Diffuse Malignant Mesothelioma

A 66-year-old man presented with recurrent, right pleural effusion. The patient underwent thoracoscopic examination, and the surgeon peeled “white-yellow fibrous plaque” from the right parietal pleura and chest wall, which was sent for frozen section. At frozen section, a diagnosis of “fibrous pleural plaques with chronic inflammation” was made, and tissue was submitted for permanent-section examination. On permanent sections, fibrous pleural tissue, adipose tissue, and skeletal muscle were present along with focal areas of proliferating epithelioid cells with round nuclei, prominent nucleoli, and relatively abundant cytoplasm consistent with mesothelial cells. Focally, the epithelioid cells were noted to infiltrate the skeletal muscle (Figure 1). A diagnosis of diffuse malignant mesothelioma (DMM) was made based on this unequivocal, focal invasion into skeletal muscle. The diagnosis of DMM had not been suspected clinically, radiographically, surgically, grossly, or initially, on frozen section. It was not until the focal invasion was discovered on permanent sections that the diagnosis of DMM could be made.

Diffuse malignant mesothelioma is an uncommon cancer, but it has great significance because (1) it is currently an incurable disease, and most patients die within months after diagnosis, thus diagnosis carries very severe implications; (2) the close association of most traditional cases with occupational asbestos exposure introduces a host of medicolegal, regulatory, environmental, and government policy issues, far beyond what is experienced with most cancers; (3) although the incidence of DMM is leveling off or decreasing in the United States and some other developed countries because of the strict control on asbestos use, DMM is increasing as a worldwide problem, both in countries that delayed asbestos control relative to the United States and in many developing countries where burgeoning industrialization is not accompanied by adequate asbestos control (there are also reports of an increase of DMM in a younger population in developed countries without evidence of asbestos exposure); and (4) DMM can be notoriously difficult to diagnose for 2 primary reasons: (1) benign, reactive mesothelial hyperplasias and organizing pleuritis can be difficult to differentiate from DMM in some settings and vice versa; and (2) the variety of DMM's histopathologic features generates an extensive list of differential diagnoses with other malignancies, particularly metastatic malignancies, which are more frequent in the pleura than are primary malignancies.^1–10 

This fourth issue—the quandaries of diagnosis—is the topic of this review. Historically, the difficulties in diagnosing DMM led to the creation, decades ago, of national panels of pathology experts for consultation, including the United States–Canadian Mesothelioma Reference Panel. The first chair of the United States–Canadian Mesothelioma Reference Panel was the late Jacob Churg, MD, and the current chair is his son, Andrew Churg, MD, of the University of British Columbia (Vancouver, British Columbia, Canada). In recent years, in addition to consulting on medical cases referred by pathologists, this group has published several articles on the differentiation of DMM from benign mimics.^11–12  Since the late 1990s, the International Mesothelioma Panel has been headed by Francoise Galateau-Sallé, MD, of Caen, France, who is also chair of the French Mesothelioma Reference Panel. This group, supported by the French Ministry of Health, is composed of experts from many countries and is particularly focused on differentiation of early, including in situ, DMM from reactive mesothelial hyperplasia.⁴  Many of us (the same experts) also serve on the pathology section of the International Mesothelioma Interest Group, headed by Aliya Husain, MD, of the University of Chicago, which has previously published a consensus guideline on diagnosis of DMM in the Archives of Pathology & Laboratory Medicine and will be publishing an update in the near future.^7,10  The publications of all of these expert panels are excellent sources for information on DMM, particularly the pathologic diagnosis.

The pleural surface is lined by a single layer of flattened, inconspicuous mesothelial cells that may show varying degrees of reactive hyperplasia when the pleura is irritated or inflamed. Reactive mesothelial hyperplasia is often, but not always, seen in association with organizing fibrinous and/or fibrous pleuritis (Figure 2). Pleuritis is often accompanied by pleural effusion. Conditions that may cause reactive mesothelial hyperplasia and/or organizing pleuritis include, but are not limited to, infections (pleural infections and subpleural pneumonias), collagen vascular diseases, pulmonary infarcts, drug reactions, pneumothorax, subpleural lung carcinomas, thoracic surgery, and chest trauma as well as nonspecific or idiopathic cases. In some cases, pleuritis with pleural thickening and pleural effusion of unknown etiology is biopsied to rule out malignancy. Whether malignancy is a clinical and radiologic consideration, the pathologist may encounter florid, reactive mesothelial hyperplasia and/or organizing pleuritis on pleural biopsy or decortication specimens, for which DMM is in the histopathologic differential diagnosis. The diagnostic question is not “what type of malignancy,” but rather, “is this cellular proliferation benign or malignant?”^1–7,9–17 

