Prospective Randomized Trial of Topical a-Interferon (a -Interferon Gels) for the Treatment of Vulvar Intraepithelial Neoplasia III
Nick M. Spirtos, M.D., Lloyd H. Smith, M.D., Ph.D., and Nelson N. H. Teng, M.D., Ph.D.
Division of Gynecologic Oncology, Department of Gynecology and Obstetrics, Stanford University Medical Center, Stanford, California 94305
Twenty-one patients were prospectively randomized into a blinded double-armed crossover study comparing a -interferon (a -IFN, 106 IU in a 3.5% aqueous methycellulose base) with and without 1% nonoxynol-9. Nine and twelve patients were randomized to arms with (+N) and without (-N) 1% nonoxynol-9, respectively. Patients applied the gel to affected areas every 8 hr. and were evaluated biweekly. Including those crossed over, 14 patients were treated with -N. Six of fourteen (43%) achieved complete responses: biopsy proven with at least 1 year follow-up (CR). One patient achieved a partial response with at least a 50% reduction in the total surface area of all lesions present (PR). Similarly, 13 patients were treated with +N. Two patients in this group were found to have invasive cancer and one to have HIV and thus were eliminated from statistical analysis. Of the remaining 10 patients, 3 had CRs (30%), 5 had PRs (50%), and 2 failed to respond. There was no significant difference in responses between the two groups. Overall, 14 of 18 (67%) patients demonstrated some response to a-IFN applied topically. These data support the conclusion that a-IFN is an active agent in the treatment of vulvar intraepithelial neoplasia III.
Severe vulvar intraepithelial neoplasia (VIN III) has become more
frequently recognized in young, sexually active women
(1,2). Management imperitives include (1) exclusion
of the existence of an invasive component and (2) removal of all lesions.
However, since such lesions are often widespread or multifocal, removal may
require resection or vaporization of substantial portions of the vulva,
resulting in unavoidable morbidity. The recognition that such lesions may
recur even after extensive therapy and that VIN lesions are classically
thought to be indolent, carrying a low probability of transition to an
invasive lesion, makes some form of conservative therapy very
Materials and MethodsBeginning on July 1, 1985 and ending on June 30, 1987, twenty-one patients were prospectively randomized into a blinded double-armed crossover study comparing the therapeutic effects of a-IFN (10-6IU/3 in 3.5% aqueous methylcellulose base) with and without 1% nonoxynol-9 (Exovir Inc., Great Neck, NY). Patients were entered into the study only after a histologic diagnosis of VIN III was confirmed in the Department of Pathology at Stanford University Hospital and informed consent obtained. Patients were then instructed to apply the gel every 8 hr. to all lesions and were seen every other week in the Gynecologic Oncology Clinic. As long as a response of consistent diminution in lesion size was observed, the patient was continued on the same drug. If cessation of clinical response was noted over an 8-week period, progression was noted, or a new lesion appeared, the patient was crossed over into the other arm of the study. Prior to the time of crossover, all patients underwent rebiopsy to confirm the diagnosis as well as to rule out an invasive lesion. In those cases in which a complete clinic response was noted, a biopsy of previously affected areas was performed 2 weeks after disappearance of the last lesion and then every 6 months. In those patients who ultimately failed both arms of the treatment protocol, management was individualized.
Of the 21 patients initially enrolled in the study, 9 were randomized to treatment with a-IFN plus nonoxynol-9 (+N) and 12 were randomized to treatment with a-IFN alone (-N). A complete history was obtained and a physical examination performed on each patient. The individual lesions were measured in two dimensions and recorded on an every-other-week basis. At these visits, all symptoms and possible side effects related to the medication were noted. A complete blood count with differential and platelet count was performed on entry into the study and again between 4 and 8 weeks after beginning therapy. A complete response (CR) was considered to have been achieved only when all clinical evidence of disease was noted to be absent and this observation was confirmed for at least 1-year in a previously affected and treated area were considered, for the purposes of this study, to have had a partial response (PR). Patients who did not achieve a CR or a PR were considered treatment failures.
Twelve patients were randomized to treatment wtih -N and 9 patients to treatment with +N. Two patients were crossed over into the -N arm and four patients were crossed over into the +N arm. The total number of patients treated with -N before or after crossover was 14. The total number of patients treated with +N was 13. Three of these patients, however, were eliminated from the statistical analysis: two because invasive lesions were found within 2 months of entry into the study and one because HIV infections was diagnosed. Characteristics and cure rates of the two groups of patients were compared using Fisher's exact test.
ResultsThe average age in the -N group was 46 years, and in the +N group, 37 years. This difference was not significant. Additionally, the initial average lesions size was 231 mm2 in the -N groups and 237 mm2 in the +N group. This was statistically indistinguishable. No differences were noted between the two groups in the following areas: (1) previous HPV infection; (2) previous gynecologic infection; (3) previous abnormal PAP smears; (4) smoking history; (5) alcohol use; (6) hormonal replacement; (7) symptoms including itching, burning, and pain; (8) previous treatment for VIN III; and (9) adverse side effects from the medication. Four of six patients (67%) in the +N groups and ten of twelve patients (83%) in the -N group presented with symptoms, the most common being itching and burning. A similar number of patients in both groups had been previously treated for VIN III. Three of six patients (50%) in the +N groups (one by cryotherapy and two by local excision) and eight of twelve (67%) in the -N groups (four by excision, and one by excision and laser ablation) had been previously treated. Nine of eleven (82%) patients previously treated for VIN III and four of six (67%) of those not previously responded to therapy.
