Placental transmogrification of the lung is a histologic pattern frequently associated with pulmonary fibrochondromatous hamartoma. (Original Articles).
Pulmonary hamartomas are common benign tumors considered to originate from primitive mesenchymal tissue. (10) Most pulmonary hamartomas are fibrochondromatous type consisting of fibroadipose tissue and cartilage with varying degrees of epithelial invagination. (10) Our hypothesis is that the epithelial invagination, if prominent, may account for the resemblance to immature placental villi, as seen in PT. In this report, we identified 6 cases of pulmonary hamartomas associated with characteristic PT. We further investigated the roles of proliferation of epithelium and a possible stromal stem cell in induction of pulmonary PT by immunostaining for proliferating index (Ki-67) and thyroid transcription factor 1 (TTF-1) (11) and c-Kit antigen (a stem cell factor receptor/mast cell growth factor receptor). (12,13)
MATERIALS AND METHODS
A total of 38 cases of pulmonary hamartomas diagnosed at the Department of Pathology, New York University Medical Center, from 1982 to 1999 were retrieved. The patients' age and sex, the site and number of hamartomas, and concurrence with other lesions were recorded.
Hematoxylin-eosin-stained slides of all 38 cases of pulmonary hamartoma were reviewed for the presence of villus papillary projections or placenta-like structures. The hamartomas with prominent papillary projections or placenta-like structure (as defined in previous publications (1-9)) were further characterized morphologically and immunohistochemically (Table 1).
Immunohistochemical analysis was performed on tissue sections prepared from formalin-fixed, paraffin-embedded tissues after heat-induced antigen retrieval according to the manufacturers' protocols. Mouse monoclonal antibodies against TTF-1 (Ab-1, dilution 1:100, Neomarkers, Union City, Calif) and Ki-67 (MIB-1, dilution 1:20, Zymed Laboratories, Inc, South San Francisco, Calif) and rabbit polyclonal antibodies against human and mouse c-Kit (dilution 1:100, Oncogene Science, Cambridge, Mass) were used as primary antibodies. Immunoreaction on slides was detected by the avidin-biotin complex and peroxidase method using a standard protocol on an automated immunostainer (Ventana Medical Systems, Tucson, Ariz), according to the manufacturer's instructions. A duplicate tissue section was also stained for naphthol-ASD-chloroacetate esterase by Leder stain. This was done in conjunction with c-Kit immunohistochemical study to detect mast cells.
Thirty-eight cases of pulmonary hamartoma resected by wedge resection or lobectomy were identified from departmental records. Most hamartomas were fibrochondromatous type (94%). Six (16%) were found to have a well-developed pattern of PT (Figure 1 and Table 1). In addition, another 14 cases (37%) showed some features of placental villuslike structures, but were insufficient to qualify as PT.
[FIGURE 1 OMITTED]
Among patients with PT associated with hamartoma, clinical presentation was similar to those without this change. Most of them were asymptomatic and found incidentally by radiography. Some of them had mild respiratory symptoms. One of the patients, however, had a non-productive cough and bronchitis for 2 weeks before admission, with a history of emphysema and pneumonia. The patients' age ranged from 50 to 72 years, with a median of 54 years. There was no predilection of sex.
The distribution of pulmonary hamartomas with PT did not appear to show lobar predilection. The macroscopic appearance was similar in most cases consisting of a tan-white, rubbery, bulging, and irregular mass without cystic change or hemorrhage. Most of the tumors had well-circumscribed borders. Infrequently, papillary projections within hamartomas could be seen grossly, resulting in a granular cut surface (Figure 1, A). Microscopically, all were to varying degrees composed of myxoid or edematous stroma. In most cases, prominent epithelial foldings formed the papillary projections, reminiscent of immature placental villi (Figure 1, B and C). The stroma contained fibroadipose tissue and blood vessels and many inflammatory cells (ie, macrophage and lymphocytes) diffusely dispersed. In a few cases, spindled cells were identified in the immature stroma. The lining epithelium consisted of either nonciliated flat (2/6) and cuboidal (2/6) or ciliated columnar cells (1/6) and mixed types (1/6) (Figure 1, D). The hamartomas consisted of fibroadipose tissue and varying proportions of cartilage, and their sizes ranged from 0.9 to 4 cm (median, 0.9 cm). The adjacent lung often showed an accumulation of clusters of intra-alveolar macrophages and/or type II pneumocytes. Slight interstitial fibrosis was noted in some cases. There was no evidence of accompanying emphysema or lipomatosis.
