Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Genital wart treatments

Search
Outcomes
Results
Imiquimod
Podophyllotoxin
Comment
DIY evidence

Anogenital warts (also called condylomas) are caused by the human papillomavirus (HPV). They occur predominantly in sexually active young adults, and have a variety of forms. They are common, with a million consultations annually in the USA, and over 50,000 newly diagnosed patients in the UK every year.

Bandolier was asked whether a recent editorial [1] suggesting that self-administered treatments were effective was correct. How can anyone tell? This seemed like a good excuse for a swift search for evidence as an exemplar of how a rapid review might be helpful and how it could be performed by a healthcare professional for, say, a health authority or primary care group wanting to make policy.

Search


Two self-administered treatments were mentioned, imiquimod and podophyllotoxin creams. The way forward seemed to be to look for high quality and relevant trials. The ideal would be randomised double blind trials of self-administered treatment versus placebo in patients with properly diagnosed anogenital warts. Therefore the search used PubMed, the names of the treatments, and the subheading of randomised controlled trials. The Cochrane Library was also searched using a similar strategy. Searches revealed no previous systematic reviews.

For each treatment there was a page or two of possibly useful references (about 30-40). Most could be dismissed by reading the abstracts online. About 20 papers seemed relevant, and were ordered from the library.

Outcomes


How does one choose an outcome in an unfamiliar area? A reasonable rule of thumb is to remember that while there can be many different outcomes, the highest hurdle for efficacy is patients cured (whatever that may be) and for harm is likely to be patients who discontinued because of adverse events. Patients with initial warts completely cleared at treatment end seemed a sensible efficacy outcome.

Results


There were five relevant papers for imiquimod [2-6] and two for podophyllotoxin [7,8]. Each was randomised and double blind, had a placebo group, enrolled patients using a sensible diagnosis of genital warts (often with HPV confirmed), and had self-administered treatments. Several other papers were likely to be relevant, and for podophyllotoxin in particular there were a number of high quality studies with only active comparators.

Details of the trials are in Table 1 for imiquimod and Table 2 for podophyllotoxin. While all were randomised none gave details of randomisation, thus scoring one point out of two for randomisation. While all stated they were double blind, only two told us categorically that active and placebo treatments were identical, the others scoring one point out of two [9]. Most discussed withdrawals.

Table 1: Randomised placebo-controlled trials of imiquimod cream for genital warts

Reference, origin and quality score Patients Treatments Design Results Adverse effects
Beutner et al, 1998a. USA R 1, DB 1, WD 1 Total 3/5 Men and women aged 18 or older with more than 2 and fewer than 50 external genital warts. HIV negative. 5% imiquimod cream 1% imiquimod cream placebo vehicle Randomised, double blind, parallel group, with weekly and later two weekly visits for a maximum of 16 weeks, with further 12 week follow up. Complete clearance 5% cream 49/94 1% cream 13/90 placebo 3/95 Local reactions included erythema, excoriation, flaking and erosion in all groups. Pain, burning and tenderness were more common with imiquimod than placebo. One patient in each imiquimod group discontinued because of skin reactions.
Beutner et al, 1998b. USA R 1, DB 2, WD 1 Total 4/5 Men and women aged 18 or older (90% men). HIV negative. 5% imiquimod cream placebo vehicle Randomised, double blind, parallel group, with weekly visits for a maximum of 8 weeks, with further 10 week follow up. Complete clearance 5% cream 18/51 placebo 0/57 Skin irritation greater with imiquimod than vehicle. Two patients withdrew because of this.
Edwards et al, 1998. USA & Canada R 1, DB 1, WD 1 Total 3/5 Men and women aged 18 or older with more than 2 and fewer than 50 external genital warts. HIV negative. 5% imiquimod cream 1% imiquimod cream placebo vehicle for 6-10 hours three times a week Randomised, double blind, parallel group, with weekly and later two weekly visits for a maximum of 16 weeks, with further 12 week follow up. Complete clearance 5% cream 54/109 1% cream 21/102 placebo 11/100 Local inflammation reactions were common, and more frequent with the 5% imiquimod cream. Two patients with the 5% cream were discontinued because of local reactions.
Syed et al, 1998. Pakistan R 1, DB 2, WD 1 Total 4/5 Women aged 18 to 45 2% imiquimod cream placebo vehicle Application twice a day for five consecutive days Randomised, double blind, parallel group, with weekly and later two weekly visits for a maximum of 16 weeks, with further 16 week follow up. Complete clearance 2% cream 25/30 placebo 1/30 Some mild adverse events, and no dropouts.
Glison et al, 1999. UK & USA R 1, DB 1, WD 1 Total 3/5 Men and women aged 18 and older (mostly men) with HIV infection, and a minimum of 2 warts. 5% imiquimod cream placebo vehicle Applied for 6-10 hours three times a week. Randomised, double blind, parallel group, with weekly and later two weekly visits for a maximum of 16 weeks, with further 16 week follow up. Complete clearance 5% cream 7/65 placebo 2/35 One patient in each group discontinued because of local skin reaction. More local reactions with imiquimod than vehicle.

