July 16, 2008
Pulmonary Sarcomatous Tumors
By Litzky, Leslie A
* Context.-Sarcomatous pulmonary neoplasms are a rare and diagnostically challenging group of tumors. Primary pulmonary sarcomas must be distinguished from the more frequent occurrence of sarcoma metastatic to the lung, primary pulmonary sarcomatoid carcinoma, and diffuse malignant mesothelioma involving the lung. In current practice, the general availability of ancillary diagnostic techniques, such as immunohistochemistry and molecular analysis, can reliably classify many sarcomatoid lesions. Objective.-To review the literature and report on additional primary material about the application of immunohistochemistry and molecular analysis in the differential diagnosis pulmonary sarcomatous neoplasms.
Conclusions.-This review discusses specific criteria for the diagnosis of primary lung sarcomas and offers a practical approach to excluding other sarcoma-like lesions involving the lung. The pathologist has an essential role in evaluating these tumors and will often be the first to suggest an unusual, alternative diagnosis, which may have significant implications for patient care, therapy, and prognosis.
(Arch Pathol Lab Med. 2008;132:1104-1117)
Sarcomatous pulmonary neoplasms are a rare but diverse and diagnostically challenging group of tumors. Primary pulmonary sarcomas must be distinguished from the more frequent occurrence of sarcoma metastatic to the lung, primary pulmonary sarcomatoid carcinoma, and diffuse malignant mesothelioma involving the lung. Spindle cell tumors of the pleura, as well as those tumors primary from the chest wall or mediastinum, may present as either localized or diffuse masses and, depending on the extent of thoracic involvement, may be difficult to distinguish from those of pulmonary origin. This review discusses specific criteria for the diagnosis of primary lung sarcomas and offers a practical approach to excluding other sarcoma-like lesions involving the lung.
HISTORICAL PERSPECTIVE AND CURRENT PRACTICE
From a clinical, radiographic, and pathologic perspective, metastases and primary pulmonary carcinomas far exceed the incidence of primary pulmonary sarcomas. In practice, a tumor that is initially thought to be a primary sarcoma of the lung can prove, on further inquiry into patient history, to be a late metastasis from a primary soft tissue tumor or, on further sampling, to be a sarcomatoid carcinoma. The widespread use of immunohistochemistry and, more recently, molecular diagnosis has altered previous diagnostic categories. In current practice, many tumors that were classified in older literature as fibrosarcoma, malignant fibrous histiocytoma, or hemangiopericytoma are now more reliably classified as other entities. The availability of multiple cytokeratin antibody tests has improved the identification of sarcomatoid carcinomas and diffuse malignant mesothelioma. The introduction of CD34 has facilitated the recognition of both benign and malignant intrathoracic, fibrous tumors. It is now possible to assemble a substantial panel of immunohistochemical markers that, in the appropriate histologic context, will reliably identify tumors of vascular, muscular, or neural origin. The wider availability of paraffin-based molecular assays for the characteristic translocation found in synovial sarcoma has led to an appreciable increase in its diagnosis. Improved radiographic techniques, including vascular imaging, have led to more precise localization of malignancies that arise in an intravascular location and allowed for further separation of pulmonary endovascular sarcomas from other parenchyma- based pulmonary sarcomas. Pulmonary artery sarcomas will not be specifically discussed, but a study of these similarly uncommon tumors has recently been published.1 Despite, these diagnostic refinements, the relative rarity of primary pulmonary sarcomas makes it difficult to assemble statistically meaningful case studies and to fully understand the natural history and biology of these tumors.
As a general rule that is true for all lung lesions, the pathologist should be certain to obtain all clinical history when confronted with a malignant spindle cell lesion in the lung. Any history of a previous malignancy or procedure is potentially relevant, including a remote history of a previously treated sarcoma, which can recur after a long latency. In particular, a spindle cell lesion of the lung in a woman should always prompt an inquiry into gynecologic history. Pertinent questions include whether there has been a hysterectomy for a "benign disease" and the date of the most recent gynecologic examination. Sarcomas metastatic to the lung outnumber primary pulmonary sarcomas and, depending on one's index of suspicion, the pathologist may need to take a more active role in prompting clinical colleagues to consider an occult or remote soft tissue primary. The pathologist should insist on radiographic correlation, particularly when handling a cytology sampling or core needle biopsy. Is the lesion clearly parenchymal, or is there pleural or bone and soft tissue involvement? Is there adenopathy or other mediastinal abnormalities? Other primary intrathoracic sites, including the heart and great vessels, mediastinum, esophagus, chest wall, and pleura, must be considered when the radiographic findings are difficult to interpret or the lesion difficult to localize.
The size of the tissue during the initial diagnostic procedure may preclude complete characterization of the lesion, but, once a hematopoietic malignancy has been excluded, that is usually of limited consequence if the patient is a candidate for surgery and a larger biopsy or attempt at resection is clinically indicated. Although most diagnostic workup on tissue from any subsequent, larger specimen can be performed reliably on appropriately formalin- fixed, paraffin-embedded tissue, it is still reasonable to procure fresh tumor tissue, if available, for future studies or for electron microscopy, which may prove useful in difficult cases. Pathologic examination of the resected specimen must include a full sampling of the lesion. Statistically, most spindle cell lesions of the lung will prove to be poorly differentiated carcinomas with sarcomatous differentiation (sarcomatoid carcinomas). Extensive sampling will often reveal recognizable foci of an epithelial tumor, particularly if any endobronchial component is thoroughly sectioned or more differentiated areas in a high-grade tumor are detected. After histologic examination, if necessary, the immunohistochemical evaluation should proceed in the manner familiar to most pathologists for any spindle cell lesion. The workup for specific pulmonary sarcomas will be reviewed in each section, but as a starting point, the diagnostic panel for a sarcomatous tumor in the lung should include multiple cytokeratins tests, as well as epithelial membrane antigen (EMA) staining, with the caveat that focal cytokeratin or EMA positivity can be problematic. Caution should be exercised to avoid overinterpreting what may be entrapped, reactive alveolar or bronchial cells. Similarly, cytokeratin positivity can be demonstrated in a variety of sarcomas. There are, however, occasional cases of sarcomatous lesions in the lung in which even extensive use of ancillary diagnostic techniques fails to provide a definitive diagnosis or only reduces the differential diagnosis to 2 entities that cannot be further distinguished. Nevertheless, as will be readily apparent in this review, these tumors are generally not distinguishable by clinical presentation, radiographic findings, or even gross appearance. The pathologist is critical to the histologic evaluation of these tumors and will most likely be the first to suggest an unusual, alternative diagnosis, which can have significant implications for patient care, therapy, and prognosis.
