Patient accountability for the combined application studies of dosing 3 times per week.
Complete clearance by most intense erythema in the 5% imiquimod cream–treated subjects. Cochran-Armitage test for trend P < .001 (2 sided). There was a trend for complete clearance to increase as the intensity of erythema increased.
Resolution of actinic keratosis (AK) lesions after treatment with 5% imiquimod cream. A, Baseline count of 5 AK lesions in treatment area. B, At treatment week 8, 6 AK lesions, mild edema, erosion/ulceration, flaking/scaling/dryness, scabbing/crusting, and moderate erythema are seen. C, At 8 weeks after treatment, complete clearance (0 lesions) and no local skin reactions are seen.
Korman N, Moy R, Ling M, Matheson R, Smith S, McKane S, Lee JH. Dosing With 5% Imiquimod Cream 3 Times per Week for the Treatment of Actinic KeratosisResults of Two Phase 3, Randomized, Double-blind, Parallel-Group, Vehicle-Controlled Trials. Arch Dermatol. 2005;141(4):467-473. doi:10.1001/archderm.141.4.467
Copyright 2005 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2005
To evaluate the efficacy and safety of 5% imiquimod cream compared with vehicle in the treatment of actinic keratosis (AK).
Two phase 3 randomized, double-blind, parallel-group, vehicle-controlled studies.
Twenty-six ambulatory care offices, including dermatologists in private practice or research centers.
Four hundred ninety-two patients, 18 years and older, with 4 to 8 AK lesions in a 25-cm2 treatment area on the face or the balding scalp were randomized; an additional 162 patients underwent screening but were ineligible.
Patients applied 5% imiquimod (Aldara) or vehicle cream to the treatment area once daily, 3 times per week, for 16 weeks, followed by an 8-week posttreatment period.
Main Outcome Measurements
Complete clearance rate (proportion of patients at the 8-week posttreatment visit with no clinically visible AK lesions in the treatment area), partial clearance rate (proportion of patients at the 8-week posttreatment visit with a ≥75% reduction in the number of baseline AK lesions in the treatment area), and frequency and severity of adverse events and local skin reactions were measured.
Complete and partial clearance rates for imiquimod-treated patients (48.3% and 64.0%, respectively) were clinically and statistically significantly higher than for vehicle-treated patients (7.2% and 13.6%, respectively). The median percentage reduction of baseline lesions was 86.6% for the imiquimod-treated group and 14.3% for the vehicle-treated group.
The 5% imiquimod cream dosed 3 times weekly for 16 weeks is safe and effective for the treatment of AK.
Actinic keratoses (AKs) are epidermal lesions consisting of dysplastic keratinocytes that generally occur in fair-skinned individuals with long-term exposure to UV radiation.1,2 Although the exact pathogenesis of AK is unknown, long-term UV radiation is known to produce local and systemic immunosuppression, mutations in the p53 tumor suppressor gene, and DNA pyrimidine covalent dimers, each of which are believed to contribute to the development of AK.3,4 The most common current therapies for AK include cryosurgery, curettage with or without electrosurgery, and topical fluorouracil. Despite the widespread use of these therapies, most have limitations such as inconvenience, pain, and scarring. Furthermore, none of the current therapies target a critical component underlying the disease pathogenesis: the suppression of the immune response.
The only treatment for AK that works by enhancing the immune response against dysplastic cells is 5% imiquimod cream (Aldara; 3M Pharmaceuticals, St Paul, Minn). Imiquimod has been shown to induce the production of interferon α, tumor necrosis factor α, and interleukin (IL) 12 (IL-12), with a resulting cytokine cascade that may induce and/or support a cytotoxic T-lymphocyte immune response.5 Ongoing research has recently shown that imiquimod may also act through Toll-like receptor 7 to stimulate rapid synthesis and release of cytokines from monocytes, macrophages, and dendritic cells. Toll-like receptors are receptors that recognize special patterns on immune cell surfaces and are considered essential components of the human innate immune response.6
Since imiquimod works through an immunological mechanism, regimens with infrequent dosing for an extended duration have been evaluated to minimize potentially intolerable adverse effects. To date, 5 large randomized, double-blind, parallel-group, vehicle-controlled studies have been conducted to evaluate the efficacy and safety of topically applied 5% imiquimod cream vs vehicle cream for the treatment of AK lesions on the face and balding scalp. Two of these studies evaluated once-daily dosing 2 times per week for 16 weeks; the other 3 studies evaluated once-daily dosing 3 times per week for 16 weeks. The complete and partial clearance rates for both studies that evaluated dosing 2 times per week and for one of those that evaluated dosing 3 times per week have been reported.7,8 The efficacy end point in that previously reported study of dosing 3 times per week included histological confirmation and clinical lesion assessments. The results of the remaining 2 studies that evaluated the regimen of dosing 3 times per week are presented herein.
