Abstract
In HPV-primary screening, sample quality significantly influences test accuracy. Unlike cytology-based screening, no consensus guidelines presently exist for sample quality assessment in HPV testing. This study aims to evaluate the impact of sample cellularity on HPV testing. A total of 37 592 liquid-based cytology (LBC) samples from women undergoing HPV-primary screening (aged 30-64, median 48; IQR: 40-56 years) were analyzed using Cobas 4800 HPV Test (Roche). Sample adequacy was assessed by the assay's β-globin internal control and by an independent quantitative cellularity assessment (OncoPredict HPV, Hiantis). HPV positivity rates (PR) were stratified according to β-globin Ct values. Among the analyzed samples, 50.0%, 47.1%, 2.3%, and 0.6% had β-globin Ct values of ≤ 28, > 28 to ≤ 32, > 32 to ≤ 34, and > 34, respectively. Overall HPV-PR was 7.7% (2891/37 592). PR reached 9.7% in samples with β-globin ≤ 28 Ct (1820/18 801), decreasing markedly to 1.4% for β-globin > 34 Ct (3/214), (p < 0.001). Quantitative analysis showed that Cobas 4800 β-globin Ct = 34 corresponds to approximately 1.5 × 10^3 nucleated cells/reaction. A subset of 195 HPV-negative samples with β-globin Ct ≥ 34 was evaluated by liquid based cytology (LBC): 19% had inadequate cellularity according to LBC guidelines, 8% were ≥ ASC-US and 73% NILMs. 65% of adequate LBC showed cellular atrophy. These findings emphasize the importance of assessing cellularity in HPV-screening to avoid potentially false-negative results due to inadequate samples. Future research should focus on establishing standardized cellularity thresholds to improve screening accuracy.