Morphometry confirms the presence of considerable nuclear size overlap between small cells and large cells in high-grade pulmonary neuroendocrine neoplasms

• Published in: American journal of clinical pathology 2001-10, Vol.116 (4), p.466-472
• Main Authors: MARCHEVSKY, Alberto M, GAL, Anthony A, SHAH, Swati, KOSS, Michael N
• Format: Article
• Language: English
• Subjects: Biological and medical sciences; Carcinoma, Neuroendocrine - ultrastructure; Carcinoma, Small Cell - ultrastructure; Cell Nucleus - ultrastructure; Diagnosis, Differential; Humans; Lung Neoplasms - ultrastructure; Lymphocytes - ultrastructure; Medical sciences; Pneumology; Tumors of the respiratory system and mediastinum
• ISSN: 0002-9173; 1943-7722
• DOI: 10.1309/h40b-8w14-4q47-03ep

We morphometrically evaluated 5-micron H&E-stained sections from 28 surgically resected high-grade pulmonary neuroendocrine neoplasms, including 16 small cell lung carcinomas (SCLCs) and 12 large cell neuroendocrine carcinomas (LCNECs). For each case, 200 tumor nuclei and 20 to 100 normal lymphocytes were measured. The frequency distributions of tumor cell/lymphocyte (TC/L) size ratios were plotted in bins ranging from 1 to 6, classified into 6 histogram types with TC/L size ratio peaks ranging from 2 to 6 (A-E) and a histogram with a wide distribution (F). SCLCs fit histograms A through E; LCNECs, A through F. Morphometry demonstrated considerable nuclear size overlap in high-grade neoplasms. Approximately one third of SCLCs exhibited considerable numbers of neoplastic cells that were larger than 3 normal lymphocytes, while 4 of 12 LCNECs had a predominant number of small cells. Ten tumors exhibited a B histogram with a "borderline" peak TC/L of 3. The rule that a TC/L size ratio larger than 3 helps distinguish "large" from "small" neoplastic cells was confirmed in only 9 of 28 cases. The use of more generic terminology such as "high-grade neuroendocrine carcinoma" or "grade III neuroendocrine carcinoma" for SCLC and LCNEC is discussed.