Pathology – Research and Practice 210 (2014) 210–216
Contents lists available at ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.com/locate/prp
Original Article
Incidence of pulmonary non-epithelial tumors: 18 years’ experience at a single institute In Ho Choi a , Dae Hyun Song a , Kang Min Han a , Yong Soo Choi b , Joungho Han a,∗ a b
Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea Department of Thoracic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
a r t i c l e
i n f o
Article history: Received 20 September 2013 Received in revised form 27 October 2013 Accepted 2 December 2013 Keywords: Lung Sarcoma Mesenchymal tumor Pulmonary sarcoma Incidence
a b s t r a c t Primary and metastatic non-epithelial tumors of the lung and pleura (NETs-LP) are presumed to represent a small proportion of pulmonary neoplasm, and their epidemiologic characterization is lacking. Moreover, classification of pulmonary mesenchymal tumor according to ‘WHO Classification of Tumors of Lung and Pleura’ (2004) is simple and necessary to give a supplementation like new ‘WHO Classification of Tumors of Soft Tissue and Bone’ (2013). Non-epithelial tumors of the lung and pleura diagnosed from January 1995 to April 2013 were retrospectively reviewed and classified in accordance with ‘WHO Classification of Tumors of Soft Tissue and Bone’ (2013). Of 681 cases represented in the study, smooth muscle tumor (13.1%) was most common, and tumor of uncertain differentiation (10.5%) and fibroblastic/myofibroblastic tumor (5.7%) were followed. Smooth muscle tumor (12.6%) was the common primary tumor, whereas chondrogenic or osteogenic tumor (21.6%) was common in metastatic ones. Pulmonary vascular sarcoma (n = 15) was the most frequent primary pulmonary sarcoma with following synovial sarcoma (n = 9) and malignant type of solitary fibrous tumor (n = 9). This study is the first trial of epidemiologic investigation in Korea about NETs-LP, indirectly representing their incidence in Korea and northeast Asia, and we hope that this presentation offers guidance in further study of NETs-LP. © 2013 Elsevier GmbH. All rights reserved.
Introduction
Materials and methods
Primary and metastatic non-epithelial tumors of the lung and pleura (NETs-LP) are presumed to represent a small proportion of pulmonary neoplasms; however, a systematic epidemiologic characterization of NETs-LP is lacking whether benign or malignant [1,9,18,19]. A classification of pulmonary non-epithelial tumors, especially mesenchymal tumor, according to the ‘WHO Classification of Tumors of Lung and Pleura’ (2004) is simple due to their rare incidence. Moreover, it is necessary to give a supplementation to catch up ‘WHO Classification of Tumors of Soft Tissue and Bone’ (2013). Here, we present the numbers and proportions of primary and metastatic pulmonary non-epithelial tumors and tumor-like lesions that were diagnosed and treated at our institute during an 18-year period using the recent concepts of classification of tumors of soft tissue.
Tumors of the lung and pleura diagnosed by histological examinations, including biopsy or resection, in patients admitted from January 1995 to April 2013, were retrospectively reviewed. Records were retrieved by searching for keywords related to non-epithelial tumors, such as sarcoma, hemangioma, melanoma and hamartoma. Names suggestive of non-epithelial tumor, but actually denoting epithelial tumors (e.g., ‘sclerosing hemangioma’) were excluded. Conversely, names suggesting epithelial tumors but in fact representing non-epithelial ones (e.g. ‘myoepithelial carcinoma’) were included. Hematologic malignancies were also excluded, because these are usually confirmed by lymph nodal examinations, and the epidemiology of hematologic malignancies is well-characterized. The identification and classification of NETs-LP were based on the ‘WHO Classification of Tumors of Soft Tissue and Bone’ (2013) and ‘WHO Classification of Tumors of Lung and Pleura’ (2004). Records for 859 patients were extracted and reviewed; and after eliminating redundant data, records for 681 patients were retained for analysis in this study. The causes for redundancy included repeated histological examinations or operations, changes of diagnosis between the preliminary biopsy and surgical resection, and
∗ Corresponding author at: Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 lrwon-dong, Gangnam-gu, Seoul 135-710, Republic of Korea. Tel.: +82 2 3410 2800; fax: +82 2 3410 0025. E-mail addresses: [email protected], [email protected] (J. Han). 0344-0338/$ – see front matter © 2013 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.prp.2013.12.002
I.H. Choi et al. / Pathology – Research and Practice 210 (2014) 210–216
diagnostic changes made by ancillary tests such as immunohistochemical staining, molecular study and electron microscopy. The Institutional Review Board of Samsung Medical Center approved this study (IRB No. 2013-08-032-001).
