Skip Navigation LinksHome > March 2013 - Volume 20 - Issue 2 > Hemangioblastoma-like Clear Cell Stromal Tumor of the Lung
Advances in Anatomic Pathology:
doi: 10.1097/PAP.0b013e318286245d
AMR Series

Hemangioblastoma-like Clear Cell Stromal Tumor of the Lung

Falconieri, Giovanni MD*; Mirra, Maurizio MD; Michal, Michal MD; Suster, Saul MD§

Free Access
Article Outline
Collapse Box

Author Information

*Department of Anatomic Pathology, General University Hospital of Udine, Udine

Department of Anatomic Pathology, General Hospital of Lodi, Lodi, Italy

Department of Pathology, Charles University, Medical Faculty of Pilsen, Pilsen, Czech Republic

§Department of Pathology, Medical College of Wisconsin, Milwaukee, WI

The authors have no funding or conflicts of interest to disclose.

Reprints: Giovanni Falconieri, MD, Department of Pathology, General University Hospital, I 33100 Udine, Italy (e-mail:

Collapse Box


The authors report 2 cases of an apparently unpublished stromal tumor of the lung characterized by a predominantly endobronchial growth pattern and benign-appearing clear cells. Both tumors were discovered incidentally in adult patients during routine workups for other medical reasons and treated with lobectomy. On gross inspection there was no evidence of infiltration of the adjacent lung tissue. Microscopically, both lesions featured monotonous oval-shaped to spindle-shaped cells growing in a vaguely nested pattern. The cytoplasm was slightly vacuolated or granular. In 1 case there was a variable admixture with mature fat. Immunohistochemistry was negative for markers of epithelial and stromal differentiation except for vimentin. A focal reaction for CD34 was seen in 1 case. No mutation of coding sequence of VHL gene was seen in one case. Medical follow-up at 1 year was negative for tumor recurrence or metastases. The broad differential diagnosis within the spectrum of stromal lung tumor is discussed. Owing to distinctive microscopic features such as the nesting of clear cells within a vascularized background, both tumors appeared similar to hemangioblastoma, although the expected immunohistochemical profile of the latter was not fully expressed. Because of pattern of growth seen in both lesions we believe that the appellation of endobronchial, hemangioblastoma-like clear cell stromal tumor may be provisionally designed.

Back to Top | Article Outline


Case 1 (#AMR60-11)

The patient is a 68-year-old woman with history of mild fever and influenza-like symptoms, who was found to have a pulmonary opacity on a CT scan. The remainder of her medical workup and past medical history was unremarkable. Right upper lobectomy was carried out. The specimen featured a well-circumscribed, expansile, partially endobronchial mass 6×5×4.5 cm in size that was yellow and gray on the cut surface (Fig. 1A). Histologic examination of H&E-stained sections revealed a cellular tumor composed of sheets of polygonal cells in close association with a large bronchus (Fig. 1B). Tumor cells were usually comparable in size, or slightly smaller than trapped alveolar cells (Fig. 1C). Cell membrane was indistinct and the cytoplasm was clear or moderately granular. Cytoplasmic vacuolization pushing or indenting the nuclei at the cell periphery was often prominent. Cell nuclei were often oval to spindle, slightly hyperchromatic, and nucleoli were most often not discernible (Fig. 1D). Mitotic figures were not recognized. A panoramic view of the peripheral part of the lesion revealed a polypoid mass with a leaf-like configuration (Fig. 2A). Scattered adipocytes could be also recognized and focal areas with preponderance of adipose tissue, sometimes with liponecrotic changes, were present (Fig. 2B). A vascular, pericytoid pattern could be occasionally recognized (Fig. 2C). Other microscopic features noted on scanning magnification were also patchy collections of lymphoid cells (Fig. 2D). An extensive panel of antibodies was applied to paraffin sections of the tumor. With the exception of vimentin, no reactivity was observed in the tumor cells for epithelial markers (including cytokeratin cocktails and EMA), neuroendocrine markers (chromogranin, synaptophysin, and CD56), TTF1, CD31, CD34, S100 protein, HMB45, Melan A, α-inhibin, CD45, CD23, CD35, GFAP, desmin, and actins. The tumor cells were PAS-negative. No further therapy was administered and the patient is free of tumor recurrence 2 years after lobectomy.

