Remember me
Abstract
Background: Carcinoma of lung is the most common cause of cancer-associated mortality worldwide. About 70% of lung cancer cases are unresectable and present in advanced stages. So, cytology and small core needle biopsy specimen are available for diagnostic as well as prognostication workup. Subtyping of non-small cell lung cancer (NSCLC) is essential for the treatment and further workup study. For this, immunocytochemistry (ICC) plays a crucial role that helps in early diagnosis. Subtyping of NSCLC from cytology samples using ICC markers like TTF-1, Napsin-A, and p63 and their clinicopathological correlation are the aims of the study. Materials and Methods: This ambispective study was conducted in the pathology department of a tertiary care hospital of eastern India for a 2-year period from 2018 to 2020. In our study, 46 cytologically diagnosed cases of NSCLC were included. Subtyping was done by cytomorphology and correlated with ICC expression, histopathology, and clinicopathological parameters. Results: In our study, adenocarcinoma (ADC) was the most common (32.61%) cancer. Most cases of ADC showed positive expression of TTF-1 and Napsin-A, and p63 was positive in most cases of squamous cell carcinoma (SCC). Concordance with cytomorphology and ICC was 87.50% and 81.81% with ADC and SCC, respectively. Cyto-ICC-histo concordance was observed in 85.51% of ADC and 66.66% of SCC cases. Sensitivity was 100%, 93.1%, and 100% for TTF-1, Napsin-A, and p63, respectively. Specificity of both TTF-1 and Napsin-A was 88.2% and for p63 was 93.8%. Conclusion: In small biopsy along with cytology samples, ICC is cost-effective and plays an important role in early diagnosis along with management of NSCLC.
Keywords: ADC, cytology, ICC, NSCLC, SCC
How to cite this article: Introduction 
Carcinoma of lungs is the most frequent cancer worldwide.[1] It is also the commonest etiology of cancer-related mortality in developed countries.[1] At present, its incidence is increasing in developing nations also.[1] Its prevalence differs according to geographic regions and ethnicity.[2] The 5-year survival rate in advanced lung cancer is poor, accounting for approximately 15% and 5% in developed countries and economically developing countries, respectively.[3] Also, 1–5 million deaths occur annually worldwide, with the death rate being 2.5 million in developing countries.[4] Approximately 63,000 lung cancer cases are reported every year in India.[5]
The World Health Organization (WHO) has classified primary malignant tumors of lung into two broad categories: small cell lung cancer (SCLC) and non-SCLC (NSCLC). NSCLC constitutes approximately 80% of all lung tumors.[6] In 2015, WHO revised the classification of lung cancer into adenocarcinoma (ADC), squamous cell carcinoma (SCC), large cell carcinoma (LCC), NSCLC-not otherwise specified (NSCLC-NOS), and others.[4] This has added knowledge regarding genetic study, benefitting patients for targeted therapy.[4] Accurate subtyping of NSCLC is essential for further targeted therapy as ADC cases have better response to therapy with a good prognosis than SCC cases.
Majority of lung carcinomas present clinically in an advanced stage which are unresectable. So, pulmonary cytology plays a challenging and key role in preliminary and early categorization of lung tumors in a short time in comparison to histopathology. It is easy, cheap, rapid, less traumatic, and can be done multiple times and utilized for further molecular marker study. Various procedures like fine needle aspiration cytology (FNAC), bronchial brush cytology, Broncho-alveolar lavage fluid, sputum, and effusion fluid are frequently available for diagnostic and further therapeutic purposes.[7] Cytomorphology with corroborative immunocytochemistry (ICC) plays a paramount and efficient role for early diagnosis, followed by further prognostic as well as predictive biomarker study with almost 100% efficacy.[8] Both aid in subsequent therapeutic intervention. Commonly used biomarkers for categorization of NSCLC are TTF-1 Thyroid Trancription Factor-1 (TTF-1), Napsin-A, p63, cytokeratins, neuroendocrine markers, and others. Different prognostication molecular indices like Anaplastic Lymphoma Kinase (ALK), ROS Proto-Oncogene 1, Receptor Tyrosine Kinase (ROS-1), Epidermal Growth Factor (EGFR), Programmed Death Ligand-1 (PDL-1), and others are also used nowadays.[9] But in contrast to so many advantages, certain preanalytic difficulties like sample adequacy and proper fixation time, along with sampling from extensive necrotic areas may cause false-negative results as well as inconclusive ICC results.