Result Of 681 patients represented in the study, 355 were male and 326 were female, ranging in age from 2 to 85 years (mean 51.1 years). Except for 40 patients who received multiple differential diagnoses compared to a definitive diagnosis and 3 patients with unknown primary site, the patients having tumors of primary origin were 467, and those with metastatic ones were 171 (Table 1). Primary NETs-LP showed a relative male predominance (M:F = 1.2:1), whereas metastatic tumor occurred more frequently in females (M:F = 1:1.3). Patients with primary NETs-LP were older than those with metastatic tumors (mean 53- vs. 45-years). Smooth muscle tumor was the most common primary NETs-LP (12.6%); however, chondrogenic/osteogenic/osteoclastic giant cell rich tumor (21.6%) was common in metastatic ones. The distribution of NETs-LP is presented in Table 2 on reference to new concepts of ‘WHO Classification of Tumors of Soft Tissue and Bone’ (2013). Apart from tumors of miscellaneous origin (53.7%), smooth muscle tumors (13.1%) were most common in this group of NETs-LP, followed by tumors of uncertain differentiation (10.5%) and fibroblastic/myofibroblastic tumors (5.7%). The fibroblastic/myofibroblastic tumor, nerve sheath tumor, and vascular tumor presented with high proportions of primary tumor and tumor-like condition (94.9%, 81.8% and 78.6%, respectively). The category of adipocytic tumor included a 23-month-old patient with lipoblastoma who suffered no apparent adverse effects after the operation. The majority of adipocytic tumors were lipoma and liposarcoma (LPS). Most of the LPS were metastatic, except for one primary myxoid liposarcoma (12 cm, high grade) that originated in the intrathoracic cavity. This tumor led to death of the patient through recurrence 35 months after en bloc resection in spite of the subsequent chemotherapy. The category of fibroblastic/myofibroblastic tumor included 23 inflammatory myofibroblastic tumors (IMT), 12 inflammatory pseudotumors (IPT) and one case each of fibromatosis, fibrosarcoma (FS), fibromyxoid sarcoma and inflammatory FS. In spite of the current definition that IMT also includes IPT and inflammatory FS, we classified each tumor in accordance with the original diagnosis. The patients with IMT included 10 men and 13 women with a mean age of 38.3 years (range, 4–75 years). Two peak ages of occurrence were noted, in the fifth (n = 8) and second (n = 4) decades. Average tumor size was 2.9 cm (range, 0.9–6.0 cm). The IPT occurred more frequently in males (n = 8) than in females (n = 4). The age was mean 51.3 (12–71) years, and the tumor size was mean 5.1 (1.2–12.8) cm. Of 89 smooth muscle tumors, 6 were metastasizing leiomyoma and 24 were metastatic leiomyosarcoma (LMS). One patient, who had primary LMS at the carina, experienced multiple metastases on gingiva and skin in spite of the postoperative radiation therapy. Most of the vascular tumors were benign except for 2 metastatic angiosarcomas (AS). Although epithelioid hemangioendothelioma (EHE) is usually regarded as benign in its behavior, one EHE developed as a metastatic tumor from the liver. Most nerve sheath tumors were benign neoplasms, and schwannoma (SCH) was the most frequent type. One of the 4 malignant peripheral nerve sheath tumors (MPNST) in this group developed from a primary neurofibroma (NF) of lobar bronchus. One patient with metastatic MPNST previously had NF, and another 6-year-old female with a pulmonary malignant triton tumor had a past history of MPNST in the pelvic cavity. Of 2 granular cell tumors, one was malignant and had metastasized from the mediastinum.
