PATIENTS AND METHODS

Sixty women with RCA entered this study at “A Sygros” Hospital, Athens Greece, from 1999 to 2002. The protocol was approved by the ethics committee of our hospital. Informed consent was obtained in each case.

The evaluation at entry included history taking, clinical examination, colposcopy, Papanicolaou smear, blood counts and biochemistry profile, and lesional biopsy. Eligibility criteria included a cytohistologically confirmed diagnosis of genital warts with at least 3 months’ duration, lesions refractory to conventional therapy, age older than 18 years, negative human immunodeficiency virus (HIV) serology, no evidence of cervical dysplasia or malignant degeneration, no contraindications for isotretinoin administration—that is, no pregnancy in course and normal medullar, renal, hepatic, and lipidaemic laboratory parameters; no history of local or systemic anti-wart therapy, or cytotoxic, or immunomodulatory treatment within 4 weeks from entry.

The patients were randomised into two treatment groups of 30 subjects each, using simple randomisation. They were assigned to receive either oral isotretinoin (group 1), or placebo (group 2) given in an equivalent number of identical in appearance capsules, on a double blind basis. Oral isotretinoin (Roaccutane, Roche) dosage was 0.5 mg/kg daily for 12 weeks. The capsules were dispensed to the patients by a third party. No other topical or systemic medication was administered during the study period. Patients were advised against pregnancy and were instructed to practise an adequate form of contraception, as well as to use condoms in order to avoid re-infection.

Response was evaluated by recording the number of lesions and by bidimensional measurements of the lesions, mapped on a schematic diagram of the cervix. This study was performed at entry, every 4 weeks thereafter, and at the end of therapy. Clinical evaluation was blinded to the treatment. Complete response was defined as the total clearance of cervical lesions; partial response was considered as a reduction of lesions equal to or greater than 50%; no response was considered to be less than 50% reduction in size. The reappearance of lesions after complete response was considered as relapse. Complete responders were followed up monthly for 12 months. Partial responders and non-responders were given other treatments.

Statistical analysis involved the χ² test (Yates’s correction included). The level of significance was fixed at α = 5%.

Toxicity was monitored, on a monthly basis, both clinically and through laboratory tests, which included complete blood counts, renal and hepatic parameters, and lipidic profile. In case of significant toxicity treatment was discontinued.

RESULTS

Sixty women aged 21–43 years (mean 32.3 (SD 4.9) years) were enrolled. The duration of lesions varied from 7–21 months. Involvement of the external genitalia was found in 40% of the patients. Both groups were well matched for age, duration of infection, and extent of involvement.

Fifty three patients were evaluable for response and toxicity. Seven patients were withdrawn from the study because of poor compliance or side effects or were lost to follow up.

The primary end point for evaluation of efficacy was at 12 weeks. The results are summarised in table 1. The therapeutic difference (percentage of responders) between patients receiving active and placebo therapy was statistically significant (χ²  = 19.35, p<0.001).

At the secondary endpoint for evaluation of efficacy (completion of 12 month follow up), out of nine complete responders eight remained in remission, whereas one (11.1%) experienced a recurrence at week 43.

Isotretinoin was generally well tolerated. Side effects were of mild or moderate severity and resolved upon completion of therapy (table 2). One patient dropped out on week 10 because of persistently elevated serum transaminases, which returned to normal a few weeks after the treatment was stopped.

DISCUSSION

Our results seem to support the feasibility and effectiveness of systemic retinoids in the treatment of RCA. Considering the potential of genital warts to resolve spontaneously in up to 20% of the cases, it cannot be ruled out that some of the responses were actually spontaneous remissions. However, this possibility seems less likely since the lesions of our patients had been resistant to previous treatments and the effectiveness of isotretinoin was documented in comparison with placebo.

Previous experience with isotretinoin in genital warts is conflicting. Olsen et al compared systemic interferon alfa-nl and isotretinoin (1 mg/kg for 6 weeks).⁶ None of the eight patients treated with isotretinoin responded. Tsambaos et al treated 56 male patients with RCA using isotretinoin 1 mg/kg for 3 months.⁷ Of these, 39.6% responded completely and 13.2% responded partially. Two recurrences (9.5%) occurred during the 1 year follow up. Cardamakis et al compared the combination of interferon α-2a and isotretinoin with isotretinoin alone (1 mg/kg for 3 months) in 57 women and in 87 men.⁸ There was no statistical difference in remission rates, but the duration of treatment was significantly shorter and the recurrence rate was lower for the combination therapy.

The mechanism through which retinoids exert their beneficial effect in HPV infection remains unknown. It has been suggested that retinoids may have some immunomodulatory activity.⁹ However, there is no evidence of restoration of the immune function in successfully treated immunosuppressed patients. The hallmark of HPV infection is epithelial hyperplasia and retinoids have an endogenous antiproliferative effect. It has been proposed that the retinoids by altering keratinisation are able to inhibit replication and assembly of the virus, which requires keratinocytes in an advanced rate of differentiation.¹⁰ An inverse relation was observed between concentration of retinoids and HPV-DNA within infected epithelial cells, suggesting a downregulation of viral replication by the retinoids.¹¹ In addition, their potent apoptotic activity may also play a part.¹²

In the present study, isotretinoin showed significantly increased efficacy compared to placebo in cervical RCA, with minor side effects and a relatively low recurrence rate. The efficacy was not significantly different compared to previous experience,^7,8 although the total dose and the cost were halved. Because of its teratogenic effect, the use of isotretinoin as first line therapy for genital warts in women of reproductive age is unacceptable. Selection of patients to be treated and strict contraceptive measures are of crucial importance. In conclusion, low dose oral isotretinoin may represent an efficacious, safe, and cost effective alternative systemic therapy for cervical RCA.

CONTRIBUTORS

SG designed and organised the study; ACK analysed the results, performed the statistical analysis, and prepared the manuscript; CG and EB were responsible for clinical evaluation an follow up and recorded the data; AM performed the gynaecological examination.