Objective: To compare the efficacy and toxicity of topical vaginal 5-fluorouracil (5-FU) maintenance therapy against the effects of observation after standard treatment for high-grade cervical dysplasia in human immunodeficiency virus (HIV)-infected women and to evaluate the association between baseline CD4 count and time to recurrence.
Methods: In a phase III unmasked, randomized, multicenter, outpatient clinical trial, 101 HIV-positive women either received 6 months of biweekly treatment with vaginal 5-FU cream (2 g) or underwent 6 months of observation after standard excisional or ablative cervical treatment for cervical intraepithelial neoplasia (CIN). Papanicolaou smears and colposcopy were scheduled at regular intervals during the ensuing 18 months, with the primary end point being the time at which CIN of any grade recurred.
Results: Thirty-eight percent of women developed recurrence: 14 (28%) of 50 in the 5-FU therapy group and 24 (47%) of 51 in the observation group. Treatment with 5-FU was significantly associated with prolonged time to CIN development (P = .04). Observation subjects were more likely to have high-grade recurrences, with 31% developing CIN 2-3 compared with 8% in the 5-FU treatment arm (P = .014), and disease recurred more quickly in observation subjects as well. Baseline CD4 count was related significantly to time to recurrence (P = .04), with 46% of subjects with CD4 counts less than 200 cells/mm3 developing recurrence compared with 33% of subjects with CD4 counts at least 200 cells/mm3. Disease recurred more slowly in subjects who had received antiretroviral therapy than in antiretroviral therapy-naive subjects. There were no instances of grade 3 or 4 toxicity, and compliance with 5-FU treatment was generally good.
Conclusion: Adjunctive maintenance intravaginal 5-FU therapy after standard surgery for high-grade lesions safely and effectively reduced recurrence of cervical intraepithelial neoplasia in HIV-infected women.