Equine melanoma occurs in approximately 80% of gray horses (1, 2). The increased incidence of melanoma in these horses and their gray coat color is associated with a gene mutation (STX17) (3–5). Melanocytic tumors may arise in this population via increased proliferation of dermal melanocytes due to duplication of the STX17 gene (3, 4) or more broadly in horses as a consequence of disrupted melanin metabolism resulting in increased formation and activity of resident melanoblasts (6, 7). Some tumors exhibit rapid growth and malignant properties immediately, however, the majority exhibit slow growth over many years with little disease progression (7). These slow growing tumors have the potential to undergo a malignant transformation followed by a sudden rapid growth phase (7, 8). While most melanomas initially present as benign (9, 10), approximately two-thirds are thought to become malignant with the ability to become systemically metastatic (5, 11, 12). It is therefore recommended that all visible and accessible masses be addressed via some form of treatment (10).

Diagnosis is typically based on the physical appearance and location of the neoplasm (6, 7) and supplementary histological analysis can be conducted to assess the degree to which malignant invasive behavior can be expected (13). Equine melanocytic tumors commonly present as solitary discrete or coalescing subcutaneous masses ranging in size from smaller than 1 cm to over 20 cm in diameter (7). They typically arise on the ventral surface of the tail, the perineal skin, and the genitalia, but may also form near the ears and eyelids, on the head and neck, or in internal locations (7). Larger tumors may ulcerate and may eventually lead to physical obstruction of the anal sphincter and genitalia causing dyschezia, dysuria, and dystocia (7).

When possible, melanomas are treated by surgical or laser excision (6, 7, 13, 14) which does not appear to affect continued growth of distal tumors (3, 15). Recurrence may occur if wide margins cannot be achieved and multiple procedures may be necessary to adequately address the disease (7). Cryotherapy (13), chemotherapeutic injections (16, 17), and immunotherapies (18, 19) are among the options when surgery is contraindicated. In cryotherapy for melanoma, the tumor is rapidly frozen via a surface applicator which has been cooled with liquid nitrogen, with applicators internally cooled with liquid or gas (20), or by directly pouring liquid nitrogen on the tumor (21). Multiple rapid freeze - slow thaw cycles are conducted to achieve maximal effect (22). Tissue injury arises from the formation of ice crystals which disrupt the cell membrane (23, 24), disruption of microvasculature (24, 25), and induction of apoptosis due to stress induced during the freezing process (26, 27). While effective for small superficial tumors, it can be challenging to adequately cool large tumors and multiple treatments may be necessary to fully address the disease.

Approaches administering cisplatin and carboplatin achieve a therapeutic effect by cross-linking DNA (3). Injections of these chemotherapeutics are typically administered over a series of four visits at a concentration of 10 mg/mL at two week intervals to administer 1 mg/cm³ of the chemotherapeutic to each tumor (16). Similarly, biodegradable beads containing cisplatin (1.6 mg/bead) can implanted into tumors, typically following surgical or laser debulking (17). Both approaches have similar efficacy achieving 100% local control up to two years following treatment (16, 17). There is some risk of acute local reactions to these treatments (16) and because these platinum based agents are carcinogenic and mutagenic, strict safety protocols must be followed during all phases of treatment and follow up (3, 28).

Immunotherapies are an emerging class of therapies which aim to reverse tumor mediated immune suppression or enhance anti-tumor activation of the immune system (3). Interleukin-12 (IL-12) and Interleukin-18 (IL-18), are a class of cytokines associated with the suppression of angiogenesis (29), production of interferon-gamma (20), induction of apoptosis in cancer cells (30), and the activation of cytotoxic T cells and natural killer cells (20, 31). Intra-tumoral injection of DNA vectors or plasmids encoding for these genes resulted in a 20–29% reduction in tumor volume with equine cytokines (32, 33) and a 59% reduction for plasmids encoding for human cytokines (34). Autologous whole-cell vaccines attempt to elicit a similar immune response by surgically implanting tumor samples which have been frozen or irradiated after excision (35) or genetically modified to produce immunogenic proteins (36) and have shown promise in early feasibility studies. Similarly, DNA vaccines encoding for human tyrosinase (e.g., Oncept) which is overexpressed in equine melanoma have been demonstrated as safe to use (37). However, there exists only anecdotal evidence of their efficacy in equine patients (38) and this approach fails to extend progression free survival, disease-free interval, or median survival times in canines (39), limiting enthusiasm for this approach (40).

Given the challenges with existing approaches, better treatment options for larger tumors and those within proximity of vital structures would be welcome. This is particularly true for treatments which are synergistic with systemic immunotherapies and those that circumvent the need for general anesthesia. Focal ablation techniques such as electrochemotherapy (ECT) (41, 42) and irreversible electroporation (IRE) (43) have the potential to fill this clinical need if clinical and technical challenges with these approaches can be addressed. ECT combines the focal application of high voltage pulsed electric fields on the order of 100 μs–10 ms (Figure 1A) and chemotherapeutic agents. In this approach, the chemotherapeutic agent is injected into the tumor, and the electrical pulses enhance cellular uptake by temporarily increasing the permeability of the cell membrane to large molecules (44). Transient reduction in vascular perfusion due to the pulsed electric fields helps retain the chemotherapeutic in the treatment zone (45) further enhancing transport of the therapeutic agents into cells. This results in the induction of a number of cell death pathways and release of a number of damage-associated molecular patterns (46, 47) which may enhance recognition of the tumor by the immune system (48). In veterinary oncology, ECT is an emerging approach for the treatment of dermal malignancies including equine skin tumors (1) along with feline and canine sarcomas (49, 50). Unlike ECT, IRE forgoes the chemotherapeutic agent and employs the pulsed electric field for a monotherapeutic lethal effect (51). In this approach, additional pulses (typically 90-100x) 80–100 μs in duration (52) are administered to permanently disrupt cell membranes and induce cell death. In veterinary studies, IRE has been used to reduce tumor volumes in canine patients with intracranial gliomas (53) and soft tissue sarcoma (54). A noted benefit of IRE is that it can be used for larger infiltrative tumors, which are difficult to treat with existing veterinary surface applicators employed during ECT. IRE is also a promising tumor ablation approach in humans with clinical trials in kidney (55), pancreatic (56), and bone tumors (55). The consensus of these trials is prolonged survival times compared to standard of care.

Despite the successful clinical utilization of electroporation-based therapies (ECT and IRE), a common challenge remains; the high voltage pulses induce intense muscle contractions (57, 58). In human clinical trials, patients receive a combination of paralytics (59) and cardiac synchronization (59, 60) to address intraoperative movement due to the intense muscle stimulation. For horses, heavy patient sedation and general anesthesia are typical (49) as the patients react unfavorably to the intense muscle stimulation. General anesthesia is inherently risky in horses (61, 62) as it increases the overall risk of morbidity and mortality from fractures (63–65), cardiovascular failure (63–65), respiratory complications (airway obstruction, poor ventilation, hypoxia, etc.) (64), postanesthetic myopathy (63, 64, 66), and abdominal complications (peritonitis, colitis, etc.) (63, 64).

To address these challenges, recent work has focused on using ultrashort (<2 μs) alternating polarity (positive and negative) electric pulses delivered in a rapid burst (67). These bi-polar pulses have demonstrated the ability to reduce muscle contractions in porcine models (58) and in human clinical trials (21). They also reduce the potential for electrical arcing (68) when compared to ECT or IRE protocols. However, early investigations indicated that these waveforms were less efficient than IRE treatments when matched electrical doses were administered (69, 70) as shorter pulses resulted in smaller treatment volumes (71). Therefore, there has been a clinical tradeoff between maximum treatment volume, which favors longer pulse widths, and reducing muscle contraction, which favors short pulse widths. Integrated-time nanosecond pulse irreversible electroporation (INSPIRE) is designed to overcome treatment volume limitations. In INSPIRE, bi-polar 500 ns to 2,000 ns pulses (Figures 1B,C) are administered until a target integrated time, calculated as the sum of all pulse durations delivered, is achieved (72). This integrated time, or electrical dose, has been shown to be a critical factor affecting the efficacy of these treatments, with shorter pulses requiring a greater integrated time to achieve similar efficacy as longer pulses (72). By tuning treatment parameters, muscle contractions can be reduced (by delivering shorter pulses) while maintaining a clinically relevant treatment volume (by increasing integrated time).

Based on in vitro data (71, 73) showing similar efficacy of INSPIRE and IRE treatments at physiological temperatures and ex vivo data (74, 75) demonstrating the ability to produce clinically relevant treatment zones, it was hypothesized that INSPIRE may be a safe and effective treatment against cutaneous tumors between 1 and 5 cm in diameter. Equine melanoma was identified as the initial clinical target for this technique based on the clinical need, ease of applying the treatment, and the potential for long-term visual follow up for treatment confirmation.

In this study, a complete response rate of 80% was found for the INSPIRE protocol with the highest dose and longest pulse durations. This single treatment protocol resulted in an 84% reduction in average tumor volume following a single treatment. While follow-up times in this study were relatively short (60 ± 17 days), these results indicate that INSPIRE may be a useful substitute for surgery in certain locations where proximity of vital structures may make surgery challenging or contraindicated. Surgical resection results in local control in approximately 93% of cases (80), however, local and distant recurrence remains a challenge (6, 15) and multiple surgeries may be necessary to adequately address melanomas (7). Local control following INSPIRE can likely be improved by administering treatments over multiple visits. Since completion of this study, it has become our standard practice to schedule three appointments at one-month intervals to repeat treatments if residual tumor tissue is present or if there is concern about microscopic disease.

The ultrashort alternating polarity pulses administered in this study enabled treatments with mild sedation and local anesthetic while horses were standing in standard surgical stocks. In most treatments, there was no observable muscle stimulation, however, this was dependent on the patient and the location of the tumor. The administration of additional local anesthetic was sufficient to alleviate local muscle activation in most animals. An epidural approach for administration of analgesia and anesthetic may additionally alleviate this response, provided that the tumors are in an appropriate anatomical location. The risks of epidural administration, such as lack of response or recumbency (81), must be weighed relative to the benefits of more widespread analgesia and desensitization as indicated for perianal and tailhead tumor treatment. In this study, we did not administer an epidural because intravenous sedation and injections of local anesthesia were more efficient and practical for an outpatient procedure and found to be sufficient to facilitate application of this therapy to the focal areas selected during the clinical trial. For other regions of the body, intravenous sedation and local or regional anesthesia are the only methods for facilitating application of INSPIRE in standing horses. Mitigating intense muscle stimulation has historically been a challenge in administering traditional electrochemotherapy in equine patients (41, 82). In our experience, traditional electrochemotherapy treatments are unlikely to be tolerated by standing horses due to the induction of intense muscle contractions and administration required complete sedation and general anesthesia to avoid issues with compliance. This muscle stimulation can be highly variable, ranging between local twitching and intense movement of adjacent limbs. Muscle stimulation associated with traditional ECT is attributed to the relatively long duration (100 μs or greater) monopolar electrical pulses administered where the electric field strength required to stimulate muscle tissue is approximately 1 V/cm (83). The need for high voltages (1,000–3,000 V) to achieve therapeutic effect results in a significant volume of normal tissue being elevated above the threshold associated with muscle stimulation. In contrast the electric field strength required to stimulate muscle stimulation with the 2,000 and 500 ns pulses used in this study is approximately 50x to 100x greater (50 to 100 V/cm (83)) resulting in a significantly smaller volume of tissue exposed to field strengths high enough to induce muscle stimulation.

While not quantitatively measured in this study, 500 ns treatments were characterized by less muscle stimulation than 2,000 ns treatments. This study highlights two potential approaches for reducing muscle stimulation: reducing treatment voltage and reducing pulse width. A significant difference was not found between voltage-matched treatments with 500 ns pulses and 2,000 ns pulses. However, a significant difference was found between treatments administered at 1,000 V and those administered at 2,000 V. If patient discomfort is observed, then an appropriate clinical strategy may be to first reduce the pulse width from 2,000 ns to 500 ns before attempting to reduce the treatment voltage. A larger study is warranted to study the effect of pulse width, voltage, and dose in more granular detail. As heating and muscle stimulation are of interest, these could be quantified via the use of thermal measurements (84) or temperature control (73) and the use of accelerometers to measure movement (43, 67) to determine if there is an optimal strategy which minimizes discomfort while achieving maximal therapeutic effect. As treatments using 2,000 V, 2,000 ns pulses and a dose of 0.02 s dramatically reduced overall tumor volumes with a single treatment, these parameters should be considered as a baseline for treating tumors moving forward.

Applicator selection is a critical factor in the success of electroporation-based treatments. The region of tissue affected using surface applicators and needle pairs can be difficult to conceptualize due to the exponential decay in electric field strength around these applicators (85). The intention of the coaxial applicator used in this study (Figure 3) was to simplify the geometry of the treatment zone. The ring component of the applicator left a visible impression on the skin following treatment which assisted in overlapping treatment areas on larger tumors (>2 cm) when multiple treatments were necessary to achieve total tumor coverage. Cells in these overlapping areas received a larger cumulative dose which has been shown to increase treatment efficacy (71, 86) and no deleterious effects were observed in regions where treatments overlapped. Tumors larger than 5 cm can be treated with this overlapping approach; In this scenario, it is recommended to address the tumor progressively over a series of 3 to 4 visits at one-month intervals. Three to five overlapping treatments should be administered each session beginning at one side of the tumor and progressively covering the entire tumor over the series of visits.

While this applicator appears to be highly effective at treating cutaneous tumors, it should be noted that it would be challenging to use for internal or deep-seated tumors. For internal tumors, two or more needle applicators can be utilized to administer INSPIRE treatments, however, additional studies are necessary to determine appropriate treatment parameters for this approach. Similarly, a single needle and grounding pad approach (74, 87, 88), akin to the approach used in radiofrequency ablation, can be used. It should be noted that this approach may stimulate muscle tissue between the applicator and grounding pad and may only be feasible under general anesthesia.

While robust systemic responses remain elusive, locally administered chemotherapeutic (16) and immunotherapeutic (34) approaches achieve partial responses in approximately 90 and 40% of cases, respectively (80). Combinatorial approaches which include INSPIRE treatment may improve overall response rates. This is of particular interest in combination with immunotherapies where antigen release by INSPIRE may enhance tumor recognition leading to a systemic immune response. An abscopal effect, where untreated tumors also shrink in size following treatment of a primary tumor, has previously been reported in both ECT (89) and IRE (90–92) studies and pulsed electric fields have been shown to stimulate adaptive immune response in vitro (93) and in small animal (94, 95) models of disease. This study observed a reduction in tumor volume in untreated tumors in 11 of 13 control tumors tracked. A more thorough experimental design and analysis are necessary before drawing conclusions regarding the mechanisms behind this observation or potential clinical implications as pulse parameters were not isolated within or across patients. However, we hypothesize that INSPIRE treatment results in the release of immunogenic tumor antigens leading to an adaptive response. INSPIRE treatments administered a higher electrical dose (0.02 s) than traditional IRE treatments (0.08–0.01 s) which may have resulted in an increase in the quantity of tumor antigens released. Additionally, the INSPIRE treatments were administered at a relatively slow rate (50 μs/s) compared to traditional cardiac synchronized IRE treatments (80–100 μs/s) which may have resulted in smaller temperature changes than would normally be expected for IRE treatments and may have prevented thermal denaturation of any released antigens. Combined, these effects may be responsible for the observed abscopal response.