211
The category of ‘tumor of uncertain differentiation’ included 3 myoepithelial neoplasms, 15 pulmonary vascular sarcomas (PVS), 23 synovial sarcomas (SS) and 7 Ewing sarcomas/primitive neuroectodermal tumors (EWS/PNET). Patients with PVS were more frequently male than female (M:F = 9:6), and the mean age at diagnosis was 49.1 years (range, 29–72 years). According to the involved vessels, 11 cases with pulmonary artery, 3 cases with pulmonary artery and vein and one case with pulmonary vein tumors were noted. Patients with SS were predominantly male (M:F = 15:8), and the mean age was 41.3 (17–70) years. Nine SS were primary and 14 were metastatic tumors. Patients with primary SS were older than patients with metastatic ones (mean 49.3 (38–70) versus 36.1 (17–54) years). The EWS/PNET occurred in 4 male and 3 female patients with 17.4 (5–27) years of mean age, and two of them were primary tumors. The category ‘tumors of miscellaneous origin’ included one each of the following primitive tumors: pulmonary pleuroblastoma, hepatoblastoma, neuroblastoma and nephroblastoma. Twenty-five patients had malignant melanoma, including one with primary MM arising in the trachea. The solitary fibrous tumor (SFT) showed a relatively equal distribution between genders (M:F = 19:17), with 53.3 (26–85) years of mean age and an average size of 7.1 (0.7–25) cm. The ratio of benign to malignant SFT was 26:10, and patients with malignant SFT were older (mean 57.9 years, 39–75) than those with benign SFT (mean 51.6 years, 26–85). In addition, tumors were grouped on the basis of their biologic behavior according to the International Classification of Diseases for Oncology (ICD-O) (Table 3). Hamartoma was the most frequent benign NETs-LP (n = 246, 72.1% of benign tumors), and 29 cases were resected at the same time with resection of other pulmonary or metastatic malignancies. Among the primary pulmonary sarcomas, the pulmonary vascular sarcoma (PVS, n = 15) was most common, followed by malignant solitary fibrous tumor, synovial sarcoma, and undifferentiated pleomorphic sarcoma (previously malignant fibrous histiocytoma). Malignant melanoma (n = 24) and leiomyosarcoma (n = 24) were the most common metastatic tumors.
Discussion NETs-LP are rare clinically, and the data concerning their incidence and behavior (benign, malignant, or tumor-like) are limited. To date, most studies of these tumors have presented small numbers of cases of primary pulmonary sarcoma or have reported clinical outcomes after treatment of metastatic pulmonary sarcoma. Moreover, mesenchymal tumors of the lung and pleura need to be reclassified, considering recent reclassification of the other soft tissue tumors. Pulmonary hamartoma is a relatively common benign neoplasm, and the majority of benign neoplasms in our study were of this type. The incidence in the general population is established as 0.25%, representing 3% of all tumors of the lung [5,14]. The incidence rates of other benign neoplasms, such as leiomyoma, schwannoma, lipoma and hemangioma, have not been determined to the best of our knowledge. About one-sixth (58/341) of the benign neoplasms in our study were leiomyoma (LM), and most of these tumors originated from the trachea and bronchus except for 4 that arose from pulmonary parenchyma. It seems to be similar to previous reports [20] in which LM represented about 2% of benign tumors of the lower respiratory tract. To confirm the diagnosis of pulmonary LM in women patients, the gynecologic and surgical records were reviewed to exclude metastasizing LM [3]. The largest investigation conducted in Korea to determine clinical outcome following resection of primary pulmonary LM (n = 16) was conducted at our hospital [16],
212
I.H. Choi et al. / Pathology – Research and Practice 210 (2014) 210–216
Table 1 Characteristics of primary or metastatic NETs-LPs. Origin
Primary
Metastatic
a
Sex (M:F)
254:213
76:98
Mean age (range)
Category of tumor
No. of cases
Total (n = 638)a
53.0 years (2–85)
Smooth muscle tumor Tumor of uncertain differentiation Fibroblastic/myofibroblastic tumor Nerve sheath tumor Vascular tumor Adipocytic tumor Undifferentiated/unclassified sarcoma Pericytic(perivascular) tumor Chondrogenic/osteogenic/osteoclastic giant cell rich tumors Miscellaneous
59 (12.6%) 40 (8.6%) 37 (7.9%) 18 (3.9%) 11 (2.4%) 10 (2.1%) 4 (0.9%) 2 (0.4%) 1 (0.2%) 285 (61.0%)
467 (100%)
45.0 years (5–83)
Chondrogenic/osteogenic/osteoclastic giant cell rich tumors Smooth muscle tumor Tumor of uncertain differentiation Undifferentiated/unclassified sarcoma Adipocytic tumor Nerve sheath tumor Skeletal muscle tumor Vascular tumor Fibroblastic/myofibroblastic tumor Miscellaneous
37 (21.6%) 30 (17.5%) 29 (17.0%) 18 (10.5%) 5 (2.9%) 4 (2.3%) 3 (1.8%) 3 (1.8%) 2 (1.2%) 40 (23.4%)
171 (100%)
Excludes 43 cases, which have no definite diagnosis but multiple differential diagnoses or have unknown origin.
