Financial and operational constraints limit low-resource countries in the screening of high-risk genital human papillomaviruses (HR-HPV), the etiological agents of cervical cancer. With its simple storage, conservation and shipping, dried cervical sample (DCS) could represent an efficient tool. The aim of the study was to evaluate the reliability of HPV genotyping from DCS. Cervical samples were obtained from 50 women infected with HIV-1 in Côte d'Ivoire. After DNA extraction from both DCS and matched liquid cervical samples (LCS), HPV genotyping was performed and the concordance of genotyping results was evaluated. HPV prevalence was 88% in LCS and 78% in DCS. Kappa statistic was 0.51 for the presence of any genotype (95% confidence interval, 0.25-0.77) and 0.73 for HR-HPV (0.45-0.99). Out of 50 samples, 45 were HPV-positive for DCS and/or LCS, and HR-HPV were detected in 37 samples (74%) with 36 HR-HPV multiple infections. Any genotype and HR genotype identification was concordant/compatible in 86% (43/50) and 88% (44/50) of samples, respectively. In most instances, kappa statistics for detection of type-specific HPV was over 0.6 (including HPV-16, -18, -31, -33). An excellent agreement (kappa statistic ≥ 0.81) was found for eight genotypes (HPV-6, -31, -35, -40, -56, -58, -66, and -82). In spite of interfering factors (multiple infections, different HPV loads, amplification competition, different inputs), DCS and LCS led to concordant/compatible results in most cases. DCS could represent an efficient tool for epidemiological field studies in resource-limited settings, and more importantly for improving the screening coverage and care management in women infected with HPV.
Keywords: HPV genotyping; cervical cancer screening; cervical sample; dried spot; human papillomavirus.
Copyright © 2013 Wiley Periodicals, Inc.