Laura Mendoza

Impact of knowledge of human papillomavirus positivity on cervical cytology performance in Latin America

Abstract Background Cervical cytology is recommended by the World Health Organization as a triage option in human papillomavirus (HPV)-based cervical cancer screening programs. We assessed the performance of cytology to detect CIN3+ without and with knowledge of HPV positivity. Methods Women were screened with cytology and HPV across ESTAMPA study centers in Latin America. Screen-positives were referred to colposcopy with biopsy and treatment as needed. Cytology was initially interpreted without knowing HPV results. A subset of cytologies from HPV-positive women were reinterpreted at the same laboratories, with knowledge of HPV status, blinded to previous cytology and histological diagnosis. Performance indicators for cytology to detect CIN3+ without and with knowledge of HPV positivity were estimated. Findings A total of 4087 women were included, of which 490 had histologically confirmed CIN3+ (455 CIN3 and 35 cancers). Cytology sensitivity without knowledge of HPV positivity for CIN3+ was 47.2% (95% CI = 42.5 to 51.9), whereas with knowledge of HPV positivity, the sensitivity was higher (58.9%, 95% CI = 54.2 to 63.5; P < .0001). The specificity without knowledge of HPV was 89.4% (95% CI = 88.2 to 90.5), whereas with knowledge of HPV positivity was 78.9% (95% CI = 77.4 to 80.4; P < .0001). Performance estimates varied by study center for cytology without knowing the HPV positivity (range = 32.8%-61.5% for sensitivity; range = 80.7%-98.6% for specificity). Similarly, performance varied with knowledge of HPV positivity (36.1%-93.4% for sensitivity; 39.6%-98.6% for specificity). Conclusion The increase in sensitivity of cytology with HPV knowledge was limited and highly variable, reinforcing the need for alternative triage methods to support cervical cancer elimination goals.

Genital infections in high-risk human papillomavirus positive Paraguayan women aged 30–64 with and without cervical lesions

Objective To determine the prevalence of genital infections (GIs), including sexual transmitted STIs: Neisseria gonorrhoeae, Chlamydia trachomatis, Mycoplasma genitalium, Trichomonas vaginalis, and opportunistic pathogens that generally do not cause STIs, non-classic STI: Ureaplasma urealyticum, Ureaplasma parvum and Mycoplasma hominis, in women with high-risk oncogenic human papillomavirus (hr-HPV) infection and their association with cervical lesions. Methods A cross-sectional study was carried out including 231 hr-HPV positive women. Of these, 46 has histologically confirmed cervical intraepithelial neoplasia 3 (CIN3) or more (including CIN3 and cervical cancer lesions-CIN3+). GIs were detected by multiplex real time PCR. Odds ratios (OR) were estimated to explore possible associations between GIs and the presence or absence of CIN3+ lesions. Additionally, we examined associations between sociodemographic, sexual, and clinical characteristics and the presence of GIs. Results In total, there were 174/231 cases of GIs corresponding to an overall prevalence of 75.3% (95%CI: 69.4–80.4), being non-classic STIs the most common (72.3%) compared to STIs (12.6%). The most prevalent non-classic STI and STI were U. parvum (49.8%) and C. trachomatis (7.4%), respectively. The odds of presenting GIs were 3 times higher in women under 46 years compared to older counterparts (OR: 3.32, 95%CI: 1.74–6.16), and in women with a normal Pap smear with inflammation compared to those without inflammation (OR: 3.31, 95%CI: 1.15–9.77). GIs were equally present in women with and without CIN3+ lesions. Conclusion We observed an association of GIs with inflammation in the Pap smear, but no association with CIN3+, as some of them are very common and likely part of the normal vaginal flora, suggesting that such infections do not appear to be cofactors in cervical carcinogenesis, although larger prospective studies are needed.