and we also included these cases and additional biopsy-proven cases in this study. After the time of the previous report of our hospital [16], the incidence of pulmonary LM has increased by about 70%; 24 cases/recent 4.5 years versus 34 cases/past 11 years before the report. We attribute this increase to recent technical improvements in detection leading to surgery for removal of the tumor. The endobronchial lipoma was known to occur approximately twice as frequently as parenchymal lipoma, accounting for 0.1–13% of benign tumors of the lung [10]. In the years of our study, lipoma accounted for 1.8% of benign lung neoplasms, including 6 cases of endobronchial lipoma (5 bronchial and 1 tracheal) and 2 cases of parenchymal lipoma. Morita et al. [13] estimated the incidence of pulmonary sarcoma as one per 2600 in males and one per 3600 in females, based on 241 autopsy reports; the relative incidence of pulmonary sarcoma per all pulmonary neoplasms was one per 240 in males and one per 170 in females. These estimates may require re-evaluation, considering the limitations of autopsy data, significant advancements of technology in tumor detection and recent recognition of new disease entities prompting reclassification of sarcomas. In a study of 26 patients with primary pulmonary sarcoma (PPS) by Keel et al. [9], they reported 7 malignant fibrous histiocytomas (MFH), 6 synovial sarcomas (SS), 3 malignant peripheral nerve sheath tumors (MPNST), 3 leiomyosarcomas (LMS), 2 angiosarcomas (AS), 2 intimal sarcomas, 2 fibrosarcomas (FS) and one epithelioid hemangioendothelioma (EHE). In the evaluation of 48 patients with PPS, Petrov et al. [17,18] reported the following composition: 15 FS, 10 fibroleiomyosarcomas, 6 LMS, 6 rhabdomyosarcomas, 3 hemagiopericytomas, 3 EHE, 3 undifferentiated sarcomas, 1 malignant schwannoma and 1 liposarcoma. Among patients in our study, pulmonary vascular sarcoma (n = 15) was the most frequent primary sarcoma in the lung and pleura, representing 9 SS, 9 malignant type of SFT, 3 UPS and 2 EWS/PNET (Table 4). This discrepancy between studies in pulmonary sarcoma subtype composition may reflect changes over time and geographic region, as well as increased diagnostic accuracy based on reclassification of soft tissue tumors. For example, FS and LMS remain relatively common primary pulmonary sarcoma (PPS), but the incidence of SS was increased based on the development of testing for t(x:18) translocation. Moreover, many tumors of soft tissue have been reclassified. For example, many tumors of previous MFH, lymphangioleiomyomatosis and HPC are diagnosed as undifferentiated
pleomorphic sarcoma (UPS), perivascular epithelioid cell tumor (PEComa) and SFT, respectively. We also reclassified these tumors as UPS, PEComa and SFT in the study. Spraker et al. [19] selected data from the Surveillance, Epidemiology, and End Results (SEER) database for 365 patients with primary pulmonary sarcoma (PPS) and reported the following distribution by histologic classification: 131 unspecified soft tissue sarcomas (36%), 59 fibroblastic and myofibroblastic tumors (16%), 53 LMS (15%), 43 SS (12%), 21 blood vessel tumors (6%), 15 nerve sheath tumors (4%), 7 miscellaneous soft tissue sarcomas (2%), 2 liposarcomas (LPS) (
Contents lists available at ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.com/locate/prp
Original Article
Incidence of pulmonary non-epithelial tumors: 18 years’ experience at a single institute In Ho Choi a , Dae Hyun Song a , Kang Min Han a , Yong Soo Choi b , Joungho Han a,∗ a b
Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea Department of Thoracic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
a r t i c l e
i n f o
Article history: Received 20 September 2013 Received in revised form 27 October 2013 Accepted 2 December 2013 Keywords: Lung Sarcoma Mesenchymal tumor Pulmonary sarcoma Incidence
