Figure 1. Clinical photographs before treatment (A) and at the end of the study (B) illustrate complete clearance of facial flat warts.
Figure 2. Wart elevation during weeks of glycolic acid use. n Indicates number of treated sites.
Moore RL, de Schaetzen V, Joseph M, Lee IA, Miller-Monthrope Y, Phelps BR, Lekalake SS, Ratcliffe SJ, Nguyen H, He Q, Rady P, Tyring S, Kovarik CL. Acquired Epidermodysplasia Verruciformis Syndrome in HIV-Infected Pediatric Patients: Prospective Treatment Trial With Topical Glycolic Acid and Human Papillomavirus Genotype Characterization. Arch Dermatol. 2012;148(1):128-130. doi:10.1001/archdermatol.2011.268
Author Affiliations: Department of Dermatology, Oregon Health and Science University, Portland (Dr Moore); Department of Dermatology and Mohs Surgery, Bright Health Physicians of Presbyterian Intercommunity Hospital, Whittier, California (Dr Moore); Institut de Recherches Interdisciplinaires et de D éveloppements en Intelligence Artificielle (IRIDIA CoDE), Universite Libre de Bruxelles, Brussels, Belgium (Dr de Schaetzen); Division of Dermatology, Department of Internal Medicine, University of Toronto, Toronto, Ontario, Canada (Drs Joseph and Miller-Monthrope); Department of Dermatology, University of California, San Francisco (Dr Lee); Botswana-Baylor Children's Clinical Center of Excellence, Gaborone, Botswana (Dr Phelps and Ms Lekalake); Department of Biostatistics & Epidemiology (Dr Ratcliffe), Department of Dermatology (Dr Kovarik), and Division of Infectious Diseases, Department of Internal Medicine (Dr Kovarik), University of Pennsylvania, Philadelphia; and Department of Dermatology, University of Texas Medical School at Houston (Drs He, Rady, and Tyring). Mr Nguyen is a Medical Student at Rice University, Houston.
Many children with human immunodeficiency virus (HIV) also have diffuse flat warts, which occur primarily on the face and dorsal surface of the hands and are often recalcitrant to multiple treatments. Alpha-hydroxy acids have demonstrated some effectiveness in the treatment of flat warts.1 Flat warts in the general population are usually caused by infection with human papillomavirus (HPV) types 3 and 10.2 However, flat warts in epidermodysplasia verruciformis (EV), a rare genodermatosis associated with a profound susceptibility to HPV infection, are caused by EV-HPV types 5, 8, 9, 12, 14, 15, 17, 19 through 25, and 96.3,4 Acquired EV-like clinical disease has been described in immunosuppressed patients, including patients with HIV,4 but it is unknown if EV-HPV types cause flat warts in this setting. We sought to characterize HPV types and evaluate the efficacy, safety, and tolerability of glycolic acid, 15%, lotion for treatment of flat warts in HIV-positive children.
Thirty-eight HIV-positive Batswana children older than 7 years with at least 15 facial and/or hand flat warts were recruited from the Botswana-Baylor Children's Clinical Center of Excellence in Gaborone, Botswana after institutional review board approval was obtained. Patients were given glycolic acid, 15%, lotion to apply daily for to the affected area(s) for 2 weeks and twice daily thereafter. Patients were evaluated every 2 weeks for 8 weeks and then monthly for 8 weeks. At each visit, warts were evaluated by number, elevation, color change, erythema, edema, and scaling. At each follow-up visit, patients were asked subjectively if the warts improved and if they had adverse effects. Background medical information was collected, including the World Health Organization (WHO) clinical HIV stage, time of antiretroviral therapy (ART) initiation, type of ART, timing of wart onset, and most recent CD4 and viral load counts prior to study enrollment.
Biopsies were performed of representative lesions in 30 patients, and specimens were analyzed for HPV using the Puregene Genomic DNA Purification Kit (Gentra Systems, Minneapolis, Minnesota).5,6
Wilcoxon rank-sum tests were used to compare the number of warts at the beginning and end of study (EOS) (Stata 11.0 MP; StataCorp LP, College Station, Texas). Characteristics potentially associated with improvement in warts were tested with Fisher exact tests.
A more detailed description of our methods is available in the eAppendix.
Baseline characteristics of our study population are listed in Table 1. Ninety-five percent (36 of 38) and 82% (31 of 38) of patients participated through weeks 12 and 16, respectively. Early withdrawal was owing largely to conflicts with transportation and school.
