A semi-automated microscopic image analysis method for scoring Ki-67 nuclear immunostaining.

Area of Science:

• Oncology
• Pathology
• Biomarker Research

Background:

• The diagnostic and prognostic significance of Ki-67 nuclear staining intensity and location (nuclear gradient, NG) remains underexplored in tumor pathology.
• Pulmonary typical carcinoid (PTC) tumors require refined prognostic markers for improved patient management.

Approach:

• A novel method was developed to classify Ki-67 NG, stratifying it into NG1, NG2, and NG3/4 categories.
• A semi-automated image analysis protocol was established to quantify Ki-67 NG in PTC, pulmonary adenocarcinoma (PAD), and breast ductal carcinoma (BDC).
• Intraobserver and interobserver agreement for Ki-67 NG determination were assessed in tumor specimens.

Key Points:

• Ki-67 NG1 and NG2 were observed at lower levels in PTC compared to PAD and BDC.
• Cox multivariate analysis revealed that Ki-67 NG1 and NG2 significantly predicted clinical outcomes in PTC, independent of age and lymph node status.
• The proposed semi-automated quantification method demonstrated high intraobserver consistency and moderate interobserver agreement.

Conclusions:

• Ki-67 nuclear gradient (NG) is a significant prognostic indicator in pulmonary typical carcinoid (PTC) tumors.
• The developed semi-automated image analysis protocol for Ki-67 NG quantification offers a valuable tool for PTC diagnosis and prognosis.
• Further validation of this method could enhance the clinical utility of Ki-67 immunohistochemistry in thoracic oncology.