HPV infections are believed to be a necessary cause of cervical cancer. Viral burden, as a surrogate indicator for persistence, may help predict risk of subsequent SIL. We used results of HPV test and cytology data repeated every 4 -6 months in 2,081 women participating in a longitudinal study of the natural history of HPV infection and cervical neoplasia in São Paulo, Brazil. Using the MY09/11 PCR protocol, 473 women were positive for HPV DNA during the first 2 visits. We retested all positive specimens by a quantitative, low-stringency PCR method to measure viral burden in cervical cells. Mean viral loads and 95% CIs were calculated using log-transformed data. RRs and 95% CIs of incident SIL were calculated by proportional hazards models, adjusting for age and HPV oncogenicity. The risk of incident lesions increased with viral load at enrollment. The mean number of viral copies/cell at enrollment was 2.6 for women with no incident lesions and increased (trend p ‫؍‬ 0.003) to 15.1 for women developing 3 or more SIL events over 6 years of follow-up. Compared to those with <1 copy per cell in specimens tested during the first 2 visits, RRs for incident SIL increased from 1.9 (95% CI 0.8 -4.2) for those with 1-10 copies/cell to 4.5 (95% CI 1.9 -10.7) for those with >1,000 copies/cell. The equivalent RR of HSIL for >1,000 copies/cell was 2.6 (95% CI 0.5-13.2). Viral burden appears to have an independent effect on SIL incidence. Measurement of viral load, as a surrogate for HPV persistence, may identify women at risk of developing cervical cancer precursors. © 2002 Wiley-Liss, Inc. Key words: human papillomavirus; cervical neoplasia; cohort study; longitudinal analysis; viral loadEstablishment of productive infections by oncogenic types of HPV is a key early event in the natural history of cervical cancer. Given the positive relationship between viral load and the likelihood of persistent HPV infection 1 and the strong relationship between the latter and risk of cervical neoplasia, 2 a single measurement of viral load in cervical specimens may be a suitable biomarker of either a transient infection, or the likelihood that an instance of HPV DNA positivity may represent a persistent infection or one that may become persistent over time and lead to the development of cervical SIL. In the latter case, little is known, however, about the relationship between level of viral burden and lesion incidence or lesion grade severity in the natural history of CIN. HPV DNA viral load has been identified as a biomarker for cervical lesions in women, 3,4 primarily with minor-grade cytologic atypia. 5 However, only a few studies have evaluated the predictive effect of viral load on the incidence of cervical cancer precursors. 3,4,6 -8 A few methods of viral load estimation have been evaluated in epidemiologic studies. Although several cross-sectional studies have examined the association between HPV load measured by HC and the presence of SIL, 9 -13 HC cannot accurately quantify HPV DNA in cervical specimens because of the variabl...

Results confirm the gender-dependence of creatinine concentrations in spot specimens and also show age-dependence, indicating the need for these aspects to be considered when the range of acceptable samples is to be set. No significant intra- or inter-day variations were observed for the two parameters. Lastly, the possibility of a comparison of differently adjusted values was indicated by a conversion formula derived from adjustments to creatinine and the corresponding specific gravity of a hypothetical urinary value, as follows: specific gravity adjusted values = 1.48 x creatinine adjusted values.