Materials and Methods This ambispective study was conducted in the pathology department of a tertiary care hospital of eastern India for a 2-year period from 2018 to 2020.
Case selection
Fifty-seven cases of radiologically detected lung mass lesions were selected at the beginning. Out of these, 11 cases were rejected due to the exclusion criteria of obscuring factors like extensive tumor diathesis and low tumor cellularity, along with procedural and processing artifacts. So, the remaining 46 cytologically diagnosed cases of NSCLC were studied. Different cytological procedures like bronchial brushings, endobronchial ultrasound (USG)-guided transbronchial needle aspiration (EBUS-TBNA), and transthoracic needle aspiration cytology (TTNA) under compute d tomography (CT) guidance were used. Clinical data of the patients, such as age, sex, smoking history, clinical presentation, and imaging studies, were recorded.
Approval of ethics committee of the institution was obtained. Informed consents of all the patients were taken before the procedures and also before the ICC analysis.
Cytology
The cytology is presented in [Figure 1],[Figure 2].
Figure 1: Distribution of different cytological procedures (N = 46). EBUS-TBNA = endobronchial ultrasound-guided transbronchial needle aspiration, TTNA = transthoracic needle aspiration cytology
Figure 2: Subtyping of NSCLC in cytomorphology (N = 46). ADC = adenocarcinoma, NSCLC = non-small cell lung cancer, NSCLC-NOS = NSCLC-not otherwise specified, SCC = squamous cell carcinomaOut of a total 46 cases in our study, 76.09% cases were bronchial brushings, 13.04% (six cases) were TTNA, and 10.87% (five cases) were EBUS-TBNA cytosmears [Figure 1]. Bronchial brushing was done from the surface of tumors and suspicious mucosal lesions present in bronchi with the help of the inbuilt brush in fiber-optic flexible bronchoscope by the pulmonologists of the Department of Pulmonary Medicine of Kalinga Institute of Medical Sciences (KIMS). Smears were prepared by detaching and rubbing these brushes on uncoated glass slides. Cytosmears of EBUS-TBNA and TTNA samples were obtained under USG and CT guidance from the Department of Radiodiagnosis. All the smears obtained by the above procedures were fixed immediately for 2 min in 95% alcohol in Coplin jars and sent to the pathology department. These smears were then stained with Papanicolaou staining followed by Dibutylphthalate Polystyrene Xylene (DPX) mounting. Subtyping of NSCLC on cytology was done according to cytological nomenclature and the WHO 2015 classification of lung cancer.[10],[11] Correlation with cytological diagnosis and clinic-radiological parameters was done.
Immunocytochemistry
Immunocytochemistry is presented in [Figure 3],[[Figure 4], [Figure]5, [Figure 6],[[Figure 7].