Figure 1
Figure 1
Image Tools
Figure 2
Figure 2
Image Tools
Back to Top | Article Outline
Case 2 (#2434/10)

The patient is a 40-year-old man with a tumor of the left lower lobe of the lung. Grossly most of the lesion was lodged within the lumen of a branch of the main bronchus. Lobectomy was carried out. The tumor measured 3 cm in largest dimension. Histologically, on scanning magnification it featured a round slightly lobulated contour (Fig. 3A) with a rather monotonous appearance, displaying mainly nests of large epithelioid cells: as for case 1, tumor cells were often arranged in small nests (Fig. 3B); nesting confluence could also be recognized. The nuclei were hyperchromatic, small to ovoid, with inconspicuous nucleoli. The cytoplasm was granular or vacuolated, with some vacuoles indenting the nuclei and imparting a vague lipoblastoid appearance (Fig. 3C). Focal fibrotic, less cellular areas could also be recognized (Fig. 3D). Tumor cells were positive for vimentin, as in case 1. Complete absence of immunohistochemical reactivity for epithelial (cytokeratins and EMA), muscle and neuroendocrine markers, and melanocytic markers (S100 protein, HMB45, and Melan A) was documented and focal positivity for CD34 was observed in clusters of neoplastic cells. Two years after lobectomy the patient is without evidence of metastatic disease or tumor recurrence. No mutation of coding sequence of VHL gene was found.

Figure 3
Figure 3
Image Tools
Back to Top | Article Outline


* Great case. Some areas of this tumor reminded me of the mesenchymal hamartoma of the liver. The immature but bland appearance of the tumor cells, distinctive capillarization, clusters of thick-walled vessels, myxoid appearance of stroma with vacuolization, as well as cystic degeneration are all also seen in mesenchymal hamartoma of the liver. I wonder if this case might represent pulmonary counterpart of that phenomenon.

* What about a CD34-negative intrapulmonary lipomatous hemangiopericytoma (solitary fibrous tumor)?

* Beautiful case.

* By default PEComa seems to be the best fit. I presume a metastasis has been excluded.

* Stromal tumor-NOS or clear cell tumor NOS. I definitely cannot be of any help or advice. Sorry. I suppose this tumor was positive at least for vimentin. Why not to try EM analysis on paraffin-embedded tissue?

* No idea.

* Don’t know what this is. I would truncate Saul’s diagnosis of “clear cell tumor of unknown etiology” to simply “tumor of unknown etiology.” I doubt that this is malignant and the funny little spindle cells have a resemblance to some of the cells seen in hamartomas and to the (usually focal) spindle cells seen in alveolar adenomas. Sometimes the cells in hamartomas will show positivity with GFAP but I certainly wouldn’t use that as any sort of specific marker.

* I have no better ideas!

* Spindle cell lipoma or a low-grade mesenchymal malignancy.

* I believe that clear cell tumor is the best diagnosis. To my opinion HMB-45 staining is less often positive in PEComas than we expect it.

* Very strange tumor, I believe most cells have a histiocytic or macrophage phenotype, my problem is how to name it.

* Difficult case. I find it difficult to accept as clear cell tumor of the lung (PECOMA) without supportive IHC evidence. I note dominating stromal differentiation including fat but I also see some epithelial ductular component in the peripheral region under the attenuated bronchiolar mucosa. The question is whether these glands are part of the tumor moving us in the direction of biphasic pulmonary tumors or salivary gland type or if there is a subtle infiltration going on by the stromal tumoral cells. I guess I am in favor of a biphasic benign tumor. Could this be an uncommon variant of pleomorphic adenoma? I would try more immunohistochemistry focused on sensitive markers of myoepithelial cells and try to characterize the ductular epithelial component. SGT’s of bronchial mucosa can be TTF1+! Any history of tumor in other organs ?

* I also think that the lesion would be a stromal tumor benign. No better diagnosis.

* Stromal tumor-NOS or clear cell tumor NOS. Very difficult case.

* I regret that I do not recognize this lesion at all—benign spindle cell neoplasm would be the best that I can do.

* I’d be concerned about a metastatic sarcoma, perhaps one resected some time ago, and would work on getting better clinical history.

* It doesn’t occur to me what could provide additional information. In case to have more material, ultrastructural study could be interesting.

* I think clear sugar tumor most likely although all immunostains were negative.

* On H&E, I would favor solitary fibrous tumor with adipocytic differentiation; you might repeat CD34 on some other blocks. I also thought about juxtaglomerular cell tumor but in view of the presence of adipocytes, it is unlikely.

* Spindle and epithelioid clear cell tumor, benign. This is not very helpful but that far I can go.

