This chapter focuses on three areas that have received considerable interest recently. The first is the use of small biopsies and cytology, which are used not only for tissue diagnosis but also for molecular studies in lung cancer. The second is the classification of lung tumors, which has changed considerably based on several consensus conferences. This includes more precise classification of lung adenocarcinomas, which frequently have a mixture of subtypes. This is important for diagnosis, predicted outcome, and therapy. The third area is the field of interstitial lung disease (ILD), which now has a firm foundation for clinical studies and evaluating new therapies. These are areas which have seen considerable advances in the past 5 years. Topics such as the pathology of asthma, chronic obstructive pulmonary disease, pneumonia, and infectious disease are addressed in other chapters of this book and in the key references.
Lung cancer is the most common cause of major cancer incidence and mortality worldwide in men, while in women it is the third most common cause of cancer incidence and the second most common cause of cancer mortality. In 2013, the American Cancer Society estimated that lung cancer would account for more than 228,190 new cases in the United States and 159,480 cancer deaths.
The pathologic diagnosis of lung cancer is established either by a histologic or cytologic approach. Major changes in the diagnosis of lung cancer were recommended by the 2011 International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society (IASLC/ATS/ERS) International Multidisciplinary Classification of Lung Adenocarcinoma ( Table 14-1 ). Changes have been made in the diagnostic approach to small biopsies and cytologies as well as to resection specimens ( eTable 14-1 ). This is of great importance because 70% of lung cancer patients present with advanced stage disease and the diagnosis is based on small specimens. There are four major histologic types of lung cancer including squamous cell carcinoma, adenocarcinoma, small cell carcinoma, and large cell carcinoma. These major types can be classified into more specific subtypes such as the lepidic predominant subtype of adenocarcinoma or the basaloid variant of squamous cell carcinoma.
|Variants of Invasive Adenocarcinoma|
|Minimally Invasive Adenocarcinoma (≤3 cm lepidic predominant tumor with ≤5 mm invasion)—nonmucinous, mucinous, mixed mucinous/nonmucinous|
|2004 WHO Classification||Small Biopsy/Cytology: IASLC/ATS/ERS|
|Morphologic adenocarcinoma patterns clearly present: |
Adenocarcinoma, describe identifiable patterns present (including micropapillary pattern not included in 2004 WHO classification)
If pure lepidic growth—mention an invasive component cannot be excluded in this small specimen.
|No 2004 WHO counterpart—most will be solid adenocarcinomas||Morphologic adenocarcinoma patterns not present (supported by special stains): |
Non–small cell carcinoma, favor adenocarcinoma
|Bronchioloalveolar carcinoma (nonmucinous)||Adenocarcinoma with lepidic pattern (if pure, add note: an invasive component cannot be excluded)|
|Bronchioloalveolar carcinoma (mucinous)||Mucinous adenocarcinoma (describe patterns present)|
|Fetal||Adenocarcinoma with fetal pattern|
|Mucinous (colloid)||Adenocarcinoma with colloid pattern|
|Signet ring||Adenocarcinoma with (describe patterns present) and signet ring features|
|Clear cell||Adenocarcinoma with (describe patterns present) and clear cell features|
|SQUAMOUS CELL CARCINOMA |
|Morphologic squamous cell patterns clearly present: |
Squamous cell carcinoma
|No 2004 WHO counterpart||Morphologic squamous cell patterns not present (supported by stains): |
Non–small cell carcinoma, favor squamous cell carcinoma
|SMALL CELL CARCINOMA||Small cell carcinoma|
|LARGE CELL CARCINOMA||Non–small cell carcinoma, not otherwise specified (NOS)|
|Large cell neuroendocrine carcinoma (LCNEC)||Non–small cell carcinoma with neuroendocrine (NE) morphology (positive NE markers), possible LCNEC|
|Large cell carcinoma with NE morphology (LCNEM)||Non–small cell carcinoma with NE morphology (negative NE markers)—see comment |
Comment: This is a non–small cell carcinoma where LCNEC is suspected, but stains failed to demonstrate NE differentiation.
|ADENOSQUAMOUS CARCINOMA||Morphologic squamous cell and adenocarcinoma patterns present: |
Non–small cell carcinoma, NOS, (comment that glandular and squamous components are present)
Comment: This could represent adenosquamous carcinoma.
|No counterpart in 2004 WHO classification||Morphologic squamous cell or adenocarcinoma patterns not present but immunostains favor separate favor glandular and adenocarcinoma component |
Non-small cell carcinoma, NOS, (specify the results of the immunohistochemical stains and the interpretation)
Comment: This could represent adenosquamous carcinoma.
|SARCOMATOID CARCINOMA||Poorly differentiated NSCLC with spindle and/or giant cell carcinoma (mention if adenocarcinoma or squamous carcinoma are present)|
Historically the most important distinction was between small cell lung carcinoma (SCLC) and non–small cell lung carcinoma (NSCLC) due to significant differences in clinical presentation, spread of tumor, and response to therapy. However, over the past decade, major changes in the approach to diagnosis and treatment of NSCLC have led to a greater need for more precise classification based on small biopsies and cytology. Several therapeutic advances have provided the need for pathologists to classify tumors more precisely because the choice of therapies and decision to perform molecular testing is based on histology. Molecular testing for epidermal growth factor receptor (EGFR) mutations and activin receptor–like kinase (ALK) rearrangements is recommended for patients with adenocarcinomas, “NSCLC, favor adenocarcinoma,” and NSCLC-not otherwise specified (NSCLC-NOS). Patients with an EGFR mutation are eligible for EGFR tyrosine kinase inhibitors and those with echinoderm microtubule-associated protein-like 4 (EML4)-ALK rearrangements are eligible for crizotinib therapy. If neither EGFR mutations nor ALK rearrangements are present, patients are eligible for either pemetrexed or bevacizumab-based regimens. However, patients with squamous cell carcinoma are not eligible for these therapies. These advances have transformed the lung cancer field and resulted in multiple paradigm shifts in the clinical practice for all specialists, including pathologists. For additional information on lung cancer, see Chapter 51 , Chapter 52 , Chapter 53 .
