Baseline screening for lung cancer of 2968 high-risk men and women utilizing HRCT enrolled in ELCAP (Early Lung Cancer Action Project) was performed between 1993-2002. Among them, 65 people had surgical resection of their screen-diagnosed lung cancer, 53 of them on the basis of a diagnosis of malignancy or atypical bronchioloalveolar proliferation (ABP) on fine needle aspiration (FNA) biopsy at Weill Medical College of Cornell University (WMC) prior to surgery. The authors compared the diagnosis obtained from the FNA with the subsequent diagnosis from the surgical specimen to assess the reliability of a cytologic diagnosis of lung cancer on FNA of these screen-diagnosed lung cancers.
The FNA biopsies were performed with a 22-gauge Wescott needle by 1 radiologist (D.Y.), with preliminary on-site as well as final diagnosis rendered by a cytologist (M.V., J.K.). These results were correlated with histologic diagnoses obtained as a result of consensus diagnosis by a panel of 5 expert pulmonary pathologists.
Of the 53 cases of lung cancer resected following FNA, 4 were diagnosed as atypical bronchioloalveolar proliferation (ABP), 14 as adenocarcinoma with bronchioloalveolar features (ADC-BAC), 28 as adenocarcinoma, not otherwise specified (ADC-NOS), 1 as squamous cell carcinoma (SQCC), 4 as nonsmall-cell carcinoma (NSCC), and 2 as typical carcinoid. In the 49 cases with a malignant cytology and 4 cases of ABP, lung cancer was confirmed histologically. The tumor sizes ranged from 4 mm to 40 mm, mean size 13mm. The final expert panel histologic diagnosis was adenocarcinoma in 47 cases; of these, 42 were invasive (mixed subtype or acinar subtype), and 5 were a noninvasive (bronchioloalveolar carcinoma, BAC). Three of the 42 invasive adenocarcinoma that had a predominant BAC component and 1 case of BAC were diagnosed as ABP on FNA; all were sampled at the periphery of the tumor. Three of 4 cases of invasive adenocarcinoma of high nuclear grade were diagnosed as NSCC, and 1 was inaccurately classified as SQCC on FNA. One case of high-grade noninvasive BAC associated with a scar was diagnosed as NSCC on FNA. Two cases classified as nonkeratinizing SQCC and 2 cases of large cell neuroendocrine carcinoma on histology were misclassified as ADC-NOS by FNA. Two cases of typical carcinoid on histology were also diagnosed as typical carcinoid on FNA.
Computed tomography (CT) screening in the Early Lung Cancer Action Program (ELCAP) starts with the initial low-dose CT test and if the result is positive a work-up proceeds along a well-defined regimen ultimately leading to diagnosis.1 An important feature of the regimen is growth assessment. If growth consistent with malignancy is documented on scans obtained after a positive baseline scan, particularly for nodules less than 15 mm in diameter, transthoracic CT-guided fine-needle aspiration (FNA) biopsy is recommended. This is the preferred method of biopsy as it is an outpatient procedure and has a high diagnostic yield for malignancy as demonstrated by meta-analyses of transthoracic needle aspiration biopsy reports.2 Its sensitivity (0.88-0.99) and specificity (0.99-1.00) are high for the diagnosis of lung cancer. It is minimally invasive, does not require hospitalization, and serious risks of FNA biopsy are uncommon, with complications generally being limited to pneumothorax and minimal hemorrhage. Typically, however, CT screen-diagnosed cancers are small and the diagnostic yield of FNA biopsy depends on the expertise, tumor size, location, and the method of tumor cell sampling. Additionally, nodules diagnosed as atypical adenomatous hyperplasia or reactive atypia of pneumocytes, a benign diagnosis, may be indistinguishable on cytology from adenocarcinoma.3, 4
In this study we focused on the reliability of a cytologic diagnosis of malignancy obtained by FNA biopsy of lung nodules measuring 4 to 40 mm (mean size, 13 mm) detected in the baseline screening at Weill Medical College (WMC) of Cornell University. In this way we focus on cytologic criteria for malignancy rather than concerns related to sampling error due to technical challenges that can occur in the course of these procedures.
MATERIALS AND METHODS
As a result of baseline CT screening from 1993 to 2002, a total of 53 people were diagnosed cytologically with findings diagnostic or suggestive of lung cancer.5 All of these patients underwent surgical resection and were evaluated for histology.