Of 14 patients in the -N group, there were 6 complete responses (43%), 1 partial response (7%), and 7 failures (50%). Among the 10 patients in the +N group there were 3 complete responses (30%), 5 partial responses (50%), and 2 failures (20%). There was no statistical difference between the number of responders in the two arms of the study. More impressive is the fact that of 18 patients treated, 14 demonstrated a significant response to a-IFN therapy. Of 4 nonresponders, 2 were not crossed over because they moved and were unable to continue in the study.
Among complete responders in the +N group all lesions had resolved by
the fourth month (Fig. 1). A similar pattern was noted in the -N groups
(Fig. 2). Two patients in the +N groups had no evidence of disease within
4 months of beginninng therapy and negative biopsies after 6 months, but
were found to have recurred at their 12-month visit. Two patients from the
+N group were crossed over into the -N group. One patient had achieved a
partial response to +N, but her response slowed, necessitating crossover,
and she ultimately failed to respond to -N. The other patient had no
response to either treatment. Four patients were crossed over from the -N
to +N group: two subsequently achieved complete responses despite failing
the -N regimen, one patient had partial responses to both regimens, and one
patient failed both regimens.
We have observed dramatic and often rapid responses to topical
a -IFN, with total disappearance of some very large, biopsy-proven
VIN III lesions. Although no untreated control group was studied, most
patients had longstanding lesions which had often failed or recurred after
prior therapy. We therefore suggest that topical a-IFN actually
mediated the disappearance of these VIN III lesions.
In this trial, a-IFN with and without nonoxynol-9 proved to be an
effective means to treat VIN III. Overall, 9 of 18 patients (50%) had a
complete, histologically proven respnse of at least 1-year duration, and 14
of 18 showed either a partial or a complete response to therapy (78%).
This compares favorably to results for topical 5-fluorouracil (5-FU) (Table
1) (1,13,14,15), an antimetabolite commonly used to
treat VIN III. However, the morbidity associated with topical a -IFN
is insignificant when compared to that associated with 5-FU. In our
series, only 5 of 18 patients (28%) described any adverse side effect
related to therapy. The complaints were transitory and did not require
interruption or discontinuation of therapy. On the other hand, Krupp and
Bohm (14) and Lifshitz and Roberts
(15) describe universal reactions to 5-FU which
include severe pruritis, pain, vulvar edema, and weeping, requiring
analgesics, sitz baths, and, in some patients, discontinuation of therapy.
In this respect it appears that a-IFN offers an improvement over
topical 5-FU. Other medical therapies reported in the literature include
topical dinitrochlorobenzene (DNCB) (16) and
bleomycin (17). Foster and Woodruff treated six
patients, only two of which had VIN III on initiation of therapy, with DNCB
and both had complete responses (16). One complained
of severe irritation requiring cessation of therapy. Roberts et al.
reported dismal results, with only 2 of 12 patients (17%) responding, using
topical and intradermal bleomycin (17). The low
morbidity associated with the use of topical a-IFN in this study and
the high response rate (78%) allow for the conclusion that a -IFN
compares favorably to other forms of medical management in the treatment of
Although we found no significant difference between the two arms of the study, overall, 50% of the patients considered evaluable had a complete response lasting at least 1 year and another 28% had partial responses. These figures are, if anything, understated since two patients who withdrew from the study were responding to the a -IFN. One had achieved a partial response in less than 2 months and the other had experienced a 39% reduction in lesion size over 3 months. An additional two patients quit and were never crossed over after failing the -N arm of the study. Nevertheless, the medication was well tolerated and patients achieving complete responses did so in less than 4 months. Given these findings, it would be reasonable to increase either the dose or the intensity of a-IFN in hopes of improving response rates.
Results of excisional and ablative techniques used to treat VIN III are difficult to compare with the results obtained in this study because the methodology and follow-up in most reports are insonsistent and incomplete. A review of the world literature through 1985 reveals 902 patients reported in 75 articles treated by excisional or ablative therapy. Within 3 years of completing therapy, 380 (42%) were lost to follow-up; 123 (14%) have documented cures; 85 (9%) have recurred; 306 (34%) were described as cures with follow-up ranging from months to years; and 8 (1%) died (18). Understandably, given this information and the relatively short follow-up in our patients, comparison of topical a-IFN to these other methods of treatment is difficult. Attempts to do so would likely be misleading.
Perhaps an even more important issue is the caution that must be exercised when any form of nonexcisional therapy is used to treat VIN III. Two of twenty-one (9%) patients initially entered into the study were found to have invasive cancer. Interestingly, both patients had been previously treated for VIN as had nine other patients. Thus, 2 of 11 (18%) previously treated patients were found to have invasive lesions. This is substantially higher than the 2% incidence noted by Friedrich et al. (19) or the 4% incidence reported by Buscema et al. (1). These findings mandate careful initial evaluation, follow-up, and excisional biopsy of any suspicious lesion. Our data would suggest this is even more important in the patient with recurrent disease, and in this setting non-excisional therapies must be closely scrutinized.
1. Buscema, J., Woodruff, J.D., Parmley, T., and Genadry, R. Carcinoma
in situ of the vulva, Obstet. Gynecol.. 55 225-230 (1980).