PRIMARY PULMONARY SARCOMATOID CARCINOMA
Tumors that have sarcoma-like elements, such as malignant spindle or giant cells, or that have a sarcomatous component that consists of a neoplastic but differentiated connective tissue phenotype, such as neoplastic bone, cartilage, or striated muscle, have been described in many primary organ sites, including the lung. The classification of these poorly differentiated primary non-small cell lung tumors, which contain a component of sarcoma or sarcoma-like elements, had been a matter of some controversy in earlier literature, and these tumors had been described under a variety of names. The 1999 and 2004World Health Organization (WHO) classifications of lung tumors set forth criteria for the classification of these tumors in an attempt to foster uniformity and to provide more refined prognostic information in the future.2,3 The 1999 revision established a minimum requirement of 10% for the certain elements, such as spindle cells or giant cells, for a tumor to be appropriately classified. The 2004 WHO revision recommended the term sarcomatoid carcinoma to categorize these tumors. As defined in the 2004 WHO criteria, sarcomatoid carcinomas are a group of poorly differentiated non-small cell lung carcinomas that contain a component of sarcoma or sarcoma-like (spindle or giant cell or both) differentiation. There are 5 recognized subgroups-pleomorphic carcinoma, spindle cell carcinoma, giant cell carcinoma, carcinosarcoma, and pulmonary blastoma-and these 5 subgroups represent a morphologic continuum.3 Tumors that are composed exclusively of spindle cells or giant cells are termed spindle cell carcinoma and giant cell carcinoma, respectively. These pure forms are extremely rare. A tumor consisting of a poorly differentiated squamous cell carcinoma, adenocarcinoma, or large cell carcinoma admixed with spindle cells or giant cells (comprising >/=10% of the tumor) is termed pleomorphic carcinoma. Carcinosarcoma is defined as a malignant tumor with a mixture of carcinoma and sarcoma containing differentiated sarcomatous elements, such as malignant cartilage, bone, or skeletal muscle. In addition to the histologic recognition of these malignant, sarcomatous elements, immunohistochemistry with epithelial markers (multiple cytokeratin antibodies, carcinoembryonic antigen [CEA], and EMA) is often used to confirm epithelial differentiation. The WHO classification relies on histologic criteria using routine light microscopy and does not require the demonstration of positive immunohistochemical staining for cytokeratin if there is a component of squamous cell carcinoma, adenocarcinoma, or large cell carcinoma present. When cytokeratin stains are negative, however, it may be difficult to separate these tumors from primary or metastatic sarcoma. The overall frequency of more specific sarcomatous differentiation is rhabdomyosarcoma, followed by osteosarcoma or chondrosarcoma or combinations of osteosarcoma and chrondrosarcoma.4 The sarcomatous elements of these tumors share the same immunohistochemical profile as their primary malignant mesenchymal counterparts, for example, the cartilage will stain positive for S100. Pulmonary blastoma is defined as a biphasic tumor containing a primitive epithelial component, which may resemble well-differentiated fetal adenocarcinoma, and a primitive mesenchymal stroma that occasionally has foci of osteosarcoma, chondrosarcoma, or rhabdomyosarcoma. This tumor presents mainly in adults and should be distinguished from pleuropulmonary blastoma, a malignant tumor of infancy and early childhood.5 Sarcomatoid carcinomas of the lung are rare, representing less than 1.5% of all patients with non-small cell lung cancer, with about 30 case reports and larger series in the literature that have variably characterized these tumors. 4,6-32 Sarcomatoid carcinomas have no distinguishing radiologic features and have been reported in both central and peripheral locations. With the exception of pulmonary blastoma, which appears to be most frequent in the fourth decade and occurs equally in women and men, the other variants tend to present in the sixth and seventh decade. In the most recent large study of sarcomatoid carcinoma, which focused on prognosis in relationship to other non-small cell lung carcinomas, men were not predominant, which had been the case in prior series.32 There is a strong association with tobacco use. Most patients with sarcomatoid carcinomas are cigarette smokers, and therefore, it is reasonable to maintain a high index of suspicion for sarcomatoid carcinoma when evaluating a malignant, sarcomatous tumor in the lung of a smoker. Most sarcomatoid carcinomas are between 3 and 11 cm, with a size range from 2 to 16 cm. The tumors are usually gray-white, and areas of hemorrhage or necrosis are frequent (Figure 1).
Careful sampling remains the practical key to distinguishing these tumors from a primary pulmonary sarcoma. Previously published reports with histologic analysis of these tumors underscore the need for careful sampling. The malignant stroma is often the predominant component in the carcinosarcoma, and the foci of carcinoma are small (Figures 2 through 5). Similarly, the poorly differentiated spindle cell component may predominate, and a careful search is required to identify the heterologous elements. Although the malignant stroma often constitutes most of the lesion, thorough sectioning of at least 1 section per centimeter of tumor diameter, particularly of the endobronchial component, if present, usually demonstrates small, identifiable foci of squamous cell carcinoma, adenocarcinoma, large cell carcinoma, or the distinctive fetaltype glands found in pulmonary blastoma. After extensive sampling of these epithelial elements, strong cytokeratin positivity is most helpful in making the diagnosis of sarcomatoid carcinoma. Even with the demonstration of cytokeratin positivity, other cytokeratin-positive sarcomas might remain within the differential diagnosis, depending on the tumor's histologic appearance. Evaluation for the characteristic (X;18)(p11.2;q11.2) translocation may prove useful in some cases when synovial sarcoma is in the differential diagnosis. Sarcomatoid carcinomas can express vimentin and smooth muscle markers. This positivity should not be used to exclude the diagnosis of sarcomatoid carcinoma. Pulmonary blastoma, in particular, may be positive for neuroendocrine markers, such as chromogranin, further confounding the differential diagnosis. It is difficult to compare the behavior of these poorly differentiated carcinomas to primary pulmonary sarcomas, but the separation of sarcomatoid carcinoma from other histologic types of non-small cell carcinoma has clinical relevance. It is generally agreed in the available literature that sarcomatoid carcinoma, stage for stage, has a higher risk for both local and distant recurrence, resulting in a poorer prognosis.32
DIFFUSE MALIGNANT MESOTHELIOMA
In overall incidence, diffuse malignant mesothelioma is more common than primary pulmonary sarcoma, and the differential diagnosis of a biopsy from an adult with an ill-defined pleuroparenchymal or mediastinal mass should routinely include diffuse malignant mesothelioma. As has been emphasized in the diffuse malignant mesothelioma literature, a history of asbestos exposure should not affect consideration or exclusion of this diagnosis.33 Sarcomatoid diffuse malignant mesothelioma is the subtype most likely to be considered in the differential diagnosis of a sarcomatous lesion and, in rare instances, a biphasic diffuse malignant mesothelioma. Sarcomatoid diffuse malignant mesotheliomas can have a variety of histologic patterns that overlap with pulmonary sarcomas. The most common pattern is a fibrosarcoma-like pattern of spindle cells arranged in storiform, fascicular, or haphazard patterns (Figure 6). Other less-common variants include a malignant fibrous histiocytoma-like tumor and diffuse malignant mesotheliomas with malignant smooth muscle, chondroid, osseous, or rhabdomyoblastic differentiation. An immunohistochemical panel for the initial evaluation of a sarcomatoid diffuse malignant mesothelioma should include cytokeratins, calretinin, and D2-40. Multiple cytokeratin antibodies, including AE1/3, CAM 5.2 (or cytokeratin [CK] 18), and CK7, should be used because cytokeratin expression can be focal, weak, or variable.34 Other affirmative markers that are used in the evaluation of epithelioid diffuse malignant mesothelioma, such as Wilms tumor (WT1) and CK5/6, as well as adenocarcinoma markers, such as BER-Ep4, CEA, and MOC-31, are unnecessary and should not be included in the workup. A histologically malignant sarcomatoid tumor that is convincingly positive for cytokeratin usually limits the differential diagnosis to sarcomatoid diffuse malignant mesothelioma, sarcomatoid carcinoma, synovial sarcoma, and