Jet-Injection Assisted Photodynamic Therapy for Superficial and Nodular Basal Cell Carcinoma: A Pilot study

Abstract

Background:

Photodynamic therapy (PDT) with topical δ-Aminolevulinic acid (ALA) has efficacy in treating basal cell carcinoma (BCC) but is limited by incomplete penetration of ALA into the deeper dermis. This prospective open-label pilot trial investigated the safety and efficacy of photosensitizer jet injection for PDT (JI-PDT) for BCC treatment. It was performed with 15 patients (n=15) with histologically confirmed, untreated, low-risk nodular BCCs at a single institution.

Methods:

For the intervention, JI-PDT patients (n=11) received two sessions of jet-injected ALA with PDT separated by four to six weeks. To further understand treatment technique, another group of patients (n=4) received jet-injected ALA followed by tumor excision and fluorescence microscopy (JI-E). Treatment tolerability was assessed by local skin responses (LSR) score at five distinct time intervals. Fluorescence microscopy assessed PpIX penetration depth and biodistribution within the tumor. At the primary endpoint, tumor clearance was evaluated via visual inspection, dermoscopy and reflectance confocal microscopy (RCM). Post-injection and post-illumination pain levels, and patient satisfaction, were scored on a 0–10 scale.

Results:

Fifteen participants with mean age of 58.3, who were 15/15 White, non-Hispanic enrolled. The median composite LSR score immediately after JI-PDT was 5 (IQR=3) which decreased to 0.5 (IQR=1) at primary endpoint (p<0.01). Immunofluorescence of excised BCC tumors with jet-injected ALA showed photosensitizer penetration into papillary and reticular dermis. Of the 13 JI-PDT tumors, 11 had tumor clearance confirmed, 1 recurred, and 1 was lost to follow-up. 1/11 patients experienced a serious adverse event of cellulitis. 70% of patients had local scarring at 3 months. Patients reported an average pain level of 5.6 (SD = 2.3) during jet injection and 3.7 (SD =1.8) during light illumination.

Conclusions:

Jet injection of ALA for PDT treatment of nodular low-risk BCC is tolerable and feasible and may represent a novel modality to improve PDT.

Trial Registration:

ClinicalTrials.gov Identifier: NCT04552990

1. Introduction

Surgical treatments are often the first-line therapy for basal cell carcinomas (BCC) [1]; however, photodynamic therapy (PDT) is an alternative treatment, well-suited for elderly patients in whom surgery is contraindicated, in poorly healing areas or in immunosuppressed patients[2]. PDT uses topical photosensitizers, which are converted into protoporphyrin IX (PpIX) in cells, to selectively destroy tumor tissue when activated by light [1]. Despite its success, PDT has limitations, particularly for deeper or more aggressive tumors e.g., thick nodular BCCs (nBCC), and has mainly been relegated to use for the superficial BCC subtype (sBCC) [3]. One limiting factor is depth of penetration.

To improve uptake of topically applied drugs, active diffusion can be applied to disrupt the skin barrier and deliver ALA to deeper dermal layers [4]. Jet injection works via active diffusion, forcing a fine stream of liquid through a specially designed precision nozzle, where the high speed and pressure of the liquid enable it to penetrate the epidermis in a web-like distribution[5, 6]. The technique has primarily been used for the treatment of skin aging and palmoplantar hyperhidrosis[7]. Results from smaller clinical trials for treatment of condyloma acuminata, keloids, atrophic scars, and acne scars, have shown promising initial results[5, 8].

This study aims to address PDT’s limitations and investigate the potential of ALA jet injection for PDT to provide a safe, tolerable, effective, and less invasive treatment option for individuals with low-risk nodular basal cell carcinoma.

Imagine App, a novel telehealth dermatology app, was also explored in this study to longitudinally record subjects’ pain and pruritus over months of study participation [9–11]. Reflectance Confocal Microscopy (RCM), a longitudinal, noninvasive, in vivo imaging method, was employed to examine participants’ skin before and after treatments.

