Číslo abstraktu: XVIII/ 255
The aim of this study is correlation of proliferation index Ki-67 and/ or mitotic count (MC) in a group of gastroenteropancreatic and lung neuroendocrine neoplasmas in the Czech Neuroendocrine Tumour Registry with median overall survival of the patients according to the origin of the primary tumour, regardles to the stage of the disease.
Materials and Methods:
In April 2014 data of 1,151 patients in the Czech Neuroendocrine Registry have been collected from hospital records, including demography, details of histology, staging, treatment and its outcome. 802 patients had information about Ki-67 and/ or MC status. They were divided into 3 groups: Ki-67/ MC < 2, Ki-67/ MC 2–20, Ki-67/ MC > 20. The retrospective analysis of proliferation index/ mitotic count influence on median overall survival (OS) according to the origin of the primary tumour was done. Results: 316 pts (39.4%) had Ki-67/ MC < 2, 385 pts (48%) had Ki-67/ MC between 2 and 20 and 101 pts (12.6%) had Ki-67/ MC > 20 (Fig. 1). According to topography Ki-67/ MC < 2 was most frequent in colorectal NENS, Ki-67/ MC between 2 and 20 was most frequent in stomach, small bowell and pancreatic NENS. The highest frequency of Ki-67/ MC > 20 was in lung and pancreatic NENS (Fig. 2). Median overall survival in all patients with Ki-67/ MC > 20 was 2.0 years (1.5; 2.6), in two other groups wasn’t reached (Fig. 3). Individual fi gures for small bowel (Fig. 4), lung (Fig. 5) and pancreatic NENS (Fig. 6) strongly support the influence of Ki-67/ MC on median OS.
Ki-67 index/ MC is an essential predicting parameter. Findings from the Czech Neuroendocrine Registry confirmed the highest frequency of Ki-67/ MC > 20 in lung and pancreatic NENs. Median OS in this group of patients in lung was 2.7 years (0.0; 5.8), median OS in pancreatic NENs 1.4 year (0.6; 2.1) in contrast to groups with Ki-67 < 20, where median OS was not reached. We strongly support the idea for using Ki-67 in classification of lung neuroendocrine tumours. Cut-off between G1 and G2 tumours in GEPNETs is also the important question for future discussion.
Datum přednesení příspěvku: 9. 4. 2015