Adenocarcinomas represent 36% of all lung cancers in the United States. The 2011 IASLC/ATS/ERS lung adenocarcinoma classification recommended multiple major changes (see Table 14-1 ). First, it is recommended to discontinue the use of the term bronchioloalveolar carcinoma (BAC) because the tumors formerly classified under this term are now classified as five different tumors. Second, there are new concepts of adenocarcinoma in situ (AIS) (see Preinvasive Lesions ) and minimally invasive adenocarcinoma (MIA). Third, it is recommended to stop using the term “mixed subtype” and to use comprehensive histologic subtyping to estimate the percentage of histologic patterns in 5% increments within a tumor with the final classification according to the predominant subtype. Fourth, tumors with a predominant component formerly called “nonmucinous BAC” should be classified as lepidic – predominant adenocarcinoma (LPA). Lepidic refers to the noninvasive growth of tumor cells along the surface of the air spaces. Fifth, micropapillary adenocarcinoma is recognized as a new subtype with a poor prognosis. Sixth, invasive mucinous adenocarcinoma is the term recommended for those tumors formerly classified as mucinous BAC. Finally, specific terminology and diagnostic criteria are proposed for tumors in small biopsies and cytology specimens along with recommendations for strategic management of tissue and EGFR mutation testing in patients with advanced adenocarcinoma.
Adenocarcinoma Classification in Resected Specimens
Classification of overtly invasive adenocarcinomas is now made according to the predominant subtype. This is best determined using comprehensive histologic subtyping to estimate the percentages of the various histologic subtypes within a tumor in a semiquantitative fashion in 5% to 10% increments. LPA consists of tumors formerly classified as mixed subtype tumors containing a predominant lepidic growth pattern of type II pneumocytes and/or club cells (Clara) (formerly known as nonmucinous BAC) that have an invasive component greater than 5 mm ( Fig. 14-1A-C ). The other major subtypes include acinar (see Fig. 14-1D ), papillary (see Fig. 14-1E ), micropapillary (see Fig. 14-1F ), and solid with mucin-predominant adenocarcinomas (see Figs. 14-1G-H ). The micropapillary-predominant subtype is a new addition due to the observation in multiple studies that it is associated with poor prognosis in early stage adenocarcinomas. It has been proposed that a cribriform pattern is associated with poor prognosis and, if this finding is validated, it may be added as a poor prognostic subtype of lung adenocarcinoma. Signet ring and clear cell carcinoma subtypes are no longer regarded as histologic subtypes, but they are now documented as cytologic features whenever present with a comment about the percentage identified. Although clear and signet ring cell cytologic changes are seen mostly in the solid subtype, they can also be seen in acinar or papillary patterns as well. There is a high correlation between the appearance on computed tomography (CT) and the pathology features on biopsy; the ground-glass component on CT tends to correlate with lepidic growth on biopsy, whereas the solid component on CT tends to correlate with invasive components on biopsy.
Lung adenocarcinoma can consist of several variants including invasive mucinous adenocarcinoma (formerly mucinous BAC), colloid adenocarcinoma, fetal adenocarcinoma, and enteric adenocarcinoma. Invasive mucinous adenocarcinomas (formerly mucinous BAC) differ from the nonmucinous invasive adenocarcinomas due to the frequent association with KRAS mutations, lack of thyroid transcription factor-1 (TTF-1), and frequent presentation with multicentric lung involvement. While historically, the lepidic pattern has been emphasized with these tumors, sometimes it is not present, and these tumors show varying amounts of other invasive patterns including acinar, papillary, or micropapillary growth. The characteristic histologic features consist of the tumor cell morphology of columnar cells with abundant apical mucin and small basally oriented nuclei ( Fig. 14-2 ). By CT, these tumors typically show localized or multifocal consolidation with air bronchograms, forming nodules and/or lobar consolidation.
Prognosis of Adenocarcinoma Subtypes in Resected Specimens.
A growing number of studies have evaluated prognosis of the adenocarcinoma subtypes according to the precise criteria and terminology of the new classification. Yoshizawa and colleagues identified three groups of tumors with different grades of clinical behavior: (1) low-grade AIS and MIA with 100% 5-year disease-free survival, (2) intermediate grade nonmucinous lepidic predominant, papillary predominant, and acinar predominant with 90%, 83%, and 84% 5-year disease-free survival, respectively, and (3) high-grade invasive mucinous adenocarcinoma, colloid predominant, solid predominant, and micropapillary predominant with 75%, 71%, 70%, and 67% 5-year disease-free survival, respectively. Generally, similar results have been demonstrated in additional independent data sets. Studies that have failed to show prognostic significance for the concepts in the new adenocarcinoma classification have either had small numbers of patients, used overall survival rather than disease-free survival, or focused on patients with advanced disease. Because most patients with stage I lung adenocarcinomas die of causes other than lung cancer, overall survival does not reflect the true biology of the tumor, and methods such as disease-free or recurrence-free survival are more clinically relevant.
Minimally Invasive Adenocarcinoma.
