Fibrosarcoma

Primary pulmonary sarcomas are uncommon neoplasms. Primary fibrosarcoma of the lung is extremely rare, and only 53 cases have been documented in the literature to date. To our knowledge, the diagnosis of primary lung fibrosarcoma by fine-needle aspiration cytology has never been reported. We report a case of pulmonary fibrosarcoma diagnosed by fine-needle aspiration cytology and core biopsy. The neoplasm consisted of interweaving fascicles of minimally atypical spindle cells with slender nuclei and scant cytoplasm. Positive immunohistochemistry for vimentin along with nonreactivity of tumor cells for keratin, SI 00 protein, desmin, alpha-smooth muscle actin, and CD34 supported the the diagnosis. The diagnosis was later confirmed by histologic and ultrastructural findings following lobectomy. A meticulous clinical search for a possible primary neoplasm elsewhere was unsuccessful, and lung was established as the primary site. Fine-needle aspiration cytology and core biopsy are reliable methods for establishing a diagnosis of fibrosarcoma

(Arch Pathol Lab Med.1999;123:731-735)

Primary sarcomas of the lung are extremely uncommon tumors,1 with fibrosarcomas accounting for about 12% of all cases.2 Since most sarcomas involving the lungs are metastatic neoplasms, a sarcoma should be regarded as primary in the lung only after a thorough clinical investigation has failed to reveal a primary site elsewhere.3 Also, the diagnosis of fibrosarcoma in the lung is largely one of exclusion,4 as it requires excluding other spindle cell proliferations and sarcomatous neoplasms having similar morphologic appearances.

Fine-needle aspiration biopsy (FNAB) is a well-established technique for the collection of cellular samples from peripheral and deep-seated lesions. Fine-needle aspiration biopsy guided by imaging modalities can yield near tissue-equivalent diagnostic material on the basis of which therapy can be instituted? Minimal morbidity, fast turnaround times, and low costa make FNAB a preferred method in the diagnostic workup of neoplasia.

To our knowledge, the diagnosis of primary fibrosarcoma of the lung (PFL) by FNAB has never been reported in the literature. A case of PFL diagnosed by FNAB as well as ancillary studies supporting this diagnosis are presented.

REPORT OF A CASE

The patient was a 39-year-old white man with a history significant for drug abuse, cigarette smoking, diabetes mellitus, infective endocarditis, congestive heart failure, and chronic obstructive pulmonary disease. He was referred to the University of Texas Medical Branch hospital for further care of his endocarditis. At the Medical Branch, the patient underwent aortic root replacement with a 27-mm aortic homograft and repair of a sinus of Valsalva aneurysm. A routine preoperative chest radiograph revealed a well-demarcated, homogenous, 1.5-cm right upper lung mass.

The patient was lost to clinical follow-up for a year, until he was readmitted to the University of Texas Medical Branch with a 4-week history of chills, fatigue, myalgia, nausea, and vomiting. Blood cultures were positive for Staphylococcus aureus, and appropriate antibiotic therapy was initiated. The right upper lobe mass was found to have increased in size to 4.0 x 4.0 x 3.3 cm as measured bv the chest roentgenogram and computed tomography (CT). A CT scan of the abdomen was unremarkable.

Computed tomography-guided FNAB of the right upper lobe mass was then performed (Figure 1). Core biopsies were also obtained to perform immunohistochemical stains. A cytologic diagnosis of spindle cell neoplasm consistent with low-grade fibrosarcoma was established. Additional imaging studies, including a bone scan and CT scan of the head, failed to demonstrate tumor elsewhere, thus establishing the pulmonary fibrosarcoma as primary in the lung. One month later, right upper lobectomy was performed.

MATERIALS AND METHODS

Fine-needle aspiration biopsy was performed under CT guidance using 22-gauge Chiba needles (Cook, Inc, Bloomington, Ind). The needles were advanced into the right upper lobe lesion via a posterior approach, and 2 aspirates were obtained. The aspirated material was expressed onto glass slides. The smears were immersed in 95% ethanol for fixation. Air-dried smears were also stained in the radiology suite with Quik-Dip (Mercedes Medical, Inc, Sarasota, Fla) for immediate assessment of specimen adequacy and for preliminary examination bv the cytopathologist. The needles were then rinsed in Hank's balanced salt solution (Gibco BRL, Grand Island, NY) for further cell recovery and concentration procedures. Cellular suspensions were concentrated by Cytospin (Shandon Corporation, Pittsburgh, Pa) and additional slide preparations were obtained.

