Efficacy and safety of electrochemotherapy in the treatment of soft tissue sarcomas in dogs

Soft Tissue Sarcomas (STS) represent up to 15% of all skin and subcutaneous tumours in dogs. Local tumour control with wide surgical margins or a combination of surgery and radiation therapy is the standard of care in the management of STS, however these approaches are not always possible due to anatomic location and/or expense of the treatment. Electrochemotherapy (ECT) is an emerging, locally ablative technique now available in the United States. ECT has been available in Europe for over a decade and is a standard therapy for cancer treatment in people and animals. The purpose of this study was to evaluate the safety and efficacy of ECT with intravenous bleomycin in the treatment of STS in dogs. Fifty-three client owned dogs with 55 STS were enrolled (two dogs each had two STS in different anatomical sites). Four dogs with STS located orally were excluded according to Bray (2016). Tumour location, size, histologic grade, electroporation parameters, recurrence rate, recurrence time, disease free interval (DFI) and treatment toxicity were evaluated. The owners elected to pursue ECT as an alternative treatment to surgery and/or radiotherapy. For all dogs, the diagnosis of STS was based on histopathology. Most dogs were staged with physical examination, fine needle aspiration of regional lymph node (if palpable) and three view thoracic radiographs. Complete blood cell count, minimal serum biochemistry profile and urinalysis were also obtained pre general anesthesia. For ECT treatment, dogs were premedicated with medetomidine (0.01 mg/kg, IM) and butorphanol (1 mg/kg, IM). Once sedated, patients were pre-oxygenated via face mask, and anesthesia was induced with intravenous alfaxalone (2 mg/kg). Dogs were intubated and anesthesia was maintained with oxygen, nitrous oxide and isofluorane. Atipamezole (0.05 mg/kg IM) was administered to reverse medetomidine at the conclusion of the procedure. Bleomycin was diluted in saline solution and administered IV at a dosage of 15,000 UI (Pharmacopeia Europe)/m2 (equivalent to 15 U-US Pharmacopeia /m2). Treatment areas were prepared for surgery, skin was clipped and aseptically prepared with povidone-iodine wash (10% iodine) and surgical spirit. All dogs received anti-inflammatory and pain medicine in the peri-operatory period with meloxicam (0.2 mg/kg SC) and butorphanol. After ECT, all patients received antibiotic coverage (Amoxicillin + Clavulanic acid 12.5 mg/kg, BID, PO or Clindamycin 5.5 mg/kg BID PO) for at least one week. Patients were assessed every 3–10 days for the initial 4 weeks after ECT treatment, then monthly for three months. Thereafter, progress was monitored until at least 2 years either by examination of the patient or by telephone contact with the owner. Any progression in the treated tissue was recorded by digital camera images. ECT was performed using Cytopulse PA4000 in 9 cases or Cytopulse Oncovet in 46 cases (Cyto Pulse Sciences, Inc., now distributed by BTX Harvard Apparatus, MA, USA), with Gehl needle electrodes of 1–2.5 cm length from the same manufacturer. The pulse pattern employed was 8 monophasic square pulses of 100 microseconds each at a frequency of 1 Hz with the PA4000, or 1 Hz or 5 kHz with the Oncovet. The electrode pattern comprised two parallel rows of six 1–2.5 cm length needles with rows 6 mm apart. The pulse amplitude was 600–720 volts (1000–1200 volts/cm). The time between bleomycin injection and pulse delivery was 8 minutes, and maximum electroporation treatment time was 20 minutes (Tozon et al. 2016). Patients were divided into three categories according to the ECT modality. (1) ECT alone, performed in smaller lesions, where STSs were clearly visible macroscopically and could be resected by conventional surgery but the owners elected to pursue ECT. (2) intraoperative ECT, performed in larger lesions. In these cases ECT was performed at the same time of the surgical cytoreduction of the tumour. During surgery the majority of the tumour was excised, with no attempt on obtaining complete margins (as standardly accepted of 2–3 cm according to Bray 2016), and ECT was applied to lateral and deep margins before wound closure; in some cases macroscopic tumour was still present. This method was used where radical excision would have resulted in functional or cosmetic compromise due to anatomic limitation. (3) post-operative (adjuvant) ECT; this method was applied when the surgical resection of the tumour was initially performed attempting wide margins but where histological examination showed inadequate margins. Adjuvant ECT was applied after a period of 2–4 weeks from the surgical intervention (mean time 22.4 days). The reason for the delay was to perform ECT treatment as early as possible after adequate wound strength had developed following surgery and the final decision was left to the surgeon’s judgement. In dogs receiving ECT alone, the tumour plus a 1–2 cm margin in all planes into grossly normal tissue were accessed by penetration of the electrode needles. The treatment was started in the margins and progressed concentrically into the tumour mass, thus avoiding transference of tumour cells from the centre to the margins. Moreover, this pattern of pulse application is reported to achieve better reduction of the blood flow to the tumour thus obtaining a better drug retention into the tumour stroma (Cemazar et