Interventions for Treating Actinic Cheilitis: A Systematic Review

ABSTRACT

Background

Actinic cheilitis (AC) is a sun‐related oral potentially malignant disorder with no established optimal treatment. This systematic review aimed to evaluate the therapeutic efficacy of available treatments for AC.

Methods

Randomized Clinical Trials (RCTs) comparing AC treatments with placebo or other interventions were included. In February 2025, a literature search was conducted in PubMed, Cochrane Central, SciELO, ClinicalTrials.gov, Scopus, and Web of Science. Risk of bias was assessed using the RoB2 tool. A qualitative synthesis was performed for each clinical outcome.

Results

Eight RCTs, involving 230 participants, were included. Treatments assessed included daylight or conventional photodynamic therapy (PDT) with or without methyl aminolevulinate (MAL), imiquimod, and CO₂ laser, among others. The highest 3‐month clearance rates were observed with MAL plus indoor daylight PDT, CO₂ laser, and electrodessication. At 6 months, the CO₂ laser remained most effective. Limitations include five trials with a high risk of bias and the inability to perform meta‐analysis due to outcome measures heterogeneity and lack of standardized RCTs.

Conclusion

Due to limitations, the most effective AC treatment remains undetermined. Future research must improve methodology, assess standardized therapies, and include long‐term follow‐up to assess recurrences and malignant transformation.

Trial Registration

PROSPERO: CRD42024561899

1. Introduction

According to the World Health Organization, actinic cheilitis (AC) is defined as “a disorder that results from the sun damage and affects exposed areas of the lips, most commonly the lower vermilion border, with a variable presentation of atrophic and erosive areas and white plaques”. Today, AC is considered an oral potentially malignant disorder (OPMD) (Warnakulasuriya et al. 2021).

The most important risk factor for this lesion is chronic exposure to ultraviolet radiation (Shah et al. 2023), which can produce genetic alterations affecting the p53 gene expression (Carneiro et al. 2023; Muse and Crane 2025). Other risk factors are light skin phototype (Muse and Crane 2025; Carneiro et al. 2023), being male and over 60 years of age. It is unclear whether alcohol or tobacco consumption influences the development of AC (Rodríguez‐Blanco et al. 2018). Immunosuppressive treatment or chronic scarring are other factors that may increase the severity of AC and its malignancy (Carneiro et al. 2023; López‐Pintor et al. 2011).

The clinical presentation of AC can vary. Lip peeling and dryness may be the only signs in mild cases. However, they may be accompanied by white lesions, and in more severe cases, crusts and erosive patches are evident (Shah et al. 2023; Carneiro et al. 2023). One of the most representative clinical characteristics is the blurring of the vermillion border between the cutaneous and the mucosal lip. Regarding the location, 98.5% of the cases are found on the lower lip (Shah et al. 2023; Carneiro et al. 2023). Most patients are usually asymptomatic or may have a “sandpapery” feel to their lips. There are cases in which symptoms such as itching, burning, or stinging have also been reported (Carneiro et al. 2023; Muse and Crane 2025).

The most frequent histopathologic features are hyperplastic or atrophic epithelium, varying degrees of keratinization or parakeratinization, disordered maturation, presence of atypia, and increased mitotic activity (Warnakulasuriya et al. 2021). The lamina propria shows solar elastosis, which is a pathognomonic sign (Seoane et al. 2021). Different degrees of inflammatory infiltrate and vasodilatation can be observed. Dysplasia may or may not be present. The degree of dysplasia, together with the clinical aspect, influences the follow‐up visits, which should be performed every 3–6 months (Seoane et al. 2021; Carneiro et al. 2023).

Malignancy rates have been reported to vary from 10% to 30% (dos Santos et al. 2003; Lupu et al. 2017; Ortiz et al. 2023). However, a recent systematic review (Dancyger et al. 2018) stated, based on one study, that the malignancy rate for AC was 3.07%. According to some studies, at least 95% of oral squamous cell carcinomas (OSCCs) occurring on the lip have their origin in a previous AC. Therefore, many professionals advise treating AC (Dancyger et al. 2018; Vasilovici et al. 2022).

