Naomi Oka a , b , Atsuko Kasajima a , c , d , * , Björn Konukiewitz c , Akira Sakurada e , Yoshinori Okada e , Toru Kameya f , Wilko Weichert c , d , Yuichi Ishikawa g , Hiroyoshi Suzuki b , Hironobu Sasano a , Günter Klöppel c
19 August 2019
The accuracy and reproducibility of the World Health Organization (WHO) 2015 classification of bronchopulmonary neuroendocrine neoplasms (BP-NENs) is disputed. The aim of this study is to classify and grade BP-NENs using the WHO 2019 classification of digestive system NENs (DiS-NEN-WHO 2019), and to analyze its accuracy and prognostic impact. Two BP-NEN cohorts from Japan and Germany, 393 tumors (88% surgically resected), were reviewed and the clinicopathological data of the resected tumors ( n = 301) correlated to patients’ disease-free survival (DFS). The DiS-NEN-WHO 2019 stratified the 350 tumors into 91 (26%) neuroendocrine tumors (NET) G1, 52 (15%) NET G2, 15 (4%) NET G3, and 192 (55%) neuroendocrine carcinomas (NEC). NECs, but not NETs, were immunohistochemically characterized by abnormal p53 (100%) and retinoblastoma 1 (83%) expression. The Ki67 index, which was on average 4 times higher than mitotic count ( p < 0.0001), was prognostically more accurate than the mitotic count. NET G3 patients had a worse outcome than NET G1 ( p < 0.01) and NET G2 patients ( p = 0.02), respectively. No prognostic difference was detected between NET G3 and NEC patients after 5 year DFS. It is concluded that stratifying BP-NEN patients according to the DiS-NEN-WHO 2019 classification results in 3 prognostically well-defined NET groups, if grading is solely based on Ki67 index. Mitotic count alone may underestimate malignant potential of NETs.