In Japan, a bivalent human papillomavirus (HPV) vaccine against carcinogenic HPV16/18 was licensed in 2009, and a quadrivalent vaccines against HPV16/18 and non-carcinogenic HPV6/11 was licensed in 2011. Recently, the next-generation 9-valent vaccine targeting HPV6/11/16/18/31/33/45/52/58 has been approved. Accurate HPV genotyping is essential for HPV vaccine research and surveillance. The Roche Linear Array (LA) has long been a standard assay for HPV genotyping, but its recent product discontinuation notice has urged us to introduce an alternative assay with comparable performance. In the present study, an in-house HPV genotyping assay that employs PCR with PGMY09/11 primers and reverse blotting hybridization (PGMY-CHUV) was compared with LA to assess genotype-specific agreement. A total of 100 cervical precancer specimens were subjected to both PGMY-CHUV and LA. For detection of genotypes included in the 9-valent vaccine, PGMY-CHUV completely agreed with LA for detection of HPV6, HPV11, HPV16, HPV18, HPV33 and HPV45, and showed near-complete agreement for HPV31 and HPV58 (98% and 99%, respectively). Moreover, PGMY-CHUV detected a significantly higher prevalence of HPV52 than LA (22% vs. 14%, P = 0.008 by McNemar's exact test), with 92.0% overall agreement, 63.6% positive agreement and a kappa value of 0.73. Most (87.5%) of HPV52 discordant cases involved mixed infections with HPV35 or HPV58. In conclusion, while the two assays present equivalent data for assessing the effectiveness of the bivalent and quadrivalent vaccines, PGMY-CHUV is more suitable for evaluating the impact of the current 9-valent vaccine because of its superior detection of HPV52 in co-infection cases.
Keywords: cervical cancer; cervical intraepithelial neoplasia; genotyping; human papillomavirus; vaccine.