Simple hyperplasia of reactive mesothelium is usually not confused with malignancy and, as the name implies, consists of a single row of regularly spaced, cytologically bland, cuboidal mesothelial cells along the pleural surface (Figure 3). Many florid, reactive mesothelial hyperplasia cases, however, can exhibit one or more features suggestive of malignancy, including high cellularity, cytologic atypia with enlarged nuclei and prominent nucleoli, and the presence of mitoses (Figure 4). In fact, some cases of florid, reactive mesothelial hyperplasia can have greater cytologic atypia than some cases of epithelioid DMM, which can be very bland cytologically. Other histopathologic features sometimes observed in florid, reactive mesothelial hyperplasia can appear even more ominous. These include architectural changes, such as simple lumens or simple papillary tufts (Figure 5); areas of necrosis (eg, leukocytoclastic necrosis occasionally seen in pleural infections); and, especially, entrapment.^{2–7,10,11,14–17} 

Entrapment (Figure 6) occurs when organizing pleuritis layers lie over a pleural surface lined by reactive mesothelial hyperplasia, giving an initial impression of invasive mesothelial cells (see below). Closer examination discloses that the hyperplastic mesothelial cells are arranged in a linear pattern (which, in fact, is along the original pleural surface). The hyperplastic mesothelial cells may exhibit the reactive changes mentioned in the previous paragraph. However, no invasion into the underlying pleural or subpleural tissues is observed, and the “lumen” or “lumens” are simple, linear spaces between mesothelial lined surfaces and not complex, branching tubules. Cytokeratin immunostain will disclose cytokeratin-positive mesothelial cells along the original pleural surface and, perhaps, scattered within the organizing pleuritis, but there should be a sharp demarcation with no cytokeratin-positive cells infiltrating the underlying pleural connective tissue or subpleural tissues (Figure 7).^{2–7,10,11,14–17} 

The proliferating fibroblasts, endothelial cells, and spindled mesothelial cells of organizing pleuritis may mimic sarcomatoid DMM, and fibrocytes and spindled mesothelial cells within mature, dense, fibrous pleuritis may mimic desmoplastic DMM. The storiform pattern that may be seen in sarcomatoid DMM is lacking in benign pleuritis. Zonation of the proliferating cells, with more toward the surface and tapering off in the deeper tissues, is characteristic of benign pleuritis and can be confirmed with cytokeratin immunostaining of the reactive spindle mesothelial cells, if needed (Figure 8). The orientation of the tissue may be important because tangential or en face sections may be misleading about the distribution of the proliferating cells.^{2–7,10,11,14–17} 

Another feature that is suggestive of organizing pleuritis, as opposed to DMM, is the presence of parallel capillaries perpendicular to the pleural surface within the granulation tissue. Infiltrating malignant cells would presumably cause disarray of the capillaries, although this is not definitive evidence of malignancy.^7,10,15,17 

A few significant features are considered confirmatory of DMM on pleural biopsy and decortication specimens and apply to both epithelioid cells and sarcomatoid cells. Invasion into underlying tissues, specifically adipose tissue and/or skeletal muscle under the parietal pleura and lung tissue under the visceral pleura, is diagnostic of DMM. Cytokeratin immunostain may be useful in confirming the infiltration of cytokeratin-positive cells within adipose tissue, skeletal muscle, or lung tissue (Figure 9). Once again, tangential or en face sections may be misleading.^{2–7,10,11,14–17}  Care should also be taken not to mistake spaces within organizing pleuritis (so-called fake fat; Figure 10) for infiltration of desmoplastic DMM around adipose tissue (the so-called Swiss cheese effect).¹² 

Proliferating cellular nodules that expand the surrounding stroma are confirmatory of a neoplasm rather than a reactive process. In addition, cancer can be diagnosed if there are unequivocal histopathologic features of malignancy, such as severe pleomorphism and atypical mitoses. A lack of zonation with high cellularity of atypical cells throughout the thickness of the pleura suggests malignancy if the specimen is properly oriented and not a tangential or en face section.^{2–7,10,11,14–17}  Bland necrosis with foci of “clean,” ischemic-type necrosis of the proliferating cells is characteristic of DMM (Figure 11) and differs from the “dirty,” leukocytoclastic necrosis often seen with infections and other types of cancer, such as colon cancer. Although bland necrosis is not diagnostic of DMM by itself, its presence in a biopsy or decortication specimen may bring attention to a desmoplastic DMM that might otherwise be overlooked.¹³ 