other metastatic carcinomas. D2-40 and calretinin have been the 2 affirmative mesothelial markers most consistently expressed in sarcomatoid diffuse malignant mesotheliomas, although the percentage of positive cases is low when compared with the epithelial subtype.35,36 In a recent, large tissue- microarray study that included more than 300 diffuse malignant mesotheliomas, calretinin expression was found in 57% of sarcomatoid tumor areas and D2-40 immunostaining was present in 30% of the sarcomatoid tumor areas.37 False positives can occur when benign, entrapped lymphatics or reactive mesothelial elements are misinterpreted. Without convincing calretinin and D2-40 positivity, it is difficult to separate the spindled cell component of a partially sampled sarcomatoid carcinoma from sarcomatoid diffuse malignant mesothelioma. Heterologous elements may be present in both tumors. A possible distinguishing feature may be the identification of areas in which the malignant cells are infiltrating through densely collagenized fibrosis, as is characteristic of desmoplastic diffuse malignant mesothelioma (Figure 7). This pattern is quite typical of diffuse malignant mesothelioma and favors that diagnosis, although ultimately, in this instance, the diagnosis may have to incorporate other gross and clinical features. Synovial sarcomas of the pleura (or primary pulmonary synovial sarcomas involving the pleura) usually present as localized solid tumors, but synovial sarcomas can present with diffuse pleural thickening that is similar to diffuse malignant mesothelioma. As will be discussed in the section on synovial sarcoma, the diagnosis of synovial sarcoma is confirmed in most instances by molecular testing. Conversely, a histologically malignant sarcomatoid tumor that is either focally positive or negative for cytokeratin should be cautiously interpreted. Focal cytokeratin positivity has been reported in many different types of sarcomas, or alternatively, focal cytokeratin positivity could represent entrapment of benign pleural elements. In this instance, if possible, additional blocks should be selected, and cytokeratin antigen-retrieval techniques, as well as antibody source and dilutions, should be considered. A vimentin stain is useful in assessing the general antigenic integrity of the tissue. Particularly in the absence of convincing cytokeratin positivity, calretinin or D2-40 positivity alone should not be interpreted as evidence of mesothelial differentiation. These markers are variably positive in many different types of sarcomas, and other immunohistochemical markers, such as CD31, CD34, desmin, myoglobin, and S100 should be added. The expanded differential diagnosis might include other sarcomas (epithelioid hemangioendothelioma, angiosarcoma, synovial sarcoma, liposarcoma, myogenic or neurogenic tumors, malignant solitary fibrous tumor) and melanoma. It should be noted that some muscle markers are positive, at least focally, and on occasion, more diffuse, in sarcomatoid diffuse malignant mesotheliomas.38 These markers include muscle-specific actin (HHF35) and alpha-smooth muscle actin. Desmin positivity in pure sarcomatoid diffuse malignant mesotheliomas is quite rare.39 In the final analysis, cytokeratin-negative sarcomatoid diffuse malignant mesotheliomas have been documented, and this seems to be particularly true of sarcomatoid diffuse malignant mesotheliomas with osteosarcomatous differentiation.34,40,41 PRIMARY LEIOMYOSARCOMA OF THE LUNG
Malignant tumors with smooth muscle differentiation are reliably diagnosed in the lung using criteria that are familiar to pathologists for similar tumors that arise in other sites, such as the uterus, gastrointestinal tract, or soft tissue. As noted in the introduction, the more important issue is usually excluding a metastasis from another site, which is impossible to do without the benefit of a careful and complete clinical history as well as thorough radiographic imaging. If metastasis can be confidently excluded, then pulmonary leiomyosarcomas can be subdivided into those that are primary within the pulmonary artery and those that originate within a bronchus or the pulmonary parenchyma. Moran et al42 reviewed 18 patients with primary leiomyosarcomas of the lung. In that study, there was a wide age range of 5 to 76 years, with a mean age of 50 years. Most patients were asymptomatic, and the lesions were discovered by routine physical and radiographic examination. As would be anticipated, only those patients with tumors that were endobronchial or bulky and locally invasive presented with symptoms, such as cough, hemoptysis, or chest pain. There was also a broad range of sizes-from 1.7 to 10 cm at greatest dimension-but size was not necessarily a reliable prognostic indicator. At least 2 lesions less than 3 cm, which were histologically high grade, proved fatal within 1 year. Grossly, all of the lesions were described as well circumscribed, grayish white, firm, and rubbery. In high-grade lesions, areas of hemorrhage and necrosis were readily identified.
The histologic diagnosis of leimoyosarcoma is prompted by the recognition of a spindle cell proliferation that is characterized by broad fascicles of tumor cells that intersect at right angles. The tumor cells typically have scant fibrillary eosinophilic cytoplasm and elongated, bluntended nuclei (Figure 8). All of these features are more readily recognizable in lower-grade lesions. Moran et al42 subdivided these tumors into low grade, intermediate grade, and high grade. Low-grade lesions showed only mild cytologic atypia and occasional mitoses (average, 1-3 per 10 high-power fields [HPFs]). Cellular pleomorphism and areas of necrosis or hemorrhage were absent. Intermediate-grade lesions retained a fascicular growth pattern, but the tumors were more cellular, and the mitotic rate increased slightly (average, 3-8 per 10 HPFs). Cellular pleomorphism was rare, and areas of hemorrhage or necrosis were not apparent. In the high-grade lesions, epithelioid morphology and a fascicular configuration were focal with more solid areas of spindle cell proliferation. Storiform and hemangiopericytoma-like patterns were also identified. The tumor cells showed marked cellular pleomorphism with frequent mitotic figures (average, 8-12 mitoses per 10 HPFs). In their series,42 15 patients had follow-up. Six patients with low- grade or intermediate-grade lesions were alive and free of disease with 2 to 12 years of follow-up. Eight patients with high-grade lesions died of disease within 24 months of diagnosis. Highgrade histology, however, was not invariably predictive of a poor prognosis. One patient with a high-grade, 7-cm mass in the right upper lobe and bronchus was alive at 12 years follow-up without evidence of metastases.
The histologic impression of smooth muscle differentiation can be confirmed by a panel of immunohistochemical markers, such as desmin and smooth muscle actin. Confirmation of diagnosis is considerably less complicated in low-grade or intermediate-grade lesions. There is a tendency, however, for these stains to be more focal and weak in higher-grade lesions. Although there are other markers of smooth muscle differentiation available (HHF35, calponin), none of these markers are significantly more sensitive in higher-grade lesions. An additional pitfall may be the presence of scattered, keratin- positive, atypical cells. The differential diagnosis in higher- grade lesions includes a malignant solitary fibrous tumor, a malignant peripheral nerve sheath tumor, synovial sarcoma, sarcomatoid carcinoma, and malignant fibrous histiocytoma (if one accepts malignant fibrous histiocytoma as a distinct entity in the lung). There are some variations in histologic growth patterns and cytologic features, which might help to suggest an alternative diagnosis. Narrower fascicular bundles and a herringbone pattern are more typical of tumors formerly considered to be fibrosarcomas. There is a general tendency for primary neurogenic sarcomas to show wavy nuclei rather than blunt-ended nuclei, but overlaps in alternating fascicular growth pattern with cellular pleomorphism may be seen in both lesions. Primary synovial sarcomas may also show an alternating fascicular growth pattern and closely resemble a high- grade leiomyosarcoma. High-grade leiomyosarcomas can have areas with storiform growth patterns similar to a malignant fibrous histiocytoma. In those instances, the absence of reactivity for smooth muscle markers, plus an affirmative marker, such as CD34, S100, or multiple cytokeratins, may help to confirm an alternative diagnosis. The diagnosis of malignant fibrous histiocytoma requires the complete absence of any differentiation markers. It is less likely that a high-grade leiomyosarcoma will be confused with a sarcomatoid carcinoma if the lesion is well sampled and multiple cytokeratin antibodies are used.