According to the available literature, there are several treatment options, which can be surgical, pharmacological, laser, or photodynamic therapies (Peralta‐Mamani et al. 2024; Arisi et al. 2022; Choi et al. 2015; de Oliveira Bezerra et al. 2019; de Godoy Peres et al. 2009; Fumagalli et al. 2024; Husein‐ElAhmed et al. 2019; Laws et al. 2000; Rossoe et al. 2011; Robinson 1989; Ayen‐Rodriguez et al. 2022; Yang et al. 2022; da Silva et al. 2024). To date, the most effective therapy and whether treatment prevents progression to OSCC is unknown (Varela‐Centelles et al. 2020), although surgical therapies have demonstrated their superiority in terms of efficacy and relapses (Trager et al. 2021; Bakirtzi et al. 2021). As new randomized clinical trials (RCTs) have been published, the aim of this systematic review was to summarize the available treatments described in the literature, to understand which treatment could be the most effective in the complete elimination of AC lesions. We also wanted to evaluate other secondary outcomes of these treatments, such as recurrences, pain, esthetic alterations, histologic changes, and adverse events.

2. Materials and Methods

This study was reported according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses checklist (Page et al. 2021). The protocol of this systematic review was registered in PROSPERO (CRD42024561899).

2.1. Focused Question

This study aimed to answer the following PICO (Population, Intervention, Comparison, Outcome) question: In patients with AC (Population), which treatment (Intervention) is more effective than placebo or another treatment (Comparison) in resolving AC (Outcome)?

2.2. Eligibility Criteria

2.2.1. Inclusion Criteria

Included studies had to be: (a) RCTs, (b) conducted in patients with a clinical and/or histological diagnosis of AC; (c) who had not received treatment for AC in the last year; (d) evaluating at least the activity or severity of AC lesions before and after treatment; (e) comparing the efficacy of the intervention with respect to a placebo or another intervention; (f) in patients of all ages; (g) without date restriction; and (h) written in any language.

2.2.2. Exclusion Criteria

The following studies were excluded: (a) those that included lesions other than AC, such as OSCC; (b) studies with fewer than 10 patients enrolled; and (c) systematic reviews, meta‐analyses, retrospective studies, quasi‐experimental studies, literature reviews, conference communications, case series, book chapters, case reports, and personal opinions or commentaries.

2.3. Sources of Information and Search Strategy

A literature search was conducted in the following databases: PubMed/MEDLINE, The Cochrane Library, Scopus, ClinicalTrials.gov, Web of Science, and SciELO. The last search was conducted on February 16, 2025. No filters were used to perform the search, and no date limit was set. The searches were conducted independently by two investigators (AF, RMLP). The search strategy used different keywords adapted to each database (Table S1).

2.4. Data Extraction

The studies found in each database were imported into the EndNote software (Clarivate Analytics). Duplicate references were identified and eliminated. Subsequently, the titles and abstracts of the articles were reviewed independently by two investigators (AF, RMLP). Studies outside the scope of the review or not designed as RCTs were excluded. In the first instance, only the title and abstract were assessed, and suitable studies were further evaluated through full‐text appraisal. The k‐agreement calculation was evaluated to rank the reviewers' agreement. A k‐agreement of 0.78 showed substantial agreement between the two reviewers (Higgins et al. 2024). A third investigator (JGS) participated in this phase to resolve discrepancies.

The following data were extracted from the included studies: (a) General characteristics of the selected studies: author, year of publication, article title, country of origin, number of AC patients, patients' age and sex; (b) Specific characteristics of the studies: inclusion and exclusion criteria, treatments applied to each study group, number of patients in each study group, outcomes, outcome measures, dropouts, and adverse effects; (c) Results of outcomes: activity or severity of AC, pain, esthetic results, adverse effects, histological changes, and recurrences at the different follow‐up periods.

2.5. Risk of Bias

The analysis of the risk of bias of the studies included was performed according to the Cochrane Risk of Bias in randomized interventional studies (RoB2) (Sterne et al. 2019). RoB2 was performed independently by two authors (JGS, VCAC), and disagreements were resolved with a third reviewer (RMLP).

3. Results

3.1. Selection of Studies

The search strategy (Figure 1) yielded 11,089 results, of which 9062 remained after removing duplicates. All titles and abstracts were screened, and 32 studies were evaluated for full‐text. Twenty‐four studies were discarded with reasons (Table S2).

FIGURE 1.

Flow diagram of the literature search, according to the preferred reporting Items for systematic reviews and meta‐analyses (PRISMA).

