Tissue microarrays for testing molecular biomarkers of cervical intraepithelial neoplasia: feasibility study

To test the possibility of creating tissue microarrays of pre-malignant lesions of the cervix. STUDY DESIGN Paraffin-embedded blocks of 240 cervical tissue specimens were sampled. Lesions from benign squamous and glandular epithelium through various grades of cervical intraepithelial neoplasia (CIN) to frank carcinoma of squamous and glandular origin were cored with a 0.6-mm needle and arrayed in 4 tissue blocks. Sections of these blocks were stained with hematoxylineosin (H-E) and evaluated as to adequacy of tissue cores, representativity of the material and correspondence to the original diagnosis. Immunohistochemical staining with p16 and a novel marker C4.8(4/2/#1) was performed.

In > 80% of cases sufficient material from the lesion could be obtained. No or inadequate material was seen in 6% of cases. The core sample did not correspond to the original diagnosis in 12% of cases. The reason was mainly a discrepancy in the grade of the CIN. Discrepancies in diagnoses occurred in only premalignant lesions. Immunohistochemical staining could reliably be performed and evaluated on all tissue cores.

Tissue microarrays of cervical intraepithelial lesions are technically feasible and can be created reliably. The key to success is a careful and repeated comparison of the tissue block with the corresponding H-E section. Tissue microarrays of preinvasive cervical lesions may allow high throughput analysis of emerging molecular biomarkers in cervical carcinogenesis.