PRIMARY PULMONARY SYNOVIAL SARCOMA
As defined by the 2004 WHO classification,3 pulmonary synovial sarcoma is a mesenchymal spindle cell tumor, which variably displays areas of cellular differentiation. Zeren et al43 were the first to report on a large series of 25 primary pulmonary synovial sarcomas, which were otherwise histologically indistinguishable from those monophasic synovial sarcomas of the soft tissues. Following this initial description, several additional series followed, and the recognition of the entity was further facilitated by the availability of molecular testing for the characteristic t(X;18) chromosomal translocation, which allows for diagnostic confirmation in more than 90% of cases.44-48 There remains a smaller subset of tumors that are more difficult to recognize when histological features unusual to synovial sarcomas, but common in other tumors, are focally present. This diagnostic dilemma was recently addressed in a large clinicopathologic study,47 which included 34 cases of pulmonary synovial sarcoma, as well as 5 mediastinal and 10 pleural synovial sarcomas. The common presenting symptoms (dyspnea, cough, chest pain, hemoptysis, or incidental radiographic finding) do not significantly differ from other lung tumors and are site-specific. In the 5 series that included primary pulmonary sarcomas, there was no gender bias, and there was a wide variation in age range with an overall median of about 40 years.43-47 Most pulmonary tumors are peripheral with occasional cases forming an endobronchial mass. At least 75% of the reported cases to date have been unencapsulated solid tumors with well-demarcated margins, but diffuse infiltration into the pleura/chest wall or mediastinum also has been described.44,46 The term pleuropulmonary synovial sarcoma may be reasonably applied in instances where it is not possible to precisely localize the tumor. Tumor size has been between 0.6 and 17 cm (with means in the 5-8-cm range). On cut surface, tumors are typically soft, tan masses, with foci of necrosis, hemorrhage, and cystic change (Figure 9). The monophasic subtype that consists solely of a spindle cell component is the most common subtype. Caution should be exercised to avoid misinterpreting entrapped pneumocytes as an epithelial component, but it is clear that there are pulmonary tumors with biphasic histology. A substantial minority of pulmonary tumors are poorly differentiated. The generally aggressive clinical behavior of pulmonary synovial sarcomas, from which most patients are dead of disease at 5 years, has been attributed to the percentage of tumors with poor differentiation, as well as the comparatively later presentation and the difficulty in obtaining a wide surgical margin when compared with its soft tissue counterpart. Tumors tend to recur locally with extension into the chest wall, pericardium, paraspinal soft tissue, diaphragm, and abdomen.
The monophasic subtype consists of densely packed, elongated spindle cells, arranged in interweaving fascicles. A prominent hemangiopericytoma-like vascular pattern is quite common as are other typical features of synovial sarcoma, such as dense hyalinized or eosinophilic stroma and focal myxoid change (Figures 10 through 12). Calcification is present in about 15% of cases. Mast cell infiltration may not be readily identified on hematoxylin-eosin (H&E) stains but can be highlighted by CD117 staining.47 In biphasic tumors, the epithelial component consists of cleftlike glandular spaces with scattered tubulopapillary differentiation. Poorly differentiated tumors can be classified according to Federation Nationale des Centres de Lutte Contre le Cancer grading, which incorporates tumor differentiation, mitotic count, and tumor necrosis. Unusual histologic features in pulmonary synovial sarcomas include Verocay body-like formations, vague rosettes, wellformed papillary structures, adenomatoid change, and rhabdoid morphology.47 The major differential diagnosis of synovial sarcoma arising in the lung includes sarcomatoid carcinoma, the biphasic or sarcomatoid subtype of pleural diffuse malignant mesothelioma, leiomyosarcoma, malignant solitary fibrous tumor or fibrosarcoma, and malignant peripheral nerve sheath tumor. It is also critical to consider metastasis from an occult extrathoracic primary because, paradoxically, metastatic disease from an extrathoracic synovial sarcoma may have a more indolent course. Almost all biphasic synovial sarcomas are positive for cytokeratin markers and EMA in the epithelial component. Positive cytokeratin and EMA staining in monophasic synovial sarcomas is still quite high and in the general range of 50% to 70% but the staining is often less intense than in the epithelial component. Cytokeratins 7 and 19 may be particularly helpful because these cytokeratins are often negative in other spindle cell sarcomas.49,50 A certain number of synovial sarcomas will be positive for S100, and this may make differentiation from a malignant peripheral nerve sheath tumor problematic.51 Synovial sarcomas are strongly and diffusely positive for Bcl-2, but that is of limited utility in distinguishing synovial sarcoma from other sarcomas. CD99, which is frequently positive in synovial sarcoma, is a potential diagnostic pitfall given the morphologic overlap with some cases of primitive neuroectodermal tumors.51 A high percentage of calretinin positivity is similarly problematic given the morphologic overlap with diffuse malignant mesothelioma.52 D2-40 positivity has also been reported in epithelioid component of synovial sarcomas.36 Some synovial sarcomas are positive for smooth muscle actin, but they should be negative for desmin. TLE1 has recently been proposed53 as another marker with high sensitivity for synovial sarcomas, although moderate staining was seen in a significant number of solitary fibrous tumors and hemangiopericytomas. When a combination of molecular approaches to detect the SS18-SSX fusion transcripts is used and interpreted along with clinical and immunohistochemical data, a very high degree of sensitivity and specificity for the diagnosis of synovial sarcoma can be achieved.48
MALIGNANT PERIPHERAL NERVE SHEATH TUMOR
Malignant peripheral nerve sheath tumors (MPNSTs) are defined as any malignant tumor arising from a peripheral nerve or showing nerve sheath differentiation, with the exception of tumors originating from the epineurium or the peripheral nerve vasculature.54 It is unlikely that diagnostic difficulties will occur in the clinical setting of a patient with known type I neurofibromatosis, but sporadic cases do occur with some frequency in the thorax, and there is a known association in sites of previous radiation. 54 There are instances when the site of origin is not clearly localized to the chest wall or to the mediastinum and when definitive diagnosis may be difficult in a tumor with infiltrative borders. If the sampling is large enough, observing tumor growth in relationship to adjacent nerves may help to support the diagnosis. Histologically, highgrade peripheral nerve sheath tumors might easily be confused with leiomyosarcoma, monophasic synovial sarcoma, sarcomatoid diffuse malignant mesothelioma, malignant solitary fibrous tumor, or fibrosarcoma. The most common pattern of MPNST is fibrosarcoma- like, where the tumor is composed of hyperchromatic spindle cells that are arranged in a fascicular pattern (Figures 13 and 14). Alternating areas of hypocellurity and hypercellurity are seen, in addition to foci of necrosis and increased mitoses. Focal benign epithelial or malignant mesenchymal differentiation has been well documented. Rhabdomyosarcomatous differentiation is most commonly seen, but osseous and chondromatous differentiation can also occur. 55 In addition to the complexities introduced by divergent differentiation, there is variability in S100 staining, such that only 50% to 70% of MPNSTs will show scattered tumor cells that are S100 positive.56 The significant phenotypic and immunophenotypic overlap between MPNSTs and synovial sarcoma extends to molecular analysis, where it has been demonstrated that the t(X;18) is not entirely specific to synovial sarcoma.57 At the current time, there is a small subset of thoracic tumors in which it is simply not possible to conclusively exclude either possibility.
