Comparision of the efficacy of photodynamic therapy and imiquimod cream in patients with actinic keratosis: a systematic review and meta-analysis.

Introduction
Actinic keratosis (AK) is a premalignant epidermal skin lesion caused by chronic sun exposure. It typically appears as rough, scaly patches on sun-exposed areas like the face, scalp, ears and upper limbs. Early treatment is important as AKs have the potential to develop into squamous cell carcinoma1–3. Among the available therapeutic options, photodynamic therapy (PDT) and imiquimod (IMQ) cream have emerged as relatively common non-invasive therapeutic options.
PDT involves the application of a photosensitizer followed by exposure to a specific wavelength of light, which generates reactive oxygen species to selectively destroy diseased cells. IMQ, an immune response modifier, stimulates the production of cytokines, thereby enhancing the body’s anti-tumor immune response and promoting the clearance of abnormal keratinocytes4. Despite their widespread use, the comparative efficacy of PDT and IMQ for AK remains controversial, with individual studies reporting varying results5–10.
Meta-analysis provides a robust statistical approach to synthesize data from multiple studies, minimizing bias and increasing the statistical power to detect true treatment effects. In this study, we aim to conduct a comprehensive meta-analysis using Review Manager (RevMan) software to compare the efficacy and safety of PDT and IMQ in treating AK, which may offer valuable insights for clinical decision-making and patient care.

Methods

Search strategy
We systematically searched databases of Pubmed, Embase, Cochrane, Web of science, CNKI for studies comparing the lesion clearance rate (CR) and adverse events of PDT and IMQ in the treatment of AK from inception to 16 May 2025, and we also searched ClinicalTrials.gov and International Clinical Trials Registry Platform (ICTRP) for additional eligible studies. The following search terms were used: ’[(Keratosis, Actinic) OR (Actinic keratoses) OR (Keratoses, Actinic) OR (Actinic keratosis)] AND (Photochemotherapy OR Photochemotherapies OR Therapy, Photodynamic OR Photodynamic Therapies OR Therapies, Photodynamic OR Photodynamic Therapy OR Blue Light Photodynamic Therapy OR Red Light Photodynamic Therapy OR Red Light PDT OR Light PDT, Red OR PDT, Red Light) AND Imiquimod.

Study selection and inclusion criteria
We included all studies that fulfill the following criteria: (1) randomized or non-randomized clinical trials; (2) patients with a confirmed diagnosis of actinic keratosis (AK); (3) reported both a PDT monotherapy group and an IMQ monotherapy group; (4) outcome measures include CR and safety outcomes; (5) studies published in English or any other language. Studies were excluded for any of the following: (1) non-clinical research (e.g., in vitro, animal, or pharmacokinetic studies); (2) meta-analysis, review, case report, case series, guideline, letter or conference abstract; (3) retrospective study designs; (4) absence of a head-to-head comparison between PDT monotherapy and IMQ monotherapy; (5) unavailable outcome data or duplicate publications. We used the EndNote 21 program (Thompson Reuter, USA) to make sure that there were no duplicated papers, and we rechecked the selected articles by titles and abstracts screening. Finally, the eligible included studies were separately evaluated by two review authors based on the above-mentioned criteria.

Data extraction
The data extraction of the included studies was made separately by two reviewers. The reviewers developed the extraction sheet using Microsoft Excel 2016. In case of discrepancies, they discussed it first, and if no agreement was reached, the third reviewer was consulted. We abstracted data by two independent reviewers and double-checked by a third reviewer as follows: the first author, publication year, demographic data of enrolled patients, the total number of participants in the trial, interventions, the number of cases in each group and the number of lesions in each group. The main outcome was CR. For a given study, if a total of X AK lesions were treated and Y of these lesions achieved complete clinical clearance by the end of the trial, the lesion clearance rate for that study was calculated as Y/X. The secondary outcome was the types of adverse reactions that occurred during the study periods.