Eight RCTs published between 2000 and 2024 were included in our systematic review (Arisi et al. 2022; Choi et al. 2015; de Oliveira Bezerra et al. 2019; de Godoy Peres et al. 2009; da Silva et al. 2024; Fumagalli et al. 2024; Laws et al. 2000; Husein‐ElAhmed et al. 2019), one of them had a split‐mouth design (Laws et al. 2000). Of the 8 studies, 4 were conducted in Brazil (da Silva et al. 2024; de Godoy Peres et al. 2009; de Oliveira Bezerra et al. 2019; Fumagalli et al. 2024) and one in each of the following countries: USA (Laws et al. 2000), Korea (Choi et al. 2015), Spain (Husein‐ElAhmed et al. 2019), and Italy (Arisi et al. 2022).

Regarding the treatments applied, two studies evaluated the efficacy of photodynamic therapy (PDT) with methyl aminolevulinate (MAL) photosensitizer, either daylight photodynamic therapy (Idl‐PDT) or conventional PDT (c‐PDT) (Arisi et al. 2022; Choi et al. 2015), two evaluated CO₂ laser treatment (de Godoy Peres et al. 2009; Laws et al. 2000), one study evaluated the efficacy of diclofenac (DIC) (Husein‐ElAhmed et al. 2019) by comparing it to imiquimod (IMI) and ingenol mebutate (IM). Another study compared the combination of fludroxycortide and sunscreen against sunscreen alone (de Oliveira Bezerra et al. 2019). Another study compared fludroxycortide with IMI (Fumagalli et al. 2024). Lastly, the final study compared four treatments: IMI 5%, IMI 0.05%, IMI 0.05%‐NANO, and placebo (chitosan hydrogel) (da Silva et al. 2024).

3.2. Patients' Characteristics

Combining the samples from all the studies, a total of 230 patients were included in this systematic review. All the studies specified the age of the patients, which ranged from 26 to 86 years. With respect to sex, 158 patients were men and 72 were women. Not all studies specified age and sex by treatment group (Table 1).

TABLE 1.

Characteristics of the included studies and patients.

Authors and country of origin	Treatments and number of patients (n)	Patients' age (mean age) and gender
Fumagalli et al. (2024), Brazil	–IMI 5%: n = 13 –Fludroxycortide: n = 12	–IMI 5%: 66.69 (10.97); M = 12, F = 1 –Fludroxycortide: 70.17 (9.12); M = 10, F = 2
da Silva et al. (2024), Brazil	–IMI 5%: n = 13 –IMI 0.05%: n = 12 –IMI 0.05%‐NANO: n = 12 –Placebo: n = 11	Among all treatment groups: 63.6 (9.8); M = 31, F = 18
Arisi et al. (2022), Italy	– MAL + Idl‐PDT: n = 8 – MAL + c‐PDT: n = 8	Among all treatment groups: 64; M = 3, F = 13
Husein‐ElAhmed et al. (2019), Spain	–IMI 5%: n = 10 –DIC 3%: n = 10 –IM: n = 10	–IMI 5%: 73.2 –DIC 3%: 73.7 –IM: 77 –Among all treatment groups: M = 14, F = 16
de Oliveira Bezerra et al. (2019), Brazil	–Sunscreen: n = 8 –Fludroxycortide + sunscreen: n = 15	Among all treatment groups: 53.65; M = 19, F = 4
Choi et al. (2015), Korea	–ErYag + MAL + c‐PDT: n = 13 –MAL + c‐PDT: n = 17	ErYag + MAL + c‐PDT: 66.7 MAL + c‐PDT: 69.4 Among all treatment groups: M = 20, F = 13
de Godoy Peres et al. (2009), Brazil	–CO2 laser 250 mJ –CO2 laser 350 mJ – n = 40; sample was not specified by study group	Among all treatment groups: 42.47; M = 36, F = 4
Laws et al. (2000), USA	–CO2 laser (left side of the lip): n = 7 –Electrodessication (right side of the lip): n = 7	Among all treatment groups: 68; M = 13, F = 1

Abbreviations: Age given as mean age and if available the SD, standard deviation; c‐PDT, conventional PDT (red light); DIC, Diclofenac; F, Female; Idl‐PDT, Indoor daylight photodynamic therapy (white LED light); IM, Ingenol mebutate; IMI, Imiquimod; M, Male; MAL, Methyl aminolevulinate photosensitizer.

