Immunohistochemistry (IHC) for thyroid transcription factor-1 (TTF-1) is used to confirm the diagnosis of lung adenocarcinoma. Napsin A also has shown positivity in lung adenocarcinoma. A combined double stain for TTF-1 and napsin A has been proposed to achieve higher sensitivity and specificity. In this study, the authors evaluated the utility of this double stain in the diagnosis of lung adenocarcinoma in cell blocks of fine-needle aspirates (FNA).
The authors used a cohort comprising 35 FNA cell blocks of lung adenocarcinoma and 24 FNA cell blocks of lung squamous cell carcinoma (SqCCA). IHC was performed; expressions of TTF-1 as brown nuclear stain and of napsin A as red cytoplasmic stain were identified.
Twenty-six of 35 (74%) lung adenocarcinomas were positive for double staining with TTF-1/napsin A. Of 35 lung adenocarcinomas, only 2 (5%) were positive for TTF-1 alone and 3 (8%) were positive for napsin A alone. For the double stain TTF-1/napsin A, 3 of 24 (12%) lung SqCCAs were positive for both. Six of 24 (25%) cases were positive for TTF-1 alone, and none were positive for napsin A alone. For lung adenocarcinoma, TTF-1/napsin A has a sensitivity of 74%, specificity of 87%, accuracy of 79%, and a positive predictive value of 89%.
Lung cancer is the most common type of cancer in terms of mortality worldwide.1, 2 Nonsmall cell lung cancer (NSCLC) essentially consisting of adenocarcinoma, squamous cell carcinoma (SqCCA), and large-cell carcinoma account for approximately 80% of all lung cancers.3 The frequency of pulmonary adenocarcinoma has been increasing, and it is now the most common histologic subtype in the United States, constituting about 40% of all NSCLCs.4
In the past, pulmonary oncologists routinely distinguished only between small-cell lung cancer (SCLC) and NSCLC to select the appropriate therapeutic management; the latter were treated according to similar strategies. However, NSCLC histologic subtyping has become increasingly important. An unprecedented change in epidemiology and an increasing number of target-specific chemotherapies, with differential efficacy and toxicity according to tumor subtypes, have made it crucial to distinguish among the NSCLCs with a high degree of specificity and accuracy.5 The strong association of epidermal growth factor receptor (EGFR) gene mutation with adenocarcinoma has made it a candidate for tyrosine kinase inhibitor therapy, whereas patients without EGFR mutations respond better to chemotherapy.6 The role of bevacizumab (Avastin), in combination with carboplatin and paclitaxel, has proven to be beneficial in nonresectable cases of NSCLC.7 However, it is contraindicated in SqCCA because of fatal hemorrhagic episodes, reported to be more frequent (30%) when central cavitations are present, which is a feature of SqCCA.8
The International Association for the Study of Lung Cancer, the American Thoracic Society, and the European Respiratory Society have proposed a new International Multidisciplinary Lung Adenocarcinoma classification.9 Now, for the first time, we will have internationally developed standardized criteria and nomenclature developed in a multidisciplinary fashion for the classification of small biopsies and cytologic specimens of lung carcinoma. Immunohistochemistry (IHC) for thyroid transcription factor-1 (TTF-1) is widely used in the diagnosis of lung adenocarcinoma. It is reported positive in approximately 76% of lung adenocarcinomas.10 TTF-1 staining in primary SqCCA varies, ranging from 0% to 38%.10, 11 Increased TTF-1 positivity in lung SqCCA has been noted when SPT24 (Vector Laboratories, Burlingame, California; commercially available clone of TTF-1 monoclonal antibody) is used for IHC.12
Other IHC stains, such as surfactant protein A (SP-A), napsin A, and special stains for mucin are also used for confirming the diagnosis of adenocarcinoma. Surfactant protein A is the major protein component of the lung surfactant, which is synthesized and secreted into the alveoli of the lung by type II pneumocytes. In malignant pleural effusions, high concentrations of SP-A have been found only in patients with lung adenocarcinoma.13 Napsin A is strongly positive in up to 80% of primary lung adenocarcinomas by IHC.14 SCLC and SqCCA of the lung are negative for napsin A.15 A combination of the 2 stains (TTF-1 and napsin A) has been proposed to achieve higher sensitivity and specificity for lung adenocarcinoma, and the combination might have greater reliability.16 In this study, we evaluate the accuracy and utility of this double stain in cell blocks of fine-needle aspirations (FNA) of pulmonary adenocarcinoma compared with pulmonary SqCCA.
