ABSTRACT
High-risk human papillomavirus (hr-HPV) testing for primary cervical precancer screening offers an opportunity to improve screening in low-middle income countries (LMICs). This study aimed to compare the analytic performances of the AmpFire and MA-6000 platforms for hr-HPV DNA testing in three groups of women screened for hr-HPV types in Ghana: group 1 with 33 GeneXpert-archived ThinPrep/liquid-based samples subjected to both tests, group 2 with 50 AmpFire-archived dry brush samples subjected to MA-6000 testing, and group 3 involving 143 cotton swab samples simultaneously subjected to both tests without archiving. The overall agreement rates were 73 %, 92 %, and 84 %, for groups 1-3, respectively, and 84 % (95 % CI, 78.6-88.6) for the entire group. Neither AmpFire nor MA-6000 was more likely to test hr-HPV positive in all three groups and the combined group. Group 1 showed fair agreement without statistical significance (κ = 0.224, 95 % CI, -0.118 to 0.565), while group 3 showed significant moderate agreement (κ = 0.591, 95% CI, 0.442-0.741). Group 2 showed an almost perfect significant level of agreement (κ = 0.802; 95 % CI, 0.616-0.987). Thus, both platforms showed statistically significant moderate to near-perfect agreement for detecting hr-HPV in cervicovaginal samples, with variation according to archiving conditions and duration between sample collection and retesting. For LMICs using these platforms for COVID-19 testing, as the COVID-19 pandemic subsides, the platforms can become available for running other tests such as hr-HPV DNA testing for cervical precancer screening.