II. Methods

This prospective, open-label pilot study was conducted at a single-center tertiary academic hospital, Memorial Sloan Kettering Cancer Center in New York, New York under Institutional Review Board protocol #20–270. Investigators selected eligible patients who were 18 years or older, with at least one histologically verified, previously untreated, low-risk mixed superficial and nodular BCC(s) or nodular BCC(s) < 1 cm in diameter on scalp, extremities, or trunk (see Supplement for full inclusion and exclusion criteria). All participants provided signed, written, informed consent. The protocol was approved at the institution, and the trial was performed in accordance with the principles of the Good Clinical Practice guidelines and the Declaration of Helsinki.

Patients received jet-injected ALA to the tumor area followed by surgical excision of the tumor (JI-E) for ALA biodistribution analysis, or jet-injected ALA followed by PDT (JI-PDT), by a board-certified dermatologist investigator. ALA was applied in a novel fashion using the AirGent2.0^® device. The tumor and a 5-millimeter (mm) margin were injected with a grid of 80 microliters of 20% ALA at 5–8.5 mm between each injection (30–50% overlap). Specific pressure settings for the injection device were established based on results from our authors’ histological trial: Nozzle N2T2, pressure 30–65% (5ⁱ)[12]. The clinical endpoint of the jet injection was wheal formation. As per European Dermatology Forum guidelines on PDT, application of ALA was followed by a three-hour incubation period and PDT was carried out with a 633 nanometer (nm) red light at 37 J/cm² for 5 minutes and 51 seconds[13]. Each tumor and a surrounding five-mm margin received a grid-injection of 20% ALA solution, which, after three hours of incubation, was either excised (JI-E) or treated with red light illumination (JI-PDT). JI-PDT participants received a repeat JI-PDT session four to six weeks later. The first 4 participants to enroll had tumors excised (JI-E). The excised tumors were sent for PpIX biodistribution evaluation via fluorescence microscopy.

The primary objective was to evaluate the tolerability of this method via local skin responses (LSRs). A consistent team of investigators performed clinical evaluations on 1) the same day as treatment session (visits 2, and 4), 2) three days after each JI-PDT treatment session (visits 3 and 5), and 3) at the three-month follow-up (visit 6). Clinical evaluations were used to calculate LSR score, which encompassed ratings for erythema, edema, flaking, crusting, pustulation, and erosion/ ulceration; each were scored between 0 to 4. The combined score ranged from 0 to 24, with individual sub-category progression monitored.

Secondary objectives included evaluating pain levels, long-term cosmetic outcomes, ALA biodistribution in tumors, tumor clearance, and patient satisfaction. Pain during the ALA jet injection and pain during the PDT illumination sessions were evaluated directly after jet injection and during PDT illumination, respectively. Patients were asked, “How is your pain on a scale of 0 to 10, where 0 is no pain, and 10 is worst pain imaginable?”. Clinical assessments for long-term cosmetic outcomes(pigmentation, scarring and texture), were performed at baseline and after three months. Pigmentation and scarring were rated 0 (not present)- 3 (generalized) and texture was rated 0 (rough/rugged) to 3 (silk smooth). For the four JI-E participant excised tumor samples, ALA penetration and resultant PpIX biodistribution was evaluated using a fluorescence microscope.. Tumor clearance assessments were conducted at the three-month mark and was defined dichotomously as the absence of clinical signs of tumor via visual inspection and dermoscopy. If a definitive conclusion could not be made, non-invasive RCM imaging was used to assess tumor margins, If RCM or clinical inspection confirmed the presence of residual or recurrent tumor, , patients received a follow-up biopsy and conventional treatment in line with national guidelines. Patient satisfaction and willingness to repeat treatment were collected via survey at the primary endpoint.

Exploratory aims were evaluating the use of mobile applications for clinical trial remote “checkups”. The “Imagine - Skin Tracker” telehealth mobile application sent participants daily push notifications via their smartphone during each evaluation period to record and send their degree of pain and itching in the treatment area to a secure reporting interface.

