Limited data from hematological studies suggest that certain nutrients, including carotenoids, tocopherols, and vitamin C, may protect against malignant change in cervical tissue. Recognizing that human papillomavirus (HPV) infection induces most neoplastic transformation of cervical tissue, the authors conducted a case-control study to examine the association of plasma micronutrient concentrations with the risk of cervical dysplasia after careful adjustment for HPV infection, using a sensitive and reliable HPV detection method. The sample included 147 multiethnic women, between 18 and 65 years of age, with biopsy-confirmed squamous intraepithelial lesions (SILs) of the cervix and 191 clinic controls identified between 1992 and 1996. Cases were identified through cytology and pathology logs in three clinics on Oahu, Hawaii. Controls were selected randomly from admission logs of the participating clinics. In-person interviews were conducted in the subjects' homes, and a fasting blood sample was drawn to measure plasma levels of lutein, lycopene, cryptoxanthin, total carotene, retinol, tocopherol, ascorbic acid, and cholesterol. The presence and type of HPV was determined in exfoliated cell samples using PCR dot blot hybridization. Mean plasma lycopene, total cryptoxanthin, and alpha-cryptoxanthin levels were lower among cases than controls. We found an inverse dose-response of alpha-cryptoxanthin, total tocopherol, and alpha-tocopherol to the odds ratios for cervical SIL after adjustment for HPV and other confounders. The odds ratio among women in the highest compared with the lowest quartile was 0.3 (95% confidence interval, 0.1-0.7) for alpha-cryptoxanthin and 0.3 (95% confidence interval, 0.1-0.8) for alpha-tocopherol. Negative trends in the odds ratios were suggested for other carotenoids and vitamin C, but these were weak, and confidence intervals were wide. Our results support existing evidence that high plasma levels of antioxidants may reduce the risk of cervical SILs independent of HPV infection. These findings are significant because diet is potentially modifiable, and nutrition education and dietary intervention might be targeted at specific high-risk groups.