As an abscopal response was an unanticipated outcome, the requisite immune assays to elucidate the mechanism of this abscopal control of distant lesions were not included in the protocol. Patients with larger tumors received multiple treatments to cover those tumors; some patients had tumors treated with 500 ns and 2,000 ns pulses. Within the patient population, there was no observable evidence of a systemic response or global reversal of the disease state. The question of whether the INSPIRE anti-tumor immune response is regionally constrained is unanswered. It should also be noted that two of the eleven control tumors increased in size. While outside the scope of this study, the results indicate that improving and tuning an adaptive immune response following INSPIRE is a promising avenue for treating multifocal disease in patients. In future studies, it will be important to investigate these immunological effects in animals receiving only a single treatment. INSPIRE administered with active temperature control (71, 73–75, 94) could be used to evaluate if thermal effects have an impact on immunological responses. Similarly, untreated local and distant control tumors could be biopsied to investigate the presence and quantity of activated T-cells at intervals before and after treatments. Similarly, it may be interesting to investigate if INSPIRE can improve the efficacy of checkpoint inhibitors against equine melanoma (96–98).

This study has several important limitations. This study utilized client-owned animals, some of whom had undergone other prior treatments for melanoma. Scheduling and coordination between the clinic and owners resulted in patient-to-patient variation in the time between successive follow up appointments, and some qualitative assessments utilized client obtained photographs. Disease was confirmed via biopsy of a nearby untreated tumor. While unlikely, it is possible that the tumors included in this study were of a different pathology (e.g., nodular sarcoid, eosinophilic granulomas), future studies should consider conducting needle aspirates of all treated tumors to confirm disease pathologies prior to treatment. A small number of animals were evaluated in this study and two of the eight enrolled were lost to follow up following their first visit. Tumor volumes were tracked across multiple visits, but long term (years) responses have yet to be quantified. Finally, only a small set of possible treatment parameters were investigated (pulse width, dosage, and voltage).

Further investigation of INSPIRE parameters is necessary to optimize treatment outcomes. Treatments did not result in complete control of all tumors, as seen in previous studies utilizing IRE. However, the promising tumor control results seen in a subset of INSPIRE treatments and the positive findings of a potential abscopal effect in untread lesions substantiate further clinical use and studies of INSPIRE for the treatment of surface lesions in equine patients.

This study demonstrated INSPIRE’s ability to significantly reduce the volume of equine melanocytic tumors. Treatments administered via a coaxial ring and pin electrode were well tolerated by equine patients using a combination of standing sedation and local anesthetic. Treatment parameters consisting of 2,000 V, 2,000 ns waveforms with a treatment integrated time of 0.02 s resulted in an 88% reduction in tumor volume and 80% complete response rate. These results were achieved without the use of chemotherapeutics, the use of general anesthesia, or the need for surgical resection in regions which are challenging to keep sterile. This novel therapeutic approach has the potential to expand the role of pulsed electric fields in veterinary patients, especially when general anesthesia is contraindicated, and warrants future studies to demonstrate the efficacy of INSPIRE for tumor treatment.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal studies were approved by North Carolina State University. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent was obtained from the owners for the participation of their animals in this study.

CFo, TP, and MS conceived the experimental design. CFe, RP, MD, and MS oversaw use of INSPIRE hardware, analyzed data, and drafted the manuscript. BP, JR, EM, TM, KD, TP, and CFo conducted INSPIRE treatments. CE and BJ managed study setup, patient recruitment, and coordination. SG assisted in the drafting and editing of the manuscript. All authors contributed to the article and approved the submitted version.