Objective: Evaluation of the clinical value of the BioPerfectus multiplex real time (BMRT)-HPV for cervical cancer screening. Methods: Physician-collected specimens of 1 495 women who were positive of Cobas 4800 HPV (Cobas-HPV), HPV genotyping based on SEQ uencing (SEQ-HPV), and (or) cytology ≥low grade squamous intraepithelial lesion (LSIL) in the primary screening of Chinese Multiple-center Screening Trial (CHIMUST), and 2 990 women selected from those who were negative of primary screening in the same project through nested control randomization with age-matching were tested for BMRT-HPV, which reported type-specific viral loads/10 000 cells in each specimen. With comparing to Cobas-HPV results and taking cervical histopathological diagnosis as the endpoint, the concordance of high-risk (HR)-HPV subtypes among the three assays was explored ,and the sensitivity and specificity of BMRT-HPV for cervical cancer screening were evaluated. Results: (1) The overall agreenment of HR-HPV subtypes between BMRT-HPV and Cobas-HPV, or SEQ-HPV test sample was 94.8%, 94.4%, with Kappa values 0.827, 0.814. (2) The sensitivity and specificity for cervical intraepithelial neoplasia (CIN) Ⅱ+ of BMRT-HPV, Cobas-HPV and SEQ-HPV were 92.62%, 94.26%, 93.44% and 84.67%, 83.25%, 82.76%, respectively. There were no significant difference in sensitivity among the three HPV assays (all P>0.05), but the specificity of BMRT-HPV for CIN Ⅱ+ was higher than those of Cobas-HPV and SEQ-HPV (P<0.01). The sensitivity for CIN Ⅲ+ of three HPV assays were all 100.00%, and the specificity for CIN Ⅲ+ of BMRT-HPV was higher than those of Cobas-HPV and SEQ-HPV (83.40% vs 81.95%, 83.40% vs 81.50%; P<0.01). The number of pathological examinations of colposcopy for cervical biopsy detected in 1 case of CIN Ⅱ+ or CIN Ⅲ+ in BMRT-HPV was less than those in Cobas-HPV and SEQ-HPV (P<0.01). When using HPV 16/18 + cytology ≥atypical squamous cell of undetermined signification (ASCUS) to triage HPV positive women among three assays, there was no different in the sensitivities of detecting CIN Ⅱ+ and CIN Ⅲ+ (P>0.05). The specificity BMRT-HPV was slightly higher than those in Cobas-HPV or SEQ-HPV (all P<0.05), and the colposcopy referral rate was lower than those in Cobas-HPV and SEQ-HPV (all P<0.05). Conclusions: BMRT-HPV is as sensitive as Cobas-HPV or SEQ-HPV for primary cervical cancer screening, and has higher specificity. Therefore it could be used as a primary screening method for cervical cancer, which is worthy of clinical application.
目的： 评价核酸分型定量法HPV检测（BMRT-HPV）用于子宫颈癌筛查的临床价值。 方法： 采用巢式抽样法，在2016年9月—2018年1月中国子宫颈癌筛查多中心研究（CHIMUST）项目5个筛查点的8 856例妇女中，选取其中医生取样或自取样标本中任一检测方法HPV阳性或HPV阴性但细胞学结果≥低级别鳞状上皮内病变（LSIL）者共1 495例行BMRT-HPV检测，同时按照1∶2比例抽取年龄和参加筛查时间相匹配的HPV与细胞学结果均为阴性的2 990例为对照。CHIMUST项目中对任一HPV阳性妇女回叫行阴道镜下子宫颈病灶四象限定点活检+随机活检+子宫颈管搔刮术方案子宫颈活检，以医生取样标本的基于实时PCR技术的Cobas 4800 HPV检测（Cobas-HPV）、基于第2代基因测序技术的HPV分型检测（SEQ-HPV）为对照，以子宫颈活检病理诊断为“金标准”，分析BMRT-HPV检测的高危型HPV（HR-HPV）亚型与Cobas-HPV、SEQ-HPV检测的一致性，并比较3种HPV检测方法对子宫颈上皮内瘤变（CIN）Ⅱ及以上级别病变（CIN Ⅱ+）、CIN Ⅲ及以上级别病变（CIN Ⅲ+）的筛查效率。 结果： （1）BMRT-HPV检测方法分别与Cobas-HPV、SEQ-HPV检测方法比较，检测HR-HPV亚型整体的一致性分别为94.8%、94.4%，Kappa值分别为0.827、0.814。（2）BMRT-HPV、Cobas-HPV、SEQ-HPV 3种方法对CINⅡ+筛查的敏感度分别为92.62%、94.26%、93.44%，两两比较，差异均无统计学意义（P>0.05）；特异度分别为84.67%、83.25%、82.76%，BMRT-HPV检测对CINⅡ+筛查的特异度显著高于Cobas-HPV、SEQ-HPV检测（P<0.01）。3种HPV检测方法对CIN Ⅲ+筛查的敏感度均达到100.00%，BMRT-HPV检测对CIN Ⅲ+筛查的特异度也显著高于Cobas-HPV、SEQ-HPV检测（分别为83.40%、81.95%、81.50%，P<0.01）。检出1例CINⅡ+或CIN Ⅲ+所需要行阴道镜下子宫颈活检病理检查的患者数，BMRT-HPV检测显著少于Cobas-HPV和SEQ-HPV检测（P<0.01）。Cobas-HPV、SEQ-HPV、BMRT-HPV检测初筛阳性者分别以HPV 16和（或）18型（HPV 16/18型）阳性联合细胞学结果≥未明确诊断意义的不典型鳞状上皮细胞（ASCUS；即为方案一、二、三）进行二次分流，3种筛查方案对CIN Ⅱ+、CIN Ⅲ+筛查的敏感度分别比较均无显著差异（P>0.05），但方案三（即BMRT-HPV初筛阳性者）筛查CIN Ⅱ+、CIN Ⅲ+的特异度显著高于方案一（即Cobas-HPV初筛阳性者）和方案二（即SEQ-HPV初筛阳性者；P<0.05），而且阴道镜转诊率方案三也显著低于方案一、二（P<0.05）。 结论： BMRT-HPV检测筛查子宫颈癌具有与Cobas-HPV、SEQ-HPV检测相似的敏感度，但特异度更高，可作为子宫颈癌的初筛方法，值得临床推广使用。.
Keywords: Genotype; Human papillomavirus 16; Human papillomavirus 18; Papillomavirus infections; Uterine cervical neoplasms; Viral load.