Actinic keratosis (AK) affects approximately 1 in 6 North Americans during their lifetime with over 1 million new cases reported annually. 1 , 2 Actinic keratoses are scaly lesions, precursorous to squamous cell carcinoma, that occur on areas of sun exposure, predominantly on the face, bald scalp, hands, and forearms.3,4 Current treatments include cryotherapy, excision, and topical therapies, or a combination of these therapeutic modalities.5–8 Topical 5-fluorouracil (5-FU) is a well-known nonsurgical treatment for AK and is available in 0.5%, 1% and 5% formulations.9–15 As a schemotherapeutic agent, fluorouracil essentially destroys AK by interfering with DNA and RNA synthesis by blocking the methylation reaction of deoxyuridylic acid to thymidylic acid. This causes thymine deficiency, particularly in cells mat are growing quickly and taking up fluorouracil at a more rapid rate, leading to unbalanced growth and death of these cells.16 Recently, imiquimod (5%) cream (AldaraTM) received Canadian and U.S. Food and Drug Administration (U.S. FDA) approval for the treatment of AK. The therapeutic effect of imiquimod against AK lesions is more physiologic in the way that it mimicks the body’s own natural immune response to cells expressing tumoral antigens.

Imiquimod stimulates the innate immunity by binding to toll-like receptor (TLR) 7 and inducing synthesis and release of cytokines such as interferon-α (INF-α), interleukin-12 (IL-12), and tumor necrosis factor-α (TNF-α).17–20 It also activates Langerhans cells to process the antigens and migrate to the regional lymph nodes. The resulting influx of macrophages and activated cytotoxic T lymphocytes leads to the apoptosis of AK cells.

Since imiquimod presents a new alternative topical therapy for AK, the purpose of this study was to determine the efficacy of imiquimod 5% cream and to compare it to 5-fluorouracil (5-FU) for treatment of AKs of the face and scalp, by performing a cumulative meta-analysis of studies for each drug.

Materials and Methods

Inclusion and Exclusion Criteria

The studies included in the meta-analysis had evaluated the efficacy of imiquimod (5%) and 5-fluorouracil (0.5%, 1%, and 5%) topical creams for treating actinic keratoses of the face and scalp. A MEDLINE search (1966 to October 2004) was performed for relevant studies. To be included in the meta-analysis, studies had to meet specific criteria and contain data that could be readily extracted. Treatment areas included only the face and/or scalp. Drug regimens were as follows: 0.5% 5-FU once daily, 2–4 weeks; 1% 5-, FU twice daily, 2–6 weeks; 5% 5-FU twice daily, 2–4 weeks; and imiquimod 2–3 times weekly, up to 16 weeks. The followup period for 5-FU ranged between 4 weeks and 11 and months post-treatment, while data for imiquimod were analyzed between 2 and 8 weeks post-treatment. The primary efficacy variable was the complete clearance of all AK lesions defined as the proportion of patients at followup with no clinically visible AKs in the treatment area.

Studies were excluded when the above inclusion criteria were not met, when treatment efficacy was not assessed for AKs of the face or scalp, when the topical fluorouracil was not a 0.5%, 1%, or 5% cream, when the treatment regimen was significantly different from the one approved by Health Canada or the U.S. FDA (i.e., pulse or cycling regimen, etc.), when the study report was not in English, or when the data were duplicated from a previous publication.

Statistics

The method designed by DerSimonian and Laird21 and later modified by Velanovich22 for single-group analysis was used to combine the efficacy data for each drug. This method uses a sample-size weighted average for each efficacy rate. A 95% confidence interval (CI) (mean ± 1.96 standard error [SE]) was calculated from the derived value for the SE.

Results

Ten studies were included in this meta-analysis.9,13,15,23–29 Seventeen studies were excluded since they did not meet the inclusion criteria for various reasons (Table I).30–46

TABLE I Studies excluded from meta-analysis for treatment of AKs of the face and/or scalp using imiquimod or 5-fluorouracil (5-FU)
Full size table

Meta-analysis of 5-Fluorouracil

The efficacy rates found for treating actinic keratoses of the face and/or anterior bald scalp with topical fluorouracil are summarized in Table II.9,13,15,23–25,27 The primary efficacy variable was the complete (100%) clearance rate. Complete clearance was defined as the proportion of patients with no clinically visible AK lesions in the treatment are at the post-treatment followup visit. Patient response was quite variable among these studies. Breza et al.[23] performed a split-face trial using 1% 5-FU alone and 1% 5-FU in combination with topical triamcinolone acetonide to determine its effectiveness in suppressing the unpleasant, painful inflammatory reaction caused by topical 5-FU when used as a therapy for AK. At one month post-treatment, 70% of the subjects had complete clearance of lesions.

TABLE II Efficacy of 5-fluorouracil (5-FU) in the treatment of actinic keratoses of the face and/or scalp
Full size table

In the study by Dillaha et al.9 Various concentrations of 5-FU were applied to the entire face, neck, arms, or hands to evaluate the optimal concentration needed for treating AK lesions and to obtain data regarding systemic absorption. Only data pertaining to AKs of the head were extracted for this meta-analysis. At the 11-month follow-up appointment, complete clearance was seen in 0% and 86.4% of the 1% 5-FU and 5% 5-FU groups, respectively.