MALIGNANT SOLITARY FIBROUS TUMOR
It is now well accepted that solitary fibrous tumor is a distinctive mesenchymal tumor of probable fibroblastic origin with a range of appearances that can vary from a fibrous variant to a more cellular variant to a tumor with a prominent hemangiopericytoma- like growth pattern. The tumor commonly involves the pleura and occasionally arises within the lung or mediastinum, as well as within a number of extrathoracic sites. Following the study by van de Rijn et al in 1994,58 which documented CD34 expression by solitary fibrous tumors of the pleura, mediastinum, and lung, the diagnosis of primary pulmonary hemangiopericytoma has virtually disappeared from the pathology literature.59-61 Solitary fibrous tumors classically show a patternless architecture with both hypocellular and hypercellular areas separated by fibrous stroma having hemangiopericytoma-like, branching blood vessels. The hypercellular areas show bland spindle cells arranged in short intersecting fascicles, creating herringbone or storiform arrays (Figures 15 and 16).59 The hypocellular areas may be densely collagenized or show myxoid change. As in better-differentiated leiomyosarcomas, it is seldom difficult to confirm the diagnosis of an intrapulmonary solitary fibrous tumor, which is largely fibrous, by positive CD34 and Bcl-2 staining along with reciprocal negative staining for cytokeratin. The diagnosis becomes more problematic in more cellular or frankly malignant lesions. A hemangiopericytoma- like pattern of branching vessels is typically seen in solitary fibrous tumors. However, a hemangiopericytomatous pattern can also be seen in synovial sarcomas, malignant peripheral nerve sheath tumors, leiomyosarcomas, diffuse malignantmesotheliomas, malignant melanomas, sarcomatoid carcinomas, metastatic endometrial stromal sarcomas, and thymomas-to name other tumors likely to be encountered in the thorax. Low-grade, exclusively intraparenchymal, solitary, fibrous tumors have been reported in the lung, but the criteria predictive of malignancy have been based on pleuralbased tumors and must be extrapolated to tumors that might appear to be primarily parenchymal. These criteria include greater cellularity with an infiltrative growth pattern, moderate to marked cytologic atypia, and a high mitotic rate (>4 mitoses per 10 HPFs).62 In instances of a malignant sarcomatoid tumor that is suspected to be a malignant solitary fibrous tumor but appears to be CD34 negative, extensive sectioning may yield areas that are better differentiated and more likely to be CD34 positive (Figures 17 through 20). Otherwise, the diagnosis may rest on the workup and exclusion of other entities, such as synovial sarcoma.
PULMONARY EPITHELIOID HEMANGIOENDOTHELIOMA
Epithelioid hemangioendothelioma is a low-grade to intermediate- grade vascular tumor with a distinctive histologic appearance, which has been described in the lung as well as in numerous other sites including the liver, soft tissue, bone, mediastinum, and pleura.63 The tumor is distinctive in histologic appearance with short cords and nests of epithelioid cells associated with a myxohyaline matrix (Figure 21). In the lung, the tumor has been reported most frequently in whites, and most often in white women (80%).64-66 Most patients have been younger than 40 years but with a wider age range from 12 to 61 years. About half of patients are asymptomatic, with the remaining half presenting with cough, dyspnea, or chest pain. The radiographic presentation is typically that of multiple bilateral nodules of varying size (in the general range of 1-2 cm), with occasional calcification. Solitary lung masses have also been described. The presence of multiple lesions always raises the issue of metastasis from another site, and these findings should prompt a careful radiographic evaluation. Cases with multifocal and bilateral disease, in the absence of a documented extrathoracic primary, have been reported. Grossly, the tumor is usually circumscribed with a gray-white or gray-tan appearance. The cut surface has a cartilaginous consistency, and there may be central calcification. Of the malignant tumors in the lung, the entity that epithelioid hemangioendothelioma would be most likely confused with is chondrosarcoma. This differential diagnosis is easily sorted out by immunohistochemical stains for CD31, CD34, friend leukemia integration 1 (Fli1), and S100.67 Cytokeratin stains are not reliable discriminators because 20% to 30% of cases will show focal cytokeratin positivity.63
PRIMARY PULMONARY ANGIOSARCOMA
Primary pulmonary angiosarcomas are exceeding rare, even within the select group of primary pulmonary sarcomas or otherwise rare tumors. Patients typically present with late-stage disease, and in many instances, it is difficult to actually establish that the lung is the primary site. As a practical point, angiosarcoma in the lung is far more likely to represent metastatic disease.68 Whether primary or metastatic, epithelioid angiosarcoma can mimic a carcinoma and show cytokeratin positivity.69 The diagnosis of angiosarcoma should be considered in the immunohistochemical workup of any poorly differentiated tumor within the lung, particularly if associated with excessive hemorrhage (Figure 22). These stains should include CD31 and CD34. KAPOSI SARCOMA
For completeness' sake, Kaposi sarcoma is included in this discussion of vascular spindle cell lesions in the lung. Unless there is a complete absence of relevant history, radiographic information, and bronchoscopic findings, it is not likely that Kaposi sarcoma will be mistaken for other sarcomatoid tumors in the lung. Kaposi sarcoma typically presents as purple-red, raised nodules along the tracheobronchial tree or as diffuse bronchial wall thickening. The most common patterns of lung involvement are nodular and interstitial infiltrates, in addition to pleural effusion, rather than forming a single large mass as is characteristic of many of the other sarcomatoid lesions. The disease is uniformly associated with human herpesvirus 8 infection. 70 The spindled cells are positive for CD31 and CD34 but not factor VIII (Figure 23).
INFLAMMATORY MYOFIBROBLASTIC TUMOR
As defined by the 2004 WHO classification, inflammatory myofibroblastic tumor is a subgroup in the broad category of inflammatory pseudotumors.71 The lesion is composed of a variable mixture of collagen, inflammatory cells, and usually, cytologically bland spindle cells that show myofibroblastic differentiation. Whether this lesion is a reactive inflammatory condition or a low- grade mesenchymal malignancy remains controversial. Regardless of etiology, this lesion's clinical, radiographic, and gross presentation overlap with other pulmonary sarcomas. Inflammatory myofibroblastic tumors can occur at any age, although most occur in patients younger than 40 years old, and this tumor is the most common endobronchial mesenchymal lesion reported in childhood. Inflammatory myofibroblastic tumors occur in both endobronchial and peripheral locations. Presenting symptoms reflect the lesion's location. Patients with endobronchial lesions present with cough, wheezing, and hemoptysis as well as postobstructive pneumonia and atelectasis. Peripheral lesions may be asymptomatic, but some do invade the chest wall and cause pain. Grossly, the lesions are usually solitary, round, rubbery masses, with a wide variation in size range reported from 1 to 36 cm (average size, 3 cm). The variability in yellow to gray discoloration reflects the histiocytic component of the inflammatory infiltrate. Like other sarcomas, the lesions do not appear encapsulated, and local invasion into hilar soft tissue, pleura, or chest wall can be seen.
The identification of this lesion and its separation from other spindle cell tumors may be particularly challenging on an endobronchial biopsy or closed needle biopsy. Like other sarcomas, the spindle cells are arrayed in fascicles or may have storiform architecture. Upon closer examination (and in most cases), the spindle cells should not have obvious cytologic atypia, and mitoses should be rare. The spindle cells typically have oval nuclei, fine chromatin, inconspicuous nucleoli, and abundant lightly eosinophilic cytoplasm (Figure 24). Further consideration of this entity may be prompted by the recognition of a prominent inflammatory infiltrate that includes lymphocytes, plasma cells, and histiocytes as well as Toutonlike giant cells. Foamy histiocytes, in particular, can be quite numerous. The spindle cells show evidence of fibroblastic and myofibroblastic differentiation with expression of vimentin and smooth muscle actin. Desmin immunoreactivity is rare. The spindle cells should be negative for myogenin, myoglobin, CD117 (c-Kit), and S100. Focal cytokeratin reactivity has been reported, but it is not clear whether that represents entrapped alveolar cells. Expression of ALK-1 and p80 has been reported in about 40% of cases, but these stains are not routinely used for diagnosis. With complete excision, the prognosis for this lesion is excellent. Recurrence usually occurs with incomplete excision. About 5% of these lesions may be locally aggressive, recur, or metastasize. Poor prognostic histologic features include focal invasion, vascular invasion, high cellularity, nuclear pleomorphism with bizarre giant cells, a high mitotic rate, and necrosis.
OTHER EXCEEDINGLY RARE SARCOMAS
As with the other rare types of pulmonary sarcoma discussed, it is important to consider the possibility of metastasis from another primary site or more common diagnoses. such as carcinosarcoma, in this group of tumors.