The study population consisted of otherwise healthy men and women 18 years or older with a clinical diagnosis of 4 to 8 AK lesions within a contiguous 25-cm2 treatment area on the face or the balding scalp as previously described.7
Two independent, randomized, double-blind, parallel-group, vehicle-controlled studies collectively enrolled patients from 26 US study centers (13 centers per study). The primary and secondary objectives of these studies were to evaluate the efficacy and safety of 5% imiquimod cream, respectively, compared with vehicle cream in the treatment of AK lesions on the face or the balding scalp when the cream was applied once daily 3 times per week for 16 weeks.
The studies consisted of prestudy (≤2 weeks), treatment (16 weeks), and posttreatment (8 weeks) periods. Patients underwent assessment during treatment weeks 1, 2, 4, 6, 8, 10, 12, 14, and 16 and posttreatment weeks 4 and 8.
At the treatment initiation visit, enrolled patients were assigned the next sequential patient study number and the corresponding study cream, which was randomized and labeled, on the basis of a computer-generated randomization schedule (stratified by study center). Patients were randomized to imiquimod or to vehicle cream in a 1:1 ratio. At the treatment initiation visit, the treatment area and the location of the 4 to 8 baseline AK lesions were recorded and mapped.
Patients were instructed to apply the study cream (imiquimod or vehicle) to the entire treatment area at the same time of day (just before normal sleeping hours), and the cream was to remain in place for approximately 8 hours. Rest periods from study cream were allowed at the discretion of the investigator, but did not alter the length of the 16-week treatment period. Patients who discontinued treatment were asked to return for an assessment of their AK lesions at the 8-week posttreatment visit.
All study procedures and informed consent documents received approval from the respective institutional review boards, and all enrolled patients signed informed consent forms.
Actinic keratosis lesions were counted at the 4-, 8-, and 16-week treatment visits and the 8-week posttreatment visit. No differentiation was made between the baseline and new AK lesions.
The primary variable was the complete clearance rate, defined as the proportion of patients at the 8-week posttreatment visit with no clinically visible AK lesions in the treatment area. The secondary efficacy variable was the partial clearance rate, defined as the proportion of patients at the 8-week posttreatment visit with at least a 75% reduction in the number of baseline AK lesions in the treatment area. An additional measurement of clearance was the median percentage of reduction of baseline AK lesions at the 8-week posttreatment visit.
Safety was monitored by reviewing concomitant medication use and assessing the incidence and severity of adverse events and local skin reactions. For the purposes of these studies, local skin reactions were differentiated and prospectively collected independently from adverse events (spontaneously reported). Local skin reactions in the treatment and surrounding areas were clinically categorized as erythema, edema, erosion/ulceration, scabbing/crusting, weeping/exudation, vesicles, and flaking/scaling/dryness. The intensity of an investigator-assessed local skin reaction was rated, with 0 indicating none; 1, mild; 2, moderate; and 3, severe. Reactions within the treatment area that were not assessed as local skin reactions were reported as adverse events. At the treatment initiation and 8-week posttreatment visits, investigators performed skin quality assessments of the treatment area. Characteristics assessed included skin surface (roughness/dryness/scaliness), hyperpigmentation, hypopigmentation, mottled or irregular pigmentation (hyperpigmentation and hypopigmentation), degree of scarring, and degree of atrophy. After visual, clinical, and tactile examinations of the treatment area, the investigator coded the intensity of each characteristic (with 0 indicating none; 1, mild; 2, moderate; and 3, severe).
The primary data set analyzed was the intent-to-treat data set, which consisted of combined data from the 2 studies and included all randomized patients. For each study, 94 patients per treatment group were required to have at least 90% power to detect a difference in complete clearance rates of 35% for the imiquimod group vs 14% for the vehicle group, with a type I error rate of .05. Assuming a 20% dropout rate, 120 patients per treatment group were required, giving a total of 240 patients per study. There were no interim analyses or stopping rules.
