Ocena skuteczności 24 strategii wtórnej triażu w wykrywaniu zmian szyjki macicy u kobiet z dodatnim samodzielnie pobranym testem na wysokoonkogenne HPV

Objective: To evaluate the screening efficiency of gene DNA methylation testing, p16 staining, human papillomavirus (HPV) genotyping, and liquid-based cytology (LBC), and their combined strategies in self-collected high-risk HPV (HR-HPV) positive individuals. Methods: This study analyzed a subset of data from a free self-sampling cervical cancer screening program conducted in Baise, Guangxi Zhuang Autonomous Region (from May 2023 to April 2024). HR-HPV positive self-collected cases with complete data and quality control were included. Using the triage strategy recommended by the 2019 American Society for Colposcopy and Cervical Pathology (ASCCP) risk-based management guideline and the Chinese cervical cancer screening guideline [ie, HPV16/18 positivity and (or) LBC of atypical squamous cell of undetermined significance (ASCUS) or worse] as reference standard, the sensitivity, specificity, colposcopy referral rate, and the number needed to colposcopy (NNC) of various secondary triage approaches were assessed. Results: (1) A total of 777 HR-HPV positive women with complete data were analyzed. (2) Methylation positivity was significantly higher in high-risk group HPV16/18 (20.5%, 25/122) and high-intermediate-risk group HPV31/33/35/45/52/58 (17.2%, 66/381) than that in the low-risk group HPV39/51/56/59/66/68 (9.6%, 26/371; χ²=8.85 and χ²=7.59, P=0.003 and P=0.006). (3) Methylation positivity of cervical intraepithelial neoplasia (CIN)Ⅲ and cervical cancer cases (67.5%, 27/40) was significantly higher than CINⅠ cases (12.9%, 31/240), and CINⅡ cases (16.4%, 12/73), with statistically significant differences (P<0.05). The p16 positivity increased with lesion severity in all pairwise comparisons (P≤0.005 after correction). (4) When the guideline-recommended triage strategy [HPV16/18 positivity and (or) LBC≥ASCUS] was applied to this study population, using colposcopy-directed biopsy pathology as the reference standard, the sensitivity and specificity for detecting CINⅡ+ and CINⅢ+ were 69.03%, 65.21%, and 95.00%, 63.23%, respectively; the colposcopy referral rate was 39.77% (309/777), and NNC was 3.96 for CINⅡ+ and 8.13 for CINⅢ+. (5) Methylation performance: for CINⅡ+, sensitivity was 34.51% and specificity was 88.25%; for CINⅢ+, sensitivity was 67.50% and specificity was 87.79%. (6) p16 performance: the sensitivity for detecting CINⅡ+ and CINⅢ+ were 72.57% and 92.50%, respectively, which did not differ significantly from the guideline-recommended strategy (all P>0.05), whereas the specificity for CINⅡ+ and CINⅢ+ were 68.67% and 65.60%, respectively, similar to the guideline-recommended strategy (all P>0.05). The colposcopy referral rate was lower with p16 testing alone compared with the guideline-recommended strategy [37.32% (290/777) vs 39.77%; P>0.05]. (7) Combined strategy 3a achieved a sensitivity of 63.72% for detecting CINⅡ+, which was similar to the guideline-recommended strategy (P=0.109), and a sensitivity of 95.00% for detecting CINⅢ+, identical to that of the guideline-recommended strategy. The specificity of strategy 3a for CINⅡ+and CINⅢ+ were 70.18% and 68.52%, respectively, both significantly higher than those of the guideline-recommended strategy (χ²=9.75 and 12.56, all P<0.01). The colposcopy referral rate for strategy 3a was significantly lower than that of the guideline-recommended strategy [34.75% (270/777) vs 39.77%; P<0.001], with corresponding NNC of 3.75 for CINⅡ+ and 7.11 for CINⅢ+. Strategy 5a showed a sensitivity of 64.60% for detecting CINⅡ+, which was not significantly different from that of the guideline-recommended strategy (64.60% vs 69.03%; χ²=3.20, P=0.074), and a sensitivity of 95.00% for CINⅢ+, identical to that of the guideline-recommended strategy. The specificity of strategy 5a for CINⅡ+ and CINⅢ+ were 75.30% and 73.00%, respectively, both significantly higher than the guideline-recommended strategy (both P<0.001). NNC for strategy 5a was 3.25 for CINⅡ+ and 6.24 for CINⅢ+, both lower than those of the guideline-recommended strategy (3.96 and 8.13, respectively). The colposcopy referral rate for strategy 5a was also significantly lower than that of the guideline-recommended strategy [30.50% (237/777) vs 39.77%; χ²=70.01, P<0.001]. Conclusions: Among self-collected, high-risk HPV-positive individuals, combined triage strategies that integrate HPV genotyping with cytology, p16 staining and methylation testing, specifically strategy 3a and 5a, offer comparable sensitivity with superior specificity and lower referral rates versus co-testing for detecting CINⅡ+ and CINⅢ+.