Figure 3: (a) Coplin jar containing 95% ethanol as fixative, (b) Diamond pen used to mark back side of slide, (c) Pap pen used to mark area on smears for ICC, (d) Artificial well created by pap pen for ICC
Figure 4: (ADC): (a) Pap stained bronchial brush cytosmear showing malignant cells in clusters and scattered singly(200X), (b) Pleomorphic tumor cells in dyscohesive clusters and acinar pattern(400X), (c) Positive external control of TTF-1 of known case of papillary carcinoma of thyroid showing nuclear positivity(400X), (d) ICC of TTF-1 in ADC (200x), (e) TTF-1 positivity in ADC(400X), (f) TTF-1 positivity in ADC(1000X)
Figure 5: Distribution of types of NSCLC by ICC (N = 46). ADC = adenocarcinoma, ICC = immunocytochemistry, NSCLC = non-small cell lung cancer, SCC = squamous cell carcinoma
Figure 6: (ICC of Napsin-A in ADC): (a) Positive external control of a known case of clear cell carcinoma kidney showing cytoplasmic granular positivity(400X), (b) Positive external control of a known case of LADC from pleural fluid cell block(200X), (c)Tumor cells of ADC showing granular cytoplasmic expression of Napsin-A (200X), (d) Tumor cells of ADC showing granular cytoplasmic expression of Napsin-A (400X), (e) Cytoplasmic granular positivity with positive internal control in alveolar macrophage(1000X)
Figure 7: (SCC): (a) Pap stained bronchial brush cytosmear showing malignant cells in dyscohesive clusters and singly(200X), (b) Tadpole cells, fibre cells and dyskeratotic tumor giant cell(400X), (c) Positive external control of p63 showing nuclear positivity(400X), (d) p63 expression in tumor cells showing nuclear positivity(400X), (e) p63 expression in tumor cells showing nuclear positivity(1000X)The areas showing tumor cells on the smears were marked on the opposite side of the slides with the help of a diamond marker. Then the smears were kept in xylene for 3–72 h for removal of the cover slips. Artificial wells were created using pap pen on marked sites of the slides. ICC evaluation of TTF-1, Napsin-A and p63 was done over the precreated wells by using indirect polymer two-step method. Mouse monoclonal antibody (Ab) for TTF-1 from Master Diagnostica and rabbit polyclonal Abs for Napsin-A and p63 from Quartett were used. Correlation between cytomorphological diagnoses and immunocytochemical results was done.
Histopathology
[Figure 8] shows the histopathology results of the study.
Figure 8: Histopathology (H&E 400x): (a) [ADC]-Tumor cells in acinar pattern, (b) [SCC]-Pleomorphic tumor cells in diffuse patternThe core needle biopsy samples of radiologically detected lung mass lesions were received in 10% buffered formalin at the pathology department. These were routinely processed for 16 h by an automated Leica tissue processor. Tissue blocks were made after embedding with paraffin. Then 3–4 μm tissue sections were obtained with the help of Leica microtome. These slides were stained by hematoxylin and eosin. Subtyping of NSCLC was done as per the 2015 classification of lung cancer by WHO.[4] Histopathologic diagnosis was correlated with cytological diagnosis in 25 cases.
Finally, cytological, histomorphological, and ICC correlation was done in 25 cases only. Clinico-radiological correlation was also done.
Quality control
[Figure 4], [Figure 5], [Figure 6] and [Figure 7] show external and internal controls for TTF1, Napsin and p63. External positive and negative control slides were used with each batch of ICC staining. For external positive controls for TTF-1, Napsin-A, and p63, cytosmears of known cases of papillary thyroid carcinoma, renal cell carcinoma, and SCC were taken respectively. Negative control involved omission of primary Ab and incubation in phosphate-buffered saline (PBS) in a test case in each batch of staining. Type 2 pneumocytes and alveolar macrophages in cytosmears were taken as the internal control for TTF-1 and Napsin-A, respectively.[12],[13]
TTF-1 and p63 show strong nuclear positivity, whereas Napsin-A shows coarse granular cytoplasmic positivity. The test was considered positive for all the immunomarkers if ≥5% of tumor cells showed appropriate staining patterns.
Results In our study, 50–60 years was the common age group with male preponderance. Smoking was more associated with SCC. In most cases, tumor was peripherally located in ADC cases and centrally located in SCC cases.
Cytological subtyping of NSCLC
[Figure 2] and [Figure 5] show the cytological subtyping of NSCLC.
On cytomorphology, of the 46 cases of NSCLC, the most common cancer was ADC comprising 32.61% (15 cases). This was followed by SCC in 23.91% (11 cases), NSCLC-favoring ADC in 17.39% (eight cases), and NSCLC-NOS in 8.70% (four cases). Three cases (6.52%) could not be subtyped on cytomorphology, so suspicious of malignancy was given. Two cases (4.35%) were NSCLC-favoring SCC and one case (2.17%) was lepidic pattern ADC. In the remaining two cases (4.35%), only a few atypical cells were seen cytologically for which diagnosis could not be done and presence of atypical cells was mentioned in the reports.
Subtyping of NSCLC by ICC in cytology
[Table 1] and [Figure 5], [Figure 6], [Figure 7] show subtyping of NSCLC by ICC in cytology.