MIA was introduced to describe a lepidic-predominant tumor measuring 3 cm or less that has 5 mm or less of an invasive component ( Fig. 14-3 ). Limited data suggest that patients with MIA will have almost a 100% 5-year disease-free survival. Most of these are nonmucinous, but rarely some of these are mucinous cases. On chest CT, nonmucinous MIA typically shows ground-glass opacity with a solid component measuring 5 mm or less, whereas mucinous MIA typically presents as a solid nodule.
Atypical adenomatous hyperplasia (AAH) was previously the only preinvasive lesion for lung adenocarcinoma but now AIS has been added.
Atypical Adenomatous Hyperplasia.
AAH is an atypical pneumocyte proliferation that resembles but falls short of criteria for nonmucinous AIS (see Fig. 14-1 ). AAH is typically found as an incidental histologic finding in a lung cancer resection specimen.
The incidence of AAH varies from 6% to 21% depending on the extent of the search and the criteria used for the diagnosis. Most lesions of AAH are less than 5 mm in diameter and frequently they are multiple. Histologically, AAH consists of a focal proliferation of slightly atypical cuboidal to low columnar epithelial cells along alveoli and respiratory bronchioles ( Fig. 14-4 ). Slight thickening of alveolar septa may be present.
AAH must be distinguished from a variety of lesions; the most important of which is the nonmucinous AIS, MIA, or lepidic-predominant adenocarcinoma. This distinction can be difficult because there is considerable overlap in the morphologic features between AAH and the lepidic pattern of adenocarcinoma. Currently, there are no data to show that patients with lung cancer and AAH have any different prognosis from those without AAH.
Adenocarcinoma In Situ.
In the new IASLC/ATS/ERS adenocarcinoma classification, AIS is defined as a glandular proliferation measuring 3 cm or less that has pure lepidic growth lacking invasion ( Fig. 14-5 ). In most cases the tumor cells are nonmucinous, with a proliferation of type II pneumocytes or club cells, but rarely they are mucinous, consisting of tall columnar goblet cells having abundant apical mucin. If these lesions are completely resected, patients have been reported to have 100% 5-year disease-free survival. On chest CT, these lesions typically consist of a ground-glass opacity if nonmucinous and a solid nodule of mucinous AIS.
TNM Staging: Potential Changes According to New Classification
Several aspects of TNM ( tumor, node, metastasis ) staging may be modified based on the 2011 lung adenocarcinoma classification. In the setting of multiple nodules of lung adenocarcinoma, the distinction between metastases or primary cancers that are either synchronous (presenting within 2 months of each other) or metachronous (presenting more than 2 months apart) can be helped by the use of comprehensive histologic subtyping and the analysis of other morphologic features including cytologic and stromal characteristics. These morphologic features have been shown to correlate highly with molecular and clinical outcomes. The decision to classify a second tumor as an intrapulmonary metastasis or a separate primary has a great impact on TNM staging and patient management (see Chapter 55 ).
In the future, tumor size for T-factor staging may be measured by invasive size rather than total size. Pathologically, comprehensive histologic subtyping can aid in determining the size of the invasive component by subtracting the percentage of the lepidic component. Several studies have shown that the invasive size is an independent prognostic factor. These data suggest that, like in breast cancer, the T factor for early lung adenocarcinomas may be best determined by the size of the invasive component rather than the total tumor size. On chest CT, the solid versus ground-glass component generally corresponds to the invasive versus lepidic pattern seen by histologic examination. Initial data also suggest that the size of the solid component rather than the total size including the ground-glass component is a better predictor of prognosis. Because CT assessment is used for clinical staging, hopefully sufficient data can be accumulated before the next TNM revision to address this issue.
Adenocarcinoma Classification in Small Biopsies and Cytology
The IASLC/ATS/ERS lung adenocarcinoma classification provides criteria for diagnosis of lung cancer in small biopsies and cytology (see eTable 14-1 ). This is of great importance because 70% of lung cancer patients present with advanced stage disease and the diagnosis is based on small specimens. The new therapeutic implications based on histology provide the rationale for pathologists to distinguish adenocarcinoma from squamous cell carcinoma. Patients with advanced stage tumors who have the pathologic diagnosis of adenocarcinoma, NSCLC, favor adenocarcinoma, or NSCLC-NOS are eligible for three therapeutic options that are not available to patients with squamous cell cancer. Patients with adenocarcinoma may exhibit EGFR mutations and, if so, EGFR tyrosine kinase inhibitor therapy has benefit for response and progression-free survival. In addition, patients with adenocarcinoma are responsive to pemetrexed, whereas those with squamous cell carcinoma show little response. Finally, patients with adenocarcinoma may respond to the anti-vascular endothelial growth factor agent bevacizumab, whereas those with squamous cell carcinoma treated with bevacizumab have experienced life-threatening hemorrhage. These clinically important differences between adenocarcinoma and squamous cell carcinoma make pathologic distinction essential.
In the 2011 lung adenocarcinoma classification, tumors that show clear morphologic features of adenocarcinoma or squamous cell carcinoma are classified with these standard terms. However, if the tumor only shows a carcinoma with no clear squamous or glandular features (NSCLC-NOS), a minimal immunohistochemical workup is recommended using a single adenocarcinoma marker and squamous marker, which should allow for classification of most tumors. At the moment, the best markers for adenocarcinoma and squamous cell carcinoma are TTF-1 and p63, respectively. In a tumor that shows no clear squamous or glandular morphology, but the staining results favor adenocarcinoma (i.e., TTF-1 positive, p63 negative), the tumor should be classified as NSCLC, favor adenocarcinoma ( Fig. 14-6 ). Likewise, if the stains in such a tumor favor squamous cell carcinoma, the diagnosis would be NSCLC, favor squamous cell carcinoma ( Fig. 14-7 ). Then, for tumors in which there is clear differentiation by light microscopy or special stains or if the results are conflicting, the diagnosis remains NSCLC-NOS. Cytology is another powerful tool in subclassifying poorly differentiated NSCLC. In some cases, it may be easier to classify the tumor based on cytology than on biopsy. It is recommended to avoid use of the term “nonsquamous carcinoma” and state the specific diagnosis in precise terms as outlined earlier. Also, use of the term NSCLC should be minimized and instead the specific diagnosis (adenocarcinoma or squamous cell carcinoma) should be used when possible.
