Surfactant Protein A Detection in Primary Pulmonary Adenocarcinoma without Bronchioloalveolar PatternUzaslan E.a · Stuempel T.a · Ebsen M.a · Freudenberg N.b · Nakamura S.d · Costabel U.c · Guzman J.a
aGeneral and Experimental Pathology, University of Bochum, Bochum, bDepartment of Cytology, Institute of Pathology, University of Freiburg i. Br., Freiburg i. Br., cDepartment of Pneumology and Allergy, Ruhrlandklinik, Essen, Germany; dDepartment of General Medicine and Clinical Investigation, Nara Medical University, Nara, Japan
Background: Immunohistochemical studies in human lung carcinoma reported positive staining of tumor cells for surfactant protein A (SP-A), especially in peripheral airway cell carcinoma, which include bronchioloalveolar carcinoma and in some reports also papillary subtypes. Objective: The purpose of this study was to determine the SP-A expression in tumor cells of lung adenocarcinoma without a bronchioloalveolar pattern, classified according to the WHO. Methods: In total, 169 primary adenocarcinomas of the lung (109 acinar, 32 solid with mucin, 24 papillary and 4 mucinous) were examined by immunohistochemistry for SP-A expression. Results: Twenty-five percent of acinar, 38% of papillary and 3% of solid adenocarcinoma with mucin showed a positive intracytoplasmic SP-A reaction of the tumor cells. None of the mucinous adenocarcinomas stained for SP-A. This study included the largest number of acinar adenocarcinomas and solid adenocarcinomas with mucin studied for SP-A. We clearly demonstrated that also primary lung adenocarcinoma without a bronchioloalveolar pattern can express SP-A. A positive staining of hyperplastic type II cells surrounding the tumors or entrapped in the tumor could clearly be differentiated from the SP-A-positive stain of tumor cells. Conclusion: These results support the theory that SP-A-producing cells may generate not only bronchioloalveolar and papillary carcinoma, but also other subtypes of lung adenocarcinoma.
Copyright © 2005 S. Karger AG, Basel
Immunohistochemical studies performed in human lung carcinoma reported a positive staining of tumor cells for surfactant protein A (SP-A) in part of adenocarcinoma of the lung [1, 2, 3, 4], especially in peripheral airway cell carcinoma, which include bronchioloalveolar carcinoma and in some reports also papillary subtypes [5, 6, 7, 8, 9, 10, 11, 12]. Ueno et al.  indicated that more than 80% of bronchioloalveolar carcinoma and 50–60% of grade I–III invasive adenocarcinoma of the lung stain positive with SP-A, but they did not further classify their adenocarcinoma according to histological subtypes. Mizutani et al.  concluded that only tumors with alveolar type II and/or Clara cell type are SP-A positive.
However, Hirata et al.  indicated in a study of 33 cases of bronchial gland cell type of adenocarcinoma that 68% of these cases were SP-A positive. Ten Have-Opbroek et al.  showed in a canine bronchogenic carcinoma model that alveolar type II tumor stem cells may generate not only adenocarcinoma with bronchioloalveolar and papillary patterns but also adenocarcinoma with other growth patterns. SP-A mRNA expression was also described in solid and glandular lung adenocarcinoma . Different percentages of SP-A-positive tumors were found in other subtypes of adenocarcinoma (without bronchioloalveolar or papillary patterns), without further subclassifications of the tumor [9, 15], or in mixed acinar/papillary  or acinar/solid groups . Only Sugiyama et al.  clearly classified 54 primary lung adenocarcinomas according to the WHO classification and found SP-A expression in tumors without a bronchioloalveolar pattern.
Therefore, the purpose of this study was to determine the SP-A expression in a large group of lung adenocarcinomas without bronchioloalveolar patterns, classified according to the WHO , because the presence of SP-A in these tumors may indicate that they also originate from type II or Clara cells.
Materials and Methods
All primary adenocarcinomas of the lung without bronchioloalveolar patterns surgically resected by either lobectomy or pneumonectomy in the Ruhrlandklinik in the years 1992–1997 were included in this study. All cases were reclassified by three pathologists in consensus according to the latest edition of the WHO histological typing of lung tumors . Bronchioloalveolar adenocarcinoma and mixed subtypes with a bronchioloalveolar pattern were excluded from this analysis.
In total, 169 primary adenocarcinomas of the lung, namely 109 acinar (ICD-0-8550/3), 32 solid with mucin (ICD-0-8230/3), 24 papillary (ICD-0-8260/3) and 4 mucinous ‘colloid’ adenocarcinomas (ICD-0-8480/3), were available.