* For me the lesion looks like an unusual stromal tumor with some mature adipocytic cells and scattered inflammatory cells. A PEComa-like lesion is not present in my opinion.

* Unclassified, gave some consideration into an intrapulmonary solitary fibrous tumor variant.

* I have no idea what this is but it looks benign.

* I do not know either. My first impression was a benign mesenchymal tumor and I was wondering about some subtypes of hamartomatous tumors. It would be relevant to know the follow-up.

* I don’t know what this lung lesion is.

* No idea. The best we could come up with was PEComa versus inflammatory myofibroblastic tumor.

* I too wondered about clear cell sugar tumor but it is does not seem to be that. I have no idea what it is. Histiocytic/dendritic markers all negative?

* Hemangioblastoma of the lung?

* I still don’t know what this is. In looking at this a second time, it does look rather similar to case no. 5 (note of the authors: hemangioblastoma of subcutaneous soft tissues) in the same seminar.

* No clue.

Back to Top | Article Outline


These 2 cases represent bronchial tumors featuring a polypoid, endobronchial growth and composed of epithelioid cells with granular to clear/vacuolated cells growing either in a diffuse sheet-like pattern or sometimes arranged in small nests. Lipoblastoid cells were also seen. In case 1, mature adipocytes either in small or in larger clusters were present or were sometimes preponderant, and were likely responsible for the prevalent yellowish discoloration of tumor cut surfaces. In addition, some remnants of epithelial elements, likely preexisting glands or alveolar structures could be recognized focally which were focally reactive for TTF1. Tumor cell nuclei were bland, oval-shaped to spindle-shaped, or indented by cytoplasmic droplets. Although the overall features indicate that these are benign tumors of stromal origin, neither the histologic nor the immunohistochemical features are consonant with any known entities. In fact, as reflected by the comments circulated by the club members, there is substantial agreement on what case 1 is not, but no consensus on any category to which these tumors could be assigned. A number of “look-alikes” have been included in the differential diagnosis, including solitary fibrous tumor, mesenchymal hamartoma, PEComa, spindle cell lipoma, or even a low-grade mesenchymal malignancy or a reactive proliferation. A subsequent similar case with comparable histologic and immunohistochemical features was then identified from the personal files of one of the club members (Michal Michal, M.D.). The latter case shared the architectural and cytologic features of the original tumor but lacked mature adipocytes. In addition there were focally pronounced fibrotic changes, perhaps suggesting scarring or regression.

Although some microscopic features of these tumors were reminiscent of well-known microscopic entities that can occur in the lung,1,2 as pointed out by the club members, the microscopic appearance and immunophenotype in case 1 were inconsistent with any of these lesions. The same considerations may be applied to case 2. The tumor cells lacked phenotypical evidence of any given stromal cell line, and their immunoreactivity was basically negative with the exception of vimentin. Bronchopulmonary stromal tumors may sometimes exhibit an epithelioid morphology, but they are expected to retain either morphologic and/or immunohistochemical clues of their origin. Expanding further the spectrum of differential diagnosis, we also considered paraganglioma, which may notably occur in the lung.3 However, we could not recognize the fairly organoid pattern observed in those tumors, which typically features nests of epithelioid cells with clear to granular cytoplasm. In contrast to paraganglioma, in addition, we could not identify any population of sustentacular cells in our cases, and the tumor cells lack any immunoreactivity for neuroendocrine markers. Solitary fibrous tumor is a ubiquitous neoplasm that can rarely occur in the lung4; adipocytic differentiation may be observed in some cases. Solitary fibrous tumor shows a variegated proliferation of spindle cells enmeshed in a fibrous, often sclerotic matrix, and features short spindle cells that are often arranged in a cart-wheel or hemangiopericytic pattern. The overall microscopic features (epithelioid rather than spindle) along with the lack of reactivity for any of the markers that are usually expressed in such tumors including bcl2 and CD99, argued against that interpretation, although it may argued that a pericytic pattern reminiscent of that occasionally seen in cellular solitary fibrous tumor was present in case 1, and case 2 was focally positive for CD34. Clear cell “sugar” tumor was another important consideration. According to the classic description, these rare neoplasms of uncertain histogenesis are of small size, occur in the lung proper, and feature epithelioid cells with clear or eosinophilic cytoplasm.5,6 In contrast to our cases, however, sugar tumors show strong positivity for PAS owing to their abundant glycogen content. Furthermore, we could not observe any reactivity of the tumor cells for HMB45 or actin, which is notably expressed by clear cell tumors as part of the spectrum of PEComa/sugar cell tumor. Metastasis from a keratin-negative low-grade clear cell carcinoma of the kidney could be reasonably excluded on a clinical basis. Finally, tumors with chondroid differentiation were also considered but they usually have a more abundant, lightly basophilic ground substance and lack the striking organoid architecture that was consistently seen in our 2 cases.

