Evaluation of p16INK4a as a diagnostic tool in the triage of Pap smears demonstrating atypical squamous cells of undetermined significance

Abstract

P16(INK4a) (p16) has emerged as a biomarker for the detection of high-risk human papillomavirus (HR-HPV) in Papanicolaou (Pap) smears. Many studies have confirmed a strong correlation between p16 immunohistochemical positivity and high-grade squamous intraepithelial lesions (HSIL) of the cervix. Because p16 is predictive of HR-HPV and HSIL, it seems plausible that p16 could be used as a diagnostic tool to triage atypical squamous cells of undetermined significance (ASCUS) Pap smears. In this way, Pap smears with no p16 staining could be recategorized as negative for intraepithelial lesion or malignancy (NILM) before final case disposition, thus preventing unnecessary and costly follow-up. p16 immunostains were performed on 178 ThinPrep (Cytyc, Marlborough, Mass) Pap smears signed out as ASCUS among 5 cytopathologists. p16 stains were independently scored between 0 (no staining) and 4 (staining in cells with nuclear aberration) by either 2 or 3 pathologists. The p16 score was compared with both Hybrid Capture 2 (hc(2)) (Digene, Gaithersburg, Md) and follow-up (Pap smear and tissue) results. The sensitivity and specificity of p16 immunohistochemistry compared with both hc(2) and follow-up were not statistically significant, with both data subsets having P-values greater than .05. Statistical significance was not demonstrated in any of the data subsets, indicating that the p16 score alone cannot be used to recategorize Pap smears from ASCUS to NILM as a means to prevent unnecessary and expensive follow-up. Although not meeting criteria for statistical significance, the sensitivity and positive predictive value of p16 scores versus tissue follow-up only were more statistically favorable, suggesting that p16 has better correlation with tissue follow-up than results of hc(2). In addition, p16 staining was identified consistently in atrophic Pap smears, including 23 of 25 additional NILM atrophic smears stained, indicating that p16 cannot be used as a marker to triage atypical atrophic smears.