a b s t r a c t Primary and metastatic non-epithelial tumors of the lung and pleura (NETs-LP) are presumed to represent a small proportion of pulmonary neoplasm, and their epidemiologic characterization is lacking. Moreover, classification of pulmonary mesenchymal tumor according to ‘WHO Classification of Tumors of Lung and Pleura’ (2004) is simple and necessary to give a supplementation like new ‘WHO Classification of Tumors of Soft Tissue and Bone’ (2013). Non-epithelial tumors of the lung and pleura diagnosed from January 1995 to April 2013 were retrospectively reviewed and classified in accordance with ‘WHO Classification of Tumors of Soft Tissue and Bone’ (2013). Of 681 cases represented in the study, smooth muscle tumor (13.1%) was most common, and tumor of uncertain differentiation (10.5%) and fibroblastic/myofibroblastic tumor (5.7%) were followed. Smooth muscle tumor (12.6%) was the common primary tumor, whereas chondrogenic or osteogenic tumor (21.6%) was common in metastatic ones. Pulmonary vascular sarcoma (n = 15) was the most frequent primary pulmonary sarcoma with following synovial sarcoma (n = 9) and malignant type of solitary fibrous tumor (n = 9). This study is the first trial of epidemiologic investigation in Korea about NETs-LP, indirectly representing their incidence in Korea and northeast Asia, and we hope that this presentation offers guidance in further study of NETs-LP. © 2013 Elsevier GmbH. All rights reserved.
Introduction
Materials and methods
Primary and metastatic non-epithelial tumors of the lung and pleura (NETs-LP) are presumed to represent a small proportion of pulmonary neoplasms; however, a systematic epidemiologic characterization of NETs-LP is lacking whether benign or malignant [1,9,18,19]. A classification of pulmonary non-epithelial tumors, especially mesenchymal tumor, according to the ‘WHO Classification of Tumors of Lung and Pleura’ (2004) is simple due to their rare incidence. Moreover, it is necessary to give a supplementation to catch up ‘WHO Classification of Tumors of Soft Tissue and Bone’ (2013). Here, we present the numbers and proportions of primary and metastatic pulmonary non-epithelial tumors and tumor-like lesions that were diagnosed and treated at our institute during an 18-year period using the recent concepts of classification of tumors of soft tissue.
Tumors of the lung and pleura diagnosed by histological examinations, including biopsy or resection, in patients admitted from January 1995 to April 2013, were retrospectively reviewed. Records were retrieved by searching for keywords related to non-epithelial tumors, such as sarcoma, hemangioma, melanoma and hamartoma. Names suggestive of non-epithelial tumor, but actually denoting epithelial tumors (e.g., ‘sclerosing hemangioma’) were excluded. Conversely, names suggesting epithelial tumors but in fact representing non-epithelial ones (e.g. ‘myoepithelial carcinoma’) were included. Hematologic malignancies were also excluded, because these are usually confirmed by lymph nodal examinations, and the epidemiology of hematologic malignancies is well-characterized. The identification and classification of NETs-LP were based on the ‘WHO Classification of Tumors of Soft Tissue and Bone’ (2013) and ‘WHO Classification of Tumors of Lung and Pleura’ (2004). Records for 859 patients were extracted and reviewed; and after eliminating redundant data, records for 681 patients were retained for analysis in this study. The causes for redundancy included repeated histological examinations or operations, changes of diagnosis between the preliminary biopsy and surgical resection, and
∗ Corresponding author at: Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 lrwon-dong, Gangnam-gu, Seoul 135-710, Republic of Korea. Tel.: +82 2 3410 2800; fax: +82 2 3410 0025. E-mail addresses: [email protected], [email protected] (J. Han). 0344-0338/$ – see front matter © 2013 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.prp.2013.12.002
I.H. Choi et al. / Pathology – Research and Practice 210 (2014) 210–216
diagnostic changes made by ancillary tests such as immunohistochemical staining, molecular study and electron microscopy. The Institutional Review Board of Samsung Medical Center approved this study (IRB No. 2013-08-032-001).