Immunohistochemical analyses revealed that a major portion of the lining epithelium of PT was positive for TTF-1 and focally epithelial cells were also stained with anti-Ki-67 (MIB-1) antibody, indicating that the epithelium was of lung origin and proliferating (Figure 2). In contrast, very few stromal cells were positive for these 2 markers. Conversely, many stromal cells, especially the spindled cells, were immunoreactive for c-Kit, a stem cell or mast cell marker (Figure 3, A and B). Further study by Leder stain for naphthol-ASD-chloroacetate esterase demonstrated that the c-Kit-positive cells were mast cells and not stem cells (Figure 3, C, and Table 2).
[FIGURES 2-3 OMITTED]
Placental transmogrification associated with pulmonary fibrochondromatous hamartoma slightly differs from that associated with bullous emphysema or lipomatosis clinically. The former often presents in patients who are relatively older, who are asymptomatic, or who have only mild respiratory symptoms and no sex predilection. Histopathologically, however, PT associated with pulmonary hamartomas is similar to PT associated with other conditions. (1-9)
The mechanism by which PT develops from lung parenchyma is still unknown. Since it was originally reported to be associated with emphysema, many believe that PT develops from bullous emphysema (3,4,6) or is reactive to emphysema. (9) Some investigators further speculate that PT may be related to a lymphatic or vascular abnormality, based on the observation of the frequent lymphatic or vascular proliferation and dilation of vascular channels and its unusual distribution within the stroma of PT. (4,6) The stroma of PT, however, consists of myxoid fibroadipose tissue, a variable number of inflammatory cells, and, in some cases, smooth muscle, in addition to relatively prominent blood or lymphatic channels. (1-9) These additional components cannot be simply attributed to lymphatic or vascular malformation. Some authors have hypothesized that the process of PT is likely a congenital malformation with slow progression and not a sequela of preexisting lung emphysema. (6) A case of PT associated with lipomatosis has also been reported where the authors claim that lipomatosis is related to severe emphysema or represents a special type of emphysema. This conclusion, however, is not supported by histologic presentation in their report. They contend that lipomatosis may be differentiated from a pulmonary hamartoma. (8) McChesney (1) has also thought that PT might represent a unrecognized pulmonary hamartoma. The exact relationship between the PT and hamartoma, however, has not been clearly elucidated by systematic examination of a series of cases.
We herein describe a clear association of PT with pulmonary hamartoma. Interestingly, all cases of PT in our study are associated with the fibrochondromatous type of pulmonary hamartoma. This may merely reflect the fact that most pulmonary hamartomas are of this type as demonstrated in this study (34/38) and by others. (10) According to Gjievre et al, (10) the general population of patients with pulmonary fibrochondromatous hamartomas comprise 71.2% of all lung hamartomas and show a wide age range (20 to 90 years; mean, 61.7 years). Men predominate over women by a ratio of 2:1. These findings are different from those of our patients who are relatively younger and have an equal sex distribution. In some cases, a focally papillary growth pattern has been noted within or near the hamartomas, but none have related this phenomenon to the placenta-like change. (10) This papillary growth pattern may well be an early form of villuslike projection. In our study, we found that all pulmonary fibrochondromatous hamartomas show varying degrees of development of papillary structures, most often lined by a single layer of epithelial cells or with epithelial invagination into stroma. Based on the size and extent of papillary growth, hamartomas can be roughly categorized into 3 groups: those with minimal epithelial clefting (44%), those with a small number of epithelial projections (39%), and those with a large number of tall villuslike projections resembling placenta or PT (17%). The epithelium covering the papillary projections may be flat, cuboidal, or columnar type, and some of the cells are ciliated. Morphologically and immunohistochemically, these cells originate from respiratory epithelium or represent an extension from adjacent lung epithelium. The epithelial cells show frequent immunoreactivity for the proliferative index marker Ki-67. In contrast, only rare stromal cell are ever positive for this marker or TTF-1. Our data, therefore, confirm that the epithelium lining the placenta-like projections represents respiratory epithelium that is mitotically active. We hypothesize that the greater proliferation of epithelial lining cells over relatively inactive stromal cells results in glandular inclusions or placental villuslike structures. This exaggeration of epithelial clefting is a likely explanation for the papillary pattern. Prominent lymphatic or vascular channel formation and dilation, as frequently noted in PT associated with emphysema, were not impressive in our cases. It is possible, therefore, that the origin and pathogenesis of PT associated with bullous emphysema may be different from PT associated with pulmonary hamartomas.
