Ki-67 Immunoreactivity in the Differential Diagnosis
The aim of the present study was to evaluate the usefulness of immunohistochemical markers in the differential diagnosis of pulmonary neuroendocrine tumors with particular emphasis on the preservation of immunoreactivity in areas showing crush artifacts. Specimens from 9 carcinoid tumors (CTs) and 13 small cell carcinomas (SCCs) with crush artifact were stained with antibodies to Ki-67, chromogranin A, synaptophysin, and cytokeratin. The immunoreactivity was well preserved in the crushed areas. Ki-67 was expressed in the crushed areas of all SCCs. Reactivity was diffuse or at least present in 25% of the crushed areas. In contrast, the immunoreactive areas in CTs never exceeded 10%. Immunoreactivity for Ki-67, synaptophysin, chromogranin A, and cytokeratin is well preserved in tissue with crush artifacts and can be interpreted reliably. The diagnosis of SCC should be questioned if fewer than 25% of cells show reactivity for Ki-67.
Neuroendocrine proliferations of the lung encompass a spectrum of nonneoplastic lesions and low-, intermediate-, and high-grade tumors. The former includes diffuse neuroendocrine cell hyperplasia and possibly carcinoid tumorlets, and the latter includes typical carcinoid tumor (CT), atypical carcinoid (AC), large cell neuroendocrine carcinoma, and small cell neuroendocrine carcinoma (SCC), in increasing order of biologic aggressiveness. Low- and intermediate-grade neuroendocrine neoplasms (NENs) usually are treated with surgery, whereas high-grade NENs are treated with chemotherapy or combined-modality treatment. The 10-year survival rates for CT and AC are 90% and 50%, respectively. For SCC and large cell neuroendocrine carcinoma, the survival rates are 5% and 9%, respectively. The distinction of CT and AC from SCC is critical because of major differences in management and prognosis.
The 1999 WHO classification of lung tumors provides the criteria to distinguish low-, intermediate-, and high-grade neuroendocrine tumors on the basis of their microscopic appearance on H&E-stained slides. CT has a bland population of cells and infrequent mitoses (<2/10 high-power fields [hpf]). AC displays pleomorphism, focal necrosis, and fewer than 10 mitotic figures per 10 hpf. High-grade neuroendocrine tumors and SCCs have high nuclear/cytoplasmic ratios, nuclear molding, and more than 10 mitoses per hpf and frequently display crush artifacts. Because the latter artifacts have been emphasized only in descriptions of SCCs and lymphomas, not in other lung tumors, their presence coupled with the demonstration of epithelial and neuroendocrine differentiation by immunohistochemical analysis has been regarded as diagnostic of SCC. However, specimens of CT obtained by endobronchial biopsy can show extensive crush artifact. Indeed, during the last 5 years we have seen 3 cases of CT that were misdiagnosed as SCC on these bases.
It has been recognized for a long time that crush artifact is a characteristic feature of SCC of all body sites. This artifact has been observed in tissue sections, bronchial washing specimens, brushing specimens, aspirates, and sputum specimens and is believed to result from mechanical crushing of fragile nuclei with little cytoplasmic support. Biopsy studies of pulmonary SCC are not uncommonly interpreted as nondiagnostic owing to extensive crush artifact obscuring morphologic details. In many cases, cytologic examination of sputum, bronchial brushing, and washing specimens optimizes the diagnostic yield of endobronchial biopsies. Compared with SCC, CT has a lower cellular density and the cells have more cytoplasm that prevents extensive nuclear distortion. However, artifacts occasionally happen and can be misleading.
Immunohistochemical analysis has little established use in the grading of NENs of the lung. CTs usually stain diffusely and strongly positive for synaptophysin and chromogranin A. The immunoreactivity for these markers generally is not as strong or universal in SCC; however, considerable overlap occurs. Both tumors express cytokeratins. Other markers that might be valuable in grading include p53 protein, which commonly is overexpressed, and Rb protein, which commonly is lost in high-grade NENs while low- and intermediate-grade NENs usually express it.
Previous studies have demonstrated a strong correlation between the proliferative index, as detected by Ki-67 labeling by MIB-1 or Ki-67 antibodies, and grade and prognosis. In the articles in which Ki-67 labeling indices were stated numerically, most intermediate-grade NENs had a proliferative index of less than 25%, most CTs of less than 10%, and most SCCs of substantially more than 25%.
The purpose of the present study was to evaluate the usefulness of immunoperoxidase staining for Ki-67, cytokeratin, chromogranin A, and synaptophysin in the differential diagnosis of CTs and SCCs with particular emphasis on the preservation of immunoreactivity in areas with crush artifact.