Data analysis
All statistical analyses were conducted by Review Manager (RevMan) 5.4 software. For continuous variables, the built-in analysis method of the software was employed to calculate the pooled effect size. The fixed-effect model was selected under “Fixed efficacy” to combine the results of included studies. Heterogeneity among studies was assessed using the I2 statistic and Cochran’s Q test. The results of the heterogeneity test showed that the I2 value was 0.43, indicating a moderate level of heterogeneity (I2 < 0.5), and the P-value of the Cochran’s Q test was 0.12, suggesting that the observed heterogeneity was not statistically significant. In addition, we selected a portion of the included literature for the analysis of treatment-emergent adverse events (TEAEs). We used the binary variable method to analyze the three common adverse reactions (pain, erythema, and ulceration/crusting) that occurred during PDT and IMQ treatment for AK.

Results

Search results
We found 1144 studies and 11 reports initially from PubMed (n = 60), Embase (n = 651), Cochrane (n = 20), CNKI (n = 267), and web of science (n = 146), ClinicalTrials.gov (n = 5) and ICTRP (n = 6). 990 studies and 1 reports remained after removing the duplicates. Screening of titles and abstracts led to the exclusion of 982 studies for the following reasons: meta-analysis, review, case report, case series, guideline, letter, conference abstract, and animal studies (n = 428); registry entries without reported outcomes (n = 5); retrospective cohort studies (n = 2); duplicate publications (n = 4); and studies not reporting on PDT or IMQ for actinic keratosis (AK) (n = 543). 8 studies proceeded to full-text assessment. After detailed review, 3 studies were excluded due to irrelevant outcomes, resulting in 5 eligible studies being included in the analysis.
Furthermore, 11 records were retrieved from clinical trial registries. 1 report could not be retrieved. Among the assessable reports, eight were excluded due to irrelevant topics, and two were excluded for not reporting outcomes of interest, and 1 report were evaluated for eligibility. Finally, 5 studies and 1 report were included. All the literature and trial report screening processes were strictly conducted in accordance with the PRISMA standards (Fig. 1).

Study characteristics
Among the six included studies, a total of 147 patients with 2,453 lesions were analyzed. One study was conducted in Italy, one in Germany, one in the US, one in Greece, one in Japan, and one in Denmark and Sweden. IMQ concentrations were 5% in five studies and 3.75% in one study. The photosensitizer precursors used in the PDT of the three studies were 5-aminolevulinic acid (ALA) and methyl aminolevulinate (MAL). In the six studies, the majority of subjects were male. With the exception of one study in which the proportion of male subjects was 41.7%, the proportion in the remaining studies exceeded 50%, reaching up to 100%. Among the six studies, only one did not report the age of the participants. In the other five studies, the average age of the participants was above 60 years. Specifically, the lowest average age was 63 years, while the highest was 80 years. A comprehensive overview of the features of the included studies is provided in Table 1.

Risk of bias assessment
The Cochrane Collaboration risk of bias tool was used. Two authors separately ensured the quality of all included studies. The six domains that were used to assess for risk of bias were sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting and other issues and they were rated as “high risk of bias”, “low risk of bias”, or “unclear risk of bias”11. Any disagreement between the assessing reviewers was resolved by discussion or after consulting another reviewer (Fig. 2).

Lesion clearence rate (CR)
The six studies included in this meta-analysis provided data on the mean, standard deviation (SD), and sample size for both PDT and IMQ groups. The data we extracted has undergone some processing12–15. The sample size for both groups was 135 each. The number of lesions in the PDT group was approximately 1,244, and in the IMQ group, approximately 1,209. The mean difference (MD) in the CR between PDT and IMQ, using a fixed-effects model, was 10.65 (95% CI: [7.11, 14.19]). In conclusion, this meta-analysis suggests that PDT has a statistically significant higher CR compared to IMQ in the treatment of AK. The test for overall effect showed a Z-value of 5.89 (P < 0.00001), indicating a statistically significant difference in favor of PDT. The heterogeneity statistic Chi² was 8.84 with df = 5 (P = 0.12) and I² = 43%, suggesting a moderate level of heterogeneity among the studies. Subgroup analysis revealed a notable difference based on the photosensitizer used: MAL-PDT studies showed no significant benefit over IMQ (MD -8.77, 95% CI: [-24.73, 7.18]). ALA-PDT studies demonstrated a clear and significant advantage for PDT (MD 11.66, 95% CI: [8.02, 15.29]). The test for subgroup differences was significant (P = 0.01), suggesting the type of PDT agent is a source of heterogeneity. In conclusion, PDT, particularly ALA-PDT, is significantly more effective than IMQ in CR for AK (Fig. 3).

