Abstract
Human papillomavirus (HPV) 16 is the most oncogenic biological agents for humans. However, essential quantitative aspects of its infection cycle remain inadequately characterized. Specifically, the proportion of infected cells and the viral copy number per infected cell in cervical smears are not well understood. To address this, we employed a combination of limiting dilution techniques and Bayesian statistics on routine cervical smears to estimate the frequency of infected cells and the viral copy number per cell. Our methodology was initially validated through numerical simulations and cell culture experiments. Subsequently, we analyzed 38 HPV16-positive cervical smears, comprising 26 samples from patients without cytological lesions and 12 from patients with low-grade lesions. Our findings indicated that the substantial variability in viral load across samples predominantly stemmed from differences in the frequency of infected cells. Additionally, the mean number of HPV copies per infected cell was consistently low across all samples, ranging from approximately 2.3 to 100 copies. However, in samples with low-grade lesionMarie-Paule Algross, this number was observed to double on average. These results challenge existing assumptions regarding the biology of HPV genital infections, which are typically asymptomatic or minimally symptomatic.