The FNA biopsy in all cases was performed using a 22-gauge Westcott needle.6 Once CT documented the needle as being inside the nodule, a 10 cc syringe was attached and a vacuum was created while the needle was moved back and forth within the nodule to aspirate a cytologic sample. After the radiologist obtained the aspirate, the syringe with needle attached was given to the cytologist for preparation of smears. Air-dried smears were stained by the Diff-Quik method (Dade AG, Dudingen, Switzerland) for immediate assessment. Smears were also submerged in 95% alcohol for staining by the Papanicolaou method and blood clots were submitted in formalin for cell block preparation. The needle was rinsed in Cytolyt (Cytyc, Marlboro, Mass) for ThinPrep slides and for special stains. On-site cytology review was performed and a preliminary cytologic diagnosis was given. In some cases, when the cytologist found that the specimen was not adequate for diagnosis, additional needle passes were requested. The final diagnosis was made based on all of the cytologic preparations utilizing standard cytologic criteria.7, 8 The initial concern was the identification of cancer and, if possible, its classification as small-cell or nonsmall-cell carcinoma (Fig. 1). A diagnosis of adenocarcinoma was subclassified whenever possible utilizing a combination of nuclear and cytoplasmic features9-13 (Table 1; Figs. 2, 3).
Table 1. Criteria for the Cytologic Diagnosis of Adenocarcinoma/ABP on Transthoracic FNA
I. Monolayered clusters of cuboidal to dome shaped epithelial cells with fine chromatin, pinpoint nucleoli and fine cytoplasmic vesicles.
II. Clusters of bland epithelial cells with round nuclei showing nuclear clearing, intranuclear inclusions and pinpoint nucleoli (Fig. 3). The cytoplasm may be columnar with apocrine snouts.
With mucinous BAC features
Glands with basally placed nuclei and clear to orangeophilic columnar cytoplasm with luminal margins. The nuclei have evenly dispersed fine chromatin, nuclear grooves and pinpoint nucleoli.
Atypical bronchioloalveolar proliferation (ABP)
Atypical bronchioloalveolar cells with nuclear clearing, intranuclear inclusions and pinpoint nucleoli accompanied by a histiocytic infiltrate (Fig. 4).
The cytologic diagnoses were reviewed separately by 2 cytologists (M.V., J.K.) according to the ELCAP pathology protocol14, 15 and their final consensus diagnosis is used here. This pathology protocol included a new diagnostic category designated “atypical bronchioloalveolar cell proliferation” (ABP) when the findings were suspicious for, but not diagnostic of, adenocarcinoma with bronchioloalveolar features (nonmucinous subtype). This diagnosis was rendered when there were scant clusters of small bland, yet atypical, bronchioloalveolar cells with uniform round nuclei having pinpoint nucleoli and, frequently, intranuclear grooves, intranuclear chromatin clearing, and intranuclear inclusions (Fig. 4). It should be emphasized that ABP is a cytologic classification distinct from a histologic diagnosis of atypical adenomatous hyperplasia (AAH) and was initially utilized in the ELCAP protocol to identify adenocarcinomas with a specific low-grade cytomorphology, which may mimic benign or borderline bronchioloalveolar cell proliferations.16
Nodule consistency on CT was described as being solid, part-solid, and nonsolid. A nodule was considered to be solid if the nodule obscures the entire lung parenchyma within it, part-solid if it obscures part of the lung parenchyma within it, and nonsolid if it obscures none of the parenchyma within it.17 The location of the needle tip in the solid or nonsolid part of the lesion was reviewed by the radiologist who performed the biopsy (D.Y.) together with a cytologist (M.V.).
