Objective Feasibility of a paired-sample study comparing cervical screening, and urine self-sampling, at 6- and 12-weeks postnatal. Design Paired-sample feasibility study. Setting Acute hospital. Population or Sample Females within 6-weeks of childbirth. Methods Paired-sample study comparing cervical screening and high-risk human papillomavirus (HPV) testing from clinician-taken samples and urine self-samples at 6- and 12-weeks postnatal. Main Outcome Measures Uptake rates, patient-reported outcomes, adverse events, HPV detection, sensitivity and specificity. Results Of 245 potential participants, 115 (47%) consented, 102 (89%) and 96 (83%) attended their 6- and 12-week week screening visits, respectively. Median pain scores for cervical screening did not differ (6-weeks = 1 (range 0-8); 12-weeks = 1 (range 0-7); p = 0.76). Most would be happy for future screening at 6-week postnatal (97/102 (95%) and 85/95 (89%) when asked at 6- and 12-week study visits, respectively; p = 0.11). Agreement rate of clinician-taken HPV tests between 6- and 12-weeks was 94.8% (91/96, 95% confidence interval (CI), 88.4%-97.8%), with no inadequate combination HPV/cytology tests (sensitivity 80% (95%CI) 44.4-97.5; specificity 96.5% (95%CI 90.1-99.3); negative predictive value (NPV) 97.6% (95%CI 91.8-99.7). Relative to 12-week clinician-taken samples, urine self-sampling may have lower sensitivity (e.g., 6-week: sensitivity 60.0% (95%CI 26.2-87.8); specificity 96.5% (95%CI 90.1-99.3); NPV 95.4% (95%CI 88.6-98.7)). Conclusions A paired-sample study of cervical screening at 6- and 12-weeks is feasible. Inadequacy rates of cervical screening were low, with high agreement. Urine self-sampling may be less sensitive in this population. An adequately powered study is needed to test for non-inferiority.

Effective Cervical screening involves a complex sequence of interactions between women, clinical staff and systems. Even in well organised screening programmes, achieving high population coverage is challenging. The transition to using Human Papilloma Virus (HPV) testing as the primary screening modality, rather than cytology, has introduced the possibility of using a self-collected sample for the primary HPV test. This has been shown to be a reliable method of detecting CIN2+ (Polman et al, Lancet Oncology 2019;20(2):229-38).Regardless of the way the sample is collected, the low specificity of a positive HPV test means that positive results must be triaged. In all cases where self-sampling has been integrated into a cervical screening programme to date, this triage is by cytology on a subsequent, clinician taken sample. Other molecular tests which can be performed on the original sample may in the future be an alternative, but are not yet fully evaluated.It has generally been assumed that a self-collected sample would not be suitable for cytology because the cervix is unlikely to be thoroughly (if at all) sampled.However, the authors have shown that using a well-established self-sampling device, 95.8% samples had at least 5000 cells, the usual threshold for adequacy for cytology. (ThinPrep LBC method). Results were highly specific, and in fact the positive predictive value (PPV) for high grade disease was higher on the self-collected specimens than on the subsequent professionally taken samples (Loopik et al. BJOG 2020 xxxx).As expected, the sensitivity of the test was lower than for a professionally taken sample, but it still detected 29.4 % of CIN2+. In this study, an impressive 91.9% of women returned for their reflex cytology test. This is higher than in other studies, possibly because the study was not focussed on women who have previously not responded to invitation.It is likely that a significant proportion of women needing treatment could be reliably identified by self-sampling alone by this method.It is worth noting that despite implementation in several organised screening programmes, self-sampling tests for HPV have yet to achieve regulatory approval, and the process for doing so for cytology from self-collected samples is likely to require studies with more extended monitoring of clinical outcomes.Self-sampling is not yet widely implemented internationally and the pathways can be challenging. There is a concern about how to address loss to follow up, design of multistep pathways for management, and potential delays in diagnosis. A pathway which could reliably prioritise women for referral for immediate colposcopy, with a high PPV for CIN2+, without requiring a face to face clinical interaction would be extremely valuable. This is potentially even more topical in the post COVID-19 world where there are new challenges in risks of consultations and in capacity of health care delivery.No disclosures: A completed disclosure of interest form is available to view online as supporting information.