Regarding the diagnosis, five studies required prior histological confirmation of patients with AC for inclusion (da Silva et al. 2024; de Godoy Peres et al. 2009; Choi et al. 2015; Laws et al. 2000) and three studies did not specify the need for this inclusion criterion (Fumagalli et al. 2024; de Oliveira Bezerra et al. 2019; Husein‐ElAhmed et al. 2019).

3.3. Main Findings

3.3.1. Clinical Activity‐Severity

The clinical response results for each of the studies are summarized in Table 2. The included studies assessed the clinical activity of AC using different outcome measures, as detailed in Table 2. Some studies defined complete resolution as the disappearance of 100% of the lesions, while others considered partial lesion reduction as complete resolution. In some cases, only overall improvement was reported.

TABLE 2.

Results related to outcome variables: clinical activity and severity.

Study	Group	End of treatment	1–1.5 months after treatment	3 months after treatment	6 months after treatment	12 months after treatment	Outcome measures to evaluate the clinical activity‐severity
Fumagalli et al. (2024), Brazil	IMI 5%	NA	NA	NA	15.38% complete response	NA	–Complete clearance: absence of dryness, scaling, whitish plaques, erosions or ulcers –Partial clearance: reduction in the severity grade of AC –Unchanged –Worsening: increase in the severity of AC
	Fludroxycortide cream	NA	NA	NA	0% complete response	NA
da Silva et al. (2024), Brazil	IMI 0.05%‐NANO	66.67% Grade 2	75% Grade 2	NA	50% Grade 2	NA	–Grade 0: worsening –Grade 1: no improvement –Grade 2: improvement –Grade 3: side effects at the moment of evaluation
	IMI 0.05%	25% Grade 2	25% Grade 2	NA	25% Grade 2	NA
	IMI 5%	38.4% Grade 2	84.6% Grade 2	NA	54.5% Grade 2	NA
	Placebo (chitosan hydrogel)	27.7% Grade 2	36.4% Grade 2	NA	20% Grade 2	NA
Arisi et al. (2022), Italy	MAL + Idl‐PDT	NA	NA	100% complete response	NA	NA	Complete response: > 65% reduction of the lesions
	MAL + c‐PDT	NA	NA	87% complete response	NA	NA
Husein‐ElAhmed et al. (2019), Spain	IMI 5%	NA	NA	NA	50% complete response	NA	Complete response: 100% clearance of the lesions
	DIC 3%	NA	NA	NA	20% complete response	NA
	IM	NA	NA	NA	40% complete response	NA
de Oliveira Bezerra et al. (2019), Brazil	Sunscreen	NA	12.5% complete response	NA	NA	NA	Complete response: 100% clearance of the lesions
	Fludroxycortide + sunscreen	NA	33.33% complete response	NA	NA	NA
Choi et al. (2015), Korea	ErYag + MAL + c‐PDT	NA	NA	92% complete response	NA	85% complete response	Complete response: 100% clearance of the lesions
	MAL + c‐PDT	NA	NA	59% complete response	NA	50% complete response
de Godoy Peres et al. (2009), (Brazil)	CO2 laser single‐pass 250 mJ	NA	NA	NA	87.5% complete response	NA	Complete response: 100% clearance of the lesions Minimum clinical follow‐up period varied from 6 to 30 months.
	CO2 laser single‐pass 350 mJ	NA	NA	NA	87.5% complete response	NA
Laws et al. (2000), USA	CO2 laser single‐pass 360 mJ	NA	NA	100% improved	NA	NA	NA
	Electrodessication	NA	NA	100% improved	NA	NA

Abbreviations: AC, Actinic cheilitis; c‐PDT, conventional PDT; DIC, Diclofenac; Idl‐PDT, Indoor daylight photodynamic therapy; IM, Ingenol mebutate; IMI, Imiquimod; MAL, Methyl aminolevulinate photosensitizer; NA, Not available.

We analyzed the improvement in clinical activity after treatment, considering the variability in the duration of follow‐up among the studies. Two studies (de Oliveira Bezerra et al. 2019; da Silva et al. 2024) evaluated treatment response at 1–1.5 months. These two studies reported the highest improvements with IMI 5% (84.6%) and IMI 0.05%‐NANO (75%), while lower efficacy was observed with chitosan hydrogel (36.04%), fludroxycortide cream (33.3% complete resolution), and IMI 0.05% (25%).