The Cochran-Mantel-Haenszel test was used to compare the treatment groups (imiquimod vs vehicle) with respect to complete clearance rate; the test was performed at the .05 level of significance. The difference in complete clearance rates (imiquimod minus vehicle) with an associated 95% confidence interval (CI) was calculated. The same analyses were performed on partial clearance as described for complete clearance. The Cochran-Armitage test for trend in proportions (2 sided) examined relationships between complete clearance and the most intense local skin reaction experienced by patients.
Adverse events were compared between treatment groups using the Fisher exact test. Wilcoxon rank sum tests were used to compare treatment groups with respect to local skin reaction assessments. For skin quality assessments, Wilcoxon signed rank tests were used to assess the significance of within-treatment changes from baseline, and Wilcoxon rank sum tests were used to assess the significance of between-treatment changes from baseline.
The combined studies enrolled 492 patients; 242 patients were randomized to 5% imiquimod cream and 250 patients were randomized to vehicle cream (Figure 1). Enrollment for both studies began in August 2001, and all study procedures were completed by August 2002.
The median age of the patients was 67 years. All were white; 61 (12.4%) were female; and 431 (87.6%) were male. The sex distribution observed in this study was consistent with previous epidemiological data.9 There were no distribution differences for sex, age, race, or skin type between treatment groups (Table 1).
Complete clearance occurred in 117 (48.3%) of 242 patients in the imiquimod group and 18 (7.2%) of 250 patients in the vehicle group (P<.001). The difference in complete clearance rates was 41.1% (95% CI, 34.1%-48.2%). Partial clearance occurred in 155 (64.0%) of imiquimod-treated patients and 34 (13.6%) of vehicle-treated patients (P<.001).
Differences for the complete and partial clearance rates in the imiquimod groups were observed between the 2 studies; complete clearance rates were 56.4% vs 40.8% and partial clearance rates were 71.8% vs 56.8%. No differences in baseline characteristics or demographics of the patients were noted between the 2 studies (data not shown).
For the imiquimod group, there was a statistically significant observed association between complete and partial clearance rates and the intensity of local skin reactions. Specifically, clearance rates tended to increase as the intensity of erythema increased (Figure 2).
The number of patients who had an increase in AK lesion count at any point in the treatment period was higher in the imiquimod group (103/242 [42.6%]) compared with the vehicle group (55/250 [22.0%]). The complete clearance rate was slightly higher for imiquimod-treated patients who had an increase in AK lesion count over baseline (56/103 [54.4%]) compared with imiquimod-treated patients who did not have an increase in lesion count (61/139 [43.9%]).
At the 8-week posttreatment visit, the median percentage of reduction in AK lesion count was 86.6% for the imiquimod group and 14.3% for the vehicle group. This means that half of the patients in the imiquimod group had at least an 86.6% reduction in the number of baseline AK lesions.
Adverse events occurred in both the imiquimod (176/242 [72.7%]) and vehicle (149/250 [59.6%]) treatment groups. At least 1 adverse event considered by the investigator as possibly or probably related to the study drug was reported by 104 (43.0%) of imiquimod-treated patients and 22 (8.8%) of vehicle-treated patients (P<.001). The most frequently reported adverse event was application site reaction, with “itching at target site” reported most often. Application site reactions were reported by 94 (38.8%) of imiquimod-treated patients and 18 (7.2%) of vehicle-treated patients (P<.001). The most frequently reported application site reactions that were considered possibly or probably related to study cream are listed by type in Table 2.
Other adverse events showing statistically significant differences between the imiquimod and vehicle groups were fever (7 patients [2.9%] vs 1 patient [0.4%]; P = .04) and postoperative pain (9 patients [3.7%] vs 2 patients [0.8%]; P = .03). The adverse events of postoperative pain reported by patients included references to dental procedures (5 patients), unspecified operation (2 subjects), and squamous cell carcinoma removal from the ear of a subject randomized to the treatment of the scalp, hammer toe surgery, arthroscopic knee surgery, and kidney surgery (1 patient each). None of the adverse events of postoperative pain were considered by the investigator to be related to study drug.