The approach to interpretation of small biopsies and cytology must include considerations of diagnoses other than NSCLC, such as neuroendocrine tumors (carcinoid, small cell carcinoma, or large cell neuroendocrine carcinoma) as well as metastatic tumors including metastatic malignant melanoma, breast cancer, or prostate cancer. Therefore, if the initial evaluation does not clearly point to adenocarcinoma or squamous cell carcinoma, some of these other diagnoses may need to be considered.
The diagnosis of NSCLC-NOS was encouraged by previous World Health Organization classifications when there was no clinical value in being more precise. In studies of advanced NSCLC, this diagnosis was made in 20% to 40% of cases and some data suggest its use has been increasing. However, with the new IASLC/ATS/ERS criteria and utilization of immunohistochemistry as well as cytology correlation, the percentage of NSCLC diagnosed as NSCLC-NOS, should be less than 5% of cases.
EGFR Mutation Testing
In the new IASLC/ATS/ERS Lung Adenocarcinoma Classification, there is a clinical recommendation that EGFR mutation testing should be performed in advanced lung adenocarcinomas because of the predictive benefit of EGFR mutations with treatment by EGFR tyrosine kinase inhibitors as described earlier. EGFR mutation testing should be performed for all patients with a pathologic diagnosis of (1) adenocarcinoma, (2) NSCLC, favor adenocarcinoma, and (3) NSCLC-NOS. This recommendation has major implications for tissue management and pathologic diagnosis.
Multidisciplinary Strategy Needed to Obtain and Process Small Biopsies and Cytology
Each institution needs to develop a multidisciplinary strategy to manage these small pieces of tissue at each stage of handling: (1) obtaining the specimen, (2) processing it in the pathology laboratory, (3) providing material to the molecular diagnostic laboratory, and (4) documenting the results in a pathology report and the medical record. This process requires ongoing communication between specialists to ensure optimal management of tissues and efficient reporting of results. One of the central aspects of this process that impacts radiologists, pulmonologists, and surgeons is the need to obtain sufficient tissue not only for diagnosis, but also for molecular studies. To that end, biopsy procedures should be designed to result in either a core biopsy or a cell block from tissue samples obtained for cytology. Cytology specimens such as pleural fluids should also be processed to generate cell blocks such that immunostaining and molecular studies can be performed.
Use of Minimal Stains to Maximize Tissue for Molecular Testing
Pathologists should minimize the amount of tissue used for making the diagnosis, including use of as few special stains as possible. This is necessary to preserve as much tissue as possible for molecular testing. One helpful approach is to cut multiple unstained slides from the block after initial review in cases that are potential candidates for molecular testing, so that the block is cut only once and valuable tissue is not lost during the process of facing the block multiple times; facing is the process of shaving tissue from the surface to obtain full cuts across the tissue block. This would include tumors that are either clearly adenocarcinoma or those with NSCLC-NOS patterns that will require special stains. If adenocarcinoma is suspected, a single stain for TTF-1, if positive, would confirm not only the adenocarcinoma diagnosis but also a pulmonary origin. If by morphology the tumor could be either adenocarcinoma or squamous cell carcinoma, it may be best to perform one adenocarcinoma (i.e., TTF-1) and one squamous (i.e., p63) marker as recommended in the new classification. Limited additional stains may be considered for the small percentage of cases that cannot be classified after this initial panel. Molecular testing guidelines for lung adenocarcinoma: Utility of cell blocks and concordance between fine-needle aspiration cytology and histology samples.
Squamous Cell Carcinoma
Squamous cell carcinoma accounts for approximately 20% of all lung cancers in the United States. Historically, two thirds of squamous cell carcinomas presented as central lung tumors, but one third were peripheral. However, recent reports document an increasing percentage of squamous cell carcinomas in the periphery, exceeding 50% in some studies. The morphologic features that suggest squamous differentiation include intercellular bridging, squamous pearl formation, and individual cell keratinization ( Fig. 14-8 ). In well-differentiated tumors these features are readily apparent; however, in poorly differentiated tumors, they are difficult to find. Squamous cell carcinoma arises most often in segmental bronchi and involves the lobar and main-stem bronchus by extension. According to the 2004 World Health Organization classification, squamous cell carcinoma can have papillary, clear cell, small cell, and basaloid subtypes. However, this subtyping needs updating because it does not address the morphologic spectrum of appearances of lung squamous cell carcinoma, and it does not allow for meaningful correlations with clinical, prognostic, or molecular features. For example, the small cell variant probably should be dropped because most of these cases would be better classified as basaloid variants, and the term “small cell” creates confusion with true small cell carcinoma.
Several papers have proposed alternative approaches to subclassifying pulmonary squamous cell carcinomas. These include recognition of an alveolar space-filling variant, which corresponds to a favorable prognosis. However, this pattern is seen in the minority of cases, more often it is seen only focally and, in a study from North America, prognostic significance could not be demonstrated. In another study of pulmonary squamous cell cancer, minimal tumor cell nests were defined as large (more than six tumor cells), small (two to five cells), and single cell; the single cell infiltrating tumors had the worst prognosis. Also, tumors associated with a background of usual interstitial pneumonia (UIP) and lymph node metastases had a poor prognosis. Further work is needed to develop a more practical approach to subclassification of squamous cell carcinoma and to identify better histologic predictors of prognosis.
