Study protocol of the ACCESS trial: a randomised trial to evaluate the effectiveness of human papillomavirus testing by self-sampling in cervical cancer screening uptake and precancer detection

Effectiveness of self-sampling human papillomavirus test on precancer detection and screening uptake in Japan: The ACCESS randomized controlled trial

Japan is lagging in cervical cancer prevention. The effectiveness of a self-sampling human papillomavirus (HPV) test, a possible measure to overcome this situation, has not yet been evaluated. A randomized controlled trial was performed to evaluate the effectiveness of a self-sampling HPV test on detection of cervical intraepithelial neoplasia grade 2 or worse (CIN2+) and screening uptake. Women between 30 and 58 years old who did not participate in the cervical cancer screening program for ≥3 years were eligible and assigned to the intervention group (cytology or self-sampling HPV test) or control group (cytology). Participants assigned to the intervention group were sent a self-sampling kit according to their ordering (opt-in strategy). A total of 7337 and 7772 women were assigned to the intervention and control groups, respectively. Screening uptake in the intervention group was significantly higher than that in the control group (20.0% vs. 6.4%; risk ratio: 3.10; 95% confidence interval [CI]: 2.82, 3.42). The compliance rate with cytology triage for HPV-positive women was 46.8% (95% CI: 35.5%, 58.4%). CIN2+ was detected in five and four participants in the intervention and control groups, respectively; there was no difference for intention-to-screen analysis (risk ratio: 1.32; 95% CI: 0.36, 4.93). Self-sampling of HPV test increased screening uptake; however, no difference was observed in the detection of CIN2+, probably due to the low compliance rate for cytology triage in HPV-positive women. Efforts to increase cytology triage are essential to maximize precancer detections.

Acceptability of self-sampling human papillomavirus test for cervical cancer screening in Japan: A questionnaire survey in the ACCESS trial

PURPOSE

In terms of medical policy for cervical cancer prevention, Japan lags far behind other industrialized countries. We initiated a randomized controlled trial to evaluate the self-sampling human papillomavirus (HPV) test as a tool to raise screening uptake and detection of pre-cancer. This study was conducted to explore the acceptability and preference of self-sampling using a subset of the data from this trial.

METHODS

A pre-invitation letter was sent to eligible women, aged 30-59 years who had not undergone cervical cancer screening for three or more years. After excluding those who declined to participate in this trial, the remaining women were assigned to the self-sampling and control groups. A second invitation letter was sent to the former group, and those wanting to undergo the self-sampling test ordered the kit. A self-sampling HPV kit, consent form, and a self-administered questionnaire were sent to participants who ordered the test.

RESULTS

Of the 7,340 participants in the self-sampling group, 1,196 (16.3%) administered the test, and 1,192 (99.7%) answered the questionnaire. Acceptability of the test was favorable; 75.3-81.3% of participants agreed with positive impressions (easy, convenient, and clarity of instruction), and 65.1-77.8% disagreed with negative impressions (painful, uncomfortable, and embarrassing). However, only 21.2% were confident in their sampling procedure. Willingness to undergo screening with a self-collected sample was significantly higher than that with a doctor-collected sample (89.3% vs. 49.1%; p<0.001). Willingness to undergo screening with a doctor-collected sample was inversely associated with age and duration without screening (both p<0.001), but that with a self-collected sample was not associated.

CONCLUSIONS

Among women who used the self-sampling HPV test, high acceptability was confirmed, while concerns about self-sampling procedures remained. Screening with a self-collected sample was preferred over a doctor-collected sample and the former might alleviate disparities in screening rates.

Implementation of a self-sampling HPV test for non-responders to cervical cancer screening in Japan: secondary analysis of the ACCESS trial

A self-sampling human papillomavirus (HPV) test could improve the morbidity and mortality of cervical cancer in Japan. However, its effectiveness and feasibility have not been demonstrated sufficiently. Hence, we launched a randomized controlled trial, which is ongoing, and report the results of a secondary analysis. To ensure autonomous participation with a minimum selection bias, opt-out consent was obtained from women who met the inclusion criteria, and written consent was obtained from those who underwent a self-sampling test. The number of women who met the inclusion criteria was 20,555; 4283 and 1138 opted out before and after the assignment, respectively. Of the 7340 women in the self-sampling arm, 1372 (18.7%) ordered and 1196 (16.3%) underwent the test. Younger women in their 30 s and 40 s tended to undertake the test more frequently than older women in their 50 s (P for trend < 0.001). Invalid HPV test results were rare (1.3%), and neither adverse events nor serious complaints were reported. Despite adopting the opt-out procedure, more women than expected declined to participate, suggesting the need for a waiver of consent or assignment before consent to reduce selection bias. A self-sampling HPV test can be implemented in Japan and would be more accessible to young women, the predominant group affected by cervical cancer.