Actinic keratoses (AKs) are common skin lesions resulting from cumulative exposure to UV radiation and are associated with an increased risk for invasive squamous cell carcinoma1; therefore, diagnosis and treatment are important.2 Individual AKs are most frequently treated with cryosurgery, while topical agents including ingenol mebutate gel are used as field treatments on areas of confluent AKs of sun-damaged skin.2,3 Studies have shown that rates of complete clearance with topical therapy can be improved with more than a single treatment course.4-6
Although the mechanisms of action of ingenol mebutate on AKs are not fully understood, studies indicate that it induces cell death in proliferating keratinocytes, which suggests that it may act preferentially on AKs and not on healthy skin.7 The field treatment of AKs of the face and scalp using ingenol mebutate gel 0.015% involves a 3-day regimen,8 and clearance rates are similar to those observed with topical agents that are used for longer periods of time.3,9,10 Local skin reactions (LSRs) associated with application of ingenol mebutate gel 0.015% on the face and scalp generally are mild to moderate in intensity and resolve after 2 weeks without sequelae.3
The presumption that the cytotoxic actions of ingenol mebutate affect proliferating keratinocytes preferentially was the basis for this study. We hypothesized that application of a second sequential cycle of ingenol mebutate during AK treatment should produce lower LSR scores than the first application cycle due to the specific elimination of transformed keratinocytes from the treatment area. This open-label study compared the intensity of LSRs during 2 sequential cycles of treatment on the same site of the face or scalp using ingenol mebutate gel 0.015%.
Eligible participants were adults with 4 to 8 clinically typical, visible, nonhypertrophic AKs in a 25-cm2 contiguous area of the face or scalp. Inclusion and exclusion criteria were the same as in the pivotal studies.3 The study was approved by the institutional review board at the Icahn School of Medicine at Mount Sinai (New York, New York). Enrollment took place from March 2013 to August 2013.
All participants were treated with 2 sequential 4-week cycles of ingenol mebutate gel 0.015% applied once daily for 3 consecutive days starting on the first day of each cycle (day 1 and day 29). Participants were evaluated at 11 visits (days 1, 2, 4, 8, 15, 29, 30, 32, 36, 43, and 56) during the 56-day study period (Figure 1). Eligibility, demographics, and medical history were assessed at day 1, and concomitant medications and adverse events (AEs) were evaluated at all visits. Using standardized photographic guides, 6 individual LSRs—erythema, flaking/scaling, crusting, swelling, vesiculation/pustulation, and erosion/ulceration—were assessed on a scale of 0 (none) to 4 (severe), with higher numbers indicating more severe reactions. For each participant, a composite score was calculated as the sum of the individual LSR scores.3 Throughout the study, 3 qualified evaluators assessed AK lesion count and graded the LSRs. The same evaluator assessed both treatment courses for each participant for the majority of assessments.
|
| Figure 1. Time course of the composite local skin reaction (LSR) scores during cycle 1 (A) and cycle 2 (B) following initiation of a 3-day treatment course (indicated by arrow) with ingenol mebutate gel 0.015% (N=17 for days 2, 30, 32, 36, and 43; N=18 for days 4, 8, 15, 29, and 56). Error bars indicate standard deviation (SD). |
The primary end point of the study was to evaluate the degree of irritation in each of the 2 sequential cycles of ingenol mebutate treatment by assessing the mean area under the curve (AUC) of the composite LSR score over time following each of the 2 applications. Actinic keratoses were counted at baseline and at the end of each treatment cycle. The paired t test was used to compare AUCs of the composite LSR scores of the 2 cycles and to compare the changes in lesion counts from baseline to day 29 and from baseline to day 56. The complete clearance rates (number of participants with no AKs) at the end of cycles 1 and 2 were compared using a logistic regression model. Participant-perceived irritation and treatment satisfaction were evaluated using a 0 to 100 visual analog scale (VAS), with higher numbers indicating greater irritation and higher satisfaction. Participant-reported scores were summarized.