At the 3‐month follow‐up, three studies (Arisi et al. 2022; Choi et al. 2015; Laws et al. 2000) assessed clinical activity. The most favorable results were achieved with MAL + ldl‐PDT (100% complete clinical response), CO₂ laser (100% clinical improvement), electrodessication (100% clinical improvement), ErYag + MAL + c‐PDT (92% complete response), MAL + c‐PDT (87% complete response), and another MAL + c‐PDT protocol (59% complete response).

Four studies (Fumagalli et al. 2024; da Silva et al. 2024; Husein‐ElAhmed et al. 2019; de Godoy Peres et al. 2009) reported outcomes at the 6‐month follow‐up. Results at this time point were generally less favorable than those at 3 months, with evidence of recurrence in several cases. The most effective treatments, ranked from highest to lowest response, were as follows: CO₂ laser single‐pass at 250 mJ and 350 mJ (both with 87.5% complete response), IMI 5% (50% complete response in Husein‐ElAhmed et al. 2019; 54.5% improvement in da Silva et al. 2024; and 15.38% improvement in Fumagalli et al. 2024), IMI 0.05%‐NANO (50% improvement), IM (40% complete response), IMI 0.05% (25% improvement), diclofenac 3% (20% complete response), chitosan hydrogel (20% improvement), and fludroxycortide cream (0%).

Only one study (Choi et al. 2015) conducted follow‐up at 12 months, reporting complete resolution in 85% of patients treated with ErYag + MAL + c‐PDT, and in 50% of those treated with MAL + c‐PDT.

3.3.2. Pain

Pain (Table 3) was assessed in four studies (Arisi et al. 2022; Choi et al. 2015; da Silva et al. 2024; de Godoy Peres et al. 2009). In the majority of the studies, the VAS (0–10) scale was employed to quantify pain intensity either during (Choi et al. 2015; da Silva et al. 2024) or after treatment (Arisi et al. 2022; da Silva et al. 2024; de Godoy Peres et al. 2009).

TABLE 3.

Results related to pain, adverse effects, aesthetic, and histological findings.

Study	Treatment	Pain	Adverse effects	Esthetic	Histological findings
Fumagalli et al. (2024), Brazil	IMI 5%	NA	14–21 days: erythema, scaling, edema, erosions, ulcers, crusts	NA	NA
	Fludroxycortide cream		Not observed
da Silva et al. (2024), Brazil	IMI 0.05%‐NANO	During treatment: VAS = 0 End of treatment: VAS = 0	14 days: erythema (16.66%), desquamation (8.33%), ulceration (8.33%) and burning (8.33%) End of the treatment: 8.3% had side effects: erythema (16.66%), desquamation (8.33%) and ulceration (8.33%) 1 month: 0% had side effects 6 months: 0% had side effects	NA	NA
	IMI 0.05%	During treatment: VAS = 0 End of treatment: VAS = 0	14 days: desquamation (16.66%) and burning (8.33%) End of the treatment: 0% had side effects 1 month: 0% had side effects 6 months: 0% had side effects
	IMI 5%	During treatment: 69.23% End of treatment: 53.84%	14 days: 92.3% with at least 1 side effect (redness and ulceration in 84.6%; also swelling, burning, desquamation) End of treatment: 46% had side effects: erythema (84.64%), desquamation (61.53%), ulceration (69.23%), swelling (53.84%), itching (23.07%) and burning (61.53%) 1 month: 0% had side effects 6 months: 9.1% had side effects
	Placebo (chitosan hydrogel)	During treatment: 9.09% End of treatment: VAS = 0	14 days: Erythema (18.18%), desquamation (27.27%), ulceration (9.09%), swelling (18.08%), itching (27.27%), burning (18.18%), pain (9.09%) End of treatment: 0% had side effects 1 month: 0% had side effects 6 months: 0% had side effects
Arisi et al. (2022), Italy	MAL + Idl‐PDT	After exposure: VAS = 0		Excellent*: 62.5%	NA
	MAL + c‐PDT	After exposure: VAS = 4	Swelling and pain	Excellent*: 25%
Husein‐ElAhmed et al. (2019), Spain	IMI 5%	NA	Not show the number of patients suffering each adverse effect. Appearance of erythema, scabs, swelling and burning was common for the 3 groups. Highest rate of side effects in the IM group	NA	NA
	DIC 3%
	IM
de Oliveira Bezerra et al. (2019), Brazil	Sunscreen	NA	No adverse effects	NA	NA
	Fludroxycortide + sunscreen		Burning sensation: 1 patient
Choi et al. (2015), Korea	ErYag + MAL + c‐PDT	During treatment: VAS = 5.6 (1.5)	Erythema: 100% of the patients Burning sensation: 100% of the patients Swelling: 39% of the patients Hemorrhagic crusting: 23% of the patients Blisters: 15% of the patients	Excellent**: 45.4%	3 months: 92% had a complete histological clearance 12 months: 85% had a complete histological clearance
	MAL + c‐PDT	During treatment: VAS = 5.4 (1.4)	Erythema: 100% Burning sensation: 100% Swelling: 35% Hemorrhagic crusting: 12% Blisters: 6%	Excellent**: 40%	3 months: 59% have a complete histological clearance 12 months: 50% have a complete histological clearance
de Godoy Peres et al. (2009), Brazil	CO2 laser One pass 250 mJ	Immediately post‐treatment: VAS = 5 1st day post‐treatment: VAS = 3 2nd day post‐treatment: VAS = 0	NA	NA	At day 168.35 (27.11) (range 109–240): No dysplasia: 53.8% Elastosis: 100%
	CO2 laser One pass 350 mJ	Immediately post‐treatment: VAS = 5 1st day post‐treatment: VAS = 3 2nd day post‐treatment: VAS = 0			At day 168.35 (27.11) (range 109–240): No dysplasia: 61.5% Elastosis: 100%
Laws et al. (2000), USA	CO2 laser One pass 360 mJ	18 h post‐treatment: 100% minimal or no pain	3–6 h post‐treatment: Burning sensation (85.71%)	3 months: one patient suffered an hypertrophic scar in the midline of the lip	3 months: 5/14 biopsies 2 patients: residual AC 2 patients: resolved AC 1 patient: OSCC
	Electrodessication

