AbstractTo identify new personal biomarkers for the improved diagnosis, prognosis and biological follow-up of human papillomavirus (HPV)-associated carcinomas, we developed a generic and comprehensive Capture-HPV method followed by Next Generation Sequencing (NGS). Starting from biopsies or circulating DNA samples, this Capture-NGS approach rapidly identifies the HPV genotype, HPV status (integrated, episomal or absence), the viral-host DNA junctions and the associated genome rearrangements. This analysis of 72 cervical carcinomas identified five HPV signatures. The first two signatures contain two hybrid chromosomal–HPV junctions whose orientations are co-linear (2J-COL) or non-linear (2J-NL), revealing two modes of viral integration associated with chromosomal deletion or amplification events, respectively. The third and fourth signatures exhibit 3–12 hybrid junctions, either clustered in one locus (MJ-CL) or scattered at distinct loci (MJ-SC) while the fifth signature consists of episomal HPV genomes (EPI). Cross analyses between the HPV signatures and the clinical and virological data reveal unexpected biased representation with respect to the HPV genotype, patient age and disease outcome, suggesting functional relevance(s) of this new classification. Overall, our findings establish a facile and comprehensive rational approach for the molecular detection of any HPV-associated carcinoma and definitive personalised sequence information to develop sensitive and specific biomarkers for each patient.