Forty-two serious adverse events were reported by 20 patients (12 imiquimod- and 8 vehicle-treated patients); none of the serious adverse events were considered by the investigator to be related to study drug. No deaths occurred in these studies.
Local skin reactions were common and occurred in both treatment groups. Both imiquimod- and vehicle-treated patients experienced severe local skin reactions; the rate of severe local skin reactions was higher in the imiquimod group. Table 3 includes the most intense investigator-assessed local skin reactions in the treatment area during the course of the study (excluding the treatment initiation visit), the percentage of patients with any reaction at any time after treatment initiation, the number of patients with no reaction at treatment initiation, and the number of patients with no reaction at the 8-week posttreatment visit. Overall, local skin reactions were well tolerated and generally resolved or diminished in intensity after cessation of treatment (Figure 3).
Patients could have discontinued (withdrawn) from the treatment period, the posttreatment period, or both. Patients who withdrew from the treatment period did not restart treatment, but were asked to return for posttreatment procedures. During the treatment period, 32 (13.2%) of 242 imiquimod-treated patients and 13 (5.2%) of 250 vehicle-treated patients discontinued from the study. Fifteen imiquimod-treated patients and 15 vehicle-treated patients discontinued during the posttreatment period. Table 4 summarizes the patients who discontinued during the treatment and posttreatment periods, stratified by reason for the discontinuation.
Rest periods were taken by 99 (40.9%) of 242 imiquimod-treated patients and 2 (0.8%) of 250 vehicle-treated patients. Of the patients taking rest periods, 86 (86.9%) of the 99 imiquimod-treated patients and 2 (100%) of the 2 vehicle-treated patients resumed treatment after their last rest period.
In general, the 8-week posttreatment assessments for each skin quality category indicated more patients with less intense scores than those with more intense scores after treatment with imiquimod.
Scores for scarring and skin surface categories showed statistically significant within-treatment shifts from baseline to 8 weeks after treatment for the imiquimod group. Statistical significance was also found when scores for skin surface were compared between imiquimod and vehicle groups. Each significant shift was due to imiquimod-treated patients having less intense scores at 8 weeks after treatment than at baseline. Of the 226 imiquimod-treated patients with both initiation and 8-week posttreatment skin surface assessments, 9 (4.0%) had an increase and 115 (50.9%) had a decrease in intensity at the 8-week posttreatment assessment compared with baseline. Because of inconsistencies in the investigators’ interpretation of hyperpigmentation, hypopigmentation, and mottled pigmentation, these data are not presented.
The results from these large clinical studies and previously published studies confirm that 5% imiquimod cream is a safe and effective treatment option for AK. Unlike other treatments for AK, imiquimod is an immune-based therapy. Imiquimod has been shown to stimulate the immune system by activating antigen-presenting cells such as monocytes/macrophages and dendritic cells to produce interferon and other cytokines and chemokines. Imiquimod appears to mediate these cellular effects through a Toll-like receptor, leading to the activation of transcription factor nuclear factor κB, which then culminates in the transcription of a number of cytokines including tumor necrosis factor α, IL-1, IL-6, IL-8, and IL-12.10- 12 These cytokines stimulate several other aspects of the innate immune response and are important for directing the adaptive immune response. The enhancement of the immune response is an ideal therapeutic approach for AK. The pathogenesis of AK involves suppression of the immune response against the abnormal cells that can be caused by long-term UV light exposure. Imiquimod treatment can reverse this immunosuppression, which potentially could reduce the rate of recurrences and the possibility of malignant transformation.
The complete and partial clearance rates for the imiquimod-treated patients were 48.3% and 64.0%, respectively. The clinical response with imiquimod dosing 3 times per week was higher than the response seen with dosing 2 times per weekg7; however, the magnitude of the increase was not as large as expected for a dosing regimen that was 50% higher. The modest difference in the clinical response between the 2 dosing regimens may be partly related to the low complete clearance seen in one of the studies evaluating dosing 3 times per week (40.8% vs 56.4%). No methodological explanation for this difference was apparent. The difference was most likely the result of random variation seen with clinical studies. However, the complete clearance rate from the third study that also evaluated dosing 3 times per week was 57.1%.8 These data suggest that the 56% clearance rate is more likely the true efficacy for this dosing regimen. If the actual efficacy rate for dosing 3 times per week is closer to the 56%, the incremental benefit of higher doses of imiquimod may be clinically justified in some patients.
