Cervical Cancer Screening (PDQ®): Screening - Health Professional Information [NCI] - Description of the Evidence
Testing for HPV DNA as a primary screening test has been FDA approved only in conjunction with cervical cytology and only in women aged 30 years and older. Women who are negative by cytology and HPV testing are at extremely low risk of CIN 3+ and therefore may be screened less often. A prospective cohort study of nearly 332,000 U.S. women aged 30 years and older undergoing HPV DNA and cervical cytology co-testing every 3 years found that the cumulative incidence of CIN 3+ in women negative for both tests at baseline was 0.047% at 3 years and 0.16% at 5 years. Screening more frequently than every 3 years would not improve sensitivity significantly but would increase costs and overtreatment.[51,52]
Numerous studies have demonstrated that, compared with cytology, HPV DNA testing is more sensitive for identifying women who have CIN 2+ (range of sensitivities 84%–97%).[32,53,54,55,56,57,58] In one randomized trial using both Pap and HPV testing in random order among women aged 30 to 69 years, sensitivity of HPV was 95% compared with 55% for Pap cytology. The combination of HPV and cytology had 100% sensitivity and a referral rate of 7.9%.
The lower specificity of HPV DNA testing compared with cytology is a consideration. Among women older than 30 years, cytology had a specificity of 97% compared with 94% for HPV testing. The specificity of HPV DNA testing would likely be even lower among women younger than 30 years, who have more transient HPV infection that is of little consequence. Thus, detecting such women would potentially increase the number of follow-up diagnostic workups. Potential approaches to minimize over-referral with HPV DNA testing and improve specificity include: (1) triage HPV-positive results with cytology  or another more specific molecular assay; and (2) trigger further workup only after two sequential positive HPV test results because it is the persistence of carcinogenic HPV that confers the greatest risk of CIN 2–3.[60,61]
An Italian population-based randomized, controlled trial of HPV DNA testing versus cervical cytology performed at 3-year intervals in approximately 94,000 women aged 25 to 60 years found a statistically significant decrease in the number of invasive cervical cancer cases diagnosed in the HPV DNA arm at the second round of screening (0 cases vs. 9 cases; P = .004). However, about 48% of individuals in the HPV DNA arm also received conventional cytology testing at the first screening round, making it impossible to discern whether the observed difference resulted from the use of a combined testing strategy or HPV DNA testing alone. Of note, many more women in the HPV DNA arm were referred to colposcopy for abnormal findings than in the cytology-alone arm (4,436 women vs. 1,416 women), prompting the authors to conclude that if the HPV DNA test is used as a primary screening strategy, women with positive test results should be triaged by cytology before referral.