Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



Research Advisor

Donna Hathaway, PhD


Barbara Benstein, PhD Patricia A. Cowan, PhD Jim Y. Wan, PhD Nadeem Zafar, MD


cancer; cytology; HPV; precancer; screening; Vulva


Introduction: Colposcopy and tissue biopsy remain the gold standard for diagnosing vulvar intraepithelial neoplasia (VIN). Vulvar colposcopy is fairly nonspecific. As a result, many women undergo unwarranted painful biopsies. Vulvar cytology, which is relatively painless, inexpensive and allows a larger area to be sampled, may likely reduce false negative diagnoses. However, previous cytological studies that used conventional methodologies were largely unsuccessful in diagnosing VIN. In this study, liquid based cytology and HPV typing by PCR have been assessed as possible alternatives to biopsy for follow-up surveillance of women treated for VIN.

Methods: Women with a history of VIN and a control group were recruited from a colposcopy clinic. Clinically suspicious lesions and normal vulvar tissue were vigorously brushed and the sample collected in PreservCyt® fluid for cytologic examination and for HPV-typing with a multiplex PCR assay, using primers designated PGMY09/11. Samples (N=82) were obtained from 52 lesions clinically suspicious for VIN, 15 controls from the same women in areas of the vulva with no clinical abnormality, and 15 controls from women with no current clinical evidence or past history of VIN. Concurrent tissue biopsies were obtained, immediately after brushing, from the 52 clinically suspicious VIN lesions. A single pathologist read and interpreted the biopsy samples. To ensure unbiased testing, cytology and HPV analyses were performed at separate independent laboratories by professionals blinded to clinical findings and biopsy results. Cytology was coded as negative, ASCUS, VIN I, VIN 2, or VIN 3. Specificity, sensitivity, and predictive values of cytology and HPV for VIN were calculated. Fisher’s exact was used to determine associations between HPV and cytology. Logistic regression was done to determine if cytology or HPV predicted tissue biopsy results.

Results: Vulvar samples (N=82) were collected from 48 women aged 19-65 who participated in this study. Histology results of the 52 lesions clinically suspected as VIN were reported as follows: VIN I (n=33), VIN 2/3 (n=13), benign (n=4), contact dermatitis and condyloma (n=1 each). Ninety percent of the vulvar samples were adequate for cytologic evaluation, but only 72% of samples had adequate cellularity for HPV testing. Sensitivity, specificity and positive predictive value (PPV) of vulvar cytology for recurrent VIN were 95%, 15% and 65% respectively. PCR for HPV, when independently correlated with histology, had, 62% sensitivity, 85% specificity and 89% PPV for VIN. No significant associations were found between cytology and HPV (p=0.3559). Neither cytology nor HPV predicted pathological diagnosis of VIN.

Conclusion: By vigorous brushing, it is possible to obtain an adequate cellular sample from the vulva for cytologic and/or molecular evaluation for HPV.The specificity for VIN at cytology was not satisfactory for use as a clinical alternative to biopsy to detect recurrent VIN. This may be due to the small sample size, difficulty in accurately grading vulvar dysplasia at cytology, and possible differences in cytomorphologic criteria for diagnosing dysplasia in the vulva as compared to the cervix. This study will be further refined with the development of more reproducible consensus criteria for cytologic evaluation of VIN. A larger number of participants could shed more light on the significance of cytology with or without HPV testing as an alternative to tissue biopsy for follow up of patients with VIN.