Several lines of evidence have already shown how dietary habits might affect the risk of cancer in women [42
]. With respect to cervical cancer, however, persistent hrHPV infection is an essential factor for the progression of precursor lesions to cancer [4
]. For this reason, there is also the need for evaluating other factors, in particular the association between diet and hrHPV infection. In a previous study, we demonstrated the dual effect of diet on the risk of hrHPV infection—on one hand, the risk increased among women with high adherence to a Western dietary pattern, on the other hand, it decreased among those who adhered to the Mediterranean diet [5
]. In the current study on the same population, we intended to deepen our analysis evaluating whether dietary antioxidant intake was associated with hrHPV status among women with normal cervical cytology. As summarized by Garcıa-Closas, in fact, antioxidants might exert a protective effect against cervical cancer progression by modulating immune response, viral replication, and gene expression [25
]. In line with this, a recent review by Chih and colleagues suggested that high levels of plasma antioxidants might enhance the clearance of hrHPV [44
]. By contrast, previous studies reported inconsistent results about the effect of dietary antioxidant intake, probably due to differences in study design, populations, and methods used for dietary assessment. For instance, a case-control study by Giuliano and colleagues found that dietary intakes of b-cryptoxanthin, lutein, zeaxanthin, and vitamin C were negatively associated with the risk of HPV persistence [45
]. Similarly, a cohort study reported a borderline significant association between dietary lutein intake and HPV persistence [46
]. By contrast, in our study, no differences in dietary intake of carotenoids were evident between hrHPV-positive and -negative women. However, we observed a lower dietary intake of zinc, manganese, and vitamin A and C among women with hrHPV infection. Among these, dietary zinc intake maintained a negative association with the risk of hrHPV-positive status when all antioxidants were considered simultaneously. This finding supports the notion that zinc is involved in the immunomodulatory pathways—specifically, zinc intake and its deficiency might be associated with hypoplasia of lymphoid tissues followed by a reduction in the number and activity of T-helper and natural killer cells, antibody production, cell mediated immunity, and phagocytosis [47
]. In line with this, a pilot study demonstrated that a zinc-based compound via intra-vaginal infusion might help the clearance of hrHPV from the uterine cervix among women with no evidence of high grade squamous intraepithelial lesions [52
To our knowledge, our study is also the first that has evaluated the influence of a combination of dietary antioxidants on hrHPV status using the CDAI. This index—developed by Wright and colleagues—allows for the ranking of dietary antioxidant intake in relation to population norms and the evaluation of the overall impact of antioxidants on health outcomes [34
]. Notably, in our study, women with higher dietary antioxidant intake had lower odds of being infected by hrHPV than those who reported a lower intake. The current lack of evidence supporting our findings, however, raises the need for further research on large and prospective cohorts. In particular, it is necessary to understand to what extent dietary intake of antioxidants could prevent HPV infection and promote its clearance. Indeed, HPV survives in a cellular oxidative environment and even leads to neoplastic changes by increasing oxidative stress and DNA damage [16
]. For this reason, dietary antioxidant intake could exert a protective effect against HPV infection and persistence [25
]. However, further investigations on the biological mechanisms underpinning the antioxidant effect against HPV should be encouraged. In this scenario, our study adds to the current knowledge about the association between dietary antioxidant intake and the risk of HPV infection. Our findings, if confirmed by large-scale prospective research, might inform novel preventive strategies against HPV infection on the basis of dietary recommendations, especially in regions where vaccination is not implemented yet or vaccination coverage levels are not sufficient for prevention of infections. It is important to clarify that our study does not intend to give recommendations regarding the use of antioxidant supplements. In fact, there is the need for RCTs to evaluate benefits and drawbacks of antioxidant supplementation in women at risk for cervical cancer.
Our study had some limitations that should be considered when interpreting our findings. First, the cross-sectional design and the absence of information about timing of infection did not allow us to determine the causality of the relationship, nor did it allow us to discriminate between hrHPV infection and persistence. Second, dietary assessment did not preclude measurement error and inaccuracy. Third, no information about dosage regimen of antioxidant supplementation was available. To partially address this issue, we adjusted our analysis for the use of supplements, which could also include antioxidants. We found no significant differences in the use of supplements according to hrHPV status and CDAI tertiles. In line with this, our findings remained significant after adjusting for the use of supplements and other covariates. However, we cannot completely exclude the effect of unmeasured residual factors that potentially affect dietary antioxidant intake and/or hrHPV status.
In conclusion, our study showed that dietary intake of zinc was negatively associated with the risk of hrHPV infection, probably due to its immunomodulatory properties. Moreover, to our knowledge, we demonstrated for the first time that the cumulative intake of antioxidants might be associated with a decreased risk of hrHPV infection. However, further large-scale prospective studies are encouraged to understand whether antioxidants exert their effect on hrHPV infection, rather than on its persistence.