MATERIALS AND METHODS
A total of 59 lung cancers (35 adenocarcinomas and 24 SqCCA) were selected from the cytology records of the Emory University Hospital Anatomic Pathology Department for this study. The cases were selected after reviewing cytomorphologic features on smears from FNA cases. FNAs were diagnosed as adenocarcinoma with smears showing crowded syncytial aggregates of malignant cells with ill-defined cell borders. The cytoplasm ranged from homogeneous to foamy. The nuclei were enlarged with irregular nuclear membranes and fine to coarse chromatin. Prominent macronucleoli were also present. SqCCA diagnosis was based on presence of cohesive groups of malignant cells forming disorderly sheets, with intermediate to large cells, and defined cell boundaries. The cytoplasm ranged from dense to cyanophilic, with occasional refractile rings of keratinization. The nuclei were enlarged, hyperchromatic, and had coarse chromatin. Nucleoli were conspicuous but not prominent. Formalin-fixed, paraffin-embedded, hematoxylin and eosin (H&E)-stained, cell-block sections of these FNA specimens were then examined for adequacy of cellular diagnostic tissue and were selected for this study. The final FNA diagnosis was also confirmed in surgical pathology reports of the corresponding biopsy of resected tumors.
Sections (5 microns) of formalin-fixed, paraffin-embedded, cell-block tissue were tested for the presence of 2 primary antibodies by using Bond Polymer Refine Detection Kit (DAB chromogen, brown) and Bond Polymer Refine Red Detection Kit (alkaline phosphatase chromogen; Leica Microsystems, Bannockburn, Illinois). The detection systems avoid the use of streptavidin and biotin and, therefore, eliminate nonspecific staining as a result of endogenous biotin. The bond polymer refine detection is a biotin-free, polymeric, horseradish peroxidase (HRP)-linker antibody conjugate system for the detection of tissue-bound mouse IgG, rabbit IgG, and some mouse IgM primary antibodies. Bond polymer refine red detection is a biotin-free, polymeric alkaline phosphatase (AP)-linker antibody conjugate system for the detection of tissue-bound mouse IgG, rabbit IgG, and some mouse IgM primary antibodies.
All steps were performed on the Leica Bond III automated system (Leica Microsystems). Specimens were deparaffinized and antigen was retrieved on the instrument. All slides were incubated with the first primary antibody TTF-1 (RTU, clone SPT24, Leica Microsystems) for 15 minutes, with post primary polymer for 8 minutes, blocked with 3% hydrogen peroxide for 5 minutes and 3,3-diaminobenzidine (DAB, brown chromogen) for 10 minutes. The second primary antibody napsin (Leica Microsystems; IP64 clone; 1:400) was incubated for 15 minutes, post primary AP for 15 minutes, polymer AP for 15 minutes, mixed red refine (alkaline phosphatase, red chromogen) for 10 minutes, and hematoxylin as counterstain for 5 minutes. These incubations were performed at room temperature. Between incubations sections were washed with Tris-buffered saline (bond wash solution). Cover-slipping was performed by using the Tissue-Tek SCA (Sakura Finetek, Torrance, California) coverslipper.
Positive controls of known lung adenocarcinoma and negative controls with primary antibody replaced with Tris buffer were run with the study slides. An expert pathologist evaluated immunoreactivity. Expressions of TTF-1 as a brown nuclear stain and napsin A as red cytoplasmic stain were identified easily in tumor cells and quantified as positive or negative. TTF-1–stained alveolar lining cells and napsin A-stained cytoplasm of alveolar macrophages were internal positive controls.
Twenty-six of 35 (74%) lung adenocarcinomas (Figs. 1, 2, 3, 4, 1-4) were positive for the double stain TTF-1/napsin A (Table 1). Two (5%) were positive for TTF-1 alone, and 3 (8%) were positive for napsin A alone. With the double stain TTF-1/napsin A, 21 of 24 SqCCAs were negative (Figs. 5 and 6), whereas 3 of 24 (12%) lung SqCCAs were positive. Six of 24 (25%) were positive for TTF-1 alone, and none were positive for napsin A alone. Thus, 28 of 35 (80%) adenocarcinomas of the lung were positive for TTF-1, and 29 of 35 (82%) were positive for napsin A. For SqCCA of the lung, 9 of 24 (37%) were positive for TTF-1, and 3 of 24 (12%) were positive for napsin A. In addition, specificity, sensitivity, positive predictive values, and negative predictive values for the immunohistochemical stain were also calculated (Table 2).