The end point of complete clearance was a rigorous study end point that underestimated the clinical effectiveness of this treatment. For example, patients who experienced resolution of 7 of 8 lesions were still considered treatment failures because they did not achieve complete clearance. A more clinically useful measurement of the benefit is the median percentage of reduction in the number of lesions counted at baseline for each patient. The median percentage of reduction in baseline lesions was greater than 86% in the studies of dosing 3 times per week. This compares favorably with reported lesion clearance rates of pharmacological and nonpharmacological treatments for AK.13
More frequent dosing (3 vs 2 times per week) was associated with more local skin and application site reactions, more rest periods, and more subjects who discontinued treatment owing to local skin reactions (10/242 [4%] for imiquimod 3 times per week).7 However, the reactions were generally well tolerated and either resolved or diminished in intensity after cessation of treatment. This pharmacodynamic response is consistent with the immune mechanism of imiquimod. Futhermore, complete and partial clearance rates tended to increase as the intensity level of erythema increased. Therefore, based on the appearance of local skin reactions and their association with clearance, local skin reactions can be considered an extension of the pharmacological effects of imiquimod cream.
This pharmacological response contributed to one of the limitations of this study. Although the study design was double blind, the pharmacodynamic effect of imiquimod treatment may have biased the investigators’ assessments; however, when there is an evident pharmacodynamic response, bias is inherent in the study.
An additional benefit from imiquimod treatment appears to be improved skin quality compared with vehicle treatment. Although this phenomenon has been observed with nonspecific treatments that induce inflammation, the cytokines observed after imiquimod treatment have been reported to mediate wound healing and tissue remodeling.14- 16 Therefore, an improved cosmetic outcome may be another potential benefit of imiquimod treatment.
Because imiquimod works by enhancing the innate and adaptive immune responses, individual patient responses will vary depending on many intrinsic and extrinsic factors, such as the number and responsiveness of Langerhans cells and extent of solar damage. Although it is not known which factors will predict a patient’s response to imiquimod, it is reassuring to clinicians that many different dosing regimens appear to be safe and effective.
Dosing with 5% imiquimod cream 3 times per week for 16 weeks was a safe and effective regimen for the treatment of AK. The overall efficacy was higher for dosing 3 times per week than for 2 times per week; however, the rate of local skin reactions was also higher. These results, in addition to results from previously published studies, suggest that many different dosing regimens are effective and that dosing could be tailored to minimize drug exposure and adverse effects.
Correspondence: Neil Korman, MD, PhD, Department of Dermatology, University Hospitals of Cleveland, 11100 Euclid Ave, Cleveland, OH 44106 (firstname.lastname@example.org).
Accepted for Publication: April 27, 2004.
Group Information: The Imiquimod AK Investigators include the following: Neil Korman, MD, PhD (coordinating investigator); Ron Moy, MD (coordinating investigator); Debra Breneman, MD; Armand Cognetta, MD; Ponciano Cruz, MD; Frank Dunlap, MD; Ruth Gilboa, MD; Marc Green, MD; Karen Harkaway, MD; Christopher Huerter, MD; Michael Jarratt, MD; Steven Kempers, MD; Debra Liu, MD; Mark Ling, MD, PhD; Keith Loven, MD; Robert Matheson, MD; Jennie Muglia, MD; Keyvan Nouri, MD; Charles Phillips, MD; Tania Phillips, MD; Elyse Rafal, MD; Paul Ratner, MD; Toivo Rist, MD; Michael Scannon, MD; Stacy Smith, MD; and James Swinehart, MD.
Acknowledgment: We thank Nanda Gosala, MD, for medical monitoring; Mary Owens, MD, for editorial contributions; Terry L. Fox, MS, for statistical analyses support; and Amy Brown for manuscript preparation.
Financial Disclosure: Drs Moy, Ling, Matheson, and Smith received support from 3M Pharmaceuticals for performing clinical trials. Drs Smith and Ling have received funding (eg, research support and honoraria) from companies with competing products. Drs Korman, Moy, and Matheson received clinical trial support and honoraria from 3M Pharmaceuticals. This research was conducted with financial support from 3M Pharmaceuticals, St Paul, Minn.