129: Benign Tumors of the Pleura

Tumors of the pleura are uncommon, and when they arise, are almost always malignant. Malignant mesothelioma and malignant pleural metastases are far more common than benign pleural tumors. Primary benign tumors of the pleura represent only 5% of all pleural tumors and include solitary fibrous tumors, lipomas and lipoblastomas, adenomatoid tumors, and calcifying fibrous tumors. Of these, solitary fibrous tumors are the most common, and will be the primary focus of this chapter. There have been fewer than a thousand cases of solitary fibrous tumor reported in the published literature since the entity was first described by Klemperer and Rabin in 1931.¹

Solitary fibrous tumor has been known for many decades, but its nomenclature has evolved. In the original description in 1931, Klemperer and Rabin classified mesothelioma as either “localized” or “diffuse.”¹ Since that time, solitary fibrous tumor has been previously described as “localized mesothelioma,” “benign mesothelioma,” “solitary fibrous mesothelioma,” or “pleural fibroma.” These terms reflected the incorrect assumption that they were related to malignant mesothelioma of the pleura. More recent research has shown that solitary fibrous tumors originate from mesenchymal cells in the subserosal or submesothelial layer of the pleura, not in the mesothelial layer, which is distinct from mesothelioma (Fig. 129-1).² Moreover, solitary fibrous tumors are not associated with asbestos exposure, further indicating a unique pathophysiologic process in the origin of these tumors.

Solitary fibrous tumors are rare. A contemporary estimate of an age-adjusted incidence indicates a rate of 0.14 per 100,000³ in comparison to mesothelioma, which has an incidence of approximately 2 per 100,000.⁴ Solitary fibrous tumors occur equally in men and women, and its incidence is highest during the fourth to seventh decades of life. It is asymptomatic in approximately half of the cases. In the other half of patients, solitary fibrous tumors can be associated with either intrathoracic or extrathoracic manifestations. Intrathoracic symptoms are related to the mass effect of the tumor itself on the lungs or chest wall, which can result in dyspnea, chest pain, or a chronic cough. Extrathoracic manifestations are paraneoplastic, and manifest as hypertrophic pulmonary osteoarthropathy or clubbing of the digits in 20% (Pierre–Marie–Bamberg syndrome), hypoglycemia in 5% (Doege–Potter syndrome), gynecomastia, or galactorrhea. Constitutional symptoms may also be present, such as fever, fatigue, and weight loss.

Solitary fibrous tumors most commonly arise on the visceral side of the pleura and are usually ovoid in shape (Figs. 129-2 and 129-3). As the name suggests, these tumors are usually isolated. The tumors usually grow in a pedunculated manner, and the pedicle can be highly vascularized. In addition, large aberrant vessels feeding into the tumor have been described. Due to the pedunculated feature, movement of the mass on x-ray with changing of the patient's position is possible.⁵ Invasion of the lung is uncommon, and such tumors are described as an “inverted” solitary fibrous tumor since the bulk of the tumor is intraparenchymal in the lung.

The differential diagnosis of solitary fibrous tumor includes mesothelioma, sarcomas (synovial sarcoma, fibrosarcoma, nerve sheath sarcoma, malignant fibrous histiocytoma), and hemangiopericytoma. The histology of solitary fibrous tumors reveals their mesenchymal origin, and typically consists of a classic “patternless pattern” of whorled, dense fibrous tissue.⁶ There are spindle cells mixed with collagen and elastin bundles. Cystic structures can be found within the tumor, as can calcification. Immunohistochemistry can be useful to distinguish them from mesothelioma, as routine histology may not give a definitive diagnosis. Staining characteristics of solitary fibrous tumor include positive staining for vimentin, CD34, and bcl-2, while being negative for keratin.⁷

There is a subset of solitary fibrous tumors that are malignant, but it is difficult to determine this preoperatively. Preoperative needle biopsy usually is not helpful in most cases due to the heterogeneous cellularity of the tumor. Approximately 12% of solitary fibrous tumors are malignant.² Furthermore, some tumors contain a mixture of benign and malignant components, indicating that malignant degeneration is possible in the natural history of the disease.⁷ There are some radiographic criteria that can help distinguish benign and malignant cases of solitary fibrous tumors. Malignant tumors are typically >10 cm, have heterogeneous low attenuation regions of necrosis, hemorrhage, cysts, or myxoid degeneration, and are associated with pleural or pulmonary metastases.⁸ Due to their larger size, malignant tumors are more likely to present with symptoms than benign tumors.⁶ PET/CT has not generally been found to be helpful for identifying malignant tumors. In one series the positive predictive value was 50% and the negative predictive value was 87.5%.⁹

