At the intersection of primary pulmonary myxoid sarcoma and pulmonary angiomatoid fibrous histiocytoma: observations from three new cases.

Sir: Twin recent reports identified a novel entity, primary pulmonary myxoid sarcoma (PPMS) and an endobronchial presentation of angiomatoid fibrous histiocytoma (AFH). PPMS were described as 20– 40-mm diameter endobronchial, lobulated neoplasms occurring in young patients (mean age 45 years), and comprising polygonal to spindled cells within a myxoid stroma evocative of extraskeletal myxoid chondrosarcoma (EMC). In contrast to EMC, seven of nine testable cases harboured EWSR1–CREB1 fusions. Nine of 10 showed remarkable associated inflammation, which was focal to follicular to lymphoplasmacytic. Two of six patients for whom followup was available developed metastases (brain and kidney), supporting classification as a low-grade sarcoma. One additional patient had no recurrence. Two cases of endobronchial AFH showed a characteristic multinodular growth pattern of bland, ovoid to spindled cells with a peritumoral lymphoplasmacytic cuff, a fibrotic capsule, and strong focal desmin expression (without classic blood-filled ‘angiomatoid’ spaces). Similarly to reports of eight AFHs occurring in unusual sites, including the lung, both endobronchial AFHs showed focal myxoid change. One lesion harboured an EWSR1–CREB1 fusion, and the other harboured an EWSR1–ATF1 fusion. Neither of the patients (males aged 61 and 64 years) has suffered recurrence. To emphasize the recent recognition of this entity, we report three additional patients with what we termed PPMS, seen in consultation, recognizing overlap with what might be considered pulmonary AFH. Patient 1 was a 66-year-old woman who presented with a 2-year history of intermittent pulmonary obstructive symptoms and pneumonia, at which time her tumour was discovered on a chest computed tomography (CT) scan. Her resected left upper lobe showed a pale 40-mm diameter endobronchial mass, with associated postobstructive consolidation. Microscopically, her endobronchial-based tumour (Figure 1A) infiltrated into peribronchial pulmonary parenchyma. Over 80% of the lesion showed polygonal to spindled cells with a reticular, myoepithelialtype appearance and mild atypia in a pink–grey myxoid stroma (Figure 1B). A peritumoral fibrous cuff was present, particularly at the interface with alveolar parenchyma (Figure 1C). The lesion was negative for pancytokeratin, p63, S100, and desmin, and showed patchy EMA-positive cells; break-apart fluorescence in-situ hybridization (FISH) was positive for both EWSR1 (Figure 1D) and CREB1 (negative for ATF1). Patient 2 was a 28-year-old man presenting with 4 months of cough with haemoptysis. Imaging showed an 85-mm diameter right lower lobe mass described grossly as obliterating the bronchus. Microscopically, this tumour was a lobular, biphasic lesion, ~40% of which was composed of myxoid pools (Figure 1E). The remainder showed exuberant fibroinflammatory reaction with confluent plasma cells. Two foci showed ‘angiomatoid’ blood-filled spaces (Figure 1F). Moderate atypia was present (more than in patient 1) among reticular and clustered cells floating in the distinctive EMC-type stroma (Figure 1G). The lesion was negative for pancytokeratin, p63 and S100, showed focal EMA positivity, but was strongly positive for desmin, visible against the inflammatory background and discontinuous peritumoral fibrous capsule (Figure 1H). FISH was positive for EWSR1 rearrangement, but not for CREB1 or ATF1 rearrangements. Patient 3 was another 28-year-old man, who presented with a 2-week history of chest pain. A CT scan showed a 60-mm diameter lesion in the right upper lobe of the lung. At low magnification, the tumour showed a destructive growth pattern, with focal necrosis and infiltration from an endobronchial base into adjacent lung parenchyma (Figure 1I). At higher power, the degree of atypia was severe (Figure 1J), and the myxoinflammatory areas (80% of the lesion) expanded to undermine the bronchial epithelium (Figure 1K). As in case 1, the lesion was negative for keratins, p63, and S100, and there were rare EMA-positive cells. FISH was negative for rearrangement at EWSR1 (Figure 1L), CREB1, and ATF1. Taken together, these lesions span a spectrum from PPMS to AFH in terms of morphology, atypia, and molecular findings. The varying proportion of the myxoid component, from 40% to 80%, is greater than that described for AFH cases in the lung. Case 2 showed a degree of plasmacytic inflammation, a fibrotic component, and desmin positivity, none of which were included in previous descriptions of PPMS, with ‘angiomatoid’ foci that have not yet been reported in endobronchial AFH or PPMS cases. All three cases showed at least a focal peritumoral fibrous 144 Correspondence