A-234 Self-collected Vaginal Swabs for High-risk HPV Testing in the us: Validation of Pre-analytical Variables

Abstract

Abstract Introduction Several clinical trials have illustrated that primary HPV screening with unsupervised self-collected vaginal swabs is clinically appropriate; however, there is little data regarding quantitative changes imposed by pre-analytical components when samples are self-collected. Here we took a detailed analysis of the variables related to self-collected vaginal swabs to calculate analytical acceptability metrics and to identify limitations that can be improved to aid in modern cervical cancer screening pathways. Methods HPV was evaluated in self-collected dry vaginal swabs or paired self-collected vaginal/provider-collected endocervical swabs using an assay that is FDA-approved for provider-collected HPV primary screening. The assay includes amplification of endogenous hemoglobin subunit beta gene (HBB;beta-globin) as an internal control. Paired provider/self-collected samples (n = 144) were compared for HPV and beta-globin detection. Self-collected dry vaginal swabs (n = 15 participants;5 swabs/participant) were used to calculate intra-individual variability in cycle threshold (Ct) between swabs. A second set of self-collected swabs (n = 20 participants;3 swabs/participant) were challenged using winter or summer simulation with temperature cycling designed to mimic extreme seasonal fluctuations. Ct values for seasonal challenges were compared to unchallenged swabs. Various approaches were used to optimize collections: (1) a time course to see if the number of hours a self-collected vaginal swab remained dry before suspending in buffer influenced detection; (2) an alternate internal control (protein only RNase P catalytic subunit gene, PRORP) was amplified to exclude sample-specific inhibitor(s); (3)participants were provided an instructional video to aid collection. Results Most samples were concordant between provider and self-collect (n = 60 HPV positive; n = 60 HPV negative). There were 20 discrepant samples: 12 negative for HPV with provider-collected, but positive for HPV using self-collected swabs and 4 positive for HPV with provider-collected, but negative for HPV using self-collected swabs. All discrepancies were due to low viral load; for provider and self-collected swabs the average Ct (SD) in discrepant samples was 31.4 (3.5) and 33.1 (2.7), respectively, for HPV and 25.8 (2.1) and 27.1 (2.6), respectively, for beta-globin. The average self-collected intra-individual % CV for HPV and beta-globin was 3.5% and 2.7%, respectively. The average HPV Ct for swabs that cycled through the summer, winter or unchallenged conditions was 28.3 (SD = 5.9), 27.2 (SD = 7.1), and 25.7 (SD = 3.1), respectively. The average beta-globin Ct for swabs that cycled through the summer, winter or unchallenged conditions was 26.8 (SD = 2.7), 26.7 (SD = 3.0), and 26.0 (SD = 2.2), respectively. The time a sample remained dry did not influence detection. RNAse P amplification indicated results with invalid beta-globin were a function of cellularity not inhibition. The instructional video improved collection by ∼8.5%; 61/64 (95.3%) 125/144 (86.8%) were valid with and without the video, respectively. Conclusions Self-collected vaginal samples are a convenient and discreet way to screen for cervical cancer. Understanding the pre-analytical/analytical limitations can improve HPV detection.