Comparing the efficacy and safety profiles of 0.025% and 0.05% tretinoin creams in treating acanthosis nigricans: a randomized double-blinded study

Acanthosis nigricans (AN) is a skin disorder with hyperpigmented and velvety plaques without a standardized treatment regimen. We aimed to compare the efficacy and safety profile of 0.025% and 0.05% tretinoin creams in managing AN. An 8-week, randomized double-blinded study was conducted in adults with AN. Participants were instructed to apply tretinoin cream on their posterior neck. Narrowband reflectance spectrophotometry was used to measure skin improvement through melanin (M) and erythema (E) indices at each follow-up visits at weeks 2, 4, and 8. Improvements in Acanthosis nigricans scoring chart (ANSC), investigator- and patient-global evaluation (IGE and PGE) and adverse cutaneous irritations were also scored. Both the 0.025% and 0.05% tretinoin creams were efficacious in acanthosis nigricans treatment with 17.1 ± 8.0% improvement and 18.4 ± 9.8% improvement after 8 weeks treatment by reflectance spectrophotometry measurement, respectively. There were generally no significant differences in efficacy, improvements in ANSC, IGE, and PGE scores, and local cutaneous irritations between the two groups. The 0.025% and 0.05% tretinoin demonstrate similar efficacy and safety profiles in the management of AN. Both concentrations are well tolerated with mild degree of local cutaneous irritation.


Introduction
Acanthosis nigricans (AN) presents with dark pigmented and velvety plaques most commonly found at the neck, intertriginous, and flexural skin surfaces [1]. It is a cutaneous manifestation of underlying systemic diseases, particularly type 2 diabetes, obesity, internal malignancy, endocrinopathies, autoimmune disease, and drug reactions [1]. With the rise of the "diabesity" epidemic, the prevalence of AN ranges from 7 to 74% depending on population anthropometric factors, such as ethnicity, age, and body mass index [2,3]. The pathogenesis of AN involves complex cytokine signaling pathways that result in aberrant epidermal proliferation. Notably, mediators such as insulin, insulin-like growth factor 1 (IGF-1), or fibroblast growth factor 21 (FGF-21) binds to receptors on epidermal keratinocytes and fibroblasts, resulting in hyperkeratosis and papillomatosis seen histologically [4,5]. While typically asymptomatic, the hyperpigmented appearance may burden patients with cosmetic concerns thus affecting qualities of life. Therefore, AN is an important skin disorder that warrants effective clinical management.
Apart from addressing the underlying etiology, there is currently no standard treatment for AN [6,7]. Topical retinoids are considered first-line treatment option for cosmetic improvement due to its pleotropic effect on epidermal proliferation and keratinization [7]. Early reports have shown that 0.1% tretinoin, a first-generation naturally occurring retinoid, resulted in subjective clearance of AN lesions at neck and axilla within 2 weeks of application despite mild cutaneous irritations [8,9]. We have previously shown that 0.025% tretinoin cream is also another promising concentration, which demonstrated a significant 20% reduction of the melanin (M) index measured via narrowband reflectance spectrophotometry after 8 weeks [10,11]. Its efficacy is comparable to treatment with 0.1% adapalene gel and is superior to 10% urea cream [10,11]. The concentration was well tolerated, as patients initially experienced mild irritation and skin peeling during the first few weeks, which later then subsided [10,11]. Nevertheless, the efficacy of tretinoin cream in reducing hyperpigmentation could still be optimized while ensuring minimal side effects. The 0.05% concentration appears a good compromise between efficacy and safety profile, but its use has not yet been characterized. Therefore, we aimed to compare the efficacy and safety profiles of 0.05% and 0.025% tretinoin creams in treating acanthosis nigricans in adults.

Study design
This study was an 8-week, randomized, double-blinded, comparative study comparing the efficacy and safety of 0.025% and 0.05% tretinoin cream for treating neck hyperpigmentation in acanthosis nigricans. Participants were recruited from a single medical center in Thailand. They were allocated to each group via simple randomization by a third investigator who was not involved in data collection and analysis. Participants were given unlabeled jars of the topical agents; they were instructed to apply 1 g of tretinoin cream to each side of their posterior neck before bedtime once every other day during the first 2 weeks, and then once daily after that. Use of other topical agents on the neck was prohibited during the study duration. Participants were instructed that if cutaneous irritations were to occur, they could discontinue the medication for 2 days before reinitiating it once every other day then every day. During their first visit, participants' baseline characteristics, including age, body weight (kg), height (cm), and Fitzpatrick skin phototype were collected. After that, follow-up visits took place at weeks 2, 4, and 8 by the same blinded investigator. Participant's jar of cream was weighted at each visit to assess compliance to treatment.
The study has been approved by the Institutional Review Board and Ethics Committee of Srinakharinwirot University. Its protocol has been registered with the Thai Clinical Trials Registry (TCTR20201124001).

