Risk of precancer and follow-up management strategies for women with human papillomavirus-negative atypical squamous cells of undetermined significance

OBJECTIVE: To investigate the relative performances of follow-up cytology and carcinogenic human papillomavirus (HPV) DNA testing among carcinogenic HPV-negative women with atypical squamous cells of undetermined significance (ASCUS), for detection of cervical precancer. METHODS: Twelve-month follow-up management strategies to detect cervical intraepithelial neoplasia grade 3 (CIN3) or worse using cytology or HPV testing or both were compared among women with HPV-negative ASCUS in the Atypical Squamous Cells of Undetermined Significance-Low-Grade Squamous Intraepithelial Lesion (ASCUS-LSIL) Triage Study. RESULTS: Overall only 22 of 1,559 (1.4%) HPV-negative ASCUS women developed CIN grade 3 or worse during follow-up compared with 269 of 1,767 (15.2%) HPV-positive ASCUS women (P<.001). Because of the low risk of disease among HPV-negative ASCUS women, only 7 cases of CIN3 were diagnosed between 12 and 24 months of follow-up, limiting power to distinguish meaningful differences in sensitivity among 12-month testing strategies. The specificity of HPV testing (84%) was significantly higher than cytology using an ASCUS threshold (71%) (P<.001). Cotesting with cytology and HPV testing at 12 months resulted in even lower specificity (61%). Because cases were uncommon, the positive predictive value for subsequent CIN3 or worse was low for cytology (2.6%), Hybrid Capture 2 (3.8%), and cotesting with cytology and HPV testing (2.2%). The negative predictive value for all three management strategies was very high (99.70%, 99.82%, and 100.0% for HPV testing, cytology, or cotesting, respectively.) CONCLUSION: Women with HPV-negative ASCUS have very low absolute risk of subsequently detected CIN3 or worse in the subsequent 2 years, similar to women with a negative cytology in the absence of HPV testing. The results suggest that women with HPV-negative ASCUS should return to routine screening intervals which may be longer than 1 year depending on age and screening history. However, if increased surveillance is chosen, a single HPV test for carcinogenic types at 12 months has significantly higher specificity and lower referrals than cytology.