Descriptive statistics (mean, standard deviation, median and interquartile range (IQR)), along with graphical methods were utilized to assess total LSR scores and LSR subdomain scores across the study evaluation time points. McNemar’s test was used to evaluate paired skin evaluation metrics between baseline and final evaluation time point. Random effects regression was used to assess the change in LSR components across the study time points. We performed an intent-to-treat analysis and analyzed the data up to the point participants were active. All analyses were performed using Stata v16.1 (Stata Corporation, College Station, TX).

III. Results

Primary objectives

200 patients were screened for the study, 30 of those were deemed eligible, and 15 patients enrolled. 15 patients with median age 59, 10 males, five females participated in the study; all were White and non-Hispanic. 11 patients with 13 tumors underwent JI-PDT, and four patients underwent JI-E. Patient demographics are summarized in Table 1. Sample size was determined appropriate for a pilot study.

Table 1.

Participant Demographics and Study Outcomes. The two interventions, as listed in the table, were Jet-injection photodynamic therapy (JI-PDT), and jet-injection followed by excision (JI-E). NED = No evidence of disease.

	Total (N-17)	JI-PDT (N=13)	Jl-Excision (N=4)
BCC Type
Superficial & Nodular	12 (70.6%)	9 (69.2%)	3 (75.0%)
Nodular	5 (29.4%)	4 (30.8%)	1 (25.0%)
BCC Location
Extremities	6 (35.3%)	6 (46.2%)	0 (0%)
Trunk	11 (64.7%)	7 (53.8%)	4 (100%)
Cure Rate, 1 – 3 months
Disease	1 (5.9%)	1 (7.7%)	0 (0%)
NED (Follow-up - 3 mo)	14 (82.4%)	10 (76.9%)	4 (100%)
NED (Follow-up −1 mo)	1 (5.9%)	1 (7.7%)	0 (0%)
Lost to follow-up	1 (5.9%)	1 (7.7%)	0 (0%)
Cure Rate, 6 −12 months (via chart review)
Disease	1 (5.9%)	1 (7.7%)	0 (0%)
NED	15 (88.2%)	11 (84.6%)	4 (100%)
Lost to follow-up	1 (5.9%)	1 (7.7%)	0 (0%)

All participants were white, non-Hispanic. Mean age 58.3, 10 males, 5 females.

Over 90% of patients (14/15) completed the study protocol; one patient did not attend the final three-month follow-up visit.

The median composite LSR score at the visit immediately following JI-PDT was 5 (IQR=3). Figure 1 contains an example of one patient’s lesion and LSR scores at pre-treatment screening (visit 1), 3 days after treatment 1 (Visit 3), 3 days after treatment 2 (Visit 5) and a clearance visit (Visit 6). By the final clearance visit, the median composite LSR score decreased to 0.5 (IQR=1). On average, participants reported a decrease in composite LSR of >1 point between each patient visit (p<0.001). Skin response categories with the highest median scores from 0–4 were erythema (median=2) and edema (median= 1). Median erythema at the visit immediately following PDT was 2, which decreased to 0.5 by the final visit. Median edema immediately following PDT was 1, which decreased to 0 by the final visit. Categories with the lowest scores were crusting (median=0), vesiculation / pustulation (median=0) and erosion / ulceration (median=0) (Figure 2).

Figure 1.

One patient’s lesion over the course of the trial. LSR scores at pre-treatment screening (visit 1), 3 days after treatment 1 (Visit 3), 3 days after treatment 2 (Visit 5) and a clearance visit (Visit 6).

Figure 2.

Local skin response grade by visit time point. Clinical evaluations were used to calculate LSR score, which includes ratings for erythema, edema, flaking, crusting, pustulation, and erosion/ulceration; each were scored between 0 to 4. The combined “Total” score ranges from 0 to 24. Erythema scale 0–4: 0 = not present 1= slightly pink <50% 3 = red, restricted to treatment area 4 = red extending outside treatment area. Flaking/Scaling scale 0–4: 0 = not present 1 = isolated scale, specific to lesions 2 = scale <50% 3 = scale >50% 4 = scaling extending outside treatment area. Crusting scale 0–4: 0 = not present 1 = isolated crusting 2 = crusting <50% 3 = crusting > 50% 4 = crusting extending outside treatment area. Vesiculation/Pustulation scale 0–4: 0 = not present 1 = vesicles only 2 = transudate or pustules, with or without vesicles <50% 3 = transudate or pustules, with or without vesicles >50% 4 = transudate or pustules, with or without vesicles extending outside treatment area. Erosion/Ulceration: 0 = not present 1 = lesion specific erosion 2 = erosion extending beyond individual lesions 3 = erosion >50% 4 = black eschar or ulceration.

One adverse event was reported among 1 patient in JI-PDT arm. Following the second JI-PDT treatment in the left lower extremity lesion for two lesions, the patient developed methicillin-sensitive staph aureus (MSSA) cellulitis requiring antibiotics.

Secondary Objectives

Mean pain during jet injection was 5.6/10 (SD 2.3). Average pain during illumination was 3.7/10 (SD 1.8). Average injection pain and illumination pain did not change significantly over the course of the two treatments (injection 1: 5.9; injection 2: 5.4, p=0.54; illumination 1: 2.9, illumination 2: 4.5, p=1.48).

Amongst the ten JI-PDT patients who returned for the three-month visit, 10 tumors were evaluated. For long-term cosmetic outcomes, , 70% experienced local scarring in the injection area (p = 0.01) (Result examples in Figure 4). 20% experienced hyperpigmentation (p = 0.16), 40% had hypopigmentation (p = 0.13) and 10% had rough / rugged skin texture between the first treatment visit and the 3-month visit (p = 0.59), approximately 2.5 months after their second treatment. Of note, 6/10 participants rated skin texture as improved from baseline; 3/10 rated it as same as their baseline (Supplement Figure 4).

Figure 4.

Cosmetic results 3 months after jet-injection photodynamic therapy (JI-PDT) for biopsy-proven BCC. Panel A contains photos of patient lesions pre-treatment, and panel B shows the lesion area 2 −3 months after their second PDT treatment. Patient 1. 2.6 months after treatments. Patient 2. 3.5 months after treatments. Patient 3. 2.4 months after treatments.

15 users generated 1,016 individual measurements of pain and pruritus over three months of the study via the telehealth mobile application. Median pain grades rated by JI-PDT and JI-E participants were equivalent (median = 0). Median (IQR) pruritus was minimal but higher in the JI-PDT (median = 1 (2)) versus the JI-E participants (median = 0 (1)) (Supplement Table 1).

For the JI-E biodistribution studies, using PpIX fluorescence as a surrogate measure for PpIX accumulation, immunofluorescence showed photosensitizer penetration into the papillary and reticular dermis in all four tumors. As expected, PpIX was slightly reduced in the deep (500–1200 um) depth compared to the superficial, however, it still fully penetrated (Figure 3; Supplement Table 2).

Figure 3.

Biofluorescence image. Biofluorescence quantified at superficial (a) and deeper (b) skin compartments. Fluorescence values after photobleaching were subtracted from incubation values. Fluorescence intensities were measured at both superficial and deep compartments.

Per protocol, all JI-PDT patients were followed-up at three months after their second treatment. One patient had declined a second treatment (for personal reasons) yet remained tumor-free at the three-month follow-up. One patient did not return for the three-month follow-up.

Five patients with 7 distinct treatment sites received RCM for tumor clearance evaluation. Five out of seven treatment sites showed complete resolution of BCC-specific features (See Supplement Table 3; example in Supplement Figure 1). The two treatment sites with persistent lesions were from a single patient with two lower extremity lesions (notably, the same patient who experienced cellulitis), demonstrating the persistence of tumor activity in the treatment area at the three-month follow-up (illustrated in Supplement Figure 2). This patient underwent a shave biopsy, was found to have BCC, superficial type on pathology and was directed to follow up with a Mohs surgeon for conclusive treatment. Thus, eight of nine patients who returned for the three-month follow-up were recurrence-free (Table 1 for demographics and outcomes). All patients except for the patient with cellulitis described above, including surgical excision patients, continued to remain recurrence-free for at least six months after second treatment. Mobile application images were successfully utilized to evaluate individual patients’ LSRs at two time points during the three-month study period. Images were readable, accessible, and clinically useful. The application was straightforward to use and useful for the study’s evaluation of patient pain and pruritus in real-time.

IV. Discussion

In this study, we piloted a novel method to treat low-risk, mixed sBCC and nBCC, or nBCC in 15 patients using jet-injected ALA combined with PDT to distribute ALA in the deeper dermis, with a focus on safety and efficacy. We found that the method was tolerable, with a median LSR immediately after JI-PDT of approximately five with significant decrease over time. Participants were also highly satisfied with the method. However, there was long-term scarring at the lesion injection site after JI-PDT. Per our clinical experience, the JI-PDT scarring appeared less severe and less atrophic than typical electrodesiccation and curettage BCC scars. During our study, there was a single failure among the JI-PDT participants, which involved cellulitis localized to the lower extremity, and persistence of the tumors. It is worth noting that this patient had a history of severe sun exposure, BCC and SCC history, and extensive BCCs on the lower legs. In addition, this patient received two injections at each treatment visit, and the lower leg inherently carries a higher risk of infection compared to other anatomical sites, even in the context of surgical procedures or biopsies. One patient was lost to follow-up, yielding a BCC recurrence rate of 8.3%.

Traditional, topical PDT has excellent cosmesis and a low adverse event rate, yet it is limited by depth of penetration of ALA, which has very low concentration below 1 mm[14, 15]. Hypo- and hyper-pigmentation is reported in 5% of patients who receive topical PDT; scarring occurs in less than 1%[16, 17]. Despite the excellent cosmesis, the recommendation for BCC treatment with topical PDT is limited to sBCC and thin (<2 mm) nBCCs, due to lower clearance rates in thick nodular or infiltrative BCCs[2]. By comparison, recurrence rates in topical PDT for BCC treatment after a year are 19 – 27%, while recurrence rates with standard surgical excision of BCCs are between 2 – 10%[18, 19]. However, surgical techniques with high clearance rates are not suitable for the elderly or patients with contraindications to surgery[20]. Intralesional PDT application has been attempted to allow deeper penetration of the medication and shorter incubation period, however, the use of needles in PDT may cause pain, and irregular distribution, in addition to increased perceived pain from the patient’s side[9].

Compared to surgical excision, our pilot JI-PDT resulted in slightly higher BCC recurrence rates (8.3% at 3 months versus average 4% at 1 year for surgical excision), but with better cosmesis. This ALA solution is also not yet specifically formulated for jet injection. Therefore, when deciding between surgical excision and our approach with jet injection PDT, technique optimization and considering patient tolerance for recurrence versus cosmesis will be important. Of note, the JI-PDT method has been reported once before in a variety of non-melanoma skin cancers[11]. They too, found a satisfactory therapeutic effect and low recurrence rate in BCC and Bowen’s Disease with JI-PDT[11]

This study establishes jet-injection settings and further histologic and biodistribution proof of the JI-PDT concept in humans. Our previous manuscript described porcine animal models of jet injection[12]. For a select group, e.g., nodular BCC lesions in the elderly, JI-PDT is tolerable and effective. However, focal scarring remains a risk for both surgery and JI-PDT.

Limitations for our study include a small sample size, short follow-up time, and its unblinded nature. Given the shorter follow-up period, future studies should analyze the long-term recurrence rates for this technique and compare it to the established recurrence rates of excision and topical PDT, as well as formulating the PDT photosensitizer for jet injection to optimize efficacy and safety. Utilizing novel technologies of mobile application tracking and RCM were strengths of this trial. The application worked to virtually capture pain and pruritus over time in this clinical trial. RCM helped detect recurrence after treatment of BCC; it is a useful method for monitoring BCC treatment efficacy, particularly for clinically ambiguous cases.

Given the substantial and ever-expanding prevalence of skin cancer, particularly among aging, sun-damaged, or immunosuppressed populations, there is a growing demand for more gentle and minimally invasive treatment options. The development of a new, minimally invasive topical treatment such as JI-PDT could potentially benefit many patients. Further research and clinical trials are warranted to validate and refine this innovative approach, potentially reshaping the landscape of BCC treatment modalities.