The surgical pathology specimens were processed according to the ELCAP pathology protocol.14, 15, 18 The histologic diagnoses were based on the consensus diagnosis rendered by a panel of 5 experts in pulmonary pathology: Dr. Darryl Carter (Yale University), chair, Dr. Elisabeth Brambilla (University Hospital Center of Grenoble), Dr. Adi Gazdar (University of Texas Southwestern Medical Center), Dr. Masayuki Noguchi (University of Tsukuba), and Dr. William Travis (Memorial Sloan Kettering).18
Table 2 gives the initial cytologic diagnosis of the 53 FNA biopsies that suggested malignancy and the consensus diagnosis based on histology. Malignancy was histologically confirmed in all cases classified as malignant or ABP by cytology. Among the 42 cases with cytologic diagnosis of adenocarcinoma with BAC features (ADC-BAC) or not otherwise specified (NOS), 38 of these had a final diagnosis of adenocarcinoma. Of the 14 identified as ADC-BAC cytologically, only 3 were confirmed to be BAC histologically. The remainder were invasive adenocarcinomas, mixed subtype with a BAC component. Out of 28 cases diagnosed cytologically as adenocarcinoma, NOS, 24 were found to be adenocarcinoma histologically, mixed or acinar subtypes. There were a total of 4 cases diagnosed cytologically as adenocarcinoma, NOS that were found to be either squamous cell carcinoma (2 cases), or large-cell neuroendocrine carcinoma (2 cases). Among the 4 cases diagnosed cytologically as ABP, all were adenocarcinoma (3 invasive, mixed, and 1 BAC).
Table 2. Cytologic-Histologic Correlation of Early Lung Cancer
Squamous cell Ca
ABP indicates atypical bronchioloalveolar proliferation; ADC-BAC, adenocarcinoma with bronchioloalveolar features; ADC-NOS, adenocarcinoma, not otherwise specified; Ca, carcinoma; BAC, bronchioloalveolar carcinoma.
Invasive, mixed or acinar
One case diagnosed cytologically as squamous cell and 4 cases diagnosed as nonsmall-cell were all found histologically to be adenocarcinoma (4 invasive and 1 BAC).
Two cases diagnosed cytologically as typical carcinoid were confirmed histologically (Fig. 5).
We found that FNA biopsy was highly reliable in the diagnosis of early lung cancer identified by CT screening. All cases found to be malignant or ABP on cytology proved to be malignant after histologic examination of the surgical specimen. Among a total of 33 cases where the cytology suggested an invasive subtype, all but 1 were found to be invasive. However, among the 18 cases where cytology suggested a BAC component (14 ADC-BAC and 4 ABP), only 4 were found to be pure BAC. Because BAC is defined as an adenocarcinoma without invasion of basement stroma, blood vessels, or pleura,16 FNA biopsy does not permit a definitive diagnosis of BAC because invasion cannot be excluded. Although the cytology and nonsolid appearance of the nodule can suggest a prominent BAC (or noninvasive) component, a microscopic focus of invasion may be identified on complete histologic examination and a diagnosis of invasive adenocarcinoma is made.16 It is a recommendation of the ELCAP pathology protocol to estimate the percentage of each component of the adenocarcinoma if it is of the mixed subtype.15 The ELCAP pathology protocol introduced the diagnostic category of ABP recognizing the difficulty in differentiating between AAH from BAC and BAC from microinvasive adenocarcinoma (invasive adenocarcinoma with predominant BAC component). Cases with this diagnosis frequently present as a nonsolid nodule on CT.19, 20 On FNA, these lesions tend to yield a population of atypical bronchioloalveolar cells with subtle nuclear features including nuclear clearing, chromatin margination, and cytoplasmic invaginations (intranuclear grooves). All of the cases diagnosed as ABP cytologically in this study had a nonsolid component.
In the 4 cases that were diagnosed as ABP by cytology, 3 had a low nuclear grade on cytology and were subsequently confirmed to be invasive adenocarcinoma with variable BAC components, whereas 1 case was confirmed as BAC. Of the 5 histologically confirmed cases of BAC, 3 were of low nuclear grade and had a cytologic diagnosis of ADC-BAC. On the other hand, 1 was of high nuclear grade BAC and this case was diagnosed as nonsmall-cell carcinoma by FNA biopsy. Although certain cytologic features have been proposed to favor a diagnosis of BAC over invasive adenocarcinoma, this diagnosis requires thorough histologic evaluation to exclude any area of invasive growth. The cells in nonmucinous BAC tend to be of low nuclear grade with homogeneous uniform, round smooth, and pale nuclei with inconspicuous nucleoli. There is an apparent overlap of cytologic features in AAH and nonmucinous BAC and currently there are no established criteria for distinguishing AAH from BAC on cytology.21–23 Of note, in this series a diagnosis of ABP had a positive predictive value (PPV) of 100% for the identification of minimally invasive adenocarcinoma, mixed subtype, and BAC. Given the Japanese studies that suggest that these lesions have a 100% survival after wedge resection,24-26 the cytologic category of ABP may serve to preoperatively identify patients that require less invasive treatments.