Note: * Excellent: No or mild redness or change in skin pigmentation; ** Excellent: only slight occurrence of redness or change in pigmentation.

Abbreviations: AC, Actinic cheilitis; c‐PDT, conventional photodynamic therapy (red light); DIC, Diclofenac; ErYag, Erbium‐doped yttrium‐aluminum‐garnet; Idl‐PDT, Indoor daylight photodynamic therapy (white LED light); IM, Ingenol Mebutate; IMI, Imiquimod; MAL, Methyl aminolaevulinate; NA, Not available; OSCC, Oral squamous cell carcinoma; PDT, Photodynamic therapy; VAS, visual analogue scale.

During the procedures, the highest VAS scores were reported by patients treated with ErYag + MAL + c‐PDT, as well as by those receiving MAL + c‐PDT alone (Choi et al. 2015). Immediately post‐treatment, the greatest pain intensity (VAS = 5) was observed in groups treated with single‐pass CO₂ laser protocols at 250 mJ or 350 mJ (de Godoy Peres et al. 2009), followed by patients treated with c‐PDT (VAS = 4) (Arisi et al. 2022). Following CO₂ laser treatment, pain resolved to VAS = 0 within 5 days (de Godoy Peres et al. 2009).

No pain was reported during or immediately after the procedure in patients treated with Idl‐PDT (Arisi et al. 2022) or in those receiving IMI 0.05% and IMI 0.05%‐NANO formulations (da Silva et al. 2024).

3.3.3. Esthetic Outcome

Esthetic outcomes (Table 3) were only evaluated in two studies (Arisi et al. 2022; Choi et al. 2015). Arisi et al. (2022) reported a higher rate of excellent esthetic results of 62.5% (defined as “no or mild redness or change in skin pigmentation”) in patients treated with Idl‐PDT compared to 25% of patients treated with c‐PDT.

Similarly, Choi et al. (2015) observed excellent esthetic results in 40% of patients (characterized by “only slight occurrence of redness or change in pigmentation”) in patients treated with ErYag + MAL + c‐PDT, or with MAL + c‐PDT.

