Cervical cancer screening: Which HPV test to choose?

Adding a high-risk human papillomavirus (HPV) test to cytology may enhance detection of high-grade cervical intraepithelial neoplasia (CIN) in primary cervical cancer screening, with the Cobas 4800 and Hybrid Capture 2 (HC2) tests showing comparable clinical performance in high-risk HPV detection in Chinese women, according to a large study by the University of Hong Kong.

Notably, the Cobas test showed significantly higher specificity than the HC2 test in identifying CIN2+ and CIN3+ cases, suggesting a potential to reduce unnecessary colposcopy referrals in a screening population. [PLoS One 2022;17:e0272721]

The study included 6,345 women aged 30–60 years in Hong Kong (mean age, 49.33 years) who had completed both the Cobas and HC2 tests in a previous randomized controlled trial comparing HPV and cytology co-testing with cytology alone for detection of high-grade CIN in primary cervical cancer screening between May 2010 and April 2014. A majority (98.64 percent; n=6,259) of the women had normal cytology. [Int J Cancer 2020;147:1152-1162]

Results showed significantly higher cumulative detection of CIN2+ and CIN3+ with HPV and cytology co-testing vs cytology alone (CIN2+: 1.01 percent vs 0.66 percent; odds ratio [OR], 1.53; 95 percent confidence interval [CI], 1.09–2.19; p=0.016) (CIN3+: 0.67 percent vs 0.39 percent; OR, 1.71; 95 percent CI, 1.10–2.69; p=0.018).

“Adding a high-risk HPV test to cytology for primary cervical screening may be more effective in reducing CIN2+ risk than cytology alone,” the researchers suggested.

Clinical performance of the real-time polymerase chain reaction (PCR)–based Cobas 4800 high-risk HPV assay was validated and compared with the standard HC2 test, a signal amplification assay that utilizes microplate chemiluminescence to detect high-risk HPV. Results showed 92.23 percent overall agreement between the two tests, with a Cohen’s kappa coefficient of 0.493 (95 percent CI, 0.454–0.534; p=0.031).

“Inter-assay agreement was correlated with the severity of underlying biology, with increasing concordance found in samples with more severe abnormalities,” the researchers reported. “Most of the discordant samples had test signal strength closer to the test limits of detection than concordant samples, reflecting a low viral load and infection of a cluster of low-risk HPV types in these samples.”

There was no significant difference between the two tests in detection of high-risk HPV in CIN2+ and CIN3+ samples (p=0.250 and 0.5, respectively). “Discrepancy between the two tests was found in women with normal histology or low-grade lesions,” the researchers noted.

However, the Cobas test demonstrated significantly higher specificity vs the HC2 test in identifying CIN2+ and CIN3+ cases (66.46 percent vs 43.67 percent and 65.42 percent vs 42.86 percent, respectively; p<0.001), with comparable sensitivity (82.14 percent vs 92.86 percent [p=0.083] and 78.95 percent vs 89.47 percent [p=0.157], respectively) in clinical evaluation.

“Increased specificity of the Cobas test would accent women with the highest risk of developing CIN2+, with the potential to reduce unnecessary colposcopy referrals in a screening population,” the researchers suggested.

“Due to the study’s limitations, the added benefit of HPV16 and HPV18 genotyping with the Cobas test has not been acknowledged and needs to be further evaluated in a screening cohort with a larger number of high-grade lesions,” they added.