Result Of 681 patients represented in the study, 355 were male and 326 were female, ranging in age from 2 to 85 years (mean 51.1 years). Except for 40 patients who received multiple differential diagnoses compared to a definitive diagnosis and 3 patients with unknown primary site, the patients having tumors of primary origin were 467, and those with metastatic ones were 171 (Table 1). Primary NETs-LP showed a relative male predominance (M:F = 1.2:1), whereas metastatic tumor occurred more frequently in females (M:F = 1:1.3). Patients with primary NETs-LP were older than those with metastatic tumors (mean 53- vs. 45-years). Smooth muscle tumor was the most common primary NETs-LP (12.6%); however, chondrogenic/osteogenic/osteoclastic giant cell rich tumor (21.6%) was common in metastatic ones. The distribution of NETs-LP is presented in Table 2 on reference to new concepts of ‘WHO Classification of Tumors of Soft Tissue and Bone’ (2013). Apart from tumors of miscellaneous origin (53.7%), smooth muscle tumors (13.1%) were most common in this group of NETs-LP, followed by tumors of uncertain differentiation (10.5%) and fibroblastic/myofibroblastic tumors (5.7%). The fibroblastic/myofibroblastic tumor, nerve sheath tumor, and vascular tumor presented with high proportions of primary tumor and tumor-like condition (94.9%, 81.8% and 78.6%, respectively). The category of adipocytic tumor included a 23-month-old patient with lipoblastoma who suffered no apparent adverse effects after the operation. The majority of adipocytic tumors were lipoma and liposarcoma (LPS). Most of the LPS were metastatic, except for one primary myxoid liposarcoma (12 cm, high grade) that originated in the intrathoracic cavity. This tumor led to death of the patient through recurrence 35 months after en bloc resection in spite of the subsequent chemotherapy. The category of fibroblastic/myofibroblastic tumor included 23 inflammatory myofibroblastic tumors (IMT), 12 inflammatory pseudotumors (IPT) and one case each of fibromatosis, fibrosarcoma (FS), fibromyxoid sarcoma and inflammatory FS. In spite of the current definition that IMT also includes IPT and inflammatory FS, we classified each tumor in accordance with the original diagnosis. The patients with IMT included 10 men and 13 women with a mean age of 38.3 years (range, 4–75 years). Two peak ages of occurrence were noted, in the fifth (n = 8) and second (n = 4) decades. Average tumor size was 2.9 cm (range, 0.9–6.0 cm). The IPT occurred more frequently in males (n = 8) than in females (n = 4). The age was mean 51.3 (12–71) years, and the tumor size was mean 5.1 (1.2–12.8) cm. Of 89 smooth muscle tumors, 6 were metastasizing leiomyoma and 24 were metastatic leiomyosarcoma (LMS). One patient, who had primary LMS at the carina, experienced multiple metastases on gingiva and skin in spite of the postoperative radiation therapy. Most of the vascular tumors were benign except for 2 metastatic angiosarcomas (AS). Although epithelioid hemangioendothelioma (EHE) is usually regarded as benign in its behavior, one EHE developed as a metastatic tumor from the liver. Most nerve sheath tumors were benign neoplasms, and schwannoma (SCH) was the most frequent type. One of the 4 malignant peripheral nerve sheath tumors (MPNST) in this group developed from a primary neurofibroma (NF) of lobar bronchus. One patient with metastatic MPNST previously had NF, and another 6-year-old female with a pulmonary malignant triton tumor had a past history of MPNST in the pelvic cavity. Of 2 granular cell tumors, one was malignant and had metastasized from the mediastinum.