The aim of the present study was to evaluate the usefulness of immunohistochemical markers in the differential diagnosis of pulmonary neuroendocrine tumors with particular emphasis on the preservation of immunoreactivity in areas showing crush artifacts. Specimens from 9 carcinoid tumors (CTs) and 13 small cell carcinomas (SCCs) with crush artifact were stained with antibodies to Ki-67, chromogranin A, synaptophysin, and cytokeratin. The immunoreactivity was well preserved in the crushed areas. Ki-67 was expressed in the crushed areas of all SCCs. Reactivity was diffuse or at least present in 25% of the crushed areas. In contrast, the immunoreactive areas in CTs never exceeded 10%. Immunoreactivity for Ki-67, synaptophysin, chromogranin A, and cytokeratin is well preserved in tissue with crush artifacts and can be interpreted reliably. The diagnosis of SCC should be questioned if fewer than 25% of cells show reactivity for Ki-67.
Neuroendocrine proliferations of the lung encompass a spectrum of nonneoplastic lesions and low-, intermediate-, and high-grade tumors. The former includes diffuse neuroendocrine cell hyperplasia and possibly carcinoid tumorlets, and the latter includes typical carcinoid tumor (CT), atypical carcinoid (AC), large cell neuroendocrine carcinoma, and small cell neuroendocrine carcinoma (SCC), in increasing order of biologic aggressiveness. Low- and intermediate-grade neuroendocrine neoplasms (NENs) usually are treated with surgery, whereas high-grade NENs are treated with chemotherapy or combined-modality treatment. The 10-year survival rates for CT and AC are 90% and 50%, respectively. For SCC and large cell neuroendocrine carcinoma, the survival rates are 5% and 9%, respectively. The distinction of CT and AC from SCC is critical because of major differences in management and prognosis.
The 1999 WHO classification of lung tumors provides the criteria to distinguish low-, intermediate-, and high-grade neuroendocrine tumors on the basis of their microscopic appearance on H&E-stained slides. CT has a bland population of cells and infrequent mitoses (<2/10 high-power fields [hpf]). AC displays pleomorphism, focal necrosis, and fewer than 10 mitotic figures per 10 hpf. High-grade neuroendocrine tumors and SCCs have high nuclear/cytoplasmic ratios, nuclear molding, and more than 10 mitoses per hpf and frequently display crush artifacts. Because the latter artifacts have been emphasized only in descriptions of SCCs and lymphomas, not in other lung tumors, their presence coupled with the demonstration of epithelial and neuroendocrine differentiation by immunohistochemical analysis has been regarded as diagnostic of SCC. However, specimens of CT obtained by endobronchial biopsy can show extensive crush artifact. Indeed, during the last 5 years we have seen 3 cases of CT that were misdiagnosed as SCC on these bases.
It has been recognized for a long time that crush artifact is a characteristic feature of SCC of all body sites. This artifact has been observed in tissue sections, bronchial washing specimens, brushing specimens, aspirates, and sputum specimens and is believed to result from mechanical crushing of fragile nuclei with little cytoplasmic support. Biopsy studies of pulmonary SCC are not uncommonly interpreted as nondiagnostic owing to extensive crush artifact obscuring morphologic details. In many cases, cytologic examination of sputum, bronchial brushing, and washing specimens optimizes the diagnostic yield of endobronchial biopsies. Compared with SCC, CT has a lower cellular density and the cells have more cytoplasm that prevents extensive nuclear distortion. However, artifacts occasionally happen and can be misleading.
Immunohistochemical analysis has little established use in the grading of NENs of the lung. CTs usually stain diffusely and strongly positive for synaptophysin and chromogranin A. The immunoreactivity for these markers generally is not as strong or universal in SCC; however, considerable overlap occurs. Both tumors express cytokeratins. Other markers that might be valuable in grading include p53 protein, which commonly is overexpressed, and Rb protein, which commonly is lost in high-grade NENs while low- and intermediate-grade NENs usually express it.
Previous studies have demonstrated a strong correlation between the proliferative index, as detected by Ki-67 labeling by MIB-1 or Ki-67 antibodies, and grade and prognosis. In the articles in which Ki-67 labeling indices were stated numerically, most intermediate-grade NENs had a proliferative index of less than 25%, most CTs of less than 10%, and most SCCs of substantially more than 25%.
The purpose of the present study was to evaluate the usefulness of immunoperoxidase staining for Ki-67, cytokeratin, chromogranin A, and synaptophysin in the differential diagnosis of CTs and SCCs with particular emphasis on the preservation of immunoreactivity in areas with crush artifact.
SHARE