Subjects
Participants included were adult more than 18 years of age who were clinically diagnosed with acanthosis nigricans at the neck. Subjects with serious comorbidities (i.e. type II diabetes mellitus, coagulopathy, liver disease), immunocompromised hosts, vulnerable skin diseases, infection, lesion, tattoos, history of applying any topical agents within 4 weeks to the studied area, or ongoing use of oral retinoid were excluded. Sample size was estimated from the expected effect size of 0.3, and alpha level and power of 0.05 and 0.8, respectively. Assuming for 10% dropout rate, 20 participants were recruited in each group.

Outcome assessment
The primary outcome was treatment efficacy in reducing skin pigmentation, as represented by changes in melanin (M) and erythema (E) indices from narrowband reflectance spectrophotometry (Mexameter MX18; Courage and Khazaka Electronic, Cologne, Germany). The M and E indices are positively correlated with darker and redder skin color, respectively, and are expressed as values ranging from 0 to 999. Participants had three readings taken consecutively from each side of their neck per visit. To ensure the same areas were sampled at each follow-up, an individualized 10 × 20 cm transparent rectangle plastic sheet with two 2 × 2 cm holes on each side was draped over the subject, with its lower border at the C7 spinous level. Once the landmark was secured, measurements were recorded at the side holes. Additionally, skin pigmentation was also quantified at the back area as a reference using a similar protocol. Spectrophotometric measurement was done by a single trained investigator to ensure controlled probe pressure under fluorescent lighting.
The secondary outcomes were improvement in scores of the acanthosis nigricans scoring chart (ANSC), overall assessment using investigator-and patient-global evaluation scales (IGE and PGE), and adverse cutaneous reactions from treatment. The acanthosis nigricans scoring chart (ANSC) is a severity assessment tool developed inhouse to evaluate clinical improvement after treatment. The ANSC involves rating of skin color and texture, and is designed to improve subjectivity across physicians by incorporating distinct descriptions and reference images in the grading scale ( Fig. 1). Individual scores for skin color (ranging from 1 to 8) and texture (ranging from 1 to 6) are added to calculate the total ANSC score, which the latter demonstrates excellent intra-and interrater reliabilities (ICC > 0.85) and strongly correlates with narrowband reflectance spectrophotometry (r > 0.6) (unpublished data). ANSC scores were rated on both sides of the neck at each visit by the same trained investigator. The global evaluation scales were evaluated at each visit via IGE and PGE scales, with scores ranging from 0 to 6 (0 = clear, 1 = almost clear or > 90% improvement; 2 = marked improvement or > 75% improvement; 3 = moderate improvement or > 50% improvement; 4 = mild improvement or > 25% improvement; 5 = no change; 6 = worsening). Adverse reactions were explored using the cutaneous irritation grading scales, in which participants rated erythema, dryness, peeling/scaling, burning/ stingling each on a scale of 0 to 4, with 0 representing no adverse effect and 4 as marked severity. The incidence of participant-reported temporary treatment discontinuation (2 days according to our protocol) was also recorded, along with the adverse effects experienced.

Statistical analysis
Analyses were done using SPSS version 26 (IBM, New York). Measurements from each side of the neck was regarded as one data set, thus each subject contributed 2 data set (left and right sides). A repeated measures ANOVA was conducted to evaluate changes in skin pigmentation between groups from week 0, 2, 4, and to 8. Paired t test was also performed to compare the differences from before and after treatment. IGE, PGE, and cutaneous reactions were compared using Kruskal-Wallis test, while incidence of patientreported discontinuation was assessed using Chi-square test.
Statistical significance was considered when p value is less than 0.05.