211
The category of ‘tumor of uncertain differentiation’ included 3 myoepithelial neoplasms, 15 pulmonary vascular sarcomas (PVS), 23 synovial sarcomas (SS) and 7 Ewing sarcomas/primitive neuroectodermal tumors (EWS/PNET). Patients with PVS were more frequently male than female (M:F = 9:6), and the mean age at diagnosis was 49.1 years (range, 29–72 years). According to the involved vessels, 11 cases with pulmonary artery, 3 cases with pulmonary artery and vein and one case with pulmonary vein tumors were noted. Patients with SS were predominantly male (M:F = 15:8), and the mean age was 41.3 (17–70) years. Nine SS were primary and 14 were metastatic tumors. Patients with primary SS were older than patients with metastatic ones (mean 49.3 (38–70) versus 36.1 (17–54) years). The EWS/PNET occurred in 4 male and 3 female patients with 17.4 (5–27) years of mean age, and two of them were primary tumors. The category ‘tumors of miscellaneous origin’ included one each of the following primitive tumors: pulmonary pleuroblastoma, hepatoblastoma, neuroblastoma and nephroblastoma. Twenty-five patients had malignant melanoma, including one with primary MM arising in the trachea. The solitary fibrous tumor (SFT) showed a relatively equal distribution between genders (M:F = 19:17), with 53.3 (26–85) years of mean age and an average size of 7.1 (0.7–25) cm. The ratio of benign to malignant SFT was 26:10, and patients with malignant SFT were older (mean 57.9 years, 39–75) than those with benign SFT (mean 51.6 years, 26–85). In addition, tumors were grouped on the basis of their biologic behavior according to the International Classification of Diseases for Oncology (ICD-O) (Table 3). Hamartoma was the most frequent benign NETs-LP (n = 246, 72.1% of benign tumors), and 29 cases were resected at the same time with resection of other pulmonary or metastatic malignancies. Among the primary pulmonary sarcomas, the pulmonary vascular sarcoma (PVS, n = 15) was most common, followed by malignant solitary fibrous tumor, synovial sarcoma, and undifferentiated pleomorphic sarcoma (previously malignant fibrous histiocytoma). Malignant melanoma (n = 24) and leiomyosarcoma (n = 24) were the most common metastatic tumors.
Discussion NETs-LP are rare clinically, and the data concerning their incidence and behavior (benign, malignant, or tumor-like) are limited. To date, most studies of these tumors have presented small numbers of cases of primary pulmonary sarcoma or have reported clinical outcomes after treatment of metastatic pulmonary sarcoma. Moreover, mesenchymal tumors of the lung and pleura need to be reclassified, considering recent reclassification of the other soft tissue tumors. Pulmonary hamartoma is a relatively common benign neoplasm, and the majority of benign neoplasms in our study were of this type. The incidence in the general population is established as 0.25%, representing 3% of all tumors of the lung [5,14]. The incidence rates of other benign neoplasms, such as leiomyoma, schwannoma, lipoma and hemangioma, have not been determined to the best of our knowledge. About one-sixth (58/341) of the benign neoplasms in our study were leiomyoma (LM), and most of these tumors originated from the trachea and bronchus except for 4 that arose from pulmonary parenchyma. It seems to be similar to previous reports [20] in which LM represented about 2% of benign tumors of the lower respiratory tract. To confirm the diagnosis of pulmonary LM in women patients, the gynecologic and surgical records were reviewed to exclude metastasizing LM [3]. The largest investigation conducted in Korea to determine clinical outcome following resection of primary pulmonary LM (n = 16) was conducted at our hospital [16],
212
I.H. Choi et al. / Pathology – Research and Practice 210 (2014) 210–216
Table 1 Characteristics of primary or metastatic NETs-LPs. Origin
Primary
Metastatic
a
Sex (M:F)
254:213
76:98
Mean age (range)
Category of tumor
No. of cases
Total (n = 638)a
53.0 years (2–85)
Smooth muscle tumor Tumor of uncertain differentiation Fibroblastic/myofibroblastic tumor Nerve sheath tumor Vascular tumor Adipocytic tumor Undifferentiated/unclassified sarcoma Pericytic(perivascular) tumor Chondrogenic/osteogenic/osteoclastic giant cell rich tumors Miscellaneous
59 (12.6%) 40 (8.6%) 37 (7.9%) 18 (3.9%) 11 (2.4%) 10 (2.1%) 4 (0.9%) 2 (0.4%) 1 (0.2%) 285 (61.0%)
467 (100%)
45.0 years (5–83)
Chondrogenic/osteogenic/osteoclastic giant cell rich tumors Smooth muscle tumor Tumor of uncertain differentiation Undifferentiated/unclassified sarcoma Adipocytic tumor Nerve sheath tumor Skeletal muscle tumor Vascular tumor Fibroblastic/myofibroblastic tumor Miscellaneous
37 (21.6%) 30 (17.5%) 29 (17.0%) 18 (10.5%) 5 (2.9%) 4 (2.3%) 3 (1.8%) 3 (1.8%) 2 (1.2%) 40 (23.4%)
171 (100%)
Excludes 43 cases, which have no definite diagnosis but multiple differential diagnoses or have unknown origin.
