Topically Applied Treatments for External Genital Warts in Nonimmunocompromised Patients

A Systematic Review and Network Meta-analysis

J.M. Jung; C.J. Jung; W.J. Lee; C.H. Won; M.W. Lee; J.H. Choi; S.E. Chang

Disclosures

The British Journal of Dermatology. 2020;183(1):24-36. 

In This Article

Results

Literature Search and Study Characteristics

We initially retrieved 1438 records from the electronic database searches. After removing duplicates and excluded studies, 41 RCTs (6371 patients) met the eligibility criteria (Figure 1).[16–56] Figure 2 shows the network plots, depicting direct evidence between treatments. In total 25 treatments were used in 41 RCTs: imiquimod 1% cream (n = 2), imiquimod 2% cream (n = 2), imiquimod 2·5% cream (n = 2), imiquimod 3·75% cream (n = 2), imiquimod 5% cream (n = 9), podophyllotoxin 0·15% cream (n = 3), podophyllotoxin 0·25% solution (n = 1), podophyllotoxin 0·3% cream (n = 2), podophyllotoxin 0·5% cream (n = 6), podophyllotoxin 0·5% solution (n = 12), podophyllotoxin 0·5% solution or cream (n = 1), podophyllin 0·5% solution (n = 2), podophyllin 2·0% solution (n = 1), podophyllin 20–25% solution (n = 8), sinecatechins 10% ointment (n = 3), sinecatechins 15% ointment (n = 3), TCA (n = 2), TCA with podophyllin 25% (n = 1), sodium nitrite with citric acid cream (n = 1), SB206 12% gel (n = 1), polyhexamethylene biguanide cream (n = 1), interferon alpha cream (n = 2), idoxuridine 0·25% cream (n = 2), idoxuridine 0·5% cream (n = 1) and cidofovir gel (n = 1). Because imiquimod 5% cream is a standard topically applied therapy for external genital warts in South Korea, this was used as the main comparator, along with placebo, to analyse treatment outcomes.

Figure 1.

The PRISMA flow diagram of the literature search and study selection RCT, randomized controlled trial.

Figure 2.

Network plots demonstrating direct evidence between the assessed topical treatments in terms of (a) complete clearance, (b) recurrence, (c) adverse events and (d) patients with severe adverse events or patients who withdrew because of treatment-related adverse events. A line between any two treatments means there is at least one trial comparing them. The line thickness is proportional to the number patients who participated in the trials between two treatments. Shading indicates a three-arm study. TCA, trichloroacetic acid. [Colour figure can be viewed at wileyonlinelibrary.com]

The baseline characteristics of the patients, detailed treatment schedules, and treatment frequencies and durations and outcomes are described in Table S2 and Table S3 (see Supporting Information). Risk of bias in the included studies is shown in Table S4 (see Supporting Information). Only eight studies showed low or unclear risk of bias for all domains,[17,25,26,29,30,42,43,53] thereby rendering a high risk of bias across trials for each outcome. Results from all comparisons regarding each outcome are shown in the Supporting Information, in Figure S1 (complete clearance), Figure S2 (recurrence), Figure S3 (adverse events) and Figure S4 (patients with SAEs or patients who withdrew because of treatment-related adverse events).

Complete Clearance

For complete clearance, 41 RCTs were included in the NMA.[16–56] All treatments were significantly more efficacious than the placebo (Figure 3a). Compared with imiquimod 5% cream, only interferon alpha cream (OR 6·22, 95% CI 1·77–21·8) and podophyllotoxin 0·5% solution (OR 1·94, 95% CI 1·02–3·71) were significantly more efficacious (Figure 3b). Imiquimod 5% cream had significantly better therapeutic efficacy than imiquimod 1% (OR, 0·21, 95% CI 0·14–0·34), imiquimod 2·5% (OR 0·25, 95% CI 0·13–0·49) and imiquimod 3·75% cream (OR 0·36, 95% CI 0·18–0·70); sinecatechins 10% (OR 0·18, 95% CI 0·11–0·30) and 15% ointment (OR 0·21, 95% CI 0·12–0·34); and sodium nitrite with citric acid cream (OR 0·27, 95% CI 0·11–0·69) (Figure 3b). Each treatment's P-score showed that idoxuridine 0·5%, interferon alpha, idoxuridine 0·25%, imiquimod 2% and polyhexamethylene biguanide creams were the most efficacious for complete clearance (Table 1).

Figure 3.

Forest plots of network meta-analysis for complete clearance: (a) others vs. placebo, (b) other vs. imiquimod 5% cream. OR, odds ratio; CI, confidence interval.

We conducted a subgroup analysis excluding the studies performed before the introduction of imiquimod.[19,22,23,26,28–32,35,37,44,46,48,53–55] The difference between the efficacy of imiquimod 5% cream and podophyllotoxin 0·5% solution was insignificant, whereas sinecatechins ointments remained inferior to imiquimod 5% cream in this analysis. When we conducted another subgroup analysis excluding the studies on unconventional modalities,[29,30,36,38,42,52] podophyllotoxin 0·5% solution continued to show better efficacy than imiquimod 5% cream. Sinecatechins remained inferior to imiquimod 5% cream in this analysis. The ranking among conventional options by P-score analysis was identical to the previous ranks shown in Table 1.