Squamous Dysplasia and Carcinoma In Situ
Squamous cell carcinoma develops through a multistep process in which the normal bronchial mucosa progresses through a series of lesions from basal cell hyperplasia to squamous metaplasia, dysplasia, and carcinoma in situ. In addition to the spectrum of histologic features, there is an accumulation of molecular events through the progression of increasing dysplasia to carcinoma in situ and invasive squamous cell carcinoma.
According to the severity of cytologic atypia and thickness of involvement of the bronchial mucosa, squamous dysplasia may be classified as mild, moderate, or severe. These changes represent a continuum of abnormalities; when there is marked cytologic atypia of the full thickness of the bronchial mucosa, the diagnosis is carcinoma in situ. Dysplasia must be distinguished from reactive atypia associated with inflammation or granulation tissue. Microinvasive squamous cell carcinoma also needs to be distinguished from carcinoma in situ with involvement of submucosal glands.
Small Cell Carcinoma
SCLC accounts for 14% of invasive lung cancers in the United States annually. Most cases present as a perihilar mass (see eFig. 53-21 ). Because most patients present in an advanced stage, the diagnosis is often made based on transbronchial biopsy and/or cytology, and these specimens are very reliable. SCLC may also present as a solitary coin lesion in up to 5% of cases, but the rarity of early stage tumors makes it unusual to encounter SCLC as a surgical specimen.
The morphology of SCLC characteristically shows tumor cells with small size, a round to fusiform shape, scant cytoplasm, finely granular nuclear chromatin, and absent or inconspicuous nucleoli ( Fig. 14-9 ). Tumor cells typically grow in diffuse sheets, but may show rosettes, peripheral palisading, and organoid nesting. Mitoses are frequent, averaging 80 per 2 mm 2 area, and necrosis is usually extensive.
SCLC can present in combination with various types of non–small cell carcinomas in less than 10% of cases, with large cell carcinoma ( Fig. 14-10 ) in about 4% to 6% of cases, and with adenocarcinoma or squamous cell carcinoma in 1% to 3% of cases. In addition, SCLC can be combined with spindle cell carcinoma, giant cell carcinoma, and carcinosarcoma. To date, compared with patients with pure SCLC, no significant difference has been demonstrated in clinical features, prognosis, or response to therapy.
A good quality hematoxylin and eosin stain is the most important stain for the diagnosis of SCLC, although immunohistochemistry is frequently used. In most tumors, a definite diagnosis can be established based on hematoxylin and eosin without immunostains. Immunohistochemistry using a pancytokeratin antibody can help confirm that the tumor is a carcinoma and make the distinction from a lymphoid lesion. A panel of neuroendocrine markers is useful, including CD56, chromogranin, and synaptophysin. TTF-1 is expressed in 70% to 80% of SCLC. However, because TTF-1 can be positive in extrapulmonary small cell carcinomas, it should not be used to determine the primary site. Ki-67 proliferation rate is very high, averaging 70% to 90%.
In up to 5% to 7% of cases, expert lung cancer pathologists disagree about the separation of SCLC and NSCLC. Agreement for the diagnosis of SCLCs for all five observers in one study was 93% and, for at least four of five observers, it was 98%. When disagreements arise, it is best to use a consensus approach among other pathology colleagues. Extramural consultation may be needed if a consensus diagnosis cannot be reached locally. Comparison of problematic biopsy specimens with any available cytology specimens can be very helpful because the morphology is often easier to assess based on cytology.
One of the reasons small biopsy interpretation can be difficult is due to the frequent finding of “crush artifact,” a loss of cellular detail usually seen from the crushing of tumor cells by the biopsy forceps. The vast majority of tumors with dense sheets of small blue cells showing crush artifact turn out to be SCLC, perhaps indicating fragility of the SCLC tumor cells. However, a similar crush artifact can be seen in carcinoid tumors, lymphocytic infiltrates, or poorly differentiated NSCLC. In the context of extensive artifact, the diagnosis of SCLC requires that some preserved tumor cells with diagnostic morphology compatible with SCLC should be seen to confirm the diagnosis. Even in crushed specimens, immunohistochemical markers can be useful, because SCLC may demonstrate positive staining for cytokeratin, chromogranin, CD56, synaptophysin, TTF-1, and a high proliferation index with Ki-67.
In the setting of keratin-negative staining in a suspected SCLC, it is important to exclude other differential diagnoses such as chronic inflammation, lymphoma, primitive neuroectodermal tumor, or small round cell sarcoma. However, the distinction of SCLC from non-SCLC is primarily based on morphology rather than any immunohistochemical or molecular marker.
Large Cell Neuroendocrine Carcinoma
In surgical series, large cell neuroendocrine carcinoma (LCNEC) comprises approximately 3% of resected lung cancers. LCNEC differs from typical and atypical carcinoid tumors in that it is a high-grade non–small cell neuroendocrine carcinoma. It is distinction from SCLC based on morphologic characteristics ( Table 14-2 and see eTable 14-1 ). The morphologic criteria include (1) neuroendocrine morphology: organoid, palisading, trabecular, or rosette-like growth patterns (see Fig. 14-10A ); (2) non–small cell cytologic features: large size, polygonal shape, low nuclear to cytoplasmic ratio, coarse or vesicular nuclear chromatin, and frequent nucleoli; (3) high mitotic rate (11 or more per 2 mm 2 ) with a mean of 60 mitoses per 2 mm 2 ; (4) frequent necrosis; (5) at least one positive neuroendocrine immunohistochemical marker or neuroendocrine granules seen on electron microscopy (see Fig. 14-10B ). The diagnosis of LCNEC based on small biopsy specimens such as needle or bronchoscopic biopsy specimens is usually very difficult because the neuroendocrine morphology is hard to appreciate without a resection specimen. Combined LCNEC is the term used for LCNECa that are combined with other histologic types of NSCLC such as adenocarcinoma or squamous cell carcinoma (see Table 14-1 ).