The cases were examined by immunohistochemistry using a modified peroxidase-antiperoxidase (PAP) technique as described before . A polyclonal antibody against pulmonary SP-A was used with diaminobenzidine as chromogen (a gift from W. Bartsch, Fraunhofer Institute, Hannover, Germany) . The quality of this antibody against SP-A was found to be comparable with commercially available antibodies. Positive and negative control sections for SP-A and for the specificity of the immunoreaction were included in each reaction. We defined as a diffusely positive immunostain when most of the tumor cells were positive at different fields in a 100× magnification and as a focally positive immunostain when few positive tumor cell groups were found disseminated or focally located in tumor tissue at different fields in a 100× magnification. The stained slides were read microscopically by three pathologists in consensus.
Twenty-five percent of acinar adenocarcinoma, 38% of papillary adenocarcinoma and 3% of solid adenocarcinoma with mucin production (fig. 1, 2a, b, 3a, b) showed a positive intracytoplasmic SP-A reaction of tumor cells. The staining patterns of the tumors were diffusely positive in 25 cases (17 acinar, 8 papillary) and focally positive in 12 cases (10 acinar, 1 papillary, 1 solid with mucin). None of the mucinous adenocarcinomas stained for SP-A (fig. 4).
Fig. 1. SP-A reaction in different subtypes of adenocarcinoma in our study.
Fig. 2.a Acinar adenocarcinoma of the lung. Hematoxylin-eosin stain. Original magnification ×50. b Acinar adenocarcinoma with positive SP-A staining. PAP method. Original magnification ×50.
Fig. 3.a Solid adenocarcinoma with mucin. PAS stain. Original magnification ×100. b Solid adenocarcinoma with mucin. Positive SP-A staining. PAP method. Original magnification ×100.
Fig. 4.a Papillary adenocarcinoma of the lung. Hematoxylin-eosin stain. Original magnification ×50. b Papillary adenocarcinoma with positive SP-A staining. PAP method. Original magnification ×50.
The positive SP-A staining of hyperplastic type II cells surrounding the tumors or in some cases entrapped between the tumor infiltrates did not disturb the recognition of SP-A-positive tumor cells.
In this study, we clearly demonstrate that also primary lung adenocarcinoma without a bronchioloalveolar pattern can express SP-A, which confirms the recently published observation by Ueno et al. . However, these authors did not perform further subclassifications of the histological subtypes, as was done in our study. We studied the largest number of acinar adenocarcinomas and solid adenocarcinomas with mucin for SP-A expression described in the literature. We found that 25% of acinar adenocarcinoma and 3% of solid adenocarcinoma with mucin showed intracytoplasmic SP-A staining of the tumor cells. In our study, the percentage of SP-A-positive cases of adenocarcinoma without bronchioloalveolar patterns is lower than in the reports by Ueno et al. , by Sugiyama et al.  and by Hirata et al. , possibly because we excluded adenocarcinoma with mixed subtypes showing bronchioloalveolar or papillary components from this study, and because we very carefully discriminated the SP-A-positive staining of hyperplastic type II cells surrounding the tumors or entrapped in the tumor from the intracytoplasmic positive SP-A stain of tumor cells , which sometimes can be difficult.
We also studied true papillary lung adenocarcinoma diagnosed according to the WHO classification and other reports . Thirty-eight percent of the true papillary carcinoma of our study were SP-A positive.
In contrast to Mizutani et al. , who concluded that only tumors with alveolar type II and/or Clara cell type are SP-A positive, our results support the theory [13, 22] that SP-A-producing cells may generate not only bronchioloalveolar and papillary carcinoma, but also other subtypes of lung adenocarcinoma such as lung acinar adenocarcinoma and solid adenocarcinoma with mucin.
This study was supported by a grant of the Ichiro Kanehara Foundation, Chihiro and Yokochi Fund.
Dr. Esra Uzaslan contributed to this study as a guest researcher at General and Experimental Pathology, University of Bochum, originally from Chest Diseases Department, Medical Faculty, University of Uludag, Bursa, Turkey.
Prof. Dr. med. J. Guzman y Rotaeche
c/o Prof. Shinobu Nakamura, Department of General Medicine and Clinical Investigation
Nara Medical University, 840 Shijo
Kashihara, Nara 634-8522 (Japan)
Tel. +81 744 29 8905, Fax +81 744 24 5739, E-Mail firstname.lastname@example.org
Received: September 3, 2003
Accepted after revision: August 25, 2004
Number of Print Pages : 5
Number of Figures : 4, Number of Tables : 0, Number of References : 22
Respiration (International Journal of Thoracic Medicine)
Vol. 72, No. 3, Year 2005 (Cover Date: May-June 2005)
Journal Editor: C.T. Bolliger, Cape Town
ISSN: 0025–7931 (print), 1423–0356 (Online)
For additional information: http://www.karger.com/res