The nesting organization with a fine capillary network and the watery perinuclear halo frequently observed in tumor cells, often attaining the size of lipid vacuoles or imparting a lipoblastoid appearance, also raised the possibility of a hemangioblastoma, as suggested by 1 member of the club. Hemangioblastomas are often negative for most immunohistochemical markers, although an erratic positivity for GFAP may be seen. Hemangioblastoma is often positive for S100 protein and α-inhibin in addition to vimentin. Although this tumor has a propensity to occur within the central or less commonly the peripheral nervous system, sporadic nonfamilial cases have been reported in the soft tissues, kidney, and skin outside the setting of von Hippel-Lindau syndrome.7–10 Outside the central nervous system, these tumors may go unrecognized because of their close similarity to paraganglioma, angiomyolipoma, or an adrenocortical neoplasm. Yet we are not aware of description of any hemangioblastoma occurring within the lung or bronchial tree in the literature. The cases at issue are in several microscopic respects comparable to hemangioblastoma, although they do not show any reactivity for the markers expected in this condition. Furthermore, molecular genetic testing carried on one of our cases showed no mutation of the coding sequence of VHL gene. Yet of this morphologic similarity, for the time being and until a better understanding of such cases is attained, we propose the terminology of “hemangioblastoma-like clear cell stromal tumor of the lung,” which to the best of our knowledge has not been described in the literature.

In conclusion, we have reported 2 unusual cases of pulmonary stromal tumors exhibiting distinctive gross and microscopic features including nested, clear tumor cells growing within a highly vascularized background—tumors phenotypically comparable to hemangioblastoma. The growth pattern in both was predominantly within the lumen of a major bronchus. Although some immunophenotypic differences exist, there are, in our opinion, enough similarities to allow a provisional designation of hemangioblastoma-like stromal tumor, until more cases are compiled and insights on their possible histogenesis can be obtained.

Back to Top | Article Outline


1. Colby TV, Koss MN, Travis WD Tumours of the Lower Respiratory Tract. 1995 Washington Armed Forces Institute of Pathology

2. Travis WDWorld Health Organization, International Agency for Research on Cancer, International Association for the Study of Lung Cancer, International Academy of Pathology. Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. 2004 Lyon Oxford IARC Press:344

3. Singh G, Lee RE, Brooks DH. Primary pulmonary paraganglioma: report of a case and review of the literature. Cancer. 1977;40:2286–2289

4. Rao N, Colby TV, Falconieri G, et al. Intrapulmonary solitary fibrous tumors: clinicopathologic and immunohistochemical study of 24 cases. Am J Surg Pathol. 2013;37:155–166

5. Gal AA, Koss MN, Hochholzer L, et al. An immunohistochemical study of benign clear cell (“sugar”) tumor of the lung. Arch Pathol Lab Med. 1991;115:1034–1038

6. Gaffey MJ, Mills SE, Askin FB, et al. Clear cell tumor of the lung. A clinicopathologic, immunohistochemical, and ultrastructural study of eight cases. Am J Surg Pathol. 1990;14:248–259

7. Michal M, Vanecek T, Sima R, et al. Primary capillary hemangioblastoma of peripheral soft tissues. Am J Surg Pathol. 2004;28:962–966

8. Nonaka D, Rodriguez J, Rosai J. Extraneural hemangioblastoma: a report of 5 cases. Am J Surg Pathol. 2007;31:1545–1551

9. Patton KT, Satcher RL Jr, Laskin WB. Capillary hemangioblastoma of soft tissue: report of a case and review of the literature. Hum Pathol. 2005;36:1135–1139

10. Fanburg-Smith JC, Gyure KA, Michal M, et al. Retroperitoneal peripheral hemangioblastoma: a case report and review of the literature. Ann Diagn Pathol. 2000;4:81–87


lung tumors; hemangioblastoma; immunohistochemistry

© 2013 Lippincott Williams & Wilkins, Inc.


Article Tools



Search for Similar Articles
You may search for similar articles that contain these same keywords or you may modify the keyword list to augment your search.