In all cases the outcome chosen was the proportion of patients with complete clearance of baseline warts. ITT analysis of all randomised patients. R is randomised, DB is double blind, WD is withdrawals and dropouts [9].

 

Table 2: Randomised placebo-controlled trials of podophyllotoxin cream for genital warts

Reference and origin Patients Treatments Design Results Adverse effects
Syed et al, 1994: Pakistan R = 1, DB = 1, WD =0 Total 2/5 Women aged 16 to 40 years. 0.5% podophyllotoxin cream 0.3% podophyllotoxin cream placebo vehicle Application twice daily for three consecutive days per week for up to four weeks Randomised, double blind, parallel group, with weekly visits for a maximum of 4 weeks, with further 16 week follow up. Complete clearance 0.5% cream 25/30 0.3% cream 16/30 placebo 0/20 Tenderness and burning sensation were reported local adverse reactions
Syed et al, 1995. Pakistan R = 1, DB = 1, WD = 1 Total 3/5 Women aged 18 to 40 years. 0.5% podophyllotoxin cream placebo vehicle Application twice daily for three consecutive days per week for up to four weeks Randomised, double blind, parallel group, with weekly visits for a maximum of 4 weeks, with further 16 week follow up. Complete clearance 0.5% cream 12/20 placebo 4/20 Not reported in detail

In all cases the outcome chosen was the proportion of patients with complete clearance of baseline warts. ITT analysis of all randomised patients. R is randomised, DB is double blind, WD is withdrawals and dropouts [9].

 


Imiquimod


Most studies used imiquimod 5% cream applied three times a week for 8-16 weeks. One [5] used a 2% cream applied twice daily for six weeks. One study [6] was specifically directed to patients with HIV infection, while the others omitted patients with HIV infection.

Results for the four studies on HIV-free patients [2-5] are shown in Figure 1 and Table 3. At the end of treatment with placebo, only 15/282 (5%) patients had their initial warts completely cleared. With imiquimod 146/284 (51%) of patients had warts completely cleared. The number needed to treat with imiquimod for six to 16 weeks to obtain one patient completely cleared who would not have been with placebo was 2.2 (95% CI 1.9 to 2.5).

Figure 1: Imiquimod (open) and podophyllotoxin (filled) RCTs versus placebo


Table 3: Imiquimod and podophyllotoxin NNTs


    Genital warts completely cleared at end of treatment with    
    Active Placebo    
Treatment Number of trials number/ total Percent (95%CI) number/ total Percent (95%CI) Relative benefit (95%CI) NNT (95% CI)
Imiquimod 4 146/284 51 (46 to 57) 15/282 5 (3 to 8) 9.1 (5.6 to 15) 2.2 (1.9 to 2.5)
Podophyllotoxin 2 37/50 74 (62 to 86) 4/40 10 (1 to 19) 7.0 (2.7 to 18) 1.6 (1.3 to 2.1)

The one study in HIV-infected patients [5] had only 7/65 (11%) patients cleared with active cream, compared with 2/35 (6%) with placebo.

Local cutaneous adverse events were common with imiquimod, notably erythema and skin irritation, with some burning and pain. Few patients discontinued therapy because of adverse events (Table 1).

