Randomized double-blind trial of beta-carotene and vitamin C in women with minor cervical abnormalities

D. Mackerras*, L. Irwig, J. M. Simpson, E. Weisberg, M. Cardona, F. Webster, L. Walton, D. Ghersi

*Corresponding author for this work

Research output: Contribution to journalArticleResearchpeer-review

48 Citations (Scopus)

Abstract

A double-blind, placebo-controlled, randomized, factorial study using a daily oral administration of 30 mg beta-carotene and/or 500 mg vitamin C was conducted in 141 women with colposcopically and histologically confirmed minor squamous atypia or cervical intraepithelial neoplasia (CIN) 1. Over approximately 2 years of follow-up, 43 lesions regressed to normal and 13 progressed to CIN II. The regression rate was slightly higher, but not significantly so, in those randomized to beta-carotene compared to no beta-carotene (hazard ratio = 1.58, 95% CI, 0.86-2.93, P = 0.14) and slightly lower, but not statistically significant, for those randomized to vitamin C compared to no vitamin C (hazard ratio = 0.65, 95% CI: 0.35-1.21, P = 0.17). In a model with no interaction, the progression rate was slightly higher in those randomized to beta-carotene (hazard ratio = 1.75, 95% CI: 0.57-5.36, P = 0.32) and also in those randomized to vitamin C (hazard ratio = 2.40, 95% CI: 0.74-7.80, P = 0.13). Neither of these were statistically significant. However, there was some evidence of an interaction effect of the two compounds on the progression rate (P = 0.052), with seven of the progressed lesions occurring in those randomized to both vitamins compared to a total of six in the three other groups. The currently available evidence from this and other trials suggests that high doses of these compounds are unlikely to increase the regression or decrease the progression of minor atypia and CIN I.

Original languageEnglish
Pages (from-to)1448-1453
Number of pages6
JournalBritish Journal of Cancer
Volume79
Issue number9-10
DOIs
Publication statusPublished - 1 Jan 1999
Externally publishedYes

    Fingerprint

Cite this