Fine-Needle Aspiration Cytology of Benign and Malignant Tumors of the Lung

The cytologic diagnosis of lung cancer can be made by the
evaluation of exfoliated cells in sputum and/or cells obtained
by transbronchoscopic techniques including bronchial washing,
bronchial brushing, bronchoalveolar lavage, and by fine-needle
aspiration biopsy (FNAB), most commonly, transthoracic
FNAB under computed tomography (CT) guidance. The sensitivity
of a single sputum specimen for the detection of lung
cancer is approximately 50% and increases with the number of
specimens examined. 1 The highest sensitivity is obtained for centrally
localized squamous cell carcinoma and the lowest for small
cell carcinoma. For endobronchial lesions that can be directly visualized
with a bronchoscope, the sensitivity of a single bronchial
washing or brushing for detecting lung cancer is about 65%,
similar to endobronchial or transbronchial biopsy. 2 Combining
sputum cytology with bronchial brush cytology increases the
sensitivity than with either method alone. 3 Bronchoalveolar
lavage utilizing narrow-diameter bronchoscopes has the higher
yield for the detection of lung cancer in peripheral lesions with a
reported sensitivity of 35% to 65%. 4 By far, the most reliable cytologic
method for diagnosing a localized peripheral lung lesion
is transthoracic CT-guided FNAB, which has a sensitivity that
exceeds 85% for the detection of lung cancer. Metaanalyses of
transthoracic needle aspiration biopsy reports show a high sensitivity
(0.88 to 0.99) and specificity (0.99 to 1.00) for the diagnosis
of lung cancer. 5 Most of the work on cytology screening
for lung cancer has been on sputum specimens for the diagnosis
of squamous cell carcinoma and its precursors. Lung cancer is
thought to arise from a series of sequential preneoplastic changes
that have been well defined for centrally arising squamous cell
carcinoma but not other lung cancer subtypes. There are different
patterns of molecular alterations in small cell carcinoma,
squamous cell carcinoma, and adenocarcinoma. For example,
Wistuba and colleagues 6 found a higher incidence of deletions at
17p13 (TP53), 13q14 (RB), 9q21 (p16 INKa), 8p21–23, and
several 3p regions in squamous cell carcinoma than in adenocarcinoma.
On the other hand, K- RAS mutations have been detected
in bronchoalveolar lavage fluids from patients with adenocarcinoma
but not in patients with other lung cancer subtypes.
It has been hypothesized that molecular analysis of sputum
for biomarkers of lung cancer may provide an effective means
of screening smokers to enable early lung cancer detection.
However, the use of biomarkers for lung cancer screening is
complicated by the fact that they can be detected in chronic
smokers long before any clinical evidence of neoplasia. For example,
p16 promotor hypermethylation and TP53 have been
detected in sputum in chronic smokers without lung cancer and
K- RAS mutations have been detected in atypical adenomatous
hyperplasia (AAH). 7 At our institution, the focus has shifted
from the use of cytology for screening for early lung cancer to the
diagnosis of CT screen–detected lung cancer by transthoracic
FNAB. It is minimally invasive, does not require hospitalization,
and serious risks are uncommon and limited to pneumothorax
and minimal hemorrhage. The FNAB is performed using a 22-
gauge Westcott needle with immediate on-site assessment of airdried
smears utilizing the Diff-Quik staining method (Dade AG,
Dudingen, Switzerland). Smears are also submerged in alcohol
for Papanicolaou staining and blood clots are submitted directly
in formalin for a cell block preparation.