On ICC evaluation, 29 cases (63.04%) and 14 cases (30.43%) were subtyped as ADC and SCC, respectively [Figure 2].
Out of 29 cases of ADC, in 26 cases (56.64%), the tumor cells showed positive expression of both TTF-1 and Napsin-A. Only TTF-1 showed positive expression in two cases (4.34%). One case (2.17%) revealed focal expression of p63 and Napsin-A in <5% of neoplastic cells, with a diffuse positivity of TTF-1.
In 12 cases (26.10%) out of 14 cases of SCC, >5% tumor cells showed strong nuclear positive expression of p53. One case (2.17%) each showed focal positivity of Napsin-A and TTF-1 in <5% of tumor cells, along with diffuse p63 expression in tumor cells.
One case (2.17%) out of 46 cases showed positive expression of all the three markers and was diagnosed as adenosquamous carcinoma. The remaining cases (4.34%) showed negative expression for the three markers and were reported as unclassified.
Our study showed significant statistical correlation of TTF-1 and Napsin-A expression in ADC and p63 expression in SCC with a P value of 0.01.
Correlation between cytology and ICC
[Table 2] shows the correlation between cytology and ICC.
Out of 16 cases of ADC, 14 cases (87.50%) showed positive correlation between cytology and ICC diagnosis. In SCC, correlation was found in nine cases (81.81%) out of 11 cases. Seven cases (15.22%) out of eight patients (17.39%) diagnosed as NSCLC-favoring ADC cytologically were labeled as ADC by ICC. One case (2.17%) out of two cases (4.35%) diagnosed as NSCLC-favoring SCC cytologically was SCC in ICC and the other case was unclassified in ICC. Out of three cases given as suspicious of malignancy in cytology, two cases were typed as ADC and one case was given as unclassified in ICC. Of the four cases reported as NSCLC-NOS cytologically, two cases each were diagnosed as ADC and SCC in ICC. Two cases given as atypical cells in cytology were proven to be ADC by ICC. Our study showed a positive correlation between cytology and ICC in 87.5% and 81.8% cases of ADC and SCC, respectively. Thus, in our study, significant statistical correlation was found between cytology and ICC with a P value of 0.01 by using Fisher exact test.
Correlation between histology and ICC
[Table 3] shows the correlation between histology and ICC.
A total of 25 biopsy cases were available for histology and ICC correlation in our study. There was correlation in 85.5% ADC and 66.66% SCC cases with a P value of 0.012 by using Fisher exact test.
Correlation between cytology, histology, and ICC
[Table 4] shows the correlation between cytology, histology, and ICC.
In our study, 25 cases were available for cytology, histology, and ICC correlation. Concordance was present in 14 cases (56%). Seven cases each (28%) of ADC and SCC showed similar results in ICC. In our study, the overall cytology–histology correlation was found in 16 cases out of 25 cases, that is, in 60% of cases. Out of 25 cases, concordance between cytology–histomorphology and ICC was found in 14 cases (56%).
Statistical analysis
[Table 5] shows the results of statistical analysis.
Table 5: Correlation of sensitivity, specificity, PPV, and NPV of TTF 1, Napsin A, and p63 with other studiesSpecificity and sensitivity along with positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated. Statistical analysis was done by using Microsoft Excel and Statistical Package for the Social Sciences (SPSS) software. A P value of < 0.05 using Chi-squared test was considered as significant.
TTF-1 and p63 were 100% sensitive, whereas Napsin-A was 83% sensitive in our study. Specificity for p63 was 93.8%. PPV value for TTF1 and p63 was 93.5% and 87.5%, respectively. NPV for both TTF-1 and Napsin-A was 100%. Accuracy was 95.65% for both TTF-1 and p63. It was 91.3% for Napsin-A.