England et al.⁶ provided a histological means of distinguishing benign from malignant solitary fibrous tumors in their landmark 1981 publication, and these criteria continue to be used today. Criteria for malignancy include (1) high mitotic rate >4 mitoses per 10 HPF; (2) high cellularity with crowding and overlapping nuclei; (3) presence of necrosis; and (4) pleomorphism. If any of these four characteristics are present in the tumor, it is classified as malignant. Immunohistochemistry has been attempted to help clarify benign from malignant cases of solitary fibrous tumor, but at this time there is no reliable marker for this purpose. Some malignant tumors lose CD34 staining but always retain bcl-2 positive staining. Ki-67 or proliferating cell nuclear antigen (PCNA) have been examined for distinguishing malignant from benign tumors, but no distinct threshold has been determined.⁷ The limitations of this method have been the relative subjectivity of interpreting immunohistochemistry slides and the heterogeneous nature of the tumors, with unequal distribution of cellularity.

De Perrot et al.⁷ have further classified solitary fibrous tumors with respect to their malignant potential by categorizing them into four distinct clinical stages (Table 129-1). This system was developed to stratify the risk of recurrence for prognosis based on 185 cases published in the literature. Tumors are classified as either benign or malignant, and either sessile or pedunculated. For benign solitary fibrous tumors, there is little prognostic difference in whether the tumor is pedunculated or sessile, with long term survival being 98% versus 92%, respectively. However, for malignant tumors, a pedunculated growth pattern was associated with much better prognosis, with long-term survival being 86% as compared to 37% for sessile tumors. Recurrence occurred in 2% of benign pedunculated tumors, 8% of benign sessile tumors, 14% of malignant pedunculated tumors, and 63% of malignant sessile tumors. Most recurrences occur in the first 24 months following resection. Size has not been shown to be an independent predictor of recurrence.

The preoperative assessment of patients with such tumors is similar to many other thoracic patients. Assessment of pulmonary function should be considered when lung resection is contemplated for tumors arising from the visceral pleura, and is mandatory if an inverted tumor necessitates a formal lobectomy for complete resection. Imaging studies typically consist of a CT scan with intravenous contrast. MRI is often useful in distinguishing soft tissue planes for involvement of the lung and assessing involvement of the chest wall or diaphragm. Preoperative biopsies are helpful when present but not required, as it is common to obtain a paucicellular specimen that is not fully informative. Very large tumors have been found to have significant vascular collaterals, and some have advocated preoperative embolization in these instances to minimize intraoperative blood loss since visualization is often poor during resection.^{10 –12}

Complete surgical resection is the mainstay of therapy for solitary fibrous tumor of the pleura. A margin of 1 to 2 cm is recommended with en bloc resection of any involved structures. Pedunculated tumors arising from the lung may include a wedge resection of the lung, but sessile or “inverted” tumors may require lobectomy. Extrapleural dissection may be necessary to resect tumors adherent to the parietal pleura, and chest wall resection is warranted if invasion is encountered. A mediastinal lymph node dissection should be done if the tumor is determined to be malignant, as nodal metastases have been reported.⁶ While a minimally invasive approach using video-assisted thoracoscopic surgery (VATS) is possible, often the tumors are too large to remove in this manner and a thoracotomy is required. Resolution of paraneoplastic hypoglycemia (Doege–Potter syndrome) and hypertrophic osteoarthropathy (Pierre–Marie–Bamberg syndrome) can be expected with complete resection.¹³

There is no standardized systemic therapy for the treatment of solitary fibrous tumor. A recent report identified the potential use of imatinib, and PDGFR pathway inhibitor, on tumors that overexpress PDGFR.¹⁴ However, at this time there is no widely accepted or conventional adjuvant therapy for the management of this tumor. Adjuvant chemotherapy and/or radiation can be considered on an individual basis. Following resection, surveillance radiographs are recommended for both benign and malignant solitary fibrous tumors, particularly in the first 24 to 36 months.

Benign pleural tumors are the exception, along with some breast and uterine tumors, to the general notion that large tumors are malignant. Large smooth-walled tumors in the pleura may only require complete excision to prevent recurrence. Pleural tumors also challenge the notion of what constitutes a “benign” tumor. Although the genetic signature of pleural tumors is likely to provide more detailed insight into potential biologic behavior, our current level of understanding is limited to counting mitoses to estimate the likelihood of local or distant recurrence.