At EOS, 10% of patients (3 of 31) had complete clearance of facial warts (Figure 1). No patient had complete clearance of hand warts. There was no significant change in wart numbers from study entry to EOS; however, there was a trend toward the warts becoming less palpable (Figure 2). At baseline, 49% of the warts overall were definitely palpable (31 of 63) compared with 8% at week 16 (4 of 49). Thirty-nine percent of treated facial sites (12 of 31) and 33% of treated hand sites (6 of 18) showed improvement in wart color.
Warts that developed after initiation of ART were significantly more likely to be nonpalpable and skin colored at EOS compared with warts that had developed prior to ART initiation (P = .045 and P = .04, respectively). When separated by wart location, patients with facial warts that had developed after the initiation of ART had over 11 times the odds of overall improvement by EOS compared with patients whose facial warts developed before ART (odds ratio [OR], 11.25; 95% CI, 1.17-108.41) (P = .03). Warts in younger patients showed statistically improved color change at EOS (OR, 0.7025; 95% CI, 0.5210-0.0947) (P = .02). Patient sex, WHO clinical stage, and CD4 counts were not associated with improvement. Subjectively, 87% of patients reported overall improvement in wart appearance by week 16 (26 of 30).
Mild scaling was seen in the minority of patients and typically resolved by week 6. By week 16, 13% (4 of 30), 3% (1 of 30) and 3% (1 of 30) of patients reported mild, moderate, and severe itching, respectively.
There were 11 known and 7 putative HPV types detected (Table 2). Overall, 90% of patients had at least one EV-related HPV type (27 of 30). Over half of patients were infected with more than 1 HPV type or isolate (17 of 30). Patients with only 1 HPV type were 17 times more likely to have an improved color at EOS compared with patients with 2 or more HPV types (OR, 16.66; 95% CI, 0.010.64) (P = .02). There was no association between number of HPV types and elevation change.
A trend toward flattening and color normalization of flat warts in HIV-infected children with glycolic acid lotion treatment was found. Complete clearance of warts may have been observed in only 10% of patients (3 of 31) owing to low concentration of the acid or short observation period. In addition, lack of response in some patients may be attributed to an attenuated immune response to HPV infection.7
Patients who developed warts after ART initiation were more likely to respond to treatment. This improved response was not correlated with the CD4 count at the time of study enrollment. One possible explanation is that patients who developed warts prior to ART initiation may have acquired the causal HPV type(s) at a point when their immune system was less intact, allowing progression of disease and rendering those lesions recalcitrant to therapy or involution.
Other factors that might influence response to therapy include duration of HIV infection and infection with multiple HPV types. We found that younger children and patients with only 1 HPV isolate were more likely to have improved color of their warts with treatment. Since the vast majority of the children in this study were infected with HIV from birth,8 age might be seen as a reflection of duration of HIV infection.
Ninety percent of our patients had EV-related HPV types or isolates identified in their warts (27 of 30), while none had infection with HPV type 3 or 10. Immunosuppression alone may account for this finding, or some patients may possess an additional genetic predisposition for EV-related HPV types.
Correspondence: Dr Kovarik, University of Pennsylvania, 2 Maloney Bldg, 3600 Spruce St, Philadelphia, PA 19104 (firstname.lastname@example.org).
Accepted for Publication: July 10, 2011.
Published Online: September 19, 2011. doi:10.1001/archdermatol.2011.268
Author Contributions: All authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Moore and Kovarik. Acquisition of data: Moore, de Schaetzen, Joseph, Lee, Miller-Monthrope, Phelps, Lekalake, Rady, Tyring, and Kovarik. Analysis and interpretation of data: Moore, Ratcliffe, Nguyen, He, Rady, Tyring, and Kovarik. Drafting of the manuscript: Moore, Rady, and Kovarik. Critical revision of the manuscript for important intellectual content: Moore, de Schaetzen, Joseph, Lee, Miller-Monthrope, Phelps, Lekalake, Ratcliffe, Nguyen, He, Tyring, and Kovarik. Statistical analysis: Moore, Ratcliffe, and Kovarik. Obtained funding: Moore and Kovarik. Administrative, technical, and material support: Moore, de Schaetzen, Joseph, Lee, Miller-Monthrope, Phelps, Lekalake, Nguyen, He, Rady, Tyring, and Kovarik. Study supervision: Tyring and Kovarik.
Financial Disclosure: None reported.
Funding/Support: This research was made possible through core services and support from the Penn Center for AIDS Research (CFAR), a program funded by the National Institutes of Health (P30 AI 045008).
Role of the Sponsors: The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; or in the preparation, review, or approval of the manuscript.