Automated assessment of Ki‐67 proliferation index in neuroendocrine tumors by deep learning
15
Citation
36
Reference
10
Related Paper
Citation Trend
Abstract:
The Ki-67 proliferation index (PI) is a prognostic factor in neuroendocrine tumors (NETs) and defines tumor grade. Analysis of Ki-67 PI requires calculation of Ki-67-positive and Ki-67-negative tumor cells, which is highly subjective. To overcome this, we developed a deep learning-based Ki-67 PI algorithm (KAI) that objectively calculates Ki-67 PI. Our study material consisted of NETs divided into training (n = 39), testing (n = 124), and validation (n = 60) series. All slides were digitized and processed in the Aiforia® Create (Aiforia Technologies, Helsinki, Finland) platform. The ICC between the pathologists and the KAI was 0.89. In 46% of the tumors, the Ki-67 PIs calculated by the pathologists and the KAI were the same. In 12% of the tumors, the Ki-67 PI calculated by the KAI was 1% lower and in 42% of the tumors on average 3% higher. The DL-based Ki-67 PI algorithm yields results similar to human observers. While the algorithm cannot replace the pathologist, it can assist in the laborious Ki-67 PI assessment of NETs. In the future, this approach could be useful in, for example, multi-center clinical trials where objective estimation of Ki-67 PI is crucial.Keywords:
Ki-67
Proliferation index
Proliferative index
Proliferation index
Ki-67
Mitotic index
Endoscopic Ultrasound
Proliferative index
Guideline
Cite
Citations (2)
The Ki-67 proliferation index (PI) is a prognostic factor in neuroendocrine tumors (NETs) and defines tumor grade. Analysis of Ki-67 PI requires calculation of Ki-67-positive and Ki-67-negative tumor cells, which is highly subjective. To overcome this, we developed a deep learning-based Ki-67 PI algorithm (KAI) that objectively calculates Ki-67 PI. Our study material consisted of NETs divided into training (n = 39), testing (n = 124), and validation (n = 60) series. All slides were digitized and processed in the Aiforia® Create (Aiforia Technologies, Helsinki, Finland) platform. The ICC between the pathologists and the KAI was 0.89. In 46% of the tumors, the Ki-67 PIs calculated by the pathologists and the KAI were the same. In 12% of the tumors, the Ki-67 PI calculated by the KAI was 1% lower and in 42% of the tumors on average 3% higher. The DL-based Ki-67 PI algorithm yields results similar to human observers. While the algorithm cannot replace the pathologist, it can assist in the laborious Ki-67 PI assessment of NETs. In the future, this approach could be useful in, for example, multi-center clinical trials where objective estimation of Ki-67 PI is crucial.
Ki-67
Proliferation index
Proliferative index
Cite
Citations (15)
The expression of the cell-cycle-associated Ki-67 antigen by MIB-1 monoclonal antibody was retrospectively assessed in 35 surgically resected neuroendocrine tumor specimens of the pancreas embedded in paraffin. The MIB-1 proliferation index was correlated with the classification of the neuroendocrine tumors of the pancreas proposed by Klöppel et al. Four prognostic factors showed a significant correlation with MIB-1: local invasion, metastases, tumor differentiation, and production of insulin. However, analysis by the Cox Proportional Hazards Regression Model showed that only local invasion was an independent predictor of outcome. Finally, our study showed a statistically significant increase in the number of deaths and a statistically significant decrease in survival time when the MIB-1 proliferation index was higher than 4%. We conclude that MIB-1 proliferation index is a simple and reliable tool to predict the clinical outcome of the neuroendocrine tumors of the pancreas. The index might be useful for determining the prognosis for an individual because of the significant decrease in survival when the index is higher than 4%.
Proliferation index
Ki-67
Proliferative index
Cite
Citations (76)
Wilms tumour
Value (mathematics)
Expression (computer science)
Cite
Citations (19)
Abstract Background Neuroendocrine tumors (NETs) constitute tumors widely distributed and with heterogeneous biological behavior. For gastrointestinal neuroendocrine tumors (GI-NETs) the following prognostic factors have been identified: location, production of hormones, size and proliferative grade. The latter must be calculated using proliferation index by the number of mitosis or the proportion of tumor cells positive for Ki67 immunostaining. The objective of this study was to use a quantitative tool to calculate the Ki67 index in GI-NETs. Material and methods We reviewed 40 cases of GI-NETs diagnosed at the Department of Pathological Sciences, Santa Casa de Misericórdia de São Paulo Hospital between 2004 and 2014 and compared the Ki67 index by manual count using scanned photomicrographs with semi-automated digital analysis (MC) and eyeball estimation (EE) of the histological slide. After Ki67 immunostaining, the slides were scanned with 3DHistech Pannoramic Scanners. Hot spots were selected and exported in a high-resolution image format and the Ki67 index was calculated with semi-automated image analysis software (AxioVision 3.0). Ki67 immunoreactivity was expressed as the percentage of tumor cells with nuclear staining (number of positive tumor cells/a minimum of 500 total tumor cells). Results We compared the classification of the neuroendocrine tumor by using the two methods in the semi-automated method 26 maintained the same grade, while 14 were re-classified, 4 being upgraded and 10 downgraded. Conclusion In the EE method there was a larger estimate of the percentage of positivity for KI67. As the Ki67 values are the criteria for the classification of neuroendocrine tumors, the semi-automated method can have less error.