The predominant component of the papillary projection is fibroadipose stroma, which is notably immature in some cases. The possibility that hamartomas arise from a mesenchymal stem cell led us to determine whether a stromal stem cell is involved in the formation of hamartoma or plays a role in PT. A proto-oncogene product, c-Kit, primarily expressed on stem cells and mast cells, has been used to identify stem cells. (12,13) Interestingly, we found many spindled stromal cells to be strongly immunoreactive for antibody against c-Kit, but a Leder stain demonstrated these c-Kit-positive cells to be mast cells and not stem cells. We, therefore, conclude that PT is not due to exuberant stromal growth and a role of stem cells cannot be implicated in this process.
The presence of a large number of mast cells within papillary stroma and other parts of hamartomas is an interesting, novel observation that raises much speculation. We believe that it cannot be attributed to an inflammatory reaction, because it localizes within the hamartoma. One study has shown that bronchial smooth muscle cells can express stem cell factor, which may attract mast cells and stimulate their growth. (14) Since pulmonary hamartomas are believed to originate from bronchial mesenchymal tissue and have a smooth muscle component in most cases, (10,15) a similar recruiting mechanism may apply in hamartomas that account for the numerous mast cells in the areas of PT. Recruited mast cells within stroma in turn may partially contribute to formation of placenta-like structures by activation of vascular endothelial cells, resulting in capillary leak phenomenon, transmigration of white blood cells, and exudation of fibrinogen (and other plasma molecules) into the tissue, with consequent edema. (16) Mast cells have also been observed in many other benign and malignant tumors, such as squamous cell carcinoma, basal cell carcinomas, mammary adenocarcinoma, and malignant melanoma. (17-20) The tumor-infiltrating mast cells have been shown to be involved in many varied processes, including tumor development, (19) immunorepression in the case of basal cell carcinoma, (18) and up-regulation of angiogenesis in squamous cell carcinoma. (17) We speculate that mast cells may be involved in the formation of pulmonary hamartomas or PT by similar mechanisms, such as producing cytokines, which in turn promote the epithelial proliferation.
In conclusion, we demonstrate that PT is a pattern frequently associated with pulmonary hamartoma. We believe that a proliferating respiratory epithelium exceeding the stromal component may induce placental papillary projections (PT). We also observe an increase in mast cells in the stroma of hamartomas. The significance of presence of mast cells in PT and hamartomas, however, awaits further evaluation.
Table 1. Clinical and Pathologic Profiles of Patients With Placental Transmogrification and Associated Pulmonary Fibrochondromatous Hamartoma Patient No./Sex/ Size, Age, y Site * cm Epithelial Type 1/M/50 LLL 0.2 Cuboidal, nonciliated 2/M/55 RUL 4.0 Flat, nonciliated 3/F/67 RUL 0.8 Cuboidal, nonciliated 4/M/72 LLL 0.9 Columnar, ciliated 5/F/53 RLL 1.0 Cuboidal-columnar, ciliated 6/M/52 RUL 4.0 Flat, nonciliated * LLL indicates left lower lobe; RUL, right upper lobe; and RLL, right lower lobe. Table 2. Immunohistochemical Studies of Placental Transmogrification * TTF-1 ([dagger]) Ki-67 ([dagger]) Patient No. Epithelium Stroma Epithelium Stroma 1 90 1 8 1 2 80 1 5 1 3 90 2 5 1 4 90 2 5 <1 5 50 0 5 <1 6 70 0 8 3 c-Kit/Leder Stain Patient No. Epithelium Stroma 1 - ++ 2 - + 3 - + 4 - ++ 5 - + 6 -/+ ++ * TTF-1 indicates thyroid transcription factor 1; minus sign, negative; plus sign, focally positive; and double plus sign, diffusely positive. ([dagger]) Percentage of immunoreactive cells.