Typical and Atypical Carcinoid
Carcinoid tumors account for 2% to 3% of all invasive lung malignancies. Patients are asymptomatic at presentation in approximately 50% of cases. The mean age of presentation for typical carcinoid (TC) and atypical carcinoid (AC) tumors is 45 to 55 years without any sex predilection ( Table 14-3 ). Carcinoids are the most frequent tumor of the lung in children. The typical presenting symptoms are hemoptysis in 18%, postobstructive pneumonitis in 17%, and dyspnea in 2% of patients. Patients may present with paraneoplastic syndromes such as the carcinoid syndrome or Cushing syndrome. Surgical resection is the primary approach to treatment. TCs have a favorable prognosis. Because 5% to 20% of TCs have regional lymph node involvement, this feature should not be used to make a distinction from AC. ACs have a larger tumor size and a higher rate of metastases than TCs, and patients have significantly worse survival; 5-year survival in AC is approximately 30%.
|Histologic or Clinical Feature||Typical Carcinoid||Atypical Carcinoid|
|Histologic patterns: organoid, trabecular, palisading, and spindle cell||Characteristic||Characteristic|
|Mitoses||Absent or < 2 per 2 mm 2 in area of viable tumor||2 to 10 per 2 mm 2 in area of viable tumor|
|Necrosis||Absent||Characteristic, usually focal or punctate|
|Nuclear pleomorphism, hyperchromatism||Usually absent, not sufficient by itself for diagnosis of atypical carcinoid||Often present|
|Regional lymph node metastases at presentation||5% to 15%||40% to 48%|
|Distant metastases at presentation||Rare||20%|
|Survival at 5 years||90% to 95%||50% to 60%|
|Disease-free survival at 10 years||90% to 95%||35%|
Carcinoid tumors may be central (see eFig. 54-7 ) or peripheral in location. Central tumors often have polypoid endobronchial growth, while peripheral carcinoids are usually subpleural. The classic histologic pattern consists of an organoid growth pattern and uniform cytologic features consisting of moderate eosinophilic, finely granular cytoplasm with nuclei possessing a finely granular chromatin pattern ( Fig. 14-11 ). Both TC and AC can show a variety of histologic patterns, including spindle cell, trabecular, palisading, papillary, sclerosing papillary, rosette-like, glandular, and follicular patterns. Cytologically, the tumor cells may have acinic cell-like, signet-ring, mucin-producing, or melanocytic features.
The diagnostic criteria for AC are a carcinoid tumor with mitoses between 2 and 10 per 2 mm 2 area of viable tumor or the presence of necrosis ( Fig. 14-12 ). In TC, mitotic figures are rare (<2 per 2 mm 2 ) and necrosis is absent. Pleomorphism, vascular invasion, and increased cellularity are not as helpful in separating TC from AC. These tumors usually stain strongly for neuroendocrine markers such as chromogranin, synaptophysin, and CD56. The proliferation rate for TC by Ki-67 is usually low (≤5%) compared with the rate for AC, which is usually between 5% and 20%. In small crushed biopsies, Ki-67 can help separate TC or AC from the high grade LCNEC or SCLC where the proliferation rates are very high.
Diffuse Idiopathic Pulmonary Neuroendocrine Cell Hyperplasia.
Diffuse idiopathic pulmonary neuroendocrine cell hyperplasia is a rare condition in which the peripheral airways are diffusely involved by neuroendocrine cell hyperplasia and tumorlets ( Fig. 14-13 ). The clinical presentation resembles ILD manifest by airway obstruction due to bronchiolar fibrosis in approximately half of patients. The remaining patients typically present with multiple incidentally discovered pulmonary nodules, often found during follow-up for an extrathoracic malignancy. Because carcinoid tumors are frequently found in patients with diffuse idiopathic pulmonary neuroendocrine cell hyperplasia and the tumors are often multiple, this is thought to represent a preinvasive lesion for carcinoid tumors. There is a distinctive CT appearance consisting of centrilobular nodules and pulmonary nodules that correspond to the tumorlets and carcinoid tumors, respectively. Furthermore, in patients who present with clinical manifestations of ILD, the chest CT can be normal or it can show mosaic perfusion from air trapping, bronchial wall thickening, and bronchiectasis.
Large Cell Carcinoma
According to the US NCI SEER data, large cell carcinoma comprises 3% of all lung carcinomas, which is a marked decrease from 9% previously reported for 1983–1987. Large cell carcinomas present most often in the lung periphery and are usually large necrotic tumors. The diagnosis of large cell carcinoma is one of exclusion, where the presence of squamous cell or glandular differentiation could not be seen by light microscopy. The microscopic appearance consists of sheets and nests of large polygonal cells with vesicular nuclei and prominent nucleoli ( Fig. 14-14 ). Because solid adenocarcinoma requires a minimum of five mucin-positive cells in at least two high-power fields, in large cell carcinoma, the number of mucin-positive cells should be less than this.