The radiologist also obtained core biopsies from the lung lesion, using 18- and 20-gauge Bard Monopty guns (CR Bard, Inc, Covington, Ga). The tissue cores were fixed in formaldehyde and embedded in paraffin. Hematoxylin-eosin-stained slides were prepared. In addition, immunochemical stains were performed on tissue sections by using an automated Ventana ES immunostainer (Ventana Medical Systems, Tucson, Ariz). A standard avidin-biotin complex technique with diaminobenzidine as the chromogen was used. Commercially available monoclonal antibodies against keratin (AE1 /AE3 keratin, dilution 1:80 with protease pretreatment, Boehringer Mannhein, Indianapolis, Ind), S100 protein (Ventana), vimentin (clone 3B4 with protease pretreatment, Ventana), desmin (clone DER11 with protease pretreatment, Ventana), a-smooth muscle actin (clone 1A4, dilution on, Dako Corporation, Carpinteria, Calif), and CD34 (dilution 1:5 with protease pretreatment, Dako) were used.

Tumor tissue from the lobectomy specimen was fixed in 2.5% glutaraldehyde, postfixed in osmium tetroxide, and embedded in epoxy resin for ultrastructural studies. The preparations were evaluated initially in toluidine blue-stained, 1-(mu)m-thick sections. Ultrathin sections were stained with uranyl acetate and lead citrate and were further evaluated using a Phillips CM-100 (Mahwah, NJ) electron microscope.

PATHOLOGIC FEATURES

Cytologic Findings

The aspirates were hypercellular and consisted of a monotonous population of neoplastic spindle cells with slender nuclei and scant cytoplasm (Figure 2). The tumor cells presented both as single cells and in fascicles with a prominent interweaving pattern. There was minimal nuclear atypia. No significant necrosis or background materialmyxoid or other-was identified.

Histologic Findings

Tissue sections from the core biopsy and the lobectomy specimen stained with hematoxylin-eosin revealed similar findings. The tumor was highly cellular and consisted of spindle cells with fusiform nuclei and pointed ends, which were arranged in a herringbone or broad fascicular pattern (Figure 3, A). Although it was unencapsulated, the neoplasm was well demarcated from the surrounding lung parenchyma. Up to 22 mitoses per 10 high-power fields were counted in the most cellular areas of the tumor. Focal necrosis was also identified.

Immunohistochemistry

The tumor cells showed strong cytoplasmic labeling with vimentin (Figure 3, B). However, no staining was evident with the following immunohistochemical markers: cytokeratin, S100 protein, desmin, alpha-smooth muscle actin, and CD34.

Gross Pathology

The resected right upper lobe of lung weighed 310 g and measured 14.5 x 11.0 x 2.5 cm. There was a wellcircumscribed, intraparenchymal, 4.0 x 3.8 x 3.7-cm, lobulated, tan-white, nodular mass. The pleural surface was uninvolved by the mass. The tumor cut surface was slightly bulging, firm, glistening, and rubbery. It was surrounded by compressed lung parenchyma.

Electron Microscopy

Ultrastructurally, the neoplasm consisted of spindleshaped cells with elongated nuclei (Figure 4). The cytoplasm contained branching rough endoplasmic reticulum with prominent dilatation and occasional pinocytotic vesicles. Rare extracellular collagen bundles were noted (Figure 4, inset).

COMMENT

Only 53 cases of PFL have been reported in the medical literature.1-3,7-16 These include both adult and childhood cases. Primary fibrosarcoma of the lung in childhood is generally considered to be a rare neoplasm of favorable prognosis, regardless of the morphologic features.11

Primary fibrosarcoma of the lung may arise from the bronchial wall or the lung parenchyma.1,8 of any lobe,3 and it frequently presents as an endobronchial mass. Primary fibrosarcoma of the lung is often asymptomatic and may be an incidental finding during diagnostic workup of unrelated conditions, as occurred in this case. However, these tumors may also present clinically with chest pain, hemoptysis, cough, fever, chills, and fatigue.

To our knowledge, PFL diagnosed by FNAB has never been reported. However, the cytologic appearance of fibrosarcoma metastatic to the lung has been described. The metastasis consisted of numerous oval to spindle-shaped, fibroblast-like cells arranged in loose aggregates, with the cells parallel to the longitudinal axes of the aggregates.'7

Morphologically, PFL has to be differentiated from other vimentin-positive spindle cell neoplasms involving the lungs, such as leiomyosarcoma, monophasic synovial sarcoma, neurofibrosarcoma, and intrapulmonary solitary fibrous tumor. Leiomyosarcomas stain positively for desmin and a-smooth muscle actin, and monophasic synovial sarcomas often react positively for cytokeratin and epithelial membrane antigen. Neurofibrosarcomas stain positively for S100 protein, and solitary fibrous tumors express CD34.s None of these markers label PFL.