(1.) McChesney TM. Placental transmogrification of the lung: a unique case with remarkable histopathologic features. Lab Invest. 1979;40:245-246.
(2.) Horsley WS, Gal AA, Mansour KA. Unilateral giant bullous emphysema with placental transmogrification of the lung. Ann Thorac Surg. 1997;64:226-228.
(3.) Fidler ME, Koom M, Sebek B, Greco MA, Rizk CC, Askin FB. Placental transmogrification of the lung, a histologic variant of giant bullous emphysema: clinicopathological study of three further cases. Am J Surg Pathol. 1995;19:563-570.
(4.) Theile A, Wex P, Muller KM. Placentoid malformation of the lung as differential diagnosis of localized emphysema [in German]. Pneumologie. 1997;51 (6): 50-54.
(5.) Hano H, Cui S, Ushigome S, Kotajima F, Sato T. Papillary adenocarcinoma arising in placentoid bullous lesion of the lung: report of a case with immunohistochemical study. Arch Pathol Lab Med. 1998;122:915-919.
(6.) Theile A, Muller KM. Placentoid malformation of the lung [in German]. Pathologe. 1998;19:134-140.
(7.) Mark EJ, Muller KM, McChesney T, Dong-Hwan S, Honig C, Mark MA. Placentoid bullous lesion of the lung. Human Pathol. 1995;26:74-79.
(8.) Hochholzer L, Moran CA, Koss MN. Pulmonary lipomatosis: a variant of placental transmogrification. Mod Pathol. 1997;10:846-849.
(9.) Kronz JD, Palmer C, Askin FB. Placental transmogrification of the lung. Arch Pathol Lab Med. 1999;123:856.
(10.) Gjievre JA, Myers JL, Prakash UB. Pulmonary hamartomas. Mayo Clin Proc. 1996;71:14-20.
(11.) Bingle CD. Thyroid transcription factor-1 [review]. Int J Biochem Cell Biol. 1997;29:1471-1473.
(12.) Natali PG, Nicotra MR, Sures I, Santoro E, Bigotti A, Ullrich A. Expression of c-kit receptor in normal and transformed human nonlymphoid tissues. Cancer Res. 1992;52:6139-6143.
(13.) Tsujimura T. Role of c-kit receptor tyrosine kinase in the development, survival and neoplastic transformation of mast cells [review]. Pathol Int. 1996;46: 933-938.
(14.) Kassel O, Schmidlin F, Duvernelle C, Gasser B, Massard G, Frossard N. Human bronchial smooth muscle cells in culture produce stem cell factor. Eur Respir J. 1999;13:951-954.
(15.) Dominguz H, Hariri J, Pless S. Multiple pulmonary chondrohamartomas in trachea, bronchi and lung parenchyma: review of the literature. Respir Med. 1996;90:111-114.
(16.) Valent P, Sillaber C, Baghestanian M, et al. What have mast cells to do with edema formation, the consecutive repair and fibrinolysis? [see comments] [review]. Int Arch Allergy Immunol. 1998;115:2-8.
(17.) Coussens LM, Raymond WW, Bergers G, et al. Inflammatory mast cells up-regulate angiogenesis during squamous epithelial carcinogenesis. Genes Dev. 1999;13:1382-1397.
(18.) Deng JS, Brod BA, Saito R, Tharp MD. Immune-associated cells in basal cell carcinomas of skin. J Cutan Pathol. 1996;23:140-146.
(19.) de Cidre LL, Eijan AM, Bertolesi G, Isturiz M, Sacerdote de Lustig E. Influence of mast cells on two murine mammary adenocarcinomas. Tumour Biol. 1996;17:345-353.
(20.) Duncan LM, Richards LA, Mihm MC Jr. Increased mast cell density in invasive melanoma. J Cutan Pathol. 1998;25:11-15.
Accepted for publication November 16, 2001.
From the Department of Pathology, New York University School of Medicine, New York, NY.
Reprints: Jonathan Melamed, MD, Department of Pathology, New York University Medical Center, 560 First Ave, New York, NY 10016 (e-mail: firstname.lastname@example.org).
|Multiple Pulmonary Nodules With Calcification.|
|Hantavirus Pulmonary Syndrome Is Distinguishable From Acute Interstitial Pneumonia.|
|A dermal vascular hamartoma in a sun conure (Aratinga solstitialis).|