A surgical resection specimen with thorough histologic sampling is required for the diagnosis of large cell carcinoma, therefore the diagnosis cannot be made based on a small biopsy or cytology specimen because an adenocarcinoma or squamous cell carcinoma component cannot be excluded. Tumors that historically have been classified as large cell carcinoma in the setting of advanced lung cancer would be best classified as NSCLC-NOS according to the 2011 IASLC/ATS/ERS lung adenocarcinoma classification. If immunostains are performed, some of these tumors might be reclassified as NSCLC, favor adenocarcinoma, or NSCLC, favor squamous cell carcinoma, and a small percent would remain as NSCLC-NOS. The adenocarcinoma phenotype cases show driver mutations in 38% of cases that are typical of adenocarcinoma, including KRAS, BRAF, ALK, EGFR, MAP2K1, and PIC3CA .
Adenosquamous carcinoma accounts for 1% of all lung cancers in the United States. The diagnosis of adenosquamous carcinoma requires the presence of at least 10% squamous cell and adenocarcinoma components as seen on light microscopy. The diagnosis of adenosquamous carcinoma may be suspected but cannot be made by small biopsy or cytology because a larger resection specimen is needed.
Carcinomas with Pleomorphic, Sarcomatoid, or Sarcomatous Elements
The rarest histologic subgroup of major lung cancers is that of sarcomatoid carcinomas, which comprise 0.5% of all invasive lung malignancies in the United States. These poorly differentiated tumors consist of a spectrum of lung carcinomas with pleomorphic, sarcomatoid, and sarcomatous elements with poor prognosis. Most pleomorphic carcinomas are large peripheral tumors that frequently invade the chest wall. The diagnosis of pleomorphic carcinomas requires the presence of at least a 10% component of a spindle cell and/or giant cell component as well as a carcinomatous component that may consist of a single or mixture of patterns of other histologic types such as adenocarcinoma and/or squamous cell carcinoma.
In small biopsies or cytology specimens, one can only suggest the diagnosis of pleomorphic carcinoma because this diagnosis requires a resection specimen. Carcinomas with a pure giant cell or spindle cell pattern are classified as giant cell or spindle cell carcinoma, respectively. Giant cell carcinomas are composed of huge bizarre pleomorphic and multinucleated tumor giant cells. The diagnosis of pleomorphic carcinoma can be made by light microscopy but immunohistochemistry, particularly for epithelial markers such as keratin, can be helpful in confirming epithelial differentiation in poorly differentiated tumor components.
Carcinosarcoma and Pulmonary Blastoma
According to the 2004 World Health Organization classification, tumors composed of a mixture of carcinoma and sarcoma that show heterologous elements such as malignant cartilage, bone, or skeletal muscle are classified as carcinosarcoma. Pulmonary blastomas are composed of a glandular component with a fetal adenocarcinoma pattern and a primitive sarcomatous component with blastomatous stroma. These tumors are now distinguished from fetal adenocarcinoma, which are classified as a variant of adenocarcinoma.
Solitary Fibrous Tumor
Solitary fibrous tumors of the pleura are localized neoplasms arising in the pleura that are usually benign, with a small percentage that are malignant. About 80% arise on the visceral pleura, with a minority arising from the parietal pleura or rarely from the pulmonary parenchyma. These tumors are not mesothelial in origin. They are derived from submesothelial connective tissue and are not related to asbestos.
The majority of tumors are large, measuring over 10 cm in diameter, and often have a pedicle. The tumors are gray-white, with a nodular, whorled, or lobulated appearance ( Fig. 14-15A ; see eFig. 56-11E ).
Histologically the tumors most often show a “patternless pattern” of disorderly or randomly arranged mixtures of spindle to oval shaped cells with a ropy collagenous stroma (see Fig. 14-15B ). Criteria for malignancy include increased cellularity, pleomorphism, necrosis, and more than four mitoses per 10 high-power fields. Immunohistochemical stains for CD34 and BCL2 are positive.
Malignant mesothelioma is a malignant neoplasm that can have an epithelioid, sarcomatoid, or biphasic appearance. These tumors have a characteristic gross appearance with diffuse pleural thickening. The tumor typically does not invade the lung parenchyma except for spread along the interlobar fissures. Microscopically, classic mesotheliomas show both epithelial and sarcomatoid patterns ( Fig. 14-16 ). The epithelioid tumors consist of glands, tubules, and solid sheets of tumor cells with abundant eosinophilic cytoplasm. The sarcomatoid tumors are composed of sheets of spindle cells that are similar to a fibrosarcoma. Biphasic mesotheliomas should have at least 10% of each component.
Immunohistochemistry plays an important role in the diagnosis of malignant mesothelioma, in particular to help make the distinction with carcinomas that metastasize to the pleura. Both mesotheliomas and carcinomas are positive for cytokeratins. Mesotheliomas are usually positive for calretinin, WT-1, and D2-40 (podoplanin), while adenocarcinomas are typically negative. Adenocarcinomas often, express carcinoembryonic antigen, Leu-M1, B72.3, and BER-EP4, while mesotheliomas are negative for these markers. Historically the absence of mucin and the presence of hyaluronic acid (positive Alcian blue staining) and long, slender microvilli seen on electron microscopy have been used to support the diagnosis of malignant mesothelioma, but immunohistochemistry has largely replaced these studies. See Chapter 82 for additional information on tumors of the pleura.
Interstitial Lung Diseases
Idiopathic Interstitial Pneumonias
The term interstitial lung disease comprises a heterogeneous group of disorders with various histopathologic features. Many different etiologies have very similar or identical pathologic ILD features and require thorough clinical and radiologic correlation to resolve. The differential diagnosis for most pathologic patterns includes collagen vascular disease, drug reaction, hypersensitivity reaction, or an idiopathic process. Therefore, the primary task of the pathologist is usually to recognize and categorize the pathologic pattern to expedite the correct clinical workup, inform treatment decisions, and guide prognosis. There may be histologic clues that can point toward an underlying etiologic process, as discussed in each subsequent section.
