Table of Contents Author Guidelines Submit a Manuscript
Cellular Oncology
Volume 29 (2007), Issue 6, Pages 483-495

Objective Assessment of Cancer Biomarkers Using Semi-Rare Event Detection

Jeroen A. W. M. van der Laak, Albertus G. Siebers, Sabine A. A. P. Aalders, Johanna M. M. Grefte, Peter C. M. de Wilde, and Johan Bulten

Department of Pathology, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands

Copyright © 2007 Hindawi Publishing Corporation and the authors. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Objective and reproducible assessment of cancer biomarkers may be performed using rare event detection systems. Because many biomarkers are not true ‘rare events’, in this study a semi-rare event detection system was developed. The system is capable of assigning a discriminant score to detected positive cells, expressing the extent and intensity of the immunocytochemical staining. A gallery image is constructed showing the diagnostically most interesting cells as well as quantitative data expressing the biomarker staining pattern. To increase scanning speed, an adaptive scanning strategy is studied in which scanning is aborted when a sufficient number of positive cells has been identified. System performance was evaluated using liquid based cervical smears, stained with an antibody directed against p16INK4a tumor suppressor protein. Overexpression of p16INK4a in cervix is related to high-risk HPV infection, which is associated with carcinogenesis. Reproducibility of the system was tested on specimens containing limited positivity. Quantitative analysis was evaluated using 10 cases within normal limits and 10 high grade lesions. The system was highly reproducible in detecting positive cells and in calculating discriminant scores (average CV 0.7%). Quantitative features were significantly increased in high grade lesions (p < 0.001). Adaptive scanning decreased scanning time with only minor impact on scanning results. The system is capable of automated, objective and reproducible assessment of biomarker expression and may be useful for a variety of applications.