Histologic Classification and Its Need for Treatment of Lung Cancer

Fig. 1.1
Photographs of H&E-stained slides diagnosed as adenocarcinoma in core biopsy specimens. (a) Adenocarcinoma with papillary pattern (original magnification × 100). (b) Adenocarcinoma with micropapillary pattern (original magnification × 200). As per current nomenclature, there is no need to perform immunohistochemical stains for further classification unless a metastatic adenocarcinoma is suspected


Fig. 1.2
Photograph of an H&E-stained slide diagnosed as squamous cell carcinoma. Note the presence of keratinization and intracellular bridges, hallmarks of keratinizing squamous cell carcinoma. As per current nomenclature, there is no need to perform immunohistochemical stains for further classification of this tumor (original magnification × 200)

Table 1.1
Proposed nomenclature for the diagnosis of NSCLC in small biopsy/cytology material

Histologic criteria

2 markers IHC profile


Presence of acinar, papillary, lepidic, or micropapillary patterns

Not recommendeda

Squamous cell carcinoma

Presence of keratinization and intracellular bridges

Not recommended

NSCC favor adenocarcinoma

Solid pattern, no other evidence of differentiation

Any TTF-1 positivity, negative for p63/p40

NSCC favor squamous cell carcinoma

Solid pattern, no other evidence of differentiation

Diffuse and strong positivity for p63 or p40. Negative for TTF-1

NSCLC not otherwise specified (NOS)

Solid pattern, no other evidence of differentiation

Double positivity for TTF-1 and p63 or p40 either focal or diffuse

Double-negative stain for TTF-1a and p40

aTTF-1 may be useful to confirm lung origin. If TTF-1 negative, markers for metastases from other sites such as the breast, gastrointestinal, or genitourinary tract may be useful if clinically indicated

On cytology specimens, a precise histologic subtype of NSCLC is feasible using well-described criteria; the issue has been addressed in several studies [4046].

As demonstrated by Rekhtman et al. [40], a subclassification of NSCLC into adenocarcinoma and squamous cell carcinoma in cytology specimens can be achieved in approximately 93 % of the cases using cytomorphologic criteria only. However, the subclassification of NSCLC can be a problem in paucicellular specimens and in poorly differentiated carcinomas where no clear histologic differentiation is present. In these situations, the distinction between adenocarcinoma and squamous cell carcinoma cannot be made by morphologic features alone. For example, solid-type adenocarcinomas can show squamoid features such as nested appearance and glassy eosinophilic cytoplasm (Fig. 1.3), although there is no clear keratinization or presence of intracellular bridges. This can be seen either on cytology or small biopsy material and can be confused with squamous cell carcinoma. Knowledge of this pitfall can greatly reduce misclassification of NSCLC.


Fig. 1.3
Photograph of a poorly differentiated carcinoma diagnosed as NSCLC favor adenocarcinoma. (a) H&E-stained section showing a solid growth pattern without evidence of glandular or squamous differentiation. Note the squamoid appearance (original magnification × 200). (b) Immunohistochemical stains show the tumor cells are diffusely positive for TTF-1, whereas negative for p40 (not showed). This pattern of immunoreactivity supports the diagnosis

It is recommended that in cases of a poorly differentiated carcinoma without any clear morphologic evidence of glandular or squamous differentiation, immunohistochemical markers should be used to help in the tumor classification.

There are many studies in the medical literature that attempted to identify the best panel of immunohistochemical stains to be used for the separation of adenocarcinoma and squamous cell carcinoma in small biopsy material [4245]. Considering the need to save tissue for molecular studies that are indicated in adenocarcinomas, a consensus for the use of a minimal panel has been recommended by the IASLC/ATS/ERS classification [15] and adopted by the World Health Organization. This minimal panel of antibodies includes 2 markers, TTF-1 (thyroid transcription factor-1) and p63 or p40 (an isoform of p63 that seems to be more specific than p63 for the detection of squamous lesions) [47, 48]. Up to 20–30 % of lung adenocarcinomas, including those positive for TTF-1, can be positive for p63, but in virtually all of these cases p40 will be negative [47, 48].