Discussion Recent advances in targeted therapy in lung cancer have increased the importance of subtyping of NSCLC into ADC and SCC. Tyrosine kinase inhibitors (TKIs) and antiangiogenic factors have been used for targeted therapy in ADC cases.[16],[17] Drugs like bevacizumab are highly contraindicated in SCC as they cause severe pulmonary hemorrhage.[18] Recent phase 3 study showed significant survival of ADC cases on cisplatin or pemetrexed combination therapy, whereas cisplatin/gemcitabine was superior in patients with SCC of lungs.[19]
More than 70% of lung cancer patients usually come in advanced stages which are unresectable. So, only cytology samples and core needle biopsy are frequently available for diagnosis as well as further prognostic marker evaluation.[10] Pulmonary cytology plays a challenging diagnostic tool in lung cancer. In this scenario, cytopathologists face an exigent role due to increasing demand for specific diagnosis from a limited available sample. However, precise judgment with presence of few tumor cells in cytology as well as core needle biopsy samples is possible with the use of immunostaining. Here, ICC has a pivotal role as it shows diagnostic, prognostic, and predictive value with limited cytology samples. In the present study, we have used commonly used ICC markers like TTF-1, Napsin-A, and p63 and also studied the sensitivity and specificity of these novel markers along with clinico-radiological correlation.
TTF-1, a novel marker for primary ADC of lung, shows positive expression in approximately 80% of primary lung carcinoma cases.[10] As a nuclear marker, it is detected easily in single isolated cells and also in clusters that are even present in a necro-hemorrhagic background of cytosmears. TTF-1 loses its immunoreactivity in poorly differentiated ADC. In these cases, Napsin-A is helpful. Napsin-A has lung specificity and is used as an adjunct to TTF-1. Both the markers are highly specific and sensitive for Lung adeno carcinoma (LADC) with the exception of mucinous carcinoma.[20],[21],[22] TTF-1 and Napsin-A show positive expression in type 2 pneumocytes as well as alveolar macrophages, which may cause false-positive interpretation. p63 plays a paramount role in the diagnosis of most cases of SCC. Focal false positivity of p63 in ADC may occur due to entrapped bronchial epithelial cells or reserve cells. However, combination of above markers plays a vital role in categorization of NSCLC in cytology samples.[23],[24]
In our study, sensitivity of TTF-1 and Napsin-A was 100% and 92%, respectively, in ADC. Specificity of both the markers was 88%. p63 was 100% sensitive with 94% specificity in cases of SCC. TTF-1, Napsin-A, and p63 had PPV of 93%, 91%, and 87%, respectively. TTF-1 and p63 had NPV of 100%, whereas it was 88% for Napsin-A. This statistical analysis was comparable to the studies done by Gurda et al.,[15] Roh et al.,[25] Sinna et al.,[26] and Van Zyl et al.,[27] and Kargi et al.[28]
On cytological evaluation, ADC was the commonest cancer followed by SCC in our study. Atta,[13] Gurda et al.,[15] Roh et al.,[25] Sinna et al.,[26] Van Zyl et al.[27] and Alekhya et al.[29] also obtained similar results as in our study.
On ICC evaluation, 63%, 31%, 2%, and 4% cases were subtyped as ADC, SCC, Adenosquamous carcinoma, and unclassified, respectively. Out of 63.05% ADC cases, 56.64% cases had positive expression of TTF-1 and Napsin-A. Of 14 cases (30.44%) of SCC, 12 cases (26.10%) showed only p63 positivity in >5% of tumor cells. Gurda et al.,[15] Roh et al.,[25] Sinna et al.[26] and Van Zyl et al.[27] also obtained similar correlation and stated that in more than 95% cases, subtyping of NSCLC can be done by a combined study of cytology and ICC. Our study showed cytology and ICC correlation in 87.50% cases of ADC and 81.81% cases of SCC. Histologically diagnosed ADC cases showed 85.71% correlation with ICC, whereas correlation for SCC was 66.66%. Thus, in our study, cytology–histology correlation was present in 60% cases. Concordance between cytology–histology and ICC was present in 14 (56%) out of 25 cases. This correlation was comparable with that obtained by Gurda et al.[15] and Pavani et al.[30]
Conclusion To conclude, ICC is a cost-effective, easy, and time-saving method in diagnostic lung cytopathology. With adequate viable tumor cells, it plays a vital and crucial role in the diagnosis and subtyping of NSCLC, which impedes further surgical intervention in advanced stages and clears the pathway of further molecular study and targeted therapy.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References 
Source of Support: None, Conflict of Interest: None
CheckDOI: 10.4103/joc.joc_5_22
Comments (0)