Digital Image Analysis
Immunostaining
Ki-67
Proliferation index
Mitotic index
Automated method
Image Analysis
Proliferative index
Cite
Citations (4)
The aim of the present study was to clarify the conflicting recorded data on the proliferative features of gastric carcinoma. The Ki67 labelling index (Ki67 LI) was evaluated using MIB-1 in 43 carcinomas (24 diffuse and 19 intestinal). In 18 cases, differential counts were performed in superficial and deep layers. In ten diffuse carcinomas with a prominent desmoplastic response, Ki67 LI was evaluated in sections double-stained with MIB-1 and CAM5.2. Flow cytometry was performed in 26 cases. Ki67 LI of diffuse carcinomas (36·3±19·0) was not significantly different from that of intestinal carcinomas (28·2±18·5). Ki67 LI was significantly higher (P=0·006) in superficial than in deep areas (41·9±22·7 and 29·7±19·7, respectively) regardless of histological tumour type. No significant relationship was observed between Ki67 LI and wall invasion, lymph node metastasis, vascular invasion or ploidy. The following conclusions were drawn: double immunostaining techniques are apparently the best way to overcome the underestimation of cell proliferation in diffuse gastric carcinomas with a prominent desmoplastic response; the diffuse and intestinal types of gastric carcinoma have proliferation rates within the same range, even when the comparison is restricted to diploid tumours; and, finally, the major pool of proliferating cells resides in the superficial areas of gastric carcinomas, regardless of the histotype, which should be taken into consideration when overall counts are performed, using either immunohistochemical markers in tissue sections or suspensions of nuclei in flow cytometry. © 1997 John Wiley & Sons, Ltd.
Ki-67
Cite
Citations (45)
Ki-67
Immunostaining
Proliferation index
Proliferative index
Cite
Citations (99)
AIMS: To determine cell proliferation in infiltrating breast carcinomas. METHODS: Using the MIB-1 monoclonal antibody, the proliferation index was measured in paraffin wax sections of 871 breast cancers. The MIB-1 proliferation index was compared with other markers of disease progression: size, lymph node status, histotype, oestrogen and progesterone receptor status, expression of p53 and Neu, and DNA ploidy. All parameters were measured using image analysis. In 347 tumours, the MIB-1 and Ki-67 proliferation indexes were compared. Follow up data were available for 170 cases (median 66.5 months). RESULTS: Of the tumours, 314 (36%) had a high proliferation index. The MIB-1 proliferation index was correlated directly with size, nodal status, overexpression of p53 and Neu, and the DNA index; and inversely with oestrogen and progesterone receptor status. The correlation between MIB-1 and Ki-67 proliferation indexes was statistically significant. In patients with pT1 tumours, a low proliferation index correlated with a longer relapse-free interval and overall survival; node negative patients with a low proliferation index had a longer overall survival. CONCLUSIONS: The MIB-1 proliferation index is a reliable, practical and useful method of measuring proliferative activity and is an important predictor of clinical behaviour.
Proliferation index
Ki-67
Proliferative index
Proliferation Marker
Progesterone receptor
Cite
Citations (65)
Objective Pancreatic neuroendocrine tumors (NETs) (pNETs) have a varied prognosis according to their grade. The European Neuroendocrine Tumor Society grading system uses assessment of the proliferation index via Ki-67 immunohistochemistry to aid prognosis. There is evidence that the proliferation index can vary significantly within a single tumor, but it is not fully understood to what extent heterogeneity occurs between the primary and metastatic sites and how this may affect the grade. The aim of this study is to determine whether the grade assigned to a pNET varies depending on which site is selected for Ki-67 immunolabeling. Methods Patients were selected from our institution's NET database. Patients were included if they had a confirmed pNETs, had multiple resection specimens, and had consented to research being performed on their specimens. Ki-67 immunohistochemistry was performed on all resected specimens meeting the inclusion criteria. Results Pancreatic neuroendocrine tumors specimens resected from 16 patients were analyzed. There was no trend to higher Ki-67 in metastatic than primary disease. Ki-67 was on average 3% higher in liver metastases than lymph node metastases ( P < 0.001). Conclusions The grade of pNETs varies according to the tumor selected for Ki-67 immunolabeling. Useful information can be gained by performing Ki-67 PI on liver metastases.
Grading (engineering)
Ki-67
Proliferation index
Proliferative index
Tumor grade
Cite
Citations (20)
Abstract Background and Objectives We designed this study to assess the biologic significance of Ki‐67 proliferation index (PI) in gastric cancer. Methods Gastric cancer tissue from 245 patients were immunostained for Ki‐67. Ki‐67 PI was defined as the percentage of tumor cells positive for Ki‐67. In addition, we have previously evaluated the expressions of nine epithelial mesenchymal transition (EMT)‐related proteins. The relationship between Ki‐67 PI and clinicopathologic parameters, patient survival, and EMT data were sought. Results Low Ki‐67 PI was correlated with poorly differentiated histology ( P = 0.034), an advanced T stage ( P < 0.001), and lymph node metastasis ( P = 0.011). Also, the low PI group was found to have a significantly worse prognosis than the high PI group ( P = 0.003, log‐rank test). Multivariate analysis revealed that Ki‐67 PI remained as an independent prognostic factor (hazard ratio (95% CI) = 0.670 (0.450–0.999)). Furthermore, greater expressional changes of EMT‐related proteins were found to be significantly associated with low Ki‐67 PI ( P = 0.025). Conclusions These findings suggest that Ki‐67 PI is an effective tool for predicting survival in gastric cancer. In addition, we found that an invasive property presented as EMT‐related protein expressional changes was inversely correlated with a proliferative activity in gastric cancer. J. Surg. Oncol. 2010;102:201–206. © 2010 Wiley‐Liss, Inc.
Ki-67
Proliferation index
Proliferative index
Histology
Cite
Citations (36)