In their split-face study, Lawrence et al.25 reported complete clearance of facial AKs in 6.7% of subjects one month following treatment with 5% 5-FU, while the same concentration of 5-FU resulted in complete clearance of lesions in 100% of subjects in the comparison study by Shuttleworth and Marks,27 six weeks following two weeks of treatment.

The 0.5% 5-FU phase III studies showed 57.8%13 and 47.5%15 of patients treated for AKs on the face and frontal scalp for four weeks had complete clearance at four weeks post-treatment followup for a combined efficacy rate of 52.7%.

The variability of efficacy rates for these studies demonstrates the need for meta-analysis. The meta-analytical average for the patient response rate following topical fluorouracil treatment was 52.2 ± 18% (mean ± 95% CI).

Meta-analysis of Imiquimod

Table III summarizes the efficacy rates for treating AKs with topical imiquimod cream.24,26,28,29 The primary efficacy variable was complete clearance rate with complete clearance defined as the proportion of subjects at the post-treatment visit with no clinically visible AK lesions in the treatment area. The number of lesions cleared included both the baseline lesions and any subclinical AK lesions that appeared during therapy.

TABLE III Efficacy of imiquimod 5% cream in the treatment of actinic keratoses of the face
Full size table

Stockfleth et al.28,29 reported two studies that met the inclusion criteria for this meta-analysis; although the initial dosing regimen in both studies was 3 times a week, it was then decreased to the approved 2-times-a-week regimen and was used continuously for no longer than 16 weeks, as per the labeling. In their initial case study,28 complete clearance of lesions was reported in each of the six patients (100%). Complete clearance in 84% of the 25 subjects treated with imiquimod cream was reported in the second study.29

The phase III trials reported by Lebwohl et al.26 resulted in 97/215 subjects having complete (100%) clearance of AK lesions at the two-month followup visit.

Szeimies et al.24 reported a phase III trial in which imiquimod was administered 3 times a week for 16 weeks on the face and scalp resulting in 57.1% of patients having complete clinical and histologic clearance.

For the studies using imiquimod to treat AKs, the average complete patient response rate (100% clearance of AKs of the face and scalp) was 70.8 ± 12.3% (mean ± 95% CI).

The results of this meta-analysis suggest that imiquimod has higher efficacy when compared with 5-fluorouracil for treatment of AKs of the face and scalp.

Discussion

Studies that determine the efficacy of a given treatment for a given disease can often report results that are not in agreement with other studies. A meta-analysis that combines the results of all studies that meet the inclusion criteria for a treatment strategy may be a helpful way of strengthening the evidence about that treatment.47–49

Although cryosurgery is the most common method of treating all types of actinic keratoses in all stages of development, it is generally limited to treating small areas and discrete lesions.5,6 Treating a large number of lesions, during any given clinical visit, may be painful or result in the development of adverse effects, e.g., blisters or pigmentary disturbance, thereby limiting the usefulness of this therapy. Thus, both imiquimod and 5-FU may have an advantage over cryotherapy especially when there is diffuse actinic damage rather than discrete AKs. Topical therapy is a useful alternative for treating this common skin condition and a number of preparations have been developed for treating AKs.50

Topical 5-FU has been a treatment for AK for almost 30 years, and it is therefore useful to compare the efficacy of 5-FU with the new nonsurgical therapy alternative, imiquimod. A number of studies have been conducted to determine the efficacy, safety, and optimal treatment regimens of 5-FU in the management of AK. 9,13,15,23,38,51–54 The mode of action of 5-FU is wellunderstood and relates to the depletion of thymidine resulting in a reduction of DNA synthesis in actively proliferating cells. This action leads to chemotoxic death of AK cells and can be associated with a local inflammatory reaction.16,55

In Canada imiquimod has been an approved treatment for external anogenital warts since 1999 and has recently received approval as treatment of AK. Imiquimod is an immune response modifier that stimulates local production of proinflammatory and immunostimulatory cytokines by innate immune cells, which in turn activates the adaptive immune response, eventually leading to apoptosis-mediated regression of the AK. This may be associated with a local inflammatory reaction which is generally less severe than that observed with 5-FU application.

This meta-analysis has several methodologic limitations. Imiquimod received U.S. and Canadian approval for treating AK in March and June 2004, respectively; therefore few studies have been published at the present time. In this study, we assessed the efficacy of topical fluorouracil and imiquimod creams for treatment of AKs on the face and scalp only. Some studies13,15,25–29 presented intent to treat results, whereby the entire initial randomized sample is included in the data presented, regardless of whether patients withdrew at any point during the trial. Others9,23,24 presented results as per protocol and excluded any data obtained from patients who withdrew from the study. The majority of studies included in the meta-analysis included any type of AK presented on the face and/or scalp, with the exception of Lebwohl et al.26 and Szeimies et al.24 who excluded hyperkeratotic and hypertrophic lesions. We may have missed some studies by not hand-searching all journals. However, a manual search was performed of all of the articles that we were able to retrieve. Studies with extractable data meeting the inclusion criteria for 0.5%, 1%, and 5% concentrations of 5-FU were pooled given the limited number of studies available for consideration for inclusion in the meta-analysis.

In summary, the results of this meta-analysis indicate that both 5-fluorouracil and imiquimod are effective methods for the treatment of actinic keratosis of the face and scalp and provide a useful alternative to cryotherapy. Our analysis suggests that imiquimod may have a higher efficacy than 5-FU, thus providing another option to dermatologists when treating AK lesions on the face and scalp.