High-sensitivity C-reactive Protein and Regression of Low-grade Squamous Intraepithelial Lesion: The Role of Low-grade Inflammation in Cervical Carcinogenesis

Abstract

Background: Inflammation is emerging as a potential mechanism of cervical carcinogenesis. However, few studies have investigated the association between host inflammatory status and the natural course of cervical precursor lesion. The aim of this study was to assess the probability of LSIL regression, associated with an inflammatory biomarker, high-sensitivity C-reactive protein (hs-CRP).

Methods: In a longitudinal cohort study, female participants were examined annually or biannually using cervical cytology between 2006 and 2015. Incident LSIL cases were included in the analysis, with regression defined as at least one consecutive normal cytologic result. A total of 520 women aged 22-64 years were followed up for LSIL regression. The multivariable-adjusted hazard ratios (HRs) for LSIL regression were estimated using a parametric proportional hazards model.

Results: During 827.5 person-years of follow-up, 486 out of 520 subjects (93.5%) showed LSIL regression. After adjusting several important potential confounders, a higher quartile of hs-CRP levels was significantly associated with a lower rate of regression (for quartile 4 vs quartile 1, inverse HR 1.33; 95% CI, 1.04-1.69; P for trend = 0.028).

Conclusions: The low rate of spontaneous regression recorded in women with higher hs-CRP lends support to the role of the perturbated host inflammatory status in cervical carcinogenesis, and suggests that hs-CRP level could help monitor LSIL.

Keywords: cervix; chronic inflammation; hs-CRP; low-grade squamous intraepithelial lesions; regression.

Figure 1.. Multivariable-adjusted hazard ratios for low-grade squamous intraepithelial lesion (LSIL) regression. Curves represent adjusted hazard ratios for LSIL regression based on restricted cubic splines with knots at the 5th, 27.5th, 50th, 72.5th, and 95th percentiles of hs-CRP distribution. Models were adjusted for age, the year of a screening exam, smoking status, alcohol intake, married, HPV infection, history of diabetes, and education level. The red line represents adjusted hazard ratio, and the dashed lines represent the confidence intervals for the spline model. The horizontal black line corresponds to the normal reference hazard ratio of 1.0.