There have been various classifications of ILD, starting in 1969 by Liebow and Carrington. In 2002, an ATS/ERS consensus classification of idiopathic interstitial pneumonias (IIP) was published. Since then, there have been numerous publications and studies that have added significantly to our understanding of this category of disease with important clinical implications for management and treatment of patients with diffuse parenchymal lung disease. The ATS/ERS IIP classification was recently updated by an international multidisciplinary panel comprising pulmonologists, pathologists, radiologists, and molecular biologists. IIPs were divided into four groups: (1) fibrosing, (2) acute/subacute, (3) smoking-related, and (4) a new subcategory of “rare IIPs” was introduced and now includes lymphocytic interstitial pneumonia (LIP) and the newly described entity pleuropulmonary fibroelastosis. The rare histologic patterns acute fibrinous and organizing pneumonia (OP) and bronchiolocentric patterns of interstitial pneumonia are recognized, but there was insufficient evidence to accept these as distinct IIPs.
In cases that prove impossible to classify, the term unclassifiable interstitial pneumonia has been coined. It accounts for cases with inadequate clinical or radiologic data or an inadequate or nondiagnostic biopsy. It is also the suggested diagnosis when there is a major discrepancy between clinical, radiologic, and pathologic findings, when previous therapy has resulted in alterations in the radiologic or histologic findings, or when there are discrepancies between histologic findings in different lobes not resolved after correlation with clinical and radiologic data. Additional information on IIPs is found in Chapter 63 .
Fibrosing Interstitial Lung Disease
Usual Interstitial Pneumonia/Idiopathic Pulmonary Fibrosis.
UIP is the most common pattern of ILD and is the pathologic hallmark of idiopathic pulmonary fibrosis (IPF). The incidence of IPF, the idiopathic form of UIP, is 7 to 20 per 100,000 people and has no geographic or ethnic predilection. The classic clinical presentation is the gradual onset of shortness of breath with cough and clubbing in up to 50% of cases. Pulmonary hypertension has been reported in up to 32% of IPF patients awaiting lung transplantation and is associated with higher mortality in comparison to patients without pulmonary hypertension. Radiographic imaging studies reveal small lung volumes with bilateral coarse reticular opacities most pronounced in the subpleural lower lobes and lower portions of the upper lobes (see Figs. 63-6 and 63-7 ). Honeycomb changes and traction bronchiectasis are typical features and become more pronounced as the disease progresses. When these typical features are present, the high-resolution CT appearance is 90% specific for UIP and surgical biopsy is often deferred (see Fig. 63-11 and eFig 63-2 , eFig 63-5 , eFig 63-6 , eFig 63-7 ). A highly probable diagnosis of IPF can be made without lung biopsy, but definitive diagnosis of IPF requires surgical lung biopsy and exclusion of other known causes of ILD. It is important for the treating clinician to perform a thorough workup, including exposure history and serologic studies, to exclude secondary causes of pulmonary fibrosis such as chronic hypersensitivity pneumonitis (HP), collagen vascular disease, drug-toxicity, asbestosis, chronic aspiration, and Hermansky-Pudlak syndrome because each of these entities can cause an UIP pattern of interstitial fibrosis. Furthermore, after a biopsy has been performed and a pathologic diagnosis rendered, it is important to correlate the radiographic features with the pathologic findings in that cases of histologic UIP might not be associated with typical high-resolution CT findings of UIP. In this setting, multidisciplinary discussion is required for some of these patients to generate the correct diagnosis of IPF.
UIP is a diffuse fibrosing lung disease with a characteristic distribution of disease (see Fig. 63-22 ). The fibrosis is worse in the lower lobes and the lower portions of the upper lobe. Grossly, the pleura is firm and thickened with a cobblestone appearance. The histologic hallmark of UIP is the heterogeneity of the interstitial fibrosis, both in space and in time. As evidence of spatial heterogeneity, there should be areas of fibrosis separated by areas of more normal lung. Areas of normal lung should be present in a biopsy specimen to support the diagnosis and exclude other interstitial diseases. The fibrosis is distributed at the periphery of the lobules and is most pronounced in the subpleural regions ( Fig. 14-17A ). As evidence of temporal heterogeneity, there are areas of long-standing older interstitial fibrosis and areas of newer fibrosis characterized by fibroblastic foci (see Fig. 14-7B-C ; see Fig. 63-24 ). Fibroblastic foci are composed of crescent shaped protrusions of fresh glycosaminoglycan-rich fibroblastic tissue composed of plump fibroblast and myofibroblasts. Fibroblastic foci are located in areas of ongoing or recent lung injury and are distinguished from older areas of denser fibrosis. Because the presence of numerous fibroblastic foci correlates with faster disease progression and decreased survival time, it is important to qualify the number of the fibroblastic foci in the surgical pathology report. The older areas of fibrosis are marked by architectural distortion of the underlying lung parenchyma with eventual development of honeycomb changes. There may be occasional aggregates of chronic inflammatory cells in the areas of older fibrosis but this does not tend to extend into normal-appearing lung parenchyma. Honeycomb changes are areas of abnormal dilatation of air spaces lined by bronchiolar epithelium (see Fig. 63-25 ). The cystic honeycomb spaces are filled with thick mucin-containing acute inflammatory cells; however, the presence of the neutrophils does not indicate an infectious process. Other characteristic features in the areas of older fibrosis and scarring include smooth muscle hyperplasia, pulmonary arterial thickening, and traction bronchiectasis. There should be no, or minimal, granulomatous inflammation, eosinophilia, acute pleuritis, or significant exogenous inorganic dust, silica, or asbestos fibers.