Several studies have shown that TTF-1, although not specific, is a good marker for adenocarcinomas of the lung. Therefore, any staining pattern of TTF-1, either focal or diffuse, is a good indication of adenocarcinoma differentiation. As per the IASLC/ATS/ERS recommendation, tumors that are poorly differentiated but show immunohistochemical features of adenocarcinoma should be classified as NSCLC favor adenocarcinoma (Fig. 1.3). In contrast, poorly differentiated tumors that show immunophenotype consistent with a squamous cell carcinoma, which is defined by a strong and diffuse positivity for P63 or P40 and negative reactivity for TTF1, should be classified as NSCLC favor squamous cell carcinoma (Fig. 1.4).


Fig. 1.4
Photograph of a poorly differentiated carcinoma diagnosed as NSCLC favor squamous cell carcinoma. (a) H&E-stained section showing a solid growth pattern without evidence of glandular or squamous differentiation. Note the squamoid appearance. (b) Immunohistochemical stains show the tumor cells are diffusely positive for p40, whereas negative for TTF-1 (not showed). This pattern of immunoreactivity supports the diagnosis (original magnification × 100)

Although this proposition seems straightforward, there are still some difficulties in classifying tumors that defies this binary approach. For instances, what to do with tumors that show reactivity for both markers or are negative for both markers? The 2011 classification recommended that these tumors should be classified as NSCLC-NOS meaning that the specific subtype cannot be determined in the biopsy specimen with certainty [15]. In the case of a double-negative tumor or a carcinoma with ambiguous immunophenotype based on the two markers, it is imperative to correlate the findings with clinical history. In case of a history or the possibility of a metastatic tumor, additional immunohistochemical stains should be added to the panel to rule out metastatic disease and/or a non-epithelial tumor such as melanoma, sarcoma, and lymphoma. Awareness of the limitations of the two antibody panel, clinical pitfalls, and source, specificity, and sensitivity of the clone of antibodies used are essential to avoid misclassification of lung tumors.

In studies that evaluated the immunophenotype of adenocarcinomas and squamous cell carcinomas in resections, specimens using the binary model of immunoreactivity (TTF-1/p63) have demonstrated that adenocarcinomas have great variability in their staining pattern [42], whereas squamous cell carcinomas are more homogenous with universal strong and diffuse expression of p63 and/or p40 with negative stain for TTF-1. Therefore, it can be concluded that a negative stain for p63/p40 tends to exclude the diagnosis of squamous cell carcinoma [42, 47, 48]. In addition, tumors that are double positive (TTF-1+/p63+ or p40+) or double negative (TTF-1−/p63− or p40−) have the same pattern of mutation as other adenocarcinomas that follow the more common immunoreactivity pattern (TTF-1+/p63− or p40−). In some of the cases where both TTF-1 and p40 are positive, if different populations of tumor cells are expressing these markers, a comment can be made that this could represent an adenosquamous carcinoma; however, that diagnosis requires a resection specimen. For practical purposes, the diagnosis of a NSCLC favor adenocarcinoma or NSCLC-NOS in a small biopsy material should be regarded as equivalent to the diagnosis of adenocarcinoma for therapeutic decision and or triage for molecular studies.

In patients with a resectable tumor, the diagnosis of NSCLC-NOS may correlate with the diagnosis in a resected specimen of an adenocarcinoma with a predominant solid pattern or a large cell carcinoma, if no definite differentiation is identified after examination of the entire tumor. In reality, the proposed classification for small biopsy material gives the pathologists some leeway on how to address evolving concepts, such as large cell carcinoma as a histologic independent entity and uncertainties on diagnosis due to sampling issues.

Critics of this biopsy classification have pointed out that the use of the term “favor” adenocarcinoma or a squamous cell carcinoma can generate concern and confusion among treating physicians since the term “favor” may be interpreted as doubt about the diagnosis. One possible scenario is that clinicians faced with the diagnosis of NSCLC favor adenocarcinoma may be reluctant to treat the patient appropriately with drugs that can be used in adenocarcinomas because there is the concern that there might be an underlining squamous cell carcinoma in the biopsy. Therefore, in order to apply this terminology, there must be clear communication with the treating physician on what the diagnosis implies.

Another concern for the term of “NSCLC favor” is triaging of cases for molecular diagnostic tests. This important issue has been addressed by the recent molecular testing guidelines for pulmonary adenocarcinoma recommended by the College of American Pathologists (CAP), the International Association for the Study of Lung Cancer (IASLC), and the Association for Molecular Pathology (AMP) [49]. In order to reduce the chances of missing a targetable mutation, all tumors with diagnosis of NSCLC-NOS or NSCLC favor adenocarcinoma in a small biopsy material should be submitted for molecular tests. Only tumors with the histologic diagnosis of squamous cell carcinoma and NSCLC favor squamous cell carcinoma (diffuse and strong nuclear stain for p63 or p40) should not be sent routinely for molecular diagnostic tests [49].