Primary pulmonary rhabdomyosarcomas are extremely rare but have been reported in both children and adults. The widespread use of immunohistochemistry and molecular analysis has increased the recognition of rhabdomyosarcoma in sites throughout the body and has confirmed the lung as an uncommon site. Endobronchial, intraparenchymal, and pulmonary trunk sites have all been reported. The literature has been summarized in several reviews, but it is difficult to determine the reliability of the older reports.72-75 The histologic findings of malignant cells with deeply eosinophilic cytoplasm and cross-striations may suggest the diagnosis, but these are generally difficult to find.72 Immunohistochemical stains, such as desmin and myoglobin, are useful, but staining can be focal.
Primary Chondrosarcoma of the Lung
Nearly all the older and current literature on primary chondrosarcomas of the lung consist of single-case reports with review of the literature.72,76-78 Both endobronchial and peripheral locations have been described. The central tumors tend to be discovered at an earlier stage and appear to have a relatively good prognosis until tumor size precludes complete resection.
Primary Pulmonary Osteosarcoma
There are fewer than 20 cases of primary pulmonary osteosarcoma reported in the literature.72,79,80 One of the more recent case reports during the past decade included a molecular analysis that demonstrated several specific genetic aberrations in common with other extraskeletal osteosarcomas. 80 As with other rare types of pulmonary sarcoma, it is important to consider an alternative diagnosis and this seems to be particularly true of osteosarcomatous foci. Tumors that were initially thought to be primary lung osteosarcomas proved to be carcinosarcomas upon additional sampling.79 The limited number of cases of pulmonary osteosarcoma precludes any meaningful statement on prognosis.
Primary pulmonary liposarcoma is another tumor with fewer than 20 cases reported in the literature.72,81 The more common occurrence of liposarcoma metastatic to the lung can be diagnostically challenging. The diversity in histologic appearance that may be encountered when liposarcoma metastasizes to lung has been recently reviewed.82
Primary Pulmonary Fibrosarcoma, Malignant Fibrous Histiocytoma, and Hemangiopericytoma
And what of primary pulmonary fibrosarcoma, malignant fibrous histiocytoma, and hemangiopericytoma-entities that used to be well enough represented in series of lung sarcomas?83-87 Suffice it to say that, in the recent pathology literature, to my knowledge, there have been no new reported cases.
Other Metastatic Tumors With Spindle Cell Features That May Closely Mimic Primary Pulmonary Sarcomas
Although not intended as an exhaustive list, there are several other metastatic tumors that deserve mention. Most melanomas are not entirely spindled in appearance, but metastatic melanoma may be spindled enough to mimic sarcoma in the lung. An adequate clinical history and consideration of the entity, in addition to the routine use of S100 in the workup of a spindled cell lesion in the lung, is helpful in avoiding this diagnostic pitfall (Figures 25 through 27). Similarly, an adequate clinical history and consideration of a temporally remote primary, along with use of cytokeratin stains, such as CAM 5.2, should avoid misinterpretation of a metastatic sarcomatoid renal cell carcinoma (Figures 28 and 29). Metastatic endometrial stromal sarcoma can be an overlooked consideration in the workup of a primitive-appearing mesenchymal tumor in the lung. In this instance, the more-likely uterine primary will not be radiographically obvious or clinically apparent without a thorough physical examination and an appropriate history.
This review of sarcomatous pulmonary tumors has provided an overview of primary pulmonary sarcomas as well as other thoracic lesions that may mimic them. For the most part, the clinical, radiographic, and gross presentation of these rare and extremely rare tumors does not differ significantly from the more common tumors of the lung. As a general principle, complete excision of these malignant tumors is recommended if the patient is an operative candidate but the relative infrequency of these tumors makes it difficult to predict clinical behavior in many individual circumstances. Appropriate classification ultimately requires complete sampling and a thorough microscopic evaluation. The general availability of ancillary diagnostic techniques, such as immunohistochemistry and molecular analysis, has refined and, no doubt, will continue to refine tumor identification and classification.
1. Huo L, Moran CA, Fuller GN, Gladish G, Suster S. Pulmonary artery sarcoma: a clinicopathologic and immunohistochemical study of 12 cases. Am J Clin Pathol. 2006;125:419-424.
2. Travis WD, Colby TV, Corrin B, Shimosato Y, Brambilla E, Sobin LH. Histological Typing of Lung and Pleural Tumours. 3rd ed. Berlin, Germany: Springer-Verlag; 1999.
3. Corrin B, Chang YL, Rossi G, et al. Sarcomatoid carcinoma. In: Travis WD, Brambilla E, Mu ller-Hermelink HK, Harris CC, eds. Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. Lyon, France: IARC Press; 2004: 53-58. World Health Organization Classification of Tumours; vol 10. 4. Koss MN, Hochholzer L, Frommelt RA. Carcinosarcomas of the lung: a clinicopathologic study of 66 patients. Am J Surg Pathol. 1999;23:1514-1526.
5. Manivel JC, Priest JR, Watterson J, et al. Pleuropulmonary blastoma: the socalled pulmonary blastoma of childhood. Cancer. 1986;62:1516-1526.
6. Spencer H. Pulmonary blastoma. J Pathol Bacteriol. 1961;82:161- 165.
7. Francis D, Jacobsen M. Pulmonary blastoma. Curr Top Pathol. 1983;73:265-294.
8. Davis MP, Eagan RT,Weiland LH, Pairolero PC. Carcinosarcoma of the lung: Mayo Clinic experience and response to chemotherapy. Mayo Clin Proc. 1984; 59:598-603.
9. Cabarcos A, Gomez Dorronsoro M, Beristain JL. Pulmonary carcinosarcoma: a case study and review of the literature. Br J Dis Chest. 1985;79:83-94.
10. Humphrey PA, Scroggs MW, Roggli VL, Shelburne JD. Pulmonary carcinomas with a sarcomatoid element: an immunocytochemical and ultrastructural analysis. Hum Pathol. 1988;19:155-165.
11. Ishida T, Tateishi M, Kaneko S, et al. Carcinosarcoma and spindle cell carcinoma of the lung: clinicopathologic and immunohistochemical studies. J Thorac Cardiovasc Surg. 1990;100:844- 852.
12. Yousem SA, Wick MR, Randhawa P, Manivel JC. Pulmonary blastoma: an immunohistochemical analysis with comparison with fetal lung in its pseudoglandular stage. Am J Clin Pathol. 1990;93:167- 175.
13. Matsui K, Kitagawa M. Spindle cell carcinoma of the lung: a clinicopathologic study of three cases. Cancer. 1991;67:2361-2367.
14. Moore TC. Carcinosarcoma of the lung. Surgery. 1991;50:886- 893.
15. Chejfec G, Candel A, Jansson DS, et al. Immunohistochemical features of giant cell carcinoma of the lung: patterns of expression of cytokeratins, vimentin, and the mucinous glycoprotein recognized by monoclonal antibody A-80. Ultrastruct Pathol. 1991;15:131-138.
16. Koss MN, Hochholzer L, O'Leary T. Pulmonary blastomas. Cancer. 1991; 67:2368-2381.
17. Ro JY, Chen JL, Lee JS, Sahin AA, Ordonez NG, Ayala AG. Immunohistochemical and ultrastructural studies of 14 cases. Cancer. 1992;69:376-386.
18. Eng J, Sabanathan S. Carcinosarcoma of the lung with gastrointestinal metastasis: case report. Scand J Thorac Cardiovasc Surg. 1992;26:161-162.
19. Fishback NF, Travis WD, Moran CA, Guinee DG Jr, McCarthy WF, Koss MN. Pleomorphic (spindle/giant cell) carcinoma of the lung: a clinicopathologic correlation of 78 cases. Cancer. 1994;73:2936- 2945.
20. Nappi O, Glasner SD, Swanson PE, Wick MR. Biphasic and monophasic sarcomatoid carcinoma of the lung: a reappraisal of "carcinosarcomas" and "spindle-cell carcinomas." Am J Clin Pathol. 1994;102:331-340.
