Diagnostic, Prognostic, and Predictive Role of Ki67 Proliferative Index in Neuroendocrine and Endocrine Neoplasms: Past, Present, and Future

Abstract

The introduction of Ki67 immunohistochemistry in the work-up of neuroendocrine neoplasms (NENs) has opened a new approach for their diagnosis and prognostic evaluation. Since the first demonstration of the prognostic role of Ki67 proliferative index in pancreatic NENs in 1996, several studies have been performed to explore its prognostic, diagnostic, and predictive role in other neuroendocrine and endocrine neoplasms. A large amount of information is now available and published results globally indicate that Ki67 proliferative index is useful to this scope, although some differences exist in relation to tumor site and type. In gut and pancreatic NENs, the Ki67 proliferative index has a well-documented and accepted diagnostic and prognostic role and its evaluation is mandatory in their diagnostic work-up. In the lung, the Ki67 index is recommended for the diagnosis of NENs on biopsy specimens, but its diagnostic role in surgical specimens still remains to be officially accepted, although its prognostic role is now well documented. In other organs, such as the pituitary, parathyroid, thyroid (follicular cell-derived neoplasms), and adrenal medulla, the Ki67 index does not play a diagnostic role and its prognostic value still remains a controversial issue. In medullary thyroid carcinoma, the Ki67 labelling index is used to define the tumor grade together with other morphological parameters, while in the adrenal cortical carcinoma, it is useful to select patients to treated with mitotane therapy. In the present review, the most important information on the diagnostic, prognostic, and predictive role of Ki67 proliferative index is presented discussing the current knowledge. In addition, technical issues related to the evaluation of Ki67 proliferative index and the future perspectives of the application of Ki67 immunostaining in endocrine and neuroendocrine neoplasms is discussed.

Keywords: Classification; Endocrine neoplasms; Ki67; Neuroendocrine neoplasms; Prognosis.

Fig. 1

Silent corticotroph PitNET (A) positive for TPIT (B) with a Ki67 index > 3% (C)

Fig. 2

Olfactory neuroblastoma showing the typical lobular architecture (A) and presenting a Ki67 proliferative index > 20% (B), which is associated with a worse prognosis than observed in cases with Ki67 < 20%

Fig. 3

Low-grade medullary thyroid carcinoma (A) showing a Ki67 proliferative index < 5% (B). High-grade medullary thyroid carcinoma showing necrosis (C) and a Ki67 proliferative index > 20% (D)

Fig. 4

Lung biopsies of neuroendocrine neoplasms frequently present crush artifacts creating difficulties in interpreting the morphological picture. In this context, Ki67 is useful to distinguish NET (carcinoid) from NEC. A refers to a NET, which shows low Ki67 labelling (B). (C) is a small cell carcinoma presenting high Ki67 immunolabelling (D) (courtesy of Prof. Giuseppe Pelosi, University of Milan, Milan, Italy)

Fig. 5

Lung NET G3 (atypical carcinoid with high proliferative rate). The tumor is composed of well-differentiated cells and a punctate necrosis is visible (A). The Ki67 proliferative index is > 30% (B) (courtesy of Prof. Giuseppe Pelosi, University of Milan, Milan, Italy)

Fig. 6

The Ki67 proliferative index is used to grade digestive neuroendocrine tumors (NETs). NET G1 is characterized by Ki67 index < 3% (A), NET G2 by Ki67 index > 3% but < 20% (B), while NET G3 shows a Ki67 index > 20% (C, courtesy of Prof. Silvia Uccella, Humanitas University, Milan, Italy)

Fig. 7

In A is presented an example of adrenal cortical carcinoma with Ki67 index < 10% (B), while in (C) an example of adrenal carcinoma with Ki67 > 10% (D). The Ki67 proliferative index plays a role in the selection of patients to treat with adjuvant mitotane therapy, indicated when Ki67 index is higher than 10%

Fig. 8

Pheochromocytoma (A) showing a Ki67 index of 3%, which has been proposed as the cutoff to separate patients in two different prognostic groups. In this case, a mitosis (arrow) is also observed in a Ki67 positive cell (B, courtesy of Prof. Silvia Uccella, Humanitas University, Milan, Italy)

Fig. 9

Example of Ki67 index evaluated by manually counting unlabeled and labeled nuclei on a camera-captured, printed image. In this picture, 81 Ki67-labeled cells were counted and then divided by a total of 1581 cells, resulting in a Ki67 index of 5.1% (republished with permission of Springer from the article: Klöppel et al. [90] Ki67 labeling index: assessment and prognostic role in gastroenteropancreatic neuroendocrine neoplasms. Virchows Arch 472:341–349)

Fig. 10

Example of automatic Ki67 count within a poorly differentiated thyroid carcinoma (courtesy of prof. Ozgur Mete, University Health Network, University of Toronto, Toronto, ON, Canada)