Context: Identification of early invasion in vulvar intraepithelial neoplasia 3 (VIN 3) and cervical intraepithelial neoplasia 3 (CIN 3) may be difficult with the use of routine hematoxylin-eosin staining. Presence of obscuring inflammation and tangential tissue sectioning are the most common diagnostic pitfalls.
Objective: To examine the utility of double immunostaining for cytokeratin-collagen IV or cytokeratin-laminin in the detection of early invasion in VIN 3 and CIN 3.
Design: The study group consisted of 10 cases of "VIN 3, suspicious for invasion" and 10 cases of "CIN 3, suspicious for invasion." The negative control group consisted of VIN 3 (n = 15) and CIN 3 (n = 10). The positive control group consisted of cases of invasive vulvar carcinoma (n = 11) and invasive cervical carcinoma (n = 25). All cases were double immunostained for cytokeratin and collagen IV and, in a separate reaction, for cytokeratin and laminin. The continuity of the basement membrane and the presence of stromal invasion were assessed in the stained sections.
Results: The staining for collagen IV and laminin yielded identical results. A well-defined, continuous basement membrane was visualized in all cases of VIN 3 and CIN 3. A discontinuous or absent basement membrane was observed around the malignant cells on the invasive tumor front in all cases of vulvar and cervical carcinoma. In 2 of 10 cases of VIN 3, suspicious for invasion and in 4 of 10 cases of CIN 3, suspicious for invasion definitive foci of microinvasion were identified with the use of double immunostaining. A well-defined, continuous basement membrane was present in the remaining cases "suspicious for invasion."
Conclusion: Double immunostaining for cytokeratin- collagen IV or cytokeratin-laminin is useful for evaluation of early invasion in equivocal cases of VIN 3 and CIN 3.