case of nodal disease and the survival curve in N0 patients showed that the nodal status had statistical significance in determining prognosis (p 0.001; Fig 3). Previous reports have indicated different findings about the effect of nodal metastasis on the length of survival. Some studies found that nodal status did not have a statistical effect on prognosis [9], but others found that nodal involvement shortened patient survival [10, 11, 16]. In contrast to other studies, our data showed that stage based on lymph node status was a significant prognostic variable in these patients. rate was strictly related to lymph node involvement, with significant differences between N0 and N1 and N2. On the basis these data, distinctive for poor prognosis, the question in the management of pleomorphic carcinoma of the lung is whether surgical resection should be considered. We think that, when dealing with this malignant entity, the most important issue is to obtain an accurate preoperative diagnosis and a correct clinical stage definition. Preoperative diagnosis is the most demanding item in the management of biphasic lung tumors; its correct definition requires intensive collaboration between pathologist and surgeon. Because survival curves showed that recurrences were more likely for patients with nodal involvement, our experience suggests that the goal for surgery should be complete tumor resection before nodal disease involvement occurs. We think that surgery for preoperatively proven pulmonary pleomorphic carcinoma should be restricted to N0 patients. with nodal disease should be considered for hypothetical neoadjuvant therapy, but nowadays chemotherapy is contraindicated by the generally experienced poor sensibility of these types of tumors. The clinical behavior of this biphasic malignant tumor, comparable to that of poorly differentiated carcinoma, is probably due to the presence of carcinomatous and sarcomatous elements originating from a single stem cell of multipotential lung tissue [9, 19]. In our clinical experience, we noted that the frequency of biphasic tumors (non small-cell lung carcinoma/small-cell lung cancer or squamocellular carcinoma/adenocarcinoma) [20] seems to have increased. This is disturbing, because universally adopted surgical algorithms and prognoses are actually based on monophasic tumor types. The authors are grateful for the valuable collaboration of Alberto G. L. Luporini, MD, Director of the Division of Medical Oncology, Policlinico San Donato Institute, San Donato Milanese, Milan, Italy. References 1. Ishida T, Tateuishi M, Kaneko S. Carcinoma and spindle cell carcinoma of the lung: clinicopathologic and immunohistochemical studies. J Thorac Cardiovasc Surg 1990;100: Munakata R, Cheng J, Nakajima T. Spindle cell carcinoma of the gingiva: report of an autopsy case. J Oral Pathol Med 1998;27: Torenbeek R, Blomjous CEM, Bruin P. Sarcomatoid carci-

6 Ann Thorac Surg RAVEGLIA ET AL 2004;78: PLEOMORPHIC CARCINOMA SURGICAL OUTCOMES 1747 noma of urinary bladder. Clinicopathologic analysis of 18 cases with immunohistochemical and electron microscopic findings. Am J Surg Pathol 1994;18: World Health Organization. Histological typing of lung tumors and pleural tumors. 3rd ed. Geneva: World Health Organization, Przygodzki RM, Koss MN, Moran CA, et al. Pleomorphic (giant and spindle cell) carcinoma is genetically distinct from adenocarcinoma and squamous cell carcinoma by K-ras-2 and p53 analysis. Am J Clin Pathol 1996;106: Lee YC, Chang YL, Luh SP. Significance of p53 and Rb protein expression in surgically treated non small cell lung cancers. Ann Thorac Surg 1999;68: World Health Organization. Histological typing of lung tumors. Geneva: World Health Organization, Brambilla E, Travis WD, Colby TV, Corrin B, Shimosato Y. The new World Health Organization classification of lung tumours. Eur Respir J 2001;18: Yih-Leon Chang, Yung-Chie Lee, Jin-Yuan Shih, Chen-Tu Wu. Pulmonary pleomorphic (spindle) cell carcinoma: peculiar clinicopathologic manifestations different from ordinary non small cell carcinoma. Lung Cancer 2001;34: Fishback NF, Travis WD, Moran CA, et al. Pleomorphic (spindle/giant cell) carcinoma of the lung. A clinicopathologic correlation of 78 cases. Cancer 1994;73: Rossi G, Cavazza A, Sturm N, et al. Pulmonary carcinomas with pleomorphic, sarcomatoid or sarcomatous elements: a clinicopathologic and immunohistochemical study of 75 cases. Am J Surg Pathol 2003;27: Kaplan EL, Meier P. Non parametric estimation from incomplete observations. J Am Stat Assoc 1958;53: Kazuhiro M, Masanobu K. Spindle cell carcinoma of the lung. A clinicopathologic study of three cases. Cancer 1991; 67: Haque S. Carcinosarcoma of the lung: report of a case and review of the literature. J Pak Med Assoc 1980;30: Bull JC, Grimes OF. Pulmonary carcinosarcoma. Chest 1974; 65: Nakajima M, Kasai T, Hashimoto H, Iwata Y, Manabe H. Sarcomatoid carcinoma of the lung. A clinicopathologic study of 37 cases. Cancer 1999;86: Kaiser K, Bulzebruck H, Probst G, Vogt-Moykopf I. Retrospective and prospective tumor staging evaluating prognostic factors in operated bronchus carcinoma patients. Cancer 1987;59: Lad T. Surgery for small cell lung cancer. In: Franco KL, Putnam JB Jr, eds. Advanced therapy in thoracic surgery. Hamilton, Ontario, Canada: B. C. Decker, 1998: Oscar N, Mark RW. Sarcomatoid neoplasms of the respiratory tract. Semin Diagn Pathol 1993;10: Shimizu J, Oda M, Hayashi Y, et al. A clinicopathologic study of resected cases of adenosquamous carcinoma of the lung. Chest 1996;109: