Subclassification of pulmonary non-small cell lung carcinoma in fine needle aspirates using a limited immunohistochemistry panel.

BACKGROUND: Newer treatment modalities require subtyping of non-small cell lung carcinomas (NSCLC). Morphological differentiation is often difficult and various immunohistochemical (IHC) panels have been used to maximize the proportion of accurately subtyped NSCLC. AIM: The aim of this study was to subtype NSCLC on fine needle aspirates (FNA) using a minimal antibody panel.
MATERIALS AND METHODS: Cell blocks from 23 FNA samples with a morphological diagnosis of NSCLC were taken. IHC was evaluated (blinded to clinical data) for thyroid transcription factor-1 (TTF-1), cytokeratin (CK)7, CK20, and tumor protein p63.
RESULTS: TTF-1 was positive in 14 and negative in 9 cases. The p63 was positive in two cases each of TTF-1 positive and negative tumors. CK7 was positive in 12 of the 14 TTF-1 positive tumors and 4 of the TTF-1 negative tumors. CK20 was negative in all. All the 14 TTF-1 positive tumors were primary lung tumors, 12 being NSCLC and 2 being squamous cell carcinoma. Five of nine TTF-1 negative tumors were metastatic tumors from endometrium, kidney, and head and neck region (two), and one was an unknown primary. Four of the nine TTF-1 negative tumors were morphologically NSCLC and were clinically considered to be primary lung tumors. Three of these tumors stained positive for CK7 but negative for CK20 and p63, and one case was negative for the immunomarkers.
CONCLUSION: Use of limited IHC panel helps categorize primary versus secondary tumors to the lung. The p63 is a useful marker for detecting squamous cell carcinoma. In countries where antibodies are not readily available, using a limited IHC panel of TTF-1, p63, and CK7 can help further type NSCLC lung tumors.

Introduction

Lung cancer is the most common cancer worldwide and is the leading cause of death in many countries. In the past, primary bronchopulmonary carcinomas were classified as non-small cell lung carcinoma (NSCLC) and small cell neuroendocrine carcinoma. With the introduction of new treatment modalities, it has become important to specifically classify primary NSCLC.[1] The identification of epidermal growth factor receptor (EGFR) positive NSCLC permits the use of tyrosine kinase inhibitors (TKI). Also, the recognition of squamous cell carcinoma (SCC) is important because if this subset of lung carcinoma patients is given bevacizumab, then it may lead to serious pulmonary bleeding.[2]

Most patients with lung carcinoma present with clinically advanced disease, and fine needle aspiration cytology (FNAC) may be the only available diagnostic specimen and also the only material available for molecular studies necessary for current therapeutic decision making.[234] It is well documented that cytomorphology and immunohistochemistry (IHC) are useful in further categorization of NSCLC.[5] In centers where IHC is not readily accessible, a limited panel of antibodies can be used to categorize the tumor. In this study, we used a limited panel of antibodies to classify NSCLC diagnosed based on FNA from lung lesions.

Materials and Methods

Fine needle aspirates from patients with lung carcinoma with a morphological diagnosis of NSCLC over a period of 5 years were studied. In 23 cases, adequate cell block preparations were available. Informed consent was obtained from the subjects. The clinical data were unfolded after the IHC results were analyzed. IHC was performed (blinded to the clinical data) for thyroid transcription factor-1 (TTF-1), cytokeratin 7 (CK7), cytokeratin 20 (CK20), tumor protein p63, and chromogranin A.

IHC was performed manually on representative 4-μm sections cut from formalin-fixed paraffin-embedded cell blocks using commercially available monoclonal antibodies. Dehydrated tissue sections for immunocytochemistry were treated with 3% hydrogen peroxide in methanol for 10 min to block endogenous peroxidase and heated in 0.01 M citrate buffer (pH 6.0) in a microwave for epitope retrieval. Sections as well as smears were incubated with primary antibody for 1 h at room temperature. Detection system used was Envision-Flex (DAKO, Glostrup, Denmark) according to manufacturer's instructions. Detection was achieved using diaminobenzidine (DAB+ Liquid; DAKO, Carpinteria, CA, USA). The antibodies used in the study were TTF-1 (monoclonal, 8G7G3/1; 1:50 dilution; DAKO, Carpinteria, CA, USA), CK7 (monoclonal, OV-TL 12/30; 1:50 dilution; DAKO, Glostrup, Denmark), CK20 (monoclonal, KS 20.8; 1:50 dilution; DAKO, Glostrup, Denmark), p63 (monoclonal, 4A4; 1:50 dilution; DAKO, Glostrup, Denmark), and chromogranin A (monoclonal, DAK-A3, 1:50 dilution, DAKO, Glostrup, Denmark). Standard, appropriate histologic tissue was used as positive control and the negative control was run by omission of primary antibody. They were used for each run. Staining was considered positive when the tumor cells showed a diffuse or focal staining.

A histological examination was available in two cases only.

Results

TTF-1 was positive in 14 and negative in 9 cases. The p63 was positive in two cases each of TTF-1 positive and negative tumors. CK7 was positive in 12 of the 14 TTF-1 positive tumors and 5 of the TTF-1 negative tumors. CK20 and chromogranin was negative in both the TTF-1 positive and negative tumors [Table 1]. All the 14 TTF-1 positive tumors were primary lung tumors, 12 being NSCLC and 2 being SCC. One of the two SCC tumors was confirmed on histological examination. Five of the 9 TTF-1 negative tumors were metastatic tumors from endometrium, kidney, and head and neck region (two), and in one tumor, the primary site was unknown. Four of the 9 TTF-1 negative tumors were morphologically NSCLC and were clinically considered to be primary lung tumors. Three of these tumors stained positive for CK7 but negative for CK20 and p63, and one case was negative for all the immunomarkers. A histological examination was available in this case and was reported as NSCLC. Two tumors that were positive for p63 and CK7 but negative for TTF-1 were metastatic from the head and neck region. The final diagnosis in the 23 NSCLC was primary lung NSCLC in 18 and metastatic tumor in 5. Of the 18 primary lung tumors, 16 were NSCLC and 2 were SCC.

Table 1

Immunostaining pattern of non-small cell lung carcinoma on fine needle aspirates

Discussion

The distinction of subgroups of NSCLC may be challenging on cytological material based on morphology alone.[2] This distinction is very important in the proper management as the EGFR-directed TKIs, erlotinib and gefitinib, are effective in patients with adenocarcinoma having the EGFR mutation, while patients with SCC should not be given bevacizumab because of the risk of fatal pulmonary hemorrhage.[2] The determination of the EGFR mutation is not available in all the cytology laboratories, and thus, we proposed to use a minimal panel of antibodies for further characterization of the NSCLCs.

We found that by using TTF-1, CK7, CK20, chromogranin, and p63, we were able to identify primary lung tumors and further type them as SCC. TTF-1 and p63 immunostaining has been successfully applied to cytological material to facilitate pathological differentiation between small cell carcinoma and poorly differentiated pulmonary SCC.[1] TTF-1 has been found to be the most specific marker for adenocarcinoma with 60% sensitivity and 98% specificity, in comparison to Napsin A that has a sensitivity of 83% and specificity of 98%.[36] Also, it has been shown that p63 had 95% sensitivity and 86% specificity for SCC, while cytokeratin 5/6 had 53% sensitivity and 96% specificity.[6] Napsin A and p63 have been proposed as the most cost-effective panel of antibodies for differential diagnosis of primary lung adenocarcinoma from SCC. Due to availability of TTF-1 in our laboratory, we proposed to study this marker and found it equally effective. Righi et al. (2011), by using a four-antibody panel (TTF-1, desmocollin-3, p63, and Napsin A), were able to refine lung cancer classification in FNAC cell blocks, remarkably reducing the percentage of unclassified NSCLC from 36% to 14%.

To conclude, we found that using a limited panel of antibodies, we were able to classify NSCLC. Though not ideal, this panel of antibodies offered the best discrimination and helped separate primary tumors and also classify them into adenocarcinomas or SCC.