Safety outcomes
Four studies were included for TEAEs. When conducting heterogeneity analysis, the data from the three included studies on pain showed high heterogeneity (Chi2 = 7.48, df = 1 [p = 0.006], I2 = 87%), while the data on erythema and ulceration/crusting showed low (Chi2 = 0.42, df = 2 [p = 0.81], I2 = 0%) and moderate (Chi2 = 3.42, df = 2 [p = 0.18], I2 = 42%) heterogeneity, respectively. Random-effects meta-analysis of pain (OR = 57.99, 95% CI: [0.23, 14358.18], Z = 1.44 [p = 0.15]) and fixed-effects meta-analysis of erythema (OR = 3.71, 95% CI: [0.78, 17.65], Z = 1.65 [p = 0.10]) showed no significant difference in the risk of these adverse reactions between the imiquimod treatment group and the photodynamic therapy group. However, fixed-effects meta-analysis of ulceration/crusting indicated a higher risk of this adverse reaction in the IMQ treatment group (OR = 0.15, 95% CI: [0.03, 0.70], Z = 2.40 [p = 0.02]). The forest plots of the analyses of three types of TEAEs are as follows (Figs. 4, 5 and 6).

Discussion
PDT has been extensively applied for the treatment of AK. The therapeutic mechanism involves the administration of photosensitizer precursors such as ALA or MAL, which intracellularly generate porphyrins, mainly protoporphyrin IX (PpIX). Upon exposure to light of specific wavelengths over approximately three hours, these photosensitizers are activated and generat reactive oxygen species. This process induces inflammation mediated by inflammatory factors, including prostaglandins and interleukins, ultimately leading to cell death via apoptosis and/or necrosis4.
IMQ stimulates a cell-mediated immune response by promoting the production of interferon-α (IFN-α), tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6) and interleukin-8 (IL-8), further enhanced through the recruitment of activated CD4 + T cells, CD8 + T cells and mast cells16–18. Mechanistically, IMQ activates Toll-like receptors 7 and 8 (TLR7/8) on antigen-presenting cells and induced apoptosis via cytochrome C19–23.
Our findings indicate that PDT, particularly ALA-PDT, is significantly more effective than IMQ in CR for AK. Subgroup analysis indicated that MAL-PDT showed no significant advantage over IMQ, whereas ALA-PDT demonstrated clear superiority. This discrepancy may stem from several factors: differences in photosensitizer properties; variations in study populations (MAL-PDT studies included immunosuppressed transplant recipients, while ALA-PDT studies involved immunocompetent individuals); smaller sample sizes in MAL-PDT trials, leading to limited statistical power; and heterogeneity in treatment protocols (including IMQ duration, PDT light sources, and follow-up periods). Consistent with the conclusion of Samar Salman et al. (2023)24, clinicians may improve the therapeutic effect by tailoring treatment based on lesion sensitivity to different concentrations of imiquimod cream and to PDT with various photosensitizers.
Variations in recording methodologies, diverse manifestations and potential data omission may exacerbate the heterogeneity. Consequently, the data utilized for the discussion of TEAEs was relatively restricted. This investigation focused on three TEAEs: pain, erythema and ulceration/crusting. Only the fixed-effect meta-analysis outcome of ulceration/crusting demonstrated a statistically significant effect, suggesting a higher susceptibility to adverse reactions in IMQ group. The moderate heterogeneity (I²= 43%) of CR indicates the presence of unresolved variations in study populations (e.g., age, skin type, AK severity, patient’s health condition), treatment protocols (e.g., different PDT light sources or IMQ application) and outcome assessment methods. This meta-analysis included a limited number of studies. Larger trials would enhance statistical power. As mentioned in the risk of bias assessment, inadequate blinding in some studies introduced a high risk of performance bias, which may compromise internal validity and limit generalizability of the results.

Conclusion
PDT, particularly ALA-PDT, may offer superior efficacy and enhanced safety compared to IMQ in the treatment of AK, potentially leading to improved health outcomes and quality of life for patients. However, further research is still required to more clearly establish the relative effectiveness of these two treatment modalities, as well as the impact of their combined use, in order to provide additional evidence-based guidance for the clinical management of AK.