Podophyllotoxin


Two studies examined 0.5% podophyllotoxin cream in women [7,8]. Application was twice daily for three consecutive days a week, for up to four weeks (Table 2).

Results are in Figure 1 and Table 3. At the end of treatment with placebo, only 4/40 (10%) patients had their initial warts completely cleared. With podophyllotoxin 37/50 (74%) of patients had warts completely cleared. The number needed to treat with podophyllotoxin for four weeks to obtain one patient completely cleared who would not have been with placebo was 1.6 (95% CI 1.3 to 2.1). Some local tenderness and burning was reported, but no patients discontinued because of this.

Comment


The first thing is that an NNT of 2 is usually the marker of an effective treatment. For both treatments this was obtained using the high hurdle of patients completely cleared of the warts they had initially. Using lower hurdles, like at least 50% of the warts cleared, more impressive NNTs would be likely. The analysis was an intention-to-treat analysis, in which some patients lost to follow up were regarded as treatment failures.

There were local adverse events, but not so severe as to make many participants discontinue.

There are a few other things to think about though.

  • First, especially for podophyllotoxin, there were several randomised trials with active comparators rather than placebo. Analysing these as well, though not as NNTs, would give more weight to the few patients in placebo-controlled trials.
  • Second, there are more issues over efficacy. For instance, the analysis could have included analysis on warts appearing after treatment started. Then there is the issue of warts re-appearing after the end of treatment. And, of course, a detailed analysis of adverse events might be useful in telling prospective patients what to expect.
  • Third is the issue of where these treatments are best used, in primary or secondary care? That may be more contentious, or easily incorporated into a treatment plan, according to local circumstances and budgets. An audit in Belfast [10] of 52 outpatients who had not benefited from a range of other treatments had similar outcomes to the randomised trials.

DIY evidence


Where does the itch go when you scratch it? Just one of the many difficult questions. But is the process process of answering them one that can be done only by anoraks like Bandolier , or is it do-able elsewhere? Anyone with Internet access can search PubMed. Any UK doctor can join doctors.net and have access to the Cochrane Library online. Once the search is done, a local hospital library ought to be able to obtain the papers in a week or so.

To read, digest, and follow a few simple guidelines on systematic review and data presentation is not hard. So there really is no reason why any healthcare professional should not be able to do a swift analysis from time to time. But why don't we demand that companies have to present such independent reviews and analysis before they are allowed to market the drug?

References:

  1. R Maw, G von Krogh. The management of anal warts. BMJ 2000 321: 910-911.
  2. KR Beutner et al. Imiquimod, a patient-applied immune-response modifier for treatment of external genital warts. Antimicrobial Agents and Chemotherapy 1998 42: 789-794.
  3. KR Beutner et al. Treatment of genital warts with an immune-response modifier (imiquimod). Journal of the American Academy of Dermatology 1998 38: 230-239.
  4. L Edwards et al. Self-administered topical 5% imiquimod cream for external anogenital warts. Archives of Dermatology 1998 134: 25-30.
  5. TA Syed et al. Management of female genital warts with an analog of imiquimod 2% in cream: a randomised, double-blind, placebo-controlled study. Journal of Dermatology 1998 25: 429-433.
  6. RJ Gilson et al. A randomized, controlled, safety study using imiquimod for the topical treatment of anogenital warts in HIV-infected patients. AIDS 1999 13: 2397-2404.
  7. TA Syed et al. Topical 0.3% and 0.5% podophyllotoxin cream for self-treatment of condylomata acuminata in women. Dermatology 1994 189: 142-145.
  8. TA Syed et al. Management of genital warts in women with human leukocyte interferon-alpha vs. podophyllotoxin cream: a placebo-controlled, double-blind, comparative study. Journal of Molecular Medicine 1995 73: 255-258.
  9. AR Jadad et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Controlled Clinical Trials 1996 17: 1-12.
  10. JE Maitland, R Maw. An audit of patients who have received imiquimod cream 5% for the treatment of anogenital warts. International Journal of STD & AIDS 2000 11: 268-270.
previous or next story in this issue