Results
A total of 40 participants were recruited, and all 40 completed the study. The average age of participants was 36 years, and most were female (85%) and obese (85%). All patients had Fitzpatrick skin phototype of III to IV, and were either overweight or obese. Baseline parameters, including age, gender, BMI, weight throughout follow-up visits, skin phototype and compliance of treatment are shown in Table 1.
Both 0.025% and 0.05% tretinoin creams significantly improved M index of the neck from week 0 to 8 (p < 0.001). The 0.025% concentration showed a mean reduction of 97.5   (Tables 2 and 3). Efficacy of the two concentrations was similar as there were no significant differences between groups when compared head-to-head on each follow up. Clinical photographs are shown in Fig. 2. Disease severity assessed via ANSC showed significant improvement from week 0 to 8 in both groups (p < 0.001), which is concordant to reported results from spectrophotometry. Percent changes from pre-to post-treatment scores were 28.3 ± 12.2% and 34.3 ± 11.3% in the 0.025% and the 0.05% groups, respectively. The E index of the neck and the M and E indices of the back revealed no significant changes throughout the study period.
As for overall disease assessment, the 0.025% group demonstrated better IGE scores at week 2, while the 0.05% initially either had no improvement or worsening of symptoms. However, the latter showed more clearance of lesion by week 8 (Table 4). While IGE and PGE scores continued to improve as the study progressed, significant differences were generally not observed between the groups by the end of the study. Over 70% of patients in both groups rated more than 75% improvement of their skin lesion from before treatment. Complete clearance was scored by 5% and 15% in the 0.025% and 0.05% groups, respectively.
Both concentrations of tretinoin were well tolerated by participants, and the incidence of cutaneous irritation was not significant between groups (Table 5). While the 0.025% all demonstrate either slight or no irritation, a few participants in the 0.05% experienced mild to moderate erythema, dryness, and peeling/scaling. During follow-up period, five participants in the 0.025% temporarily discontinued the treatment (2 days according to protocol) due to burning/stingling, while eight in the 0.05% group for burning/stingling and itching. The incidence of medication discontinuation was also not significant between the two groups.

Discussion
While acanthosis nigricans is a relatively common skin disorder, there is yet to be a standardized regimen for its management [7]. Tretinoin has been known alleviate the hallmark hyperpigmented plaques, but associated adverse effects may hinder long-term compliance [7,8]. Thus, ensuring the optimal concentration is challenging due to the  counterbalance between efficacy and safety profile. Herein, we sought to compare whether use of 0.025% and 0.05% would differ in terms of clinical outcomes and adverse events. Reduction in spectrophotometric readings suggests both concentrations were efficacious in treatment, but treatment outcomes were not statistically different. This is consistent with findings from ANSC, IGE, and PGE scores. The observed improvement in hyperpigmentation in this study is comparable to that of our previous report, citing an average 20.1% reduction [10]. Concordantly, previous studies in photo-aging and uncomplicated acne vulgaris show different concentrations of tretinoin may not confer statistically different clinical outcomes, but higher concentrations may result in more adverse events [12,13]. Therefore, choosing an ideal tretinoin concentration should be based on patient's tolerability to side effects as well as compliance.  While participants in the 0.025% either had slight to no irritation, a few in the 0.05% reported mild to moderate erythema, dryness, and peeling/scaling at follow-up visits. The majority of participants in both groups temporarily discontinued the treatment per our protocol due to burning/ stingling, suggesting it is possibly the most relevant adverse effect. Nevertheless, the safety profile and incidence of temporary discontinuation was not statistically different between the two groups.
While we attempted to quantify treatment outcomes practically and quantitatively, reflectance spectrophotometry only assesses changes in intensity of pigmentation, while thickness and texture, which are also disease components of AN, were disregarded. For research purposes, skin biopsy serves the gold standard to meet this end, and histopathological outcomes would be needed to validate spectrophotometric results. On the other hand, in practical settings, biopsies are not done conveniently and efficacy assessment usually relies on one's clinical judgement. Therefore, we developed ANSC to bridge this gap, and would serve as a validated tool to affirm efficacy of treatment.
As for limitations, the study only shows the efficacy of 0.025% and 0.05% tretinoin up to 8 weeks, but recurrence rate and long-term compliance have not been investigated. While we used the participant's back area as control for skin color, future study may consider including a thirdarm control group as AN participants receiving a placebo cream. Our study also only included Fitzpatrick phototype Types III to V due to ethnic reasons, therefore results may only be generalizable to this population. Additionally, previous study on healthy subjects has shown that different formulation of the same tretinoin concentrations vary in irritancy, with gels being consistently more tolerable than creams [14]. Therefore, use of tretinoin gels should also be compared to the cream formulation head-on at different concentrations.

Conclusion
While both the 0.025% and 0.05% tretinoin creams are effective in reducing hyperpigmentation in AN, they are generally well tolerated with mild degree of local irritation. Both efficacy and safety profiles are not statistically different between the two concentrations.