Persistent infection with high risk HPV is associated with increased risk of cervical cancer. Therefore understanding the predictors of persistence may provide some insights in characterizing infections that may have clinical significance.
From August 2012 to December 2013, we recruited women at our cervical cancer screening clinics in Abuja. Nurses collected ecto-cervical samples for HPV determination which was performed using Roche Linear Array (for 278 baseline samples) and SPF10 DEIA, LiPA25 version 1 for all other samples. Relative risks were estimated using Poisson regression models with robust error variance.
Of the 1020 women enrolled, (aged 18–61 years), 727 (71.1%) returned for follow up after mean (SD) 8.6 (4.0) months. Some 42.4% (432/1020) of the participants were HIV positive. Baseline prevalence of any HPV infection was 41.2% (401/973) and of these, 256 women returned for follow-up. Some 62.1% (159/256) remained persistently positive for any HPV. The RR (95% CI, P-value) for an association with prevalent any HPV were 0.99 (0.98 to 0.99, 0.02) for age, 1.23 (1.12 to 1.35, <0.001) for HIV infection, 1.26 (0.97 to 1.63, 0.08) for presence of other STIs, and 1.59 (1.28 to 1.99, <0.001) for abnormal VIA results. The RR (95% CI, P-value) for persistent infection with any HPV were 1.67 (1.39 to 2.01, <0.001) for HIV infection and 2.26 (1.64 to 3.11, <0.001) for abnormal baseline VIA.
This preliminary data suggest a high level of persistence of any HPV infection among women with prevalent any HPV infection. Significant predictors of persistence included HIV infection and an abnormal VIA result at baseline. Updated analysis, by HPV genotype, will be available in September.