- Sensitivity and specificity of self-collected cervical samples for HPV testing were similar to clinician-collected samples for cervical intraepithelial neoplasia 2 or worse (CIN2+) and CIN3+.
- Higher number of women provided self-collected samples.
Why this matters
- Adopting self-sampling in screening programs will improve screening, reduce cost.
- Randomized, noninferiority IMPROVE study.
- Women, aged 29-61 years, were randomly allocated to self-sampling (n=8212) or clinician-based sampling group (n=8198).
- Noninferiority margin of the clinical sensitivity, 90%.
- Funding: Ministry of Health, Welfare, and Sport (Netherlands); others.
- In self-collected vs clinician-collected samples:
- Number of samples collected: (93.3% vs 76.9%).
- Samples tested positive for HPV: 7.4% vs 7.2% (relative risk, 1.04; 95% CI, 0.92-1.17).
- CIN3+: incidence, 11% vs 8%; relative sensitivity, 0.99 (95% CI, 0.92-1.07); and relative specificity, 1.00 (95% CI, 0.99-1.01).
- CIN2+: incidence, 6% vs 8%; relative sensitivity, 0.97 (95% CI, 0.91-1.03); and relative specificity, 1.00 (95% CI, 0.99-1.01).
- Number of cancer cases: 3 vs 2.
- Median follow-up duration for HPV-positive women, 20 months.
- Noninferiority of the sensitivity of the HPV test on self-collected samples was observed for CIN2+ (P=.0162) and CIN3+ (P=.0064).
- Only 8.8% of invited women participated.