Risk of cervical precancer among HPV–negative women in the Netherlands and its association with previous HPV and cytology results: A follow-up analysis of a randomized screening study

BACKGROUND: Human papillomavirus (HPV)-based screening programs still use one-size-fits-all protocols but efficiency and efficacy of programs may be improved by stratifying women based on previous screening results. METHODS AND FINDINGS: We studied the association between cervical intraepithelial neoplasia grade 3 or cancer (CIN3+) and previous screening results in the Population-Based Screening Study Amsterdam (POBASCAM) trial, performed in the Netherlands in the setting of regular screening, where women aged from 29 to 61 years old were invited to cytology and HPV co-testing at enrolment in year 1999/2002 and at the next round in 2003/2007. We selected 18,448 women (9,293 from the intervention group and 9,155 from the control group) who tested HPV–negative in 2003/2007 and did not have cervical intraepithelial neoplasia grade 2 or worse (CIN2+) or hysterectomy after enrolment. Follow-up was collected until 14 years after the 2003/2007 screen, covering 4 rounds of screening. Risk of CIN3+ and CIN2+ among women with an HPV–negative test, irrespective of previous round results and stratified according to previous round HPV and cytology results, were calculated by the Kaplan–Meier method. During 14 years of follow-up, 62 CIN3+ cases (24 in the intervention group and 38 in the control group) were detected. HPV–negative women had a 14-year CIN3+ risk of 0.48% (95% confidence interval 0.37 to 0.62) and CIN2+ risk of 1.17% (0.99 to 1.38). The CIN3+ risk among HPV–negative women was increased in women with a previous positive HPV test (2.36%, 1.20 to 4.63; p < 0.001) or co-test (1.68%, 0.87 to 3.20; p < 0.001) and, equivalently, decreased in women with a previous negative HPV test (0.43%, 0.33 to 0.57) or a negative co-test (0.43%, 0.33 to 0.57). The CIN3+ risk was not influenced by the previous cytology result. The CIN3+ risk among HPV–negative women was increased after both a previous HPV16–positive test (3.90%, 1.47 to 10.12; p < 0.001) and a previous HPV16–negative/HPVother–positive test (1.91%, 0.76 to 4.74; p = 0.002). For endpoint CIN2+ (147 cases), findings were similar except that the CIN2+ risk was increased after previous abnormal cytology (4.06%, 2.30 to 7.12; p < 0.001). The presented risk estimates were calculated by tracking histological results through the Dutch nationwide pathology archive (PALGA) and were not adjusted for non-compliance with the colposcopy referral advice. CONCLUSIONS: HPV–negative women had an increased long-term risk of CIN3+ when the HPV test in the previous screening round was positive. This supports the implementation of risk-based intervals that depend on HPV results in the current and previous screening round. TRIAL REGISTRATION: POBASCAM trial, trial registration number ISRCTN20781131.