Serum and red blood cell folate levels in relation to invasive cervical cancer risk in a multicenter case-control study of United States women
The possible association between low folate status and cervical cancer risk was explored using data from a case-control study of women, aged 20-74, from five communities in the United States. Of eligible cases and controls (selected by random digit dialing and matched on age, race, and telephone exchange), 73% and 72%, respectively, were interviewed. Of these, blood samples were obtained from 51% and 68%, respectively, at least six months after completion of any treatment. Serum and red blood cell (RBC) folate levels were each measured using microbiologic and radiobinding assays, and the two assays for each measure were highly correlated (serum: r = 0.90; RBC: r = 0.76). Serum and RBC measures using the same assay were also highly correlated (microbiologic: r = 0.71; radiobinding: r = 0.62). Differences among adjusted mean folate levels in women at various disease stages who underwent surgery, radiation, and/or chemotherapy were examined using analysis of covariance. Except for chemotherapy, stage and/or treatment did not appear to reduce blood folate levels. To determine cervical cancer risk by folate status, logistic regression analyses were conducted with 203 cases of invasive, squamous cell cervical cancer, excluding cases who received chemotherapy (n = 8), and 540 controls. The four blood folate measures were analyzed separately, and, based on factor analyses, as a combined variable. Folate levels in the lowest one or two quartiles were associated with increased cervical cancer risk. The odds ratio for risk of cervical cancer in the lowest quartile was 1.6 (95% c.i. = 0.9-2.7) using the combination variable, adjusting for age, race, study site, number of sexual partners, age at first intercourse, years since last Pap smear, oral contraceptive use, smoking, number of pregnancies, education, and income). The risk associated with low folate status was not intensified among oral contraceptive users or among those with high risk of human papillomavirus (HPV) exposure, based on proxy measures. However, low folate was associated with especially strong risks (OR = 2.5-6.0) among subjects at low risk for HPV exposure. The association between low blood folate levels and increased cervical cancer risk is biologically plausible because of the role of folate in DNA repair and methylation
Thesis, Dissertation, English, 1998