Microneedle‐assisted incubation during aminolaevulinic acid photodynamic therapy of actinic keratoses: a randomized controlled evaluator‐blind trial

Dear Editor, Photodynamic therapy (PDT) is a common method of treating actinic keratoses (AKs) that compares favourably with other treatment methods.1,2 However, treatment is limited by the prolonged incubation times required for the medication to penetrate the stratum corneum.3 Microneedles (MNs) are micrometre‐scale needles that are capable of puncturing the stratum corneum with minimal pain.4,5 They have been utilized for various applications, including enhanced drug delivery.6 Few studies have evaluated MNs in conjunction with PDT for AKs.7 We sought to assess the utility of short solid MN arrays that penetrate to the epidermis in shortening the incubation time for aminolaevulinic acid (ALA) in PDT for AKs. To this end, we undertook a prospective, single‐blind, split‐face, rater‐blinded, randomized, sham‐controlled, noninferiority trial with 51 participants. MNs and sham pretreatments were randomized on the right and left foreheads and the sham‐treated sides were incubated with 5‐ALA for 60 min, which is the standard at our medical centre (Fig. 1a). MN‐pretreated sides were further randomized to 20‐, 40‐ or 60‐min 5‐ALA incubation. The primary outcome was complete response rate (CRR) of AKs at 1 month after treatment. Secondary outcomes were post‐MN treatment pain and post‐PDT treatment pain. The a priori power analysis revealed that at least 15 subjects would allow greater than 90% power to discern a change in one AK between the right and left forehead with α = 0·05 using a Wilcoxon signed‐rank evaluation of matched pairs. (a) Study flow schematic. (b) Forehead view of actinic keratoses before (I and III) and 1 month after (II and IV) photodynamic therapy (PDT) treatment with microneedle (MN) pretreatment (left forehead in all images) and with sham pretreatment (right forehead in all images). (c) Actinic keratosis complete response rate (CRR) results. Average CRR (SEM) for the 20‐, 40‐ and 60‐min treatment and control arms were 71·4% (5·8) and 68·3% (3·3); 81·1% (4·6) and 79·9% (4·5); and 72·1% (5·5) and 74·2% (6·9), respectively (n = 15, P = 0·51; n = 16, P = 0·79; n = 16, P = 0·72). (d) Pain scoring after MN application. Pain assessment using a 100‐mm visual analogue scale was performed after MN pretreatment and after PDT in all treatment groups [PDT 20, 20‐min aminolaevulinic acid (ALA) incubation; PDT 40, 40‐min ALA incubation; PDT 60, 60‐min ALA incubation]. Average pain scores after MN pretreatment (SEM) in the treatment and control arms were 13·1 mm (2·6) and 2·4 mm (0·5), respectively (N = 48, P < 0·01). Average pain scores after PDT light treatment (SEM) in the 20‐, 40‐ and 60‐min treatment and control arms were 15·7 mm (4·9) and 17·7 mm (4·9); 26·4 mm (7·5) and 24·7 mm (7·1); and 37·2 mm (6·6) and 26·0 mm (4·9), respectively (n = 15, P = 0·35; n = 16, P = 0·23; n = 17, P = 0·046). *P < 0·05, **P < 0·01. The MN device (Microchannel Skin System®, 3M Company, St. Paul, MN, U.S.A.) consisted of an array of MNs that were 690 μm in length. The sham treatment consisted of the applicator without actual MNs. Subjects received an application of 5‐ALA (Levulan® Kerastick®, Dusa Pharmaceuticals, Wilmington, MA, U.S.A.) to the entire face. At the end of the incubation period, subjects were exposed to blue light (Blu‐U®, Dusa Pharmaceuticals) with a wavelength of 417 ± 5 nm, for 8 min for a total fluence of 4·8 J cm−2. Forehead AKs were mapped prior to any intervention. The subjects were asked to rate the pain immediately after pretreatment with the MNs and sham device and also following blue‐light treatment by marking their level of pain on a 100‐mm visual analogue scale.8 All subjects returned 4 weeks after initial treatment for a follow‐up evaluation visit. Mapping of lesions was performed by two different evaluators. The primary outcome was CRR defined as the difference between the lesion count in the first and second visit normalized to the baseline count. Statistical analysis was performed using a paired Student's t‐test, Wilcoxon signed‐rank test and analysis of variance as appropriate. Data were tested for normality using the Shapiro–Wilk test. All analyses were conducted using RStudio® (RStudio Inc. Boston, MA, U.S.A.) and Excel® (Microsoft Corporation, Redmond, WA, U.S.A.). A P‐value < 0·05 was considered statistically significant. Overall, 94 patients were screened in order to enrol our target of 51 subjects. A total of 48 and 47 subjects completed all study end points for pain and AK counts, respectively. Our study population was predominantly male, older in age (average age 67·7 years, range 49–86) and 100% had Fitzpatrick skin type II (Table 1).9 Study participants and mean absolute lesion counts Total lesions analysed = 663. AK, actinic keratosis; MN, microneedle; PDT, photodynamic therapy. aP > 0·05. Study participants and mean absolute lesion counts Total lesions analysed = 663. AK, actinic keratosis; MN, microneedle; PDT, photodynamic therapy. aP > 0·05. Average CRR (SEM) for the 20‐, 40‐ and 60‐min MN‐incubation times vs. the corresponding sham MN treatment with 60‐min incubation were 71% (5·8) and 68% (3·3); 81% (4·6) and 80% (4·5); and 72% (5·5) and 74% (6·9), respectively (Table 1). There were no statistically significant differences in efficacy between the MN and sham‐device treatments (Fig. 1b, c). Average pain scores after MN pretreatment (SEM) in the treatment and sham arms were 13·1 mm (2·6) and 2·4 mm (0·5), respectively (N = 48, P < 0·01) (Fig. 1d). Average pain scores after PDT light treatment (SEM) in the 20‐, 40‐ and 60‐min treatment and control arms were 15·7 mm (4·9) and 17·7 mm (4·9) (P = 0·35); 26·4 mm (7·5) and 24·7 mm (7·1) (P = 0·23); and 37·2 (6·6) and 26·0 (4·9) mm (P = 0·046), respectively (Fig. 1d). No adverse events were reported. Our primary findings suggest that MN devices may reduce incubation periods required for PDT, while maintaining equivalent efficacy to the standard 1‐h incubation time used in our clinic. Other studies have investigated the use of MNs in combination with PDT for the treatment of AKs; however, longer MNs were used and resulted in dermal injury, which may increase patient discomfort and lead to bleeding and increased adverse events.7,10 This randomized controlled trial is distinct in that it used MNs that reach only the epidermis. Although we found a statistically significant difference between pain scores for MN application and the sham device, this difference was relatively minor. To our knowledge, this is the first assessment of the effects of short ALA incubations in PDT on human skin with AKs. Interestingly, efficacies were similar in the 20‐, 40‐ and 60‐min MN‐incubation arms. Funding sources: funding for this study was provided by a Cutting Edge Research Grant from the American Society of Dermatological Surgery (R.K.S.). Conflicts of interest: none declared.