INTRODUCTION: There have been public health concerns about a potential association between human papillomavirus (HPV) vaccines and premature ovarian failure (POF) in young women. OBJECTIVE: To identify a potential safety signal of POF after HPV vaccination using the United States (US) Vaccine Adverse Event Reporting System (VAERS) database. METHODS: We manually selected relevant MedDRA preferred terms related to POF and identified in VAERS all POF reports in women less than 40 years of age between 2 July 1990 and 14 May 2018, followed by a review of narratives to confirm the cases. We conducted descriptive analyses on age, POF type, HPV vaccine type (HPV2, HPV4, HPV9), time to onset of POF, and dose rank. We described trends in reporting over time and assessed a potential safety signal using the proportional reporting ratio (PRR). RESULTS: Of the 228,341 eligible POF reports, 281 (0.1%) were suspected to be associated with HPV vaccines. Median patient age was 15 years (range 11-39 years). POF events consisted mainly of amenorrhea (80.4%) and premature menopause (15.3%). Mean number of reported POF events significantly increased after the first HPV vaccine launch in 2006 with 22.2 POF cases/year up from 1.4 POF cases/year before the launch. PRR was 46.1 (95% confidence interval: 31.7-67.2) and sensitivity analyses yielded similar estimates. CONCLUSION: Our study suggests the presence of a potential safety signal of POF associated with HPV vaccination, which may only be partly attributed to notoriety bias. Due to the well-known limitations of spontaneous reporting data, further investigations are warranted.< Réduire