The prognosis of primary pulmonary sarcomas, particularly PFL, has been correlated with tumor size,37 histologic grade,2 mitotic count,7 and the location of the tumor (endobronchial vs intraparenchymal).7 In one case series, all patients with a PFL larger than 5 cm in diameter eventually died of tumor.3 Endobronchial fibrosarcomas occur most commonly in children and young adults, are usually detected at an earlier stage, and have a better prognosis than the parenchymal variant.7 Intraparenchymal PFLs with fewer than 8 mitoses per 10 high-power fields usually tend to grow slowly, in contrast to tumors with a higher mitotic count, which tend to metastasize.7

Surgical resection remains the main treatment modality for PFL. The prognosis after surgery is reportedly better for primary sarcomas of the lung than for bronchogenic carcinomas.1

We conclude that, in the context of consistent clinical and radiological findings, image-guide FNAB and core biopsy combine to be a reliable method for establishing a diagnosis of fibrosarcoma in the lung. However, additional studies to exclude a primary tumor site other than lung are required before a diagnosis of PFL can be established.

References

1. Cameron EWJ. Primary sarcoma of the lung. Thorax. 1975;30:516-520.

2. Gehauer C. The postoperative prognosis of primary pulmonary sarcomas. Scand I Thor Cardiovasc Surg. 1982;16:91-97.

3. Nascimento AG, Unni KK, Bernatz PE. Sarcomas of the lung. Mayo Clin Proc. 1982;57:355-359.

4. Suster S. Primary sarcomas of the lung. Semin Diagn Pathol. 1995;12:140157.

5. Logrono R, Kurtycz DFI, Sproat IA, Shalkham JE, Stewart JA, Inhorn SL. Multidisciplinary approach to deep-seated lesions requiring radiologic:ally-guided fine needle aspiration. Diagn Cytopathol. 1998:18:338-342.

6. Kaminsky DB. Aspiration biopsy in the context of the new Medicare fiscal policy Acta Cytol. 1984;28333 336.

7. Guccion JG, Rosen SH. Bronchopulmonary leiomyosarcoma and fibrosarcoma: a study of 32 cases and review of the literature. Cancer. 1972;30:836847.

8. Miller DL, Allen MS. Rare pulmonary neoplasms. Mayo Clin Proc. 1993; 68:492-498.

9. Goldthorn IF, Duncan Mt , Kosloke AM, Ball W5. Cavitating primary pulmonary fibrosarcoma in a child, / Thorac Cardiovasc Surg. 1 986;91:932-934.

10. Kamat MR, Mehta AR, Shrikhande SS. Fibrosarcoma of lung: a case report. Indian / Cancer. 173;10:382-384.

11. Pettinato G. Manivel If- Saldana Ml, Peyser J, Dehner LP. Primary bronchopulmonary fibrosarcoma of childhood and adolescence: reassessment of a low grade malignancy: clinicopathologic study of five cases and review of the literature. f lum Pithol. 1989;20:463-471.

12. Cohen MC, Kas(chula ROC. Primary pulmonary tumors in childhood: a review of 31 years' experience and the literature. Pediatr Pulmonol. 1992;14: 222-232.

13. Scully RE, Mark El, McNeely BU. Case records of the Massachusetts General Hospital, weekly clinicopathological exercises: case 7-1982. N Engl Med. 1982;306:412-420.

14. Ledet SC, Brown RW, Cagle PT. P53 immunostaining in the differentiation of inflammatory pseudotumor from sarcoma involving the lung. Mod Pathol. 1995;8:282-286.

15. Turner MA. Horne CHW. Primary fibrosarcoma of lung and diabetes mellitus. Br / Surg. 1970;57:713-715.

16. McLigeyo SO, Mbui J, Kungu A, Amayo E, Ogendo SWO. Fibrosarcoma of the lung with extrapulmonary manifestations: case report. East Atr Med 1995; 72:465-467.

17. Kim K, Naylor B, Han Ihl. Fine needle aspiration cytology of sarcomas metastatic to the long. As-ta Cytol, 1986:iO:688-694.

18. Van de Rijn M, Lombard CM, Rouse RV Expression of CD34 by solitary fibrous tumor of the pleura, mediastinum, and lung. Am I Surg Pathol. 1994;18: 814-820.

Accepted for publication january 20, 1999.

From the Department of Pathology, University of Texas Medical Branch, Galveston.

Presented as a poster at the 78th annual meeting of the Texas Society of Pathologists, Dallas, January 29-30, 1999.

Reprints: Roberto Logrono, MD, University of Texas Medical Branch, Cytopathology Division, Department of Pathology, 9.300 John Sealy Annex, Galveston, TX 77555-0548.

Copyright College of American Pathologists Aug 1999
Provided by ProQuest Information and Learning Company. All rights Reserved