• Open Access

Multicenter Prospective Trial of Hypofractionated Radiation Treatment, Toceranib, and Prednisone for Measurable Canine Mast Cell Tumors


  • Portions of this material were presented in abstract form to the 30th Annual Conference of the Veterinary Cancer Society, San Diego, CA, October 2010.

Corresponding author: Dr Douglas H. Thamm, CSU Animal Cancer Center, 300 West Drake Road, Fort Collins, CO 80523-1620; email: dthamm@colostate.edu.



Mast cell tumors (MCT) are common cutaneous tumors in dogs and when not amenable to surgical excision can present a therapeutic challenge. New treatment protocols for unresectable MCT are needed.


The combination of toceranib, prednisone, and hypofractionated radiation treatment (RT) will be well tolerated and efficacious.


Seventeen client-owned dogs with measurable MCT amenable to RT.


Prospective clinical trial. All dogs received prednisone, omeprazole, diphenhydramine, and toceranib. Toceranib was administered for 1 week before initiating RT, consisting of 24 Gy delivered in 3 or 4 fractions.


On an intent-to-treat basis, the overall response rate was 76.4%, with 58.8% of dogs achieving a complete response and 17.6% a partial response. The median time to best response was 32 days, and the median progression-free interval was 316 days. The overall median survival time was not reached with a median follow-up of 374 days. The most common toxicoses were gastrointestinal and hepatic.

Conclusions and Clinical Importance

The combination of hypofractionated RT, toceranib, and prednisone was tolerated and efficacious in the majority of dogs. Response rates and durations were higher than those reported for toceranib as a single-agent treatment for MCT. This combination is a viable treatment option for unresectable MCT.


formalin-fixed, paraffin wax-embedded


internal tandem duplication


tyrosine kinase inhibitor



Veterinary Cooperative Oncology Group Common Terminology Criteria for Adverse Events


Veterinary Radiation Therapy Oncology Group

Mast cell tumors (MCT) are the most common cutaneous tumor in the dog, accounting for between 16 and 21% of all cutaneous tumors.[1, 2] Surgery is the treatment of choice for cutaneous MCT in areas amenable to excision, but locally recurrent, large or infiltrative tumors, and those in locations not amenable to wide surgical excision represent a therapeutic challenge. External beam radiation treatment (RT) produces varying control rates when used as a primary treatment.[3-8] RT alone at total dosages between 40 and 45 Gray (Gy) results in 1-year control rates of approximately 50%. Hypofractionated RT is a treatment in which the total dose is delivered in larger doses per fraction given less than once daily. This type of treatment often is referred to as palliative RT. Reports of hypofractionated RT as a sole treatment for canine MCTs are limited.[5] The combination of hypofractionated RT and prednisolone previously was reported as an effective and well-tolerated treatment.[9] Alternative local treatments for confined MCT include intralesional brachytherapy, photodynamic treatment, intralesional corticosteroids, cryotherapy, and intralesional deionized water. None are as thoroughly investigated, clinically effective, or practical as surgery. Novel approaches for the treatment of unresectable MCT are needed.

Toceranib is a tyrosine kinase inhibitor (TKI) targeting the receptor tyrosine kinase's vascular endothelial growth factor receptor 2, platelet-derived growth factor receptor, Kit, and Flt-3, which has demonstrated encouraging single-agent antitumor activity against canine MCT.[10, 11] In a recent multicenter prospective study, 42.8% of dogs experienced an objective response with a median response duration of 18.1 weeks. Dogs with a c-kit gene activating mutation had greater objective response rates than those without the mutation.[11] Much information indicates synergy between the related TKI sunitinib and RT in preclinical models,[12-14] and at least 1 early clinical trial demonstrated an absence of acute effect potentiation with the concurrent use of sunitinib and RT.[15] Sunitinib is thought to enhance radiation-induced endothelial damage by inhibition of the PI3K/Akt signaling pathway, which then leads to apoptosis.[16, 17] In addition, activation of Kit has been demonstrated to confer intrinsic radiation resistance in vitro,[18] suggesting that its inhibition may be able to reverse this phenomenon. Toceranib is very similar in chemical structure to sunitinib, and thus may have similar radiosensitizing effects. In addition, there are anecdotal reports of toceranib used in combination with hypofractionated RT, making toceranib a good choice for the current study.

The aims of this study were to determine the tolerability, adverse event profile, and clinical activity of toceranib, prednisone, and hypofractionated RT in dogs with measurable MCT. A secondary aim of the study was to determine if there was an association between outcome and the presence of activating c-kit gene mutations.

Materials and Methods

Animal Population

We conducted a multi-institutional prospective study. Dogs with cytologically or histologically confirmed MCT that either were not amenable to surgical excision or whose owners declined surgery were eligible to participate. Before enrollment, all dogs underwent complete clinical staging, including a CBC, serum biochemistry profile, fine needle aspiration cytology of the regional lymph node if palpable or enlarged, abdominal ultrasound examination, and thoracic radiographs if indicated based on tumor location. Dogs were required to have an MCT that could be serially measured and treated with RT. Regional lymph node metastasis was allowed if the lymph node was amenable to treatment with RT. Dogs were required to have adequate diagnostic evaluation (absolute neutrophil count >1,500 cells/μL, hematocrit >25%, platelets >100,000/μL, creatinine <2.5 mg/dL, bilirubin ≤ the upper reference limit, ALT ≤3 times upper reference limit or if >3 times reference limit serum bile acids ≤ the upper limit of reference), and a VCOG performance status of 0 or 1 (0, normal activity; 1, restricted [decreased activity from predisease status]; 2, compromised [ambulatory for only vital activities, urinates and defecates in appropriate areas]; 3, disabled [requires force feeding, unable to urinate and defecate in appropriate areas]; 4, dead). No prior RT to the target lesion was allowed and a 2-week washout period from prior surgery or chemotherapy was required. Dogs were treated at Colorado State University, The Ohio State University, and Red Bank Veterinary Hospital. Treatment was begun in these dogs between March 2010 and May 2010.

Maintenance and experimental protocols followed the animal care guidelines of the Animal Care and Use Committees or Clinical Review Boards of the