Improvement in cervical dysplasia associated with folic acid therapy in users of oral contraceptives

The authors present the results of a double-blind study begun in 1976 to determine if a daily supplement of folic acid is capable of altering the course of early cervical dysplasia. Forty-seven young women with mild or moderate dysplasia of the uterine cervix (cervical intraepithelial neoplasia) diagnosed by cervical smears received oral supplements of folic acid (10 mg) or a placebo (ascorbic acid, 10 mg) daily for 3 months under double-blind conditions. All had used a combination-type oral contraceptive agent (OCA) for at least 6 months and continued it while returning monthly for followup examinations. All smears and a biopsy obtained at the end of the trial period were classified by a single observer without knowledge of treatment status using an arbitrary scoring systern (1, normal; 2, mild; 3, moderate; 4, severe; 5, carcinoma in situ). Subsequently, one of the authors also reviewed all available pretreatment and post-treatment smears blindly for features of megaloblastosis, including those described by Whitehead et al. Slides were arbitrarily classified and scored as: 1, nonmegaloblastic; 2, mild; 3, moderate; 4, marked. Seventy-eight subjects were enrolled in the study. Of this number, 47 completed the protocol and were suitable for final evaluation; nine were white and 38 non-white. The mean age was 20.6 years, range 16 to 28. Precise socioeconomic data are not available, but the population served by this clinic is generally regarded as low-income. Thirty-one subjects failed to return as scheduled for followup visits and/or final biopsy. Of the 47 patients who completed the protocol, blood vitamin analyses were available in 20 at the first visit and in 25 at the last visit. Blood vitamin analyses were also available in 14 additional OCA users who had dysplasia at their first visit but were subsequently found to be ineligible for evaluation. All subjects had at least one referral smear obtained before the first clinic visit. Fifteen subjects had smears obtained on three separate occasions before beginning therapy with folate or placebo; 30 participants had smears from two separate examinations at the start of the study. All pretreatment scores were pooled for statistical analysis and indicate that the two treatment groups were similar at the outset. The final cytology score was significantly better than the initial score in patients receiving folate (paired t test; P < 0.05). In contrast, the mean cytology scores remained essentially unchanged in the patients receiving the placebo. Treatment with folate was associated with a consistent downward trend (i.e., improvement) in cytology score (Fig. 1), although the final score was not significantly different from that of the placebo (unpaired t test). Evaluation of biopsy scores revealed a greater degree of severity in placebo subjects than in folate-treated subjects. These values are significantly different from each other (P < 0.05 using t test for unpaired data). Analysis of crude data based on routine pathology reports, performed before the single-observer review, also revealed highly significant differences (P < 0.01) between biopsies from treated and untreated subjects. Among 25 subjects receiving the placebo, there were five cases in which the biopsy was two stages more severe than the most severe cytology score at the outset. Four of these were read as carcinoma in situ. In the group of 22 subjects receiving folate, there was only one example of a two-stage difference, namely, a biopsy read as severe after four previous cytology readings of mild dysplasia. These changes have been looked upon as “progressions.” There were no cases of carcinoma in situ among subjects receiving folate. Two subjects in the folate group with moderate dysplasia at the outset demonstrated disappearance of lesions by colposcopic examination, so that the attending gynecologist did not consider a biopsy indicated. Each also had normal cytological findings on the cervical smear at this final visit. For tabulation and scoring purposes, their biopsy has been assumed to be negative. The final megaloblastosis score is significantly better (P < 0.05) in patients who received folate supplementation than in those who received placebo. While there was considerable improvement in the megaloblastosis score after folate supplementation, it did not reach statistical significance in comparison with baseline values in the same set of patients. The initial vitamin assays showed red cell folate values were lower among users than nonusers of oral contraceptives (P < 0.01) and even lower (P < 0.001) among the OCA-users classified as having cervical dysplasia. Although the mean red cell folate concentration was lower in OCA users with dysplasia than in users without dysplasia, the difference did not reach statistical significance. Plasma folate levels were also lower among users than nonusers, and were lower in subjects with dysplasia, but the differences were not as great (P < 0.05 and 0.01, respectively) as with red cell folate values. The plasma B₁₂ levels were significantly lower (P < 0.01) among the OCA-using controls than among the nonuser controls. The plasma B₁₂ value for patients with dysplasia as a group was higher (P < 0.02) than in the control group of 20 users of OCAs. As a group, the OCA users with dysplasia had plasma B₁₂ comparable to the controls who did not use OCAs. Vitamin assay results at the end of the protocol showed vitamin levels in blood remained essentially unchanged in placebo patients. Among 10 subjects assigned to treatment with folate, the mean plasma concentration increased approximately S-fc!d, and the red cell concentration increased approximately 4-fold.