Value of immunohistochemistry in crushed areas of pulmonary neuroendocrine carcinoma.

Immunohistochemical demonstration of neuroendocrine differentiation is often performed in routine diagnostic practice for lung neuroendocrine carcinoma. However, these carcinomas are often crushed, especially on small specimens. The value of immunohistochemistry on crushed areas is not known. We aimed to assess the value of immunohistochemical markers in crushed areas. We performed a retrospective study of 299 patients with a diagnosis of pulmonary neuroendocrine carcinoma. We showed that the markers TTF-1, synaptophysin, chromogranin A, CD56, and INSM1 were more often negative in crushed areas compared with well-preserved areas. The proliferation index with anti-Ki67 was decreased but remained on average around 90%. For all markers, the percentage of labeled cells was lower than in the preserved areas. Finally, we show that cases without labeling in the crushed areas and maintained labeling in the non-crushed areas have a lower percentage of labeling than cases without this labeling mismatch. Finally, there were no false positives of these stains. Neuroendocrine markers are valid in crushed areas when positive. However, the percentage of labeled cells may be lower than on preserved areas and lead to false negatives. Finally, the proliferation index, although decreased, remains close to that on preserved areas.

Proteomic Imaging of Vestibular Schwannomas and Normal Nerves. Histopathologic Correlations.

OBJECTIVES: Proteomic analysis of vestibular schwannoma (VS), non-vestibular schwannoma (NVS), and normal nerve (NN) using mass spectrometry and imaging of matrix assisted laser desorption ionization-time of flight (MALDI-TOF). METHODS: Retrospective, qualitative, and descriptive study on VS, NVS, and NN. Samples were provided by our Tumor Bank. They were analyzed histologically then sprayed by acid matrix. The laser beam of MALDI performed desorption-ionization of the sample. A mass spectrogram (MS) was drawn depending on time of flight of ionized peptides, and MALDI-imaging was obtained which is a summation color spectrum depending on sample's peptide content. The slice was reexamined histologically and results compared with MALDI-imaging. RESULTS: Fifty schwannomas were sampled, of which 27 exploitable: 22 VS (17 Antoni type A and five type B) and five NVS (all Antoni type B). Eleven NN were analyzed. Among the 22 VS, near-total correlation between MALDI-imaging and pathology was found in two cases (9.1%), partial correlation in four (18.2%), and no correlation in 16 (72.7%); correlations were more frequent in VS of the Antoni type B. MS showed a peptide spike at 2,000 m/z in 7 (31.8%) and 5,000 m/z in 21 (95.5%). Among the five NVS, near-total correlation was found in three cases (60%), partial correlation in one (20%), and no correlation in one (20%). MS showed a peptide spike at 2,000 m/z in two (40%) and 5,000 m/z in all (100%). Among the 11 NN, near-total correlation was found in nine cases (81.8%), partial correlation in one (9.1%), and no correlation in one (9.1%). MS showed no peptide spike at 2,000 or 5,000 m/z. Behind homogeneous areas on histology, there was great heterogeneity on MALDI-imaging and MS, regarding VS and NVS, but not NN. CONCLUSIONS: There was a lack of correlation between MALDI-imaging and pathology in VS (except Antoni type B) as compared with NVS and NN. The lack of correlation in VS of the type A as compared with type B VS and NVS could be attributed to the overexpression of degeneration-associated proteins/peptides in VS of the type B as well as NVS that are better correlated with histologic findings. The two peptide spikes detected in schwannoma and not in NN opens up the prospect of tumor biomarkers identifiable by sequencing. The proteomic polymorphism found in VS and NVS was absent on histology which is a new morphologic characteristic of schwannoma. Further studies should be performed in the future to confirm the benefit and usefulness of the MALDI in the analysis of VS and NVS.