21. Rainosek DE, Ro JY, Ordonez NG, Kulaga AD, Ayala AG. Sarcomatoid carcinoma of the lung: a case with atypical carcinoid and rhabdomyosarcomatous components. Am J Clin Pathol. 1994;102:360- 364.
22. Terzi A, Gorla A, Piubello Q, Tomezzoli A, Furlan G. Biphasic sarcomatoid carcinoma of the lung: report of 5 cases and review of the literature. Eur J Surg Oncol. 1997;23:457.
23. Wick MR, Ritter JH, Humphrey PA. Sarcomatoid carcinomas of the lung: a clinicopathologic review. Am J Clin Pathol. 1997;108:40- 53.
24. Nakajima M, Kasai T, Hashimoto H, Iwata Y, Manabe H. Sarcomatoid carcinoma of the lung: a clinicopathologic study of 37 cases. Cancer. 1999;86:608-616.
25. Chang YL, Lee YC, Shih JY, Wu CT. Pulmonary pleomorphic (spindle) cell carcinoma: peculiar clinicopathologic manifestations different from ordinary nonsmall cell carcinoma. Lung Cancer. 2001;34:97-91.
26. Robert J, Pache JC, Seium Y, de Perrot M, Spiliopoulos A. Pulmonary blastoma: report of five cases and identification of clinical features suggestive of the disease. Eur J Cardiothorac Surg. 2002;22:708-711.
27. Nishida K, Kobayashi Y, Ishikawa Y, et al. Sarcomatoid adenocarcinoma of the lung: clinicopathological, immunohistochemical and molecular analyses. Anticancer Res. 2002;22:3477-3483.
28. Pelosi G, Fraggetta F, Nappi O, et al. Pleomorphic carcinomas of the lung show a selective distribution of gene products involved in cell differentiation, cell cycle control, tumor growth, and tumor cell motility: a clinicopathologic and immunohistochemical study of 31 cases. Am J Surg Pathol. 2003;27:1203-1215.
29. Rossi G, Cavazza A, Sturm N, et al. Pulmonary carcinomas with pleomorphic, sarcomatoid, or sarcomatous elements: a clinicopathologic study of 75 cases. Am J Surg Pathol. 2003;27:311- 324.
30. Raveglia F, Mezzetti M, Panigalli T, et al. Personal experience in surgical management of pulmonary pleomorphic carcinoma. Ann Thorac Surg. 2004;78: 1742-1747.
31. Adluri RKP, Boddu SR, Martin-Ucar A, Duffy JP, Beggs FD, Morgan WE. Pulmonary blastoma-a rare tumor with variable presentation. Eur J Cardiothorac Surg. 2006;29:236-239.
32. Martin LW, Correa AM, Ordonez NG, et al. Sarcomatoid carcinoma of the lung: a predictor of poor prognosis. Ann Thorac Surg. 2007;84:973-981.
33. Churg A, Cagle PT, Roggli BL. Tumors of the Serosal Membranes. Washington, DC: Armed Forces Institute of Pathology; 2006. Atlas of Tumor Pathology; 4th series, fascicle 10.
34. Mayall FG, Goddard H, Gibbs AR. The diagnostic implications of variable cytokeratin expression in mesotheliomas. J Pathol. 1993;170:165-168.
35. Chu AY, Litzky LA, Pasha TL, Acs G, Zhang PJ. Utility of D2- 40, a novel mesothelial marker, in the diagnosis of malignant mesothelioma. Mod Pathol. 2005;18:105-110.
36. Ordonez NG. D2-40 and podoplanin are highly specific and sensitive markers of epithelioid malignant mesothelioma. Hum Pathol. 2005;36:372-380.
37. Hinterberger M, Reineke T, Storz M, Weder W, Vogt P, Moch H. D2-40 and calretinin-a tissue microarray analysis of 341 malignant mesotheliomas with emphasis on sarcomatoid differentiation. Mod Pathol. 2007;20:248-255.
38. Kung IT, Thallus V, Spencer EJ, Wilson SM. Expression of muscle actins in diffuse mesothelioma. Hum Pathol. 1995;26:565-570.
39. Trupiano JK, Geisinger KR, Willingham MC, et al. Diffuse malignant mesothelioma of the peritoneum and pleura, analysis of markers. Mod Pathol. 2004; 17:476-481.
40. Attanoos RL, Dojcinov SD, Webb R, Gibbs AR. Anti-mesothelial markers in sarcomatoid mesothelioma and other spindle cell neoplasms. Histopathology. 2000;37:224-231.
41. Lucas DR, Pass HI, Madan SK, Adsay NV, Wali A, Tabaczka P, Lonardo F. Sarcomatoid mesothelioma and its histologic mimics: a comparative immunohistochemical study. Histopathology. 2003;42:270- 279.
42. Moran CA, Suster S, Abbondanzo SL, Koss MN. Primary leiomyosarcomas of the lung: a clinicopathologic and immunohistochemical study of 18 cases. Mod Pathol. 1997;10:121-128.
43. Zeren H, Moran CA, Suster S, Fishback NF, Koss MN. Primary pulmonary sarcomas with features of monophasic synovial sarcoma: a clinicopathological, immunohistochemical, and ultrastructural study of 25 cases. Hum Pathol. 1995; 26:474-480.
44. Essary LR, Vargas SO, Fletcher CD. Primary pleuropulmonary synovial sarcoma: reappraisal of a recently described anatomic subset. Cancer. 2002;94:459-469.
45. Okamoto S, Hisaoka M, Daa T, Hatakeyama K, Iwamasa T, Hashimoto H. Primary pulmonary synovial sarcoma: a clinicopathologic, immunohistochemical and molecular study of 11 cases. Hum Pathol. 2004;35:850-856.
46. Begueret H, Galateau-Salle F, Guillou L, et al. Primary intrathoracic synovial sarcoma: a clinicopathologic study of 40 t(X;18)-positive cases from the French Sarcoma Group and the Mesopath Group. Am J Surg Pathol. 29:339-346.
47. Hartel PH, Fanburg-Smith JC, Frazier AA, et al. Primary pulmonary and mediastinal synovial sarcoma: a clinicopathologic study of 60 cases and comparison with five prior series. Mod Pathol. 2007;20:760-769.
48. Amary MF, Berisha F, Bernardi Fdel C, Herbert A, et al. Detection of SS18-SSX fusion transcripts in formalin-fixed embedded neoplasms: analysis of conventional RT-PCR, qRT-PCR and dual color FISH as diagnostic tools for synovial sarcoma. Mod Pathol. 2007;20:482-96.
49. Miettinen M. Keratin subsets in spindle cell sarcomas: keratins are widespread but synovial sarcoma contains a distinctive keratin polypeptide pattern and desmoplakins. Am J Pathol. 1991;138:505-513.
50. Smith TA, Machen SK, Fisher C, et al. Usefulness of cytokeratin subsets for distinguishing monophasic synovial sarcoma from peripheral malignant nerve sheath tumor. Am J Clin Pathol. 1999;112:411-414.
51. Folpe AL, Schmidt RA, Chapman D, Gown A. Poorly differentiated synovial Arch Pathol Lab Med-Vol 132, July 2008 Pulmonary Sarcomatous Tumors-Litzky 1117 sarcoma: immunohistochemical distinction from primitive neuroectodermal tumors and high-grade malignant peripheral nerve sheath tumors. Am J Surg Pathol. 1998;22:673-682.
52. Miettinen M, Limon J, Niezabitowski A, Lasota J. Calretinin and other mesothelioma markers in synovial sarcoma: analysis of antigenic similarities and differences with malignant mesothelioma. Am J Surg Pathol. 2001;25:610-617.
53. Terry J, Saito T, Subramanian S, et al. TLE1 as a diagnostic immunohistochemical marker for synovial sarcoma emerging from gene expression profiling studies. Am J Surg Pathol. 2007;31:240-246.