The cytologic diagnosis for adenocarcinoma was very reliable with 42 out of 47 cases correctly classified. Typically, adenocarcinoma on FNA biopsy presented as cohesive clusters of cells with round, ovoid, or “dome-shaped” cytoplasm, and uniformly round or ovoid nuclei with conspicuous nucleoli. The chromatin quality varied with the degree of differentiation: well-differentiated tumors have finely granular chromatin that is evenly distributed throughout the nucleus, poorly differentiated tumors have hyperchromatic nuclei with coarse chromatin. The presence of glandular and papillary forms and/or intracellular or extracellular mucin can confirm the diagnosis of adenocarcinoma (Fig. 2). Whereas 4 cases of squamous-cell and large-cell carcinoma had been classified as adenocarcinoma by cytology, it is known that the cytology of squamous cell carcinoma and large-cell carcinoma can mimic poorly differentiated adenocarcinoma. In this series we found that subclassification cannot be reliably made in cases of nonsmall-cell carcinoma (NSCC) with high nuclear grade, eccentric cytoplasm, and prominent nucleoli (Fig. 1). Based on this experience, we now only make a diagnosis of NSCC in cases with this cytomorphology unless immunohistochemistry, generally performed on cell block material (Fig. 5), supports a more specific diagnosis; for example, positive synaptophysin and/or chromogranin in large-cell neuroendocrine carcinoma.
Although we had no mucinous BAC in this series diagnosed on FNA biopsy, it typically shows a prominence of nuclear grooves and an abundance of extracellular mucin when compared with invasive adenocarcinoma.9 In 1 series, prominence of monolayered tumor cells, fine chromatin, and mild cellular pleomorphism were the cytologic features that correlated with a histologic diagnosis of nonmucinous BAC, whereas mucinous BAC showed a prominence of nuclear grooves and an abundance of extracellular mucin when compared with invasive adenocarcinoma.9
The neoplastic cells in nonmucinous BAC less than 2 cm in size are generally less pleomorphic than cells from invasive adenocarcinoma. FNA biopsy of the nonmucinous variant of BAC commonly shows monolayered clusters of cuboidal to dome-shaped epithelial cells with fine chromatin, pinpoint nucleoli, and cyanophilic cytoplasm with fine vesicles.10, 12, 13 In some cases there may be intranuclear chromatin clearing and prominence of intranuclear grooves and intranuclear inclusions as seen in papillary thyroid carcinoma (Fig. 3). The nuclear size and degree of nuclear atypia can vary.22 Some have columnar cytoplasm with apocrine snouts and appear to derive from Clara cells. There is a prominent histiocytic infiltrate that can accompany the neoplastic cells (Fig. 4). This is likely due to the accumulation of macrophages in alveolar spaces that are lined by thickened alveolar septa but without destruction of the alveolar architecture. The distinction with reactive atypia of bronchioloalveolar cells from BAC can be difficult if not impossible by FNA. Reactive pneumocytes tend to show more nuclear pleomorphism, binucleation, and enlargement than BAC and the nucleoli are also more conspicuous in reactive pneumocytes. In general, a background of acute and/or granulomatous inflammation accompanies the atypical alveolar cells in infectious processes.3
In conclusion, we found that FNA was highly reliable in the diagnosis of early lung cancer. In this series of CT-screened detected early lung cancer the diagnosis of NSCC and typical carcinoid was 100%. For the subclassification of NSCC as adenocarcinoma, the PPV was 92.3%. Features consistent with high nuclear grade on cytology were highly suggestive of an invasive type of cancer. However, misclassification as adenocarcinoma occurred in 2 cases of SQCC and 2 cases of large-cell neuroendocrine carcinoma. Additionally, with respect to adenocarcinoma, although low nuclear grade on cytology is highly predictive of a BAC component, it does not reliably predict a BAC (or noninvasive) subtype. This is primarily due to sampling, in that early adenocarcinomas showing only limited areas of microscopic invasion are likely to yield BAC cytomorphology from the noninvasive component.