3.3.4. Histological Findings

Post‐treatment biopsies (Table 3) were conducted in three studies (Choi et al. 2015; Laws et al. 2000; de Godoy Peres et al. 2009). de Godoy Peres et al. (2009) did not perform postoperative biopsies on all enrolled patients, as some declined to provide consent for the procedure. They performed a single biopsy at a mean of 5 months post‐treatment and observed elastosis in all patients treated with single‐pass CO₂ laser protocols. Furthermore, severe dysplasia was no longer evident in 53.8% and 61.5% of patients treated with the 250 and 350 mJ protocols, respectively, despite its presence prior to treatment (de Godoy Peres et al. 2009).

Choi et al. (2015) performed two post‐treatment biopsies and reported that, 3 months after treatment, nearly all patients receiving ErYag + MAL + c‐PDT achieved complete histological clearance. In contrast, 59% of patients treated with MAL + c‐PDT exhibited complete histological clearance. At the 12‐month follow‐up, both treatment groups showed similar relapse rates based on histological findings (Choi et al. 2015).

Laws et al. (2000) performed only 5 biopsies in 14 patients and found that 2 patients had resolved AC, 2 patients had residual AC, and one patient had OSCC.

3.3.5. Adverse Effects

All the studies except one (de Godoy Peres et al. 2009) recorded side effects derived from the treatments applied (Table 3). Erythema was the most common side effect when using c‐PDT, with or without ErYag (da Silva et al. 2024; Arisi et al. 2022) and topical treatments such as IMI 5%, DIC 3% and IM (da Silva et al. 2024; Husein‐ElAhmed et al. 2019; Fumagalli et al. 2024) (Table 3). Also, swelling and scabs or crusts were experienced by patients treated with this same group of therapies. A burning sensation was experienced by patients treated with c‐PDT (Choi et al. 2015), lasers such as ErYag (Choi et al. 2015) or CO₂ (Laws et al. 2000), electrodessication (Laws et al. 2000) and topical treatments like IMI (da Silva et al. 2024; Husein‐ElAhmed et al. 2019), DIC 3% and IM (Husein‐ElAhmed et al. 2019). Blisters appeared in patients treated with ErYag + MAL + c‐PDT and MAL + c‐PDT (Choi et al. 2015).

3.4. Risk of Bias

After assessing the risk of bias with the RoB2 (Sterne et al. 2019) (Table S3), we determined that two studies presented a low risk of bias (da Silva et al. 2024; Fumagalli et al. 2024), one had some concerns (Husein‐ElAhmed et al. 2019) while the other five studies had a high risk of bias (Arisi et al. 2022; Laws et al. 2000; Choi et al. 2015; de Godoy Peres et al. 2009; de Oliveira Bezerra et al. 2019).

3.5. Data Synthesis

It was not possible to perform a meta‐analysis due to the lack of uniformity of the outcomes collected, the outcome measures, the treatments applied, and the follow‐ups performed in the different studies.

4. Discussion

The present systematic review included all available RCTs (eight studies) evaluating interventions for the treatment of AC. We could not conclude which was the most effective treatment due to differences in recorded outcomes and outcome measures. However, a high level of remission was achieved after using MAL + Idl‐PDT, CO₂ laser, electrodessication, and MAL + ErYag + c‐PDT.

The most common outcome measured in the RCTs included was clinical clearance of the lesions. Most studies categorized it as a complete or incomplete response, but there were also studies that recorded a percentage of efficacy or improvement (da Silva et al. 2024; Laws et al. 2000; Fumagalli et al. 2024). Therefore, it was impossible to compare results across all studies. It should be noted that, although treatment with ErYag + MAL + c‐PDT obtained the second highest clearance rate at 3 months (Choi et al. 2015), and at 12 months, there were recurrences in 9% of the patients. Recurrences were also observed in patients treated with MAL + c‐PDT (Choi et al. 2015) and IMI 0.05%‐NANO (da Silva et al. 2024). These results are consistent with the idea that non‐surgical techniques lead to more recurrences than surgical treatments, as was highlighted in a previous systematic review (Trager et al. 2021). It is therefore important to evaluate the most effective treatment of AC over time, to define the outcome measures to evaluate the activity or severity of AC, and to establish how often patients should be reviewed and which parameters should be evaluated on each visit, since in many cases, relapses or OSCC may appear, even when treating patients with surgery.