A total of 20 participants were enrolled in the study. At the completion of the study, 2 participants withdrew consent but allowed use of data from their completed assessments. Consequently, a total of 18 patients completed the entire study. The mean age was 75.35 years (median, 77.5 years; age range, 49–87 years). Most of the participants (15/20 [75%]) were men. All participants were white, and 2 were of Hispanic ethnicity. Of the 20 participants, 19 (95%) were Fitzpatrick skin type II, and 1 (5%) was Fitzpatrick skin type I. Most of the participants (16/20 [80%]) received treatment of lesions on the face. With the exception of 2 (10%) participants, all had received prior treatment of AKs, including cryosurgery (16/20 [80%]), imiquimod (5/20 [25%]), fluorouracil (2/20 [10%]), diclofenac (2/20 [10%]), and photodynamic therapy (2/20 [10%]); 8 (40%) participants had received more than 1 type of treatment.
The time course for the development and resolution of LSRs during both treatment cycles was similar. Local skin reactions were evident on day 2 in each cycle, peaked at 3 days after the application of the first dose, declined rapidly by the 15th day of the cycle, and returned to baseline by the end of each 4-week cycle (Figure 1). The mean (standard deviation [SD]) composite LSR score at 3 days after application of the first dose was higher in cycle 1 than in cycle 2 (9.1 [2.83] vs 5.0 [3.24])(Figure 1). The composite LSR score assessed over time based on the mean (SD) AUC was significantly lower in cycle 2 than in cycle 1 (40.5 [28.05] vs 83.6 [36.25])(P=.0002)(Table). Statistical differences in scores for individual reactions between the 2 cycles were not determined because of the risk for a spurious indication of significance from multiple comparisons in such a limited patient sample.
The percentage of participants who had a score greater than 1 for any of the 6 components of the LSR assessment was lower in cycle 2 than in cycle 1 at all of the assessed time points (Figure 2). In both cycles, the percentage of participants with an LSR score greater than 1 was highest 3 days after the application of the first dose in the cycle (day 4 or day 32, respectively). Erythema, flaking/scaling, and crusting were the most freq-uently observed reactions. At day 29, there were no participants with an LSR score greater than 1 in any of the 6 components. At day 29 and day 56, 94% (17/18) and 100% (18/18) of participants, respectively, had a score of 0 for all reactions.
|
| Figure 2. Percentage of participants with an individual local skin reaction score greater than 1 in cycle 1 (A) and cycle 2 (B)(N=17 for days 2, 30, 32, 36, and 43; N=18 for days 4, 8, 15, 29, and 56). |
The photographs in Figure 3, taken 7 days after the application of the first dose of ingenol mebutate gel 0.015% in each cycle of treatment of AK lesions on the face, show that there was less flaking/scaling and crusting in cycle 2 than in cycle 1. A review of participant photographs from the third treatment day of each cycle showed that the areas of erythema were the same in both cycles. The other 5 LSRs—flaking/scaling, crusting, swelling, vesiculation/pustulation, and erosion/ulceration—were observed in different areas of the treated field in the 2 cycles when applicable.
The few AEs that were reported were considered to be mild in severity. The AEs included application-site pain (n=5), application-site pruritus (n=3), and nasopharyngitis (n=1). No serious AEs were reported. After the first treatment cycle, 1 participant experienced hypopigmentation at the treatment site that persisted as faint hypopigmentation at the last study visit (day 56).
The lesion count in all participants at baseline ranged from 4 to 8, with a mean (SD) of 5.9 (1.55). Mean lesion count was substantially reduced at the end of cycle 1 (0.9 [1.39]) and cycle 2 (0.3 [0.57]). The change in lesion count from baseline to day 56 was greater than the change from baseline to day 29 (-5.7 [1.61] vs -5.0 [1.57])(P=.0137). Complete clearance at day 29 and day 56 was achieved in 55.6% (10/18) and 77.8% (14/18) of participants, respectively. The difference in the clearance rate between day 29 and day 56 did not reach statistical significance, most likely due to the small sample size.