al. 2016). Similarly, when ECT was used intra-operatively, a 1–2 cm margin in all directions was used. In dogs treated with ECT adjuvant to surgery, where there was no macroscopic disease, the treatment margin extended at least 1.5 cm from the healed surgical scar site in all planes. Local tumour response in dogs treated with ECT alone was evaluated according to the response evaluation criteria for solid tumors established by the veterinary cooperative oncology group consensus document (Nguyen et al. 2015). A complete remission (CR) was defined as total reduction of the tumour. A partial remission (PR) was defined as ≥ 30% reduction in tumour diameter. Stable disease (SD) was defined as < 30% reduction in tumour diameter or < 20% increase in tumour diameter, and progressive disease (PD) was defined as ≥ 20% increase in tumour diameter. The presence of new lesions near to the primary tumour was considered as PD. A minimum duration of two to three weeks was required for a response to qualify as positive. In order to assess the local treatment toxicity a grading score (5-point scale) for tissue necrosis was established by the authors and previously described (Lowe et al, 2016): 0 = none, 1 = slight swelling, 2 = swelling/necrosis < 1 cm, 3 = severe swelling, 4 = deep necrosis and 5 = severe swelling and tissue loss. DFI was calculated from the date of treatment with ECT to the first recurrence or death of the animal unrelated to the tumour; or alternatively when we censored the DFI at the date we prepared this paper. The statistical analysis was performed using Chi-squared test and Fisher’s exact test. Of the 53 dogs included in the study, 22 were females (16 neutered) and 31 were males (14 neutered). The majority of dogs were pure breeds (38/53, 71.6%); Labrador retrievers (8) and Staffordshire Bull Terrier (5) were over represented. Patient age ranged from 4 to 15 years (median 9.1). Tumour grade was assessed in 46 cases, most tumours were grade I (21/46, 45.6%) and II (23/46, 50.0%), only two were grade III (4.4%). The large majority of tumours affected the limbs (42/55, 76.4%), other locations were: trunk (8/55, 14.5%) and head (5/55, 9.1%). Tumour size ranged from 0.4 to 20 cm (median 4.5 cm). Median follow-up period was 482 days (range 8–2,483). The first ECT treatment was performed as follows: ECT alone in 5 cases, intraoperative ECT in 26 and postoperative (adjuvant) ECT in 24 cases. Five dogs underwent a second treatment due to tumour recurrence, in 3 of them a further recurrence occurred and a third ECT treatment was performed. The treatment response in the ECT alone group was as follows: two CR, one PR and two SD. At the end of the observation period, twenty-two dogs had died, of these, 7 had tumour recurrence, which was the reason for euthanasia in two cases. Of the other 29 dogs, 8 had tumour recurrence and in three cases the affected limb was amputated as rescue treatment. Patient outcome was not available in four cases. Overall, tumour recurrence rate (OTRR) was 27.2% (15/55). Recurrence rate was not associated with tumour grade, anatomical site, tumour size, pulse frequency and pulse voltage (P > 0.05). For all dogs, median DFI was not reached. For dogs with tumour recurrence, median DFI was 68 days (range 0–1,025). None of our patients developed systemic adverse effects related to the ECT treatment. Local toxicity was well tolerated in 65% of cases (36/55 had toxicity score ≤2 in a scale with 5 as maximum) and was not associated with tumour grade, tumour size and pulse frequency (P > 0.05). Local toxicity was associated with higher pulse voltage (1000 vs. 1200 V), which was statistically significant (P = 0.003). The overall tumour recurrence rate (OTRR) of ECT alone or in combination with surgery (intra or post-operative) was 27.2% Adjuvant ECT with local injection of bleomycin has been successfully applied in the treatment of canine sarcomas in a previous study (Spugnini et al. 2007). However, these results cannot be compared with ours because the tumour list included neoplastic entities that are no more considered as “true canine STS” (i.e. hemangiosarcoma, malignant fibrous histiocytoma, leiomiosarcoma and neurofibrosarcoma) (Liptak & Forrest 2013). Moreover the criteria of tumour response (i.e. complete response and progressive disease) were not extensively illustrated in the paper thus making a comparison impossible. Other therapies including adjuvant radiation therapy and intralesional chemotherapy have been reported to have an OTRR of 18.0% and 16.6%, respectively (Demetriou et al. 2012, Havlicek et al. 2009). When possible, a wide margin excision still remains the best choice to treat canine STS (3.2%, Prpich et al. 2013). In this study, ECT with intravenous bleomycin was well tolerated with 65% showing mild toxicity. This finding is similar to the results previously reported by one of the authors using the same toxicity score scale (53%, Lowe et al. 2016). To our knowledge, no data has been reported regarding postsurgical time of healing or complication rate with intraoperative ECT. Local toxicity was similar in intraoperative and postoperative ECT groups, suggesting that treatment delay may be not necessary in a postoperative setting. ECT with bleomycin should be considered an option to treat STS in dogs especially when other treatment options are declined or not applicable. However, further investigation is necessary to identify ideal treatment parameters.