Randomised study of HPV prevalence and detection of CIN2+ in vaginal self‐sampling compared to cervical specimens collected by medical personnel. Issue 1 (29th October 2018)

• Record Type: Journal Article
• Authors: Gustavsson, Inger; Aarnio, Riina; Berggrund, Malin; Hedlund‐Lindberg, Julia; Sanner, Karin; Wikström, Ingrid; Enroth, Stefan; Olovsson, Matts; Gyllensten, Ulf
• Is Part Of: International journal of cancer. Volume 144:Issue 1(2019)
• Page Start: 89
• Page End: 97
• Publication Date: 2018-10-29
• Subjects: cervical cancer -- self‐sampling -- HPV test -- screening -- randomised study; Cancer -- Periodicals; Cancer -- Prevention -- Periodicals; 616.994
• DOI: 10.1002/ijc.31637
• Languages: English
• ISSNs: 0020-7136

Abstract

Abstract : We conducted a randomised study to compare vaginal self‐sampling with assisted sampling by medical personnel on the cervix for HPV testing in primary screening. The first aim was to determine if the HPV prevalence is independent of sampling location (vagina versus cervix) and the person performing the sampling. The second aim was to evaluate if the two sampling strategies differed in the detection rate of CIN2+. In total, 19, 523 women were randomised into two groups, with 9926 invited to perform self‐sampling (SS arm) using the Rover VIBA‐brush and 9597 offered assisted sampling using the cytobrush (AS arm). All samples were applied to the indicating FTA elute card and analysed for high‐risk HPV using the hpVIR real‐time PCR assay. The outcome for the first aim was HPV prevalence and for the second aim the number of CIN2+ based on histology. In the SS arm, 52.7% of invited women participated in the study, as compared to 34.2% in the AS arm. All samples contained sufficient amount of nuclear DNA for a valid HPV result, with vaginal samples having a higher DNA amount than cervical samples ( p < 4.62 × 10 −11 ). HPV prevalence was 4.6% in the SS arm and 4.1% in the AS arm ( p = 5.5 × 10 −2 ), and the distribution of HPV types similar between arms. There was no difference in the prevalence of CIN2+ per 1000 women screened between arms ( p = 0.86). The results show that vaginal self‐sampling is an equivalent alternative to sampling by medical personnel for HPV typing and identification of CIN2+. Abstract : What's new? Vaginal self‐sampling for human papillomavirus (HPV), which enables women to collect vaginal samples themselves at home, is an effective method for HPV testing in low‐resource settings. However, whether it is more effective than cervical sampling by medical personnel for the detection of high‐grade lesions remains unclear. In the present study, the number of women with histologically confirmed cervical intraepithelial neoplasia grade 2+ (CIN2+) was independent of sample site, whether vaginal or cervical. Further, vaginal self‐sampling using an appropriate sampling device and a PCR‐based HPV assay was equivalent to assisted sampling in HPV prevalence.