We recommend the term atypical mesothelial proliferation when it cannot be determined whether a mesothelial proliferation is benign or malignant in a given specimen based on the histopathologic criteria discussed above, for example, during frozen section or when a biopsy is inconclusive. No doubt, some atypical mesothelial proliferations are florid, reactive mesothelial hyperplasias, and others are DMM, but the histopathologic features in the biopsy may not be definitive for either diagnosis. Therefore, malignancy is neither confirmed nor ruled out when a diagnosis of atypical mesothelial proliferation is made. In these circumstances, the patient's treating physician may elect to observe the patient, attempt to obtain diagnostic tissue, or make another decision, based on clinical suspicion and other circumstances.^{2–7,10–17} 

Most cancers involving the pleura are metastatic cancers, and DMM, as a primary malignancy of the pleura, is actually uncommon, with fewer than 2000 cases per year in the United States. Diffuse malignant mesotheliomas have a variety of histopathologic patterns, which means that many different types of cancer potentially enter into the differential diagnosis of DMM and vice versa.

Diffuse malignant mesothelioma is divided into epithelioid (the most common), sarcomatoid, and mixed or biphasic patterns. In the mixed or biphasic pattern, both epithelioid and sarcomatoid patterns are observed in the same tumor.^{1–7,9,10,16–18} 

Epithelioid DMMs are composed of cuboidal to polygonal cells that histologically resemble mesothelial cells in well-differentiated DMM, and the correct diagnosis of DMM is strongly suspected on the basis of the hematoxylin-eosin histology, in many cases. Secondary patterns of DMM include well-differentiated solid (Figure 12), tubulopapillary (Figure 13), or acinar (Figure 14) patterns; less-frequently adenomatoid or poorly differentiated solid patterns are seen, and deciduoid, adenoid cystic, signet ring, clear cell, or small cell patterns are seen infrequently.^{1–7,9,10,16–22} 

Although most of the literature focuses on differentiation of epithelioid DMM from pulmonary adenocarcinoma, that differential, although important, is not the only differential to consider for epithelioid DMM. Metastatic adenocarcinomas from organs other than the lungs, for example, the gastrointestinal tract or kidney, may need to be considered for the solid, tubulopapillary, acinar, adenomatoid, deciduoid, adenoid cystic, signet ring, and clear cell patterns. Beyond metastatic adenocarcinomas from various sites, metastatic squamous cell carcinomas, melanomas, and urothelial carcinomas may also enter into the differential diagnosis of solid-pattern DMM, particularly, if poorly differentiated, and metastatic squamous cell carcinoma or melanoma might enter into the differential diagnosis of clear cell pattern DMM. Small cell DMMs are extremely rare, but, if encountered, must be differentiated from metastatic small cell carcinoma of the lung or other organs.^{1–7,9,10,16–28} 

Sarcomas, metastatic to the pleura and, much less-frequently, primary in the pleura, enter into the differential diagnosis of sarcomatoid DMM, as do sarcomatoid carcinomas and spindle cell melanomas. Sarcomatoid DMM may sometimes have heterologous elements (osteosarcomatous or chondrosarcomatous) or be pleomorphic.^{1–7,9,10,16–32}  The rare lymphohistiocytoid DMM (Figure 15) must be differentiated from lymphoma, in addition to inflammatory conditions.^33,34 

The differentiation of these patterns of DMM from metastatic cancers from other sites depends on the clinical history and imaging observations as well as the immunostaining patterns of the cancer. The immunohistochemical traits of the entire range of malignancies that might metastasize to the pleura and mimic DMM is beyond the scope of this review, and there are a number of other sources on this subject.^{1–7,9,10,25,27,28}  However, as noted earlier, for all the focus in the literature on pulmonary adenocarcinoma versus epithelioid DMM, the pathologist should not omit consideration of primary sites other than the lungs and cell types of cancer other than adenocarcinoma, depending on the context of the case and the histopathologic pattern observed. Because the immunostains that distinguish pulmonary adenocarcinoma from epithelioid DMM may not apply to these other cancers, other immunostains may be warranted, depending on the diagnostic considerations. Some of these other cancers may be immunopositive for markers traditionally used to separate DMM from pulmonary adenocarcinoma. For example, squamous cell carcinoma is often immunopositive for CK 5/6, which was once used as a DMM marker, and ovarian carcinoma is often immunopositive for WT1, another marker used to distinguish DMM from pulmonary adenocarcinoma.^{1–7,9,10,25,27,28}