Table 1. Immunoreactivity of TTF-1/Napsin A Double Stain, TTF-1, and Napsin A in Adenocarcinoma and Squamous Cell Carcinoma of the Lung
The World Health Organization (WHO) classification system is used to classify lung cancers usually on whole-tumor resection specimens rather than small cytologic samples.1, 2 In 2004, cytology was addressed for the first time by the WHO classification system.9 Fine-needle aspiration is frequently performed when surgery may not be the first treatment option.17 However, with cytomorphology alone, sometimes it can be very difficult to differentiate all cases of NSCLC into adenocarcinoma and SqCCA, especially when cytologic material for preparation of cell block is insufficient.18 The presence of undifferentiated areas and frequent heterogeneity in most NSCLCs can also create inherent inaccuracy in morphologic diagnosis, and IHC staining has been useful in helping differentiate between adenocarcinoma and SqCCA. Although IHC is virtually never tumor specific, with the help of relatively restricted markers, it is possible to make a distinction or at least narrow the diagnosis among different subtypes with a reasonable degree of certainty. Our current study has shown that dual TTF-1/napsin A has a sensitivity of 74% and specificity of 87% for diagnosing adenocarcinoma and, hence, is useful in differentiating adenocarcinoma from SqCCA.
Accuracy in the cytomorphologic diagnosis of adenocarcinoma is reportedly 80%, whereas diagnostic accuracy for SqCCA is reported to be 87%.19 The challenges encountered in cytomorphology diagnosis of NSCLC and separating adenocarcinoma from SqCCA are mainly due to the finding that adenocarcinoma can occasionally undergo coagulative necrosis, giving the cells pseudokeratinized appearance along with dark pyknotic nuclei, and conversely SqCCA can develop nonspecific degenerative vacuoles that can mimic secretory vacuoles of adenocarcinoma. The cytomorphologic distinction, however, between NSCLC and SCLC appears to be highly accurate, although a cytologic diagnosis of NSCLC is more reliable than a cytologic diagnosis of SCLC, with average misclassification rates of 2% and 9%, respectively.20
TTF-1 has been recognized for its usefulness in diagnosing certain types of primary lung carcinomas, particularly adenocarcinoma and SCLC. TTF-1 is a nuclear protein that is expressed almost exclusively in normal follicular cells of the thyroid, in alveolar lining cells of the lung, and in most carcinomas derived from these organs. In the normal lung, TTF-1 regulates the transcription of lung-specific genes for surfactant and Clara cell secretory proteins.21 Studies have demonstrated that TTF-1 is a valuable tool for diagnosing certain types of primary lung tumors in histologic sections of surgical specimens and cytologic cell-block sections.22 Bishop et al. found 69 of 95 (73%) adenocarcinomas to be positive for TTF-1 staining and uniformly negative (0%) for 48 SqCCAs on histologic resections.16 Moreover, in a cytologic evaluation of TTF-1, Liu et al reviewed its usefulness on FNA specimens from various types of lung tumors including 34 primary tumors (14 adenocarcinomas, 12 SqCCAs, and 8 SCLCs) and 9 metastatic tumors.22 The 9 metastatic tumors included adenocarcinoma from breast,3 colon,2 endocervix,1 and stomach,1 as well as adenoid cystic carcinoma of the tongue1 and melanoma.1 Twelve of 14 (86%) pulmonary adenocarcinomas were positive for TTF-1; of 12 pulmonary SqCCAs, only 1 (8%) was positive, and of 8 SCLCs, 6 (75%) were positive. All metastatic tumors were negative for TTF-1. Their results demonstrated that TTF-1 is both a sensitive and specific marker for adenocarcinoma and SCLC.