In the present systematic review, 5 studies evaluated pain, and only 4 studies used VAS (Arisi et al. 2022; Choi et al. 2015; da Silva et al. 2024; de Godoy Peres et al. 2009). Overall, VAS ranged from 0 to 5.6, with higher scores recorded in patients treated with CO₂ laser and with c‐PDT, either combined with or without ErYag, than in those treated with other topical treatments. The lowest VAS score obtained after the treatment was observed with MAL + Idl‐PDT, as no pain was reported (Arisi et al. 2022), which matches the results observed in a previous review regarding this type of therapy (Yang et al. 2022). We believe that although pain is a temporary symptom during or after treatment, it should be collected in future studies. This will help notify patients of its appearance and provide anesthesia or prescribe analgesics, if necessary.

Aesthetic alterations after treatment should be included as outcome in future studies, since they may give rise to complaints after treatment if the result does not meet the patient's expectations. It seems that the best aesthetic outcomes were achieved with ldl‐PDT (Arisi et al. 2022; Choi et al. 2015), since it does not produce erythematous areas or pigmentation. Although no RCT evaluating vermilionectomy treatment has been included in this systematic review, it is important to highlight that one of the main esthetic sequelae of this treatment is lip retraction (Rossoe et al. 2011).

Histological evaluation after treatment is a controversial outcome to consider in RCTs, as it implies taking a new biopsy of a tissue in which lesions may have diminished or disappeared. To evaluate possible histological changes, it is important to have a previous biopsy to compare both images, which is recommended to assess a proper diagnosis (Seoane et al. 2021). It should be noted that, despite many cases, dysplasia disappeared after treatment; this was not the case with elastosis. For example, after the use of CO₂ laser (de Godoy Peres et al. 2009), the presence of elastosis was observed in all the patients after treatment. Therefore, the usefulness of biopsy after treatment is controversial. A possible alternative, as it is a less invasive test, may be exfoliative cytology that could evaluate cellular activity before and after treatment (da Silva et al. 2024).

When selecting a treatment for AC, a thorough risk–benefit assessment should be conducted and clearly communicated to patients. Most of the adverse effects described are minor and temporary, such as erythema, burning sensation, and swelling (Arisi et al. 2022; Choi et al. 2015; da Silva et al. 2024; Husein‐ElAhmed et al. 2019; Laws et al. 2000). Though no RCTs that included vermilionectomy as AC treatment could be included in this systematic review, it is a common treatment that has permanent adverse effects such as paresthesia, functional complications, and dry lips (Robinson 1989).

None of the studies assessed the possible malignization of AC lesions as an outcome after the use of the different treatments. Only Laws et al. (2000) reported that one patient with AC, previously treated for OSCC with microsurgery, developed another in the same location after receiving electrodesiccation and CO₂ laser treatment in their study. We believe it is important to establish possible malignancy as an outcome in future studies, as well as longer follow‐up periods.

This study has several limitations. Firstly, the limited number of high‐quality RCTs evaluating AC treatment. Furthermore, of the included studies, none evaluated well‐established and frequently used techniques such as vermilionectomy. On the other hand, as discussed above, the methodology and treatments differed among the studies, so it was not possible to perform a meta‐analysis. However, this systematic review shows, as Trager et al. (2021) discussed in their previous review (Trager et al. 2021), that future RCTs are needed to evaluate the different existing treatments for AC, and efforts should be made to determine the clinical and histologic criteria for AC diagnosis (Seoane et al. 2021).

Given the limitations of the present systematic review, it is challenging to conclude which treatment for AC is the most effective. We believe it is essential to define clinical and histologic diagnostic criteria for AC. Furthermore, it is necessary to establish a future clinical outcome set to enable reliable comparisons across trials evaluating AC treatments. On the other hand, future RCTs are necessary to evaluate the long‐term efficacy of new treatments as well as the usual ones.

Author Contributions

Alejandra Fernández Herranz: data curation, investigation, resources, writing – original draft. José González‐Serrano: methodology, software, data curation, investigation, writing – review and editing, resources. Vito Carlo Alberto Caponio: software, data curation, writing – review and editing, resources, investigation. Lorenzo de Arriba: writing – review and editing. Gonzalo Hernández: supervision, writing – review and editing. Rosa María López‐Pintor: conceptualization, methodology, data curation, formal analysis, investigation, writing – original draft.

Funding

This work was supported by the Instituto de Salud Carlos III through project PI22/00905 co‐funded by the European Union received by Rosa María López‐Pintor.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Tables S1–S3: odi70162‐sup‐0001‐TablesS1‐S3.pdf.

Data Availability Statement

The data that supports the findings of this study is available in the Supporting Information of this article.