Recurrence

In total 18 RCTs were included in the NMA for recurrence.[17,18,21,23,24,26–28,31,34–36,43,44,50,53,54,56] They showed that none of the topical agents was significantly different from the placebo or imiquimod 5% cream (Figure 4a, b) or from each other in entire comparisons (Figure S2; see Supporting Information). The P-score analysis revealed that imiquimod 1% cream, placebo, sinecatechins 15% ointment, imiquimod 5% cream and sinecatechins 10% ointment were the most efficacious for lowering recurrence (Table 1).

Figure 4.

Forest plots of network meta-analysis for recurrence: (a) other vs. placebo, (b) others vs. imiquimod 5% cream. OR, odds ratio; CI, confidence interval.

Adverse Events

In total 19 RCTs were included in the NMA for analysis of adverse events.[16–22,24,26,31,32,37,38,45,46,52,53,55,56] These showed that, except for imiquimod 1% and 2% cream, interferon alpha cream, SB206 12% gel and sodium nitrite with citric acid cream, all treatment options were associated with significantly higher overall adverse event rates than placebo (Figure 5a). Podophyllin 0·5% (OR 4·71, 95% CI 1·38–16·11) and 2·0% solution (OR 5·31, 95% CI 1·56–18·06); podophyllotoxin 0·15% (OR 4·21, 95% CI 1·35–13·11) and 0·5% cream (OR 2·32, 95% CI 1·22–4·42); and podophyllotoxin 0·5% solution (OR 2·46, 95% CI 1·20–5·06) were related to significantly higher overall adverse event rates than imiquimod 5% cream (Figure 5b). Only imiquimod 1% cream (OR 0·41, 95% CI 0·23–0·73) and placebo (OR 0·25, 95% CI 0·16–0·40) showed significantly lower overall rates of adverse events than imiquimod 5% cream (Figure 5b). The P-scores indicated that, in addition to placebo, treatments with imiquimod 1% cream, sodium nitrite with citric acid cream, imiquimod 2·5% cream and imiquimod 3·75% cream showed favourable outcomes regarding adverse events (Table 1).

Figure 5.

Forest plots of network meta-analysis for adverse events: (a) others vs. placebo, (b) other vs. imiquimod 5% cream. OR, odds ratio; CI, confidence interval.

When we conducted a subgroup analysis excluding the studies performed before the introduction of imiquimod,[19,22,26,31,32,37,46,53,55] the difference in overall adverse event rate between imiquimod 5% cream and podophyllotoxin 0·5% cream became insignificant. When we conducted another subgroup analysis including only studies on imiquimod 5% cream and podophyllotoxin 0·5%,[16,18–22,24,31,32,37,46,53,55,56] podophyllotoxin 0·5% was still related to significantly higher overall adverse event rates than imiquimod 5% cream. In pairwise meta-analyses, podophyllotoxin 0·5% (OR 7·47, 95% CI 5·69–9·79; I 2 = 27%) and imiquimod 5% (OR 4·74, 95% CI 3·81–5·91; I 2 = 0%) were associated with higher overall adverse event rates than placebo.

Patients With Serious Adverse Effects or Patients who Withdrew Because of Treatment-related Adverse Events

This analysis of 17 studies[19,23,31–33,35–38,42–44,50,52,53,55,56] showed that imiquimod 5% cream (OR 8·68, 95% CI 1·01–74·4), podophyllin 2·0% solution (OR 38·4, 95% CI 1·28–1156), podophyllotoxin 0·5% cream (OR 5·98, 95% CI 1·07–33·5), polyhexamethylene biguanide cream (OR 55·9, 95% CI 3·33–938) and sinecatechins 10% (OR 8·03, 95% CI 3·97–16·2) and 15% ointment (OR 8·54, 95% CI 4·23–17·3) were associated with significantly higher numbers of patients with SAEs or patients who were lost to follow-up because of treatment-related side-effects compared with placebo (Figure 6a). Except for placebo, none of the treatments was significantly different from imiquimod 5% cream (Figure 6b). Moreover, no treatment other than placebo was significantly different from any other in entire comparisons (Figure S4; see Supporting Information). Placebo, podophyllotoxin 0·15% cream, cidofovir gel, SB206 12% gel and podophyllotoxin 0·3% cream were the top-ranked treatments (Table 1). SAEs or adverse events that led to patient withdrawal are described in Table 2. All treatment-related SAEs were limited to local reactions.

Figure 6.

Forest plots of network meta-analysis for patients with severe adverse events or patients who withdrew because of treatment-related adverse events: (a) others vs. placebo, (b) other vs. imiquimod 5% cream. OR, odds ratio; CI, confidence interval.

Heterogeneity and Consistency Analysis

Only NMA for adverse events showed moderate heterogeneity between studies. NMA analyses of the remaining three outcomes showed high homogeneity (Table S5; see Supporting Information). The results of the decomp.design function supported the global consistency for all NMAs (Table S6; see Supporting Information), indicating the reliability of the results. Only the NMA for recurrence showed local inconsistency in net heat plot and net split function analysis (Figures S5, S6; see Supporting Information).

Assessment of Publication Bias

Comparison-adjusted funnel plots (Figure 7) were symmetrical, implying no evidence of publication bias.

Figure 7.

Comparison-adjusted funnel plots of trials included in the network meta-analysis for (a) complete clearance, (b) recurrence, (c) adverse events and (d) patients with severe adverse events or patients who withdrew because of treatment-related adverse events.

processing....