54. Woodruff JM, Kourea HP, Louis DN, Sheithauer BW. Malignant peripheral nerve sheath tumor (MPNST). In: Kleihues P, Cavanee WK, eds. Pathology and Genetics of Tumors of the Nervous System. Lyon, France: IARC Press; 2000:172-174. World Health Organization Classification of Tumours; vol 7.
55. Stasik CJ, Tawfik O. Malignant peripheral nerve sheath tumor with rhabdomyosarcomatous differentiation (malignant triton tumor). Arch Pathol Lab Med. 2006;130:1878-1881.
56. Weiss SM, Langloss JM, Enzinger FM. Value of S-100 protein in the diagnosis of soft tissue tumors with particular reference to benign and malignant Schwann cell tumors. Lab Invest. 1983;49:299- 308.
57. O'Sullivan MJ, Kyriakos M, Zhu X, et al. Malignant peripheral nerve sheath tumors with t(X;18): a pathologic and molecular genetic study. Mod Pathol. 2000; 13:1336-1346. 58. van de Rijn M, Lombard CM, Rouse RV. Expression of CD34 by solitary fibrous tumors of the pleura, mediastinum, and lung. Am J Surg Pathol. 1994;18: 814-820.
59. Travis WD, Churg A, Aubry MC, et al. Solitary fibrous tumour. In: Travis WD, Brambilla E, Mu ller-Hermelink HK, Harris CC, eds. Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. Lyon, France: IARC Press; 2004:142-143. World Health Organization Classification of Tumours; vol 10.
60. Nappi O, Ritter JH, Pettinato G, Wick MR. Hemangiopericytoma: histopathological pattern or clinicopathologic entity? Semin Diagn Pathol. 1995;12: 221-232.
61. Gengler C, Guillou L. Solitary fibrous tumor and haemangiopericytoma: evolution of a concept. Histopathology. 2006;48:63-74.
62. England DM, Hochholzer L, McCarthy MJ. Localized benign and malignant fibrous tumors of the pleura: a clinicopathologic review of 223 cases. Am J Surg Pathol. 1989;13:887-899.
63. Travis WD, Tazelaar HD, Miettinen M. Epithelioid haemangioendothelioma/angiosarcoma. In: Travis WD, Brambilla E, Mu ller-Hermelink HK, Harris CC, eds. Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. Lyon, France: IARC Press; 2004:97-98. World Health Organization Classification of Tumours; vol 10.
64. Dail DH, Liebow AA, Gmelich JT, et al. Intravascular, bronchiolar, and alveolar tumor of the lung (IVBAT): an analysis of twenty cases of a peculiar sclerosing endothelial tumor. Cancer. 1983;51:452-464.
65. Eggleston JC. The intravascular bronchioloalveolar tumor and the sclerosing hemangioma of the lung: misnomers of pulmonary neoplasia. Semin Diagn Pathol. 1985;270-280.
66. Weiss SW, Ishak KG, Dail DGH, et al. Epithelioid hemangioendothelioma and related lesions. Semin Diagn Pathol. 1986;3:259-287.
67. Folpe AL, Chand EM, Goldblum JR, Weiss SW. Expression of Fli- 1, a nuclear transcription factor, distinguishes vascular neoplasms from potential mimics. Am J Surg Pathol. 2001;25:1061-1066.
68. Patel AM, Ryu JH. Angiosarcoma in the lung. Chest. 1993;103:1531-5.
69. Sheppard MN, Hansell DM, Du Bois RM, Nicholson AG. Primary epithelioid angiosarcoma of the lung presenting as pulmonary hemorrhage. Hum Pathol 1997;28:383-385.
70. Lamovec J, Knuutila S. Kaposi sarcoma. In: Fletcher CDM, Unni KK, Martens F, eds. Pathology and Genetics of Tumours of Soft Tissue and Bone. Lyon, France: IARC Press; 2002:170-172. World Health Organization Classification of Tumours; vol 5.
71. Yousem SA, Tazelaar HD, Manabe T, Dehner LP. Inflammatory myofibroblastic tumour. In: Travis WD, Brambilla E, Mu ller- Hermelink HK, Harris CC, eds. Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. Lyon, France: IARC Press; 2004:105-106. World Health Organization Classification of Tumours; vol 10.
72. Suster S. Primary sarcomas of the lung. Semin Diagn Pathol. 1995;12:140-157.
73. Przygodzki RM, Moran CA, Suster S. Koss MN. Primary pulmonary rhabdomyosarcoma: a clinicopathologic and immunohistochemical study of three cases. Mod Pathol. 1995;8:658-661.
74. Comin CE, Santucci M, Novelli L, Dini S. Primary pulmonary rhabdomyosarcoma: report of a case in an adult and review of the literature. Ultrastruct Pathol. 2001;25:269-273.
75. Gupta A, Sharma MC, Kochupillai V. Primary pulmonary rhabdomyosarcoma in adults: case report and review of literature. Clin Lung Cancer. 2007;8: 389-391.
76. Hayashi T, Tsuda N, Iseki M, Kishikawa M, Shinozaki T, Hasumoto M. Primary chondrosarcoma of the lung: a clinicopathologic study. Cancer. 1993; 72:67-74.
77. Shukla K, Jetly D, Parikh B. Primary chondrosarcoma of lung: case report and review of literature. Indian J Pathol Microbiol. 2006;49:570-573.
78. Steurer S, Huber M, Lintner F. Dedifferentiated chondrosarcoma of the lung: case report and review of the literature. Clin Lung Cancer. 2007;8:439-442.
79. Colby TV, Bilbao JE, Battifora H, Unni KK. Primary osteosarcoma of the lung: a reappraisal following immunohistologic study. Arch Pathol Lab Med. 1989;113:1147-1150.
80. Chapman AD, Pritchard SC, Yap WW, et al. Primary pulmonary osteosarcoma: case report and molecular analysis. Cancer. 2001;91:779-784.
81. Krygier G, Amado A, Salisbury S, Fernandez I, Maedo N, Vazquez T. Primary lung liposarcoma. Lung Cancer. 1997;17:271-275.
82. Nicholas M, Moran CA, Suster S. Pulmonary metastasis from liposarcoma: a clinicopathologic and immunohistochemical study of 24 cases. Am J Clin Pathol. 2005;123:265-275.
83. Attanoos RL, Appleton MA, Gibbs AR. Primary sarcomas of the lung: a clinicopathological and immunohistochemical study of 14 cases. Histopathology. 1996;29:29-36.
84. Bacha EA, Wright CD, Grill HC, et al. Surgical treatment of primary pulmonary sarcomas. Eur J Cardiothorac Surg. 1999;15:456- 460.
85. Keel SB, Bacha E, Mark EJ, Nielsen GP, Rosenberg AE. Primary pulmonary sarcoma; a clinicopathologic study of 26 cases. Mod Pathol. 1999;12:1124-1131.
86. Etienne-Mastroianni B, Falchero L, Chalabreysse L, et al. Primary sarcomas of the lung: a clinicopathologic study of 12 cases. Lung Cancer. 2002;38:283-289.
87. Petrov DB, Vlassov VI, Kalaydjiev GT, et al. Primary pulmonary sarcomas and carcinosarcomas-postoperative results and comparative survival analysis. Eur J Cardiothorac Surg. 2003;23:461- 466.
Leslie A. Litzky, MD
Accepted for publication January 11, 2008.
From the Department of Pathology and Laboratory Medicine, University of Pennsylvania Medical Center, Philadelphia.
The author has no relevant financial interest in the products or companies described in this article.
Reprints: Leslie A. Litzky, MD, Department of Pathology and Laboratory Medicine, 6 Founders Bldg, University of Pennsylvania Medical Center, 3400 Spruce St, Philadelphia, PA 19104-4283 (e- mail: [email protected]