• adenocarcinoma in situ (AIS);
  • human papillomavirus (HPV);
  • vaccine;
  • natural history


The primary objective of this report is to describe the detection of adenocarcinoma in situ (AIS) and associated human papillomavirus (HPV) type distribution that was observed in the context of two phase 3 clinical trials of a quadrivalent HPV6/11/16/18 vaccine. In this intention-to-treat analysis, we include all women who had at least one follow-up visit postenrollment. Healthy women (17,622) aged 15–26 with no history of HPV disease and a lifetime number of less than five sex partners (average follow-up of 3.6 years) were randomized (1:1) to receive vaccine or placebo at day 1, months 2, and 6. Women underwent colposcopy and biopsy according to a Papanicolaou triage algorithm. All tissue specimens were tested for 14 HPV types and were adjudicated by a pathology panel. During the trials, 22 women were diagnosed with AIS (six vaccine and 16 placebo). There were 25 AIS lesions in total, with HPV16/18 present in 96% (24 of 25 with 15 of 25 as single infections). Only two of 22 women had concomitant cytology results suggesting glandular abnormality. Colposcopic impressions (25 total) were either negative or indicated squamous lesions only. Of women with AIS, all six in the vaccine cohort and seven of 16 in the placebo cohort were infected at baseline with the same HPV type that was detected in the AIS lesion. Concurrent squamous lesions were detected in 20 of these 22 women. In summary, our findings show that AIS evades colposcopic and cervical cytologic detection. As most AIS lesions were HPV16/18-related, prophylactic HPV vaccination should reduce the incidence of invasive adenocarcinoma.