In another cytologic study, Hecht et al evaluated 122 cell blocks including 8 primary and 39 metastatic pulmonary adenocarcinomas, 11 pulmonary neoplasms of other types, 50 nonpulmonary metastatic tumors, and 14 mesotheliomas.11 TTF-1 was reactive in 42 of 47 (89%) pulmonary adenocarcinomas. Only 1 of 4 (25%) pulmonary small cell/neuroendocrine tumors was TTF-1 positive, whereas 1 of 7 (14%) SqCCA was weakly reactive. Of 50 metastatic tumors of nonpulmonary origin, focal weak reactivity was noted in only 1 (2%) metastatic ovarian carcinoma. All mesotheliomas were nonreactive. The authors concluded that in cytologic preparations, TTF-1 is a highly selective marker for pulmonary adenocarcinoma, with sensitivity and specificity of 89% and 100%, respectively. In view of the above studies, there is concordance for both surgical and cytologic evaluation of TTF-1 expression in lung adenocarcinoma.
TTF-1 expression in lung tumors may also depend on use of a particular clone, which may impact its overall use in both surgical and cytologic evaluation of NSCLC. Two commercially available clones of TTF-1 monoclonal antibodies, 8G7G3/1 and SPT24 are available and have been recently compared with each other.12 TTF-1 expression, using both clones, in primary tumors of lung, prostate, pancreas, stomach, salivary glands, breast, bladder, colon, and squamous cell carcinoma of the head and neck were compared in this study.12 The SPT24 clone detected more primary lung tumors of all histologic subtypes, including a significantly higher number of lung SqCCAs. SPT24 showed positivity for 72% of lung adenocarcinomas and 17% of lung SqCCAs, whereas clone 8G7G3/1 demonstrated positivity in 65% of lung adenocarcinomas and 1.0% of lung SqCCAs. Among nonpulmonary tumors, urothelial carcinoma (5.1%), prostate (1.2%), colon (2.5%), salivary gland (1.8%), and stomach (0.9%) were positive with both clones. Carcinomas of the breast, pancreas, and squamous cell carcinoma of the head and neck were negative with both clones. In summary, this study found TTF-1 clone SPT24 to have detected a higher number of pulmonary nonsmall cell tumors of all histologic subtypes, whereas both clones stained a similar proportion of nonpulmonary tumors. In view of this data, our current study findings, which show an increased positivity of TTF-1 for SqCCA, are probably due to the use of the SPT24 TTF-1 clone.
Napsin A is also a promising marker and has been detected in the cytoplasm of type 2 pneumocytes and alveolar macrophages. It is an aspartic protease with a molecular weight of approximately 38 kDa, which is involved in the N- and C-terminal processing of surfactant protein B.23 Napsin A has been evaluated on surgically resected lung cancers and has been found to be positive in a recent study in 79 of 95 (83%) of lung adenocarcinomas and negative in all 46 lung SqCCAs studied.16 A study conducted in Japan has also compared the usefulness of currently available IHC markers (including napsin A) specific for primary lung adenocarcinoma in cytologic materials. This study found positivity of surfactant protein A (SP-A) in all primary lung adenocarcinomas to be 68.7%, and for both TTF-1 and napsin A positivity was 76.1%. These rates increased to 73.7%, 82.5%, and 82.5%, respectively, when limited to primary lung adenocarcinomas without mucin production.24 In 2007, Dejmek et al. determined TTF-1 and napsin A immunoreactivity on formalin-fixed, paraffin-embedded, cell blocks from 50 pleural effusions. TTF-1 and napsin A were positive in 8 of 12 (66%) and 10 of 12 (83%) pulmonary adenocarcinomas, respectively. Napsin A reactivity was found in >75% of the tumor cells in 9 of 10 (90%) positive cases, whereas TTF-1 reactivity was seen in >75% of the tumor cells in 2 of 8 (25%) positive cases. Their results suggested that napsin A is an improved alternative to TTF-1 in cytologic diagnosis of effusions in which tumor cells may be scanty.25 In addition, the above studies also confirm that histologic and cytologic evaluations of lung adenocarcinoma with napsin A show a high degree of concordance between histology and cytology.
In conclusion, this study confirms that the double IHC stain TTF-1/napsin A is diagnostically useful for the identification of adenocarcinoma in FNA cell-block materials. In addition, napsin A, when used as a single IHC marker, may be superior to TTF-1 alone or the double IHC stain, TTF-1/napsin A, for the diagnosis of lung adenocarcinoma in FNA cell-block materials. These findings merit further study.