Xu Lan, Ma Chang, Cuschieri Kate, Bonde Jesper, Arbyn Marc
Department of Epidemiology and Biostatistics, School of Public Health, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
Scottish HPV Reference Laboratory, Royal Infirmary of Edinburgh, Edinburgh, Scotland, UK.
J Med Virol. 2025 Aug;97(8):e70527. doi: 10.1002/jmv.70527.
The Venus HPV assay (VenusHPV) is a real-time PCR-based human papillomavirus (HPV) test that is widely used in China but lacks extensive clinical validation. The VALidation of HPV GENotyping Tests (VALGENT) framework is an established protocol for evaluating HPV genotyping assays against a standard comparator test. This study aimed to assess the clinical accuracy and reproducibility of the VenusHPV assay following international validation criteria. The clinical performance of VenusHPV was evaluated against the GP5+/6+ PCR-based enzyme immunoassay (GP-EIA) using the VALGENT-4 panel, which included 998 consecutive routine screening samples enriched with 297 samples with abnormal cytology from the Danish cervical cancer screening program. Cases were defined as women diagnosed with histologically confirmed cervical intraepithelial neoplasia 2 or more (CIN2+), while two consecutive negative cytology results served as a proxy for nondisease. Intra- and interlaboratory reproducibility was assessed on 500 samples. Using the manufacturer-recommended cutoff, VenusHPV demonstrated noninferior sensitivity for detection of CIN2+, but its specificity for ≤ CIN1 was inferior to that of GP-EIA. Applying an optimized a posteriori cutoff improved the specificity, yielding relative specificity of 1.02 (95% CI [CI], 1.00-1.03; p noninferiority [p] < 0.0001), while maintaining a noninferior sensitivity (of 1.02; CI, 1.00-1.08; p < 0.0001). The intra- and interlaboratory reproducibility was excellent (95.2%, CI, 93.3-97.1%, Kappa [κ] = 0.87 and 94.0%, CI, 92.0%-96.0%, κ = 0.85, respectively). Notably, the reproducibility criteria were met consistently, regardless of whether the unadjusted or optimized cutoff was applied. The VenusHPV was as sensitive as the GP-EIA for detecting cervical precancer using the unadjusted cutoff but less specific. However, after cutoff optimization, VenusHPV met the international accuracy criteria for cervical cancer screening. Additionally, the assay demonstrated excellent reproducibility.
金星人乳头瘤病毒检测（VenusHPV）是一种基于实时聚合酶链反应的人乳头瘤病毒（HPV）检测方法，在中国广泛使用，但缺乏广泛的临床验证。HPV基因分型检测验证（VALGENT）框架是一种既定方案，用于对照标准比较检测评估HPV基因分型检测方法。本研究旨在根据国际验证标准评估VenusHPV检测的临床准确性和可重复性。使用VALGENT - 4样本组，对照基于GP5 + / 6 +聚合酶链反应的酶免疫测定（GP - EIA）评估VenusHPV的临床性能，该样本组包括998份连续的常规筛查样本，其中有297份来自丹麦宫颈癌筛查项目的细胞学异常样本。病例定义为经组织学确诊为宫颈上皮内瘤变2级或更高级别（CIN2 +）的女性，而连续两次细胞学阴性结果作为无疾病的替代指标。在500份样本上评估了实验室内和实验室间的可重复性。使用制造商推荐的临界值，VenusHPV在检测CIN2 +方面显示出非劣效性敏感性，但其对≤CIN1的特异性低于GP - EIA。应用优化后的后验临界值提高了特异性，相对特异性为1.02（95%置信区间[CI]，1.00 - 1.03；非劣效性p值[p]<0.0001），同时保持非劣效性敏感性（为1.02；CI，1.00 - 1.08；p<0.0001）。实验室内和实验室间的可重复性都非常好（分别为95.2%，CI，93.3 - 97.1%，kappa[κ]=0.87和94.0%，CI，92.0% - 96.0%，κ = 0.85）。值得注意的是，无论应用未调整的还是优化后的临界值，均可始终满足可重复性标准。使用未调整的临界值时，VenusHPV在检测宫颈上皮内瘤变方面与GP - EIA一样敏感，但特异性较低。然而，经过临界值优化后，VenusHPV符合宫颈癌筛查的国际准确性标准。此外，该检测方法显示出出色的可重复性。