and we also included these cases and additional biopsy-proven cases in this study. After the time of the previous report of our hospital [16], the incidence of pulmonary LM has increased by about 70%; 24 cases/recent 4.5 years versus 34 cases/past 11 years before the report. We attribute this increase to recent technical improvements in detection leading to surgery for removal of the tumor. The endobronchial lipoma was known to occur approximately twice as frequently as parenchymal lipoma, accounting for 0.1–13% of benign tumors of the lung [10]. In the years of our study, lipoma accounted for 1.8% of benign lung neoplasms, including 6 cases of endobronchial lipoma (5 bronchial and 1 tracheal) and 2 cases of parenchymal lipoma. Morita et al. [13] estimated the incidence of pulmonary sarcoma as one per 2600 in males and one per 3600 in females, based on 241 autopsy reports; the relative incidence of pulmonary sarcoma per all pulmonary neoplasms was one per 240 in males and one per 170 in females. These estimates may require re-evaluation, considering the limitations of autopsy data, significant advancements of technology in tumor detection and recent recognition of new disease entities prompting reclassification of sarcomas. In a study of 26 patients with primary pulmonary sarcoma (PPS) by Keel et al. [9], they reported 7 malignant fibrous histiocytomas (MFH), 6 synovial sarcomas (SS), 3 malignant peripheral nerve sheath tumors (MPNST), 3 leiomyosarcomas (LMS), 2 angiosarcomas (AS), 2 intimal sarcomas, 2 fibrosarcomas (FS) and one epithelioid hemangioendothelioma (EHE). In the evaluation of 48 patients with PPS, Petrov et al. [17,18] reported the following composition: 15 FS, 10 fibroleiomyosarcomas, 6 LMS, 6 rhabdomyosarcomas, 3 hemagiopericytomas, 3 EHE, 3 undifferentiated sarcomas, 1 malignant schwannoma and 1 liposarcoma. Among patients in our study, pulmonary vascular sarcoma (n = 15) was the most frequent primary sarcoma in the lung and pleura, representing 9 SS, 9 malignant type of SFT, 3 UPS and 2 EWS/PNET (Table 4). This discrepancy between studies in pulmonary sarcoma subtype composition may reflect changes over time and geographic region, as well as increased diagnostic accuracy based on reclassification of soft tissue tumors. For example, FS and LMS remain relatively common primary pulmonary sarcoma (PPS), but the incidence of SS was increased based on the development of testing for t(x:18) translocation. Moreover, many tumors of soft tissue have been reclassified. For example, many tumors of previous MFH, lymphangioleiomyomatosis and HPC are diagnosed as undifferentiated
pleomorphic sarcoma (UPS), perivascular epithelioid cell tumor (PEComa) and SFT, respectively. We also reclassified these tumors as UPS, PEComa and SFT in the study. Spraker et al. [19] selected data from the Surveillance, Epidemiology, and End Results (SEER) database for 365 patients with primary pulmonary sarcoma (PPS) and reported the following distribution by histologic classification: 131 unspecified soft tissue sarcomas (36%), 59 fibroblastic and myofibroblastic tumors (16%), 53 LMS (15%), 43 SS (12%), 21 blood vessel tumors (6%), 15 nerve sheath tumors (4%), 7 miscellaneous soft tissue sarcomas (2%), 2 liposarcomas (LPS) (