Alternatively, the diagnosis of adenocarcinoma can be rendered in the cases of NSCLC favor adenocarcinoma (no histologic evidence of gland formation and immunohistochemical profile of TTF-1+/P63-/p40-) and NSCLC-NOS (no histologic evidence of gland formation and double-negative immunoreactivity profile TTF-1-/p63-or p40) based on the variability of reactivity patterns seen in adenocarcinoma [42], the molecular profile similarities to that of adenocarcinoma [50], and the levels of confidence of the pathologists rendering the diagnosis.

Conciliation of the Histologic Classification of Adenocarcinoma in Resected Material and in Biopsy Specimens

The 2011 IASLC/ATS/ERS classification of pulmonary adenocarcinoma recognizes that adenocarcinomas are histologic heterogeneous and that different histologic patterns are associated in prognostic significance [15]. Therefore, the recommendation indicates that an invasive adenocarcinoma should be classified based on the predominant histologic pattern. All other histologic patterns present in the tumor should be indicated in the report (Table 1.2).

Table 1.2
New proposed classification of pulmonary adenocarcinoma and its implication to small biopsy specimens


Small biopsy


Adenocarcinoma in situ (AIS)

Lepidic patterna

Flat sheetsa (strips on cell blocks [37])

Minimally invasive adenocarcinoma (MIA)

Lepidic patterna

Flat sheetsa (strips on cell blocks [37])

Lepidic predominant (LPA)

Lepidic pattern may be present

Flat sheets (strips on cell blocks [37])

Papillary predominant

Papillary pattern may be present

Papillary clusters may be present

Acinar predominant

Acini may be present

Acini may be present

Solid predominant

Solid pattern may be present

Micropapillary predominant

Micropapillae may be present

Small clusters may be presentb

aAIS and MIA cannot be diagnosed in a small biopsy material or cytology. A comment about the differential diagnosis (AIS, MIA, LPA) can be mentioned

bSmall tight clusters have been reported as cytologic correlates for micropapillary pattern; however, this feature has low sensitivity and specificity for the diagnosis of micropapillary pattern in resection specimen

In patients with stage 1 disease, invasive adenocarcinomas with a predominant lepidic pattern have a good prognosis with approximately 92 % 5-year recurrence-free survival. Tumors with predominant papillary and acinar pattern have an intermediate recurrence-free survival of approximately 84 % in 5 years, and those with predominant solid or micropapillary patterns have approximately 73 % recurrence-free survival at 5 years. The latter shows approximately 30 % change of recurrence or death of disease in 5 years [5153]. Some studies have also suggested that the prognostic stratification by histologic pattern is also effective in patients with advanced stage [5355] and in patients receiving adjuvant therapy. These results are still preliminary and there is no current recommendation for a differential treatment of lung adenocarcinoma with systemic therapy based on predominant histologic pattern or tumor grade.

The new classification also recognizes two new categories, adenocarcinoma in situ and minimally invasive adenocarcinoma. Both tumors in these categories have excellent prognosis with 100 % recurrence-free survival in 5 years if the tumor is completely resected [51]. Both tumors are characterized by a predominant lepidic growth pattern without significant invasion. However, these tumors cannot be diagnosed in biopsy material, since it is required by definition that the entire tumor must be examined in order to render the diagnosis.

The next challenge to the pathologists is how to address the issue of determining the predominant type or subtypes of adenocarcinoma in invasive tumors on small biopsies and cytology. There are no studies on a direct correlation of predominant histologic types between small biopsy material and resection specimen. The difficulty in these studies may rest on sampling issues, since a biopsy may not contain all growth patterns present in the tumor and may not be representative of the predominant type either. Although no formal recommendation exists, it is possible that, similarly to the resection specimen, the pathologist should list the growth patterns seen in the biopsy. This is particularly important for histologic patterns such as micropapillary, where even a small proportion may be an indication of a high risk for recurrence [56].

Only gold members can continue reading. Log In or Register to continue

Sep 27, 2017 | Posted by in PATHOLOGY & LABORATORY MEDICINE | Comments Off on Histologic Classification and Its Need for Treatment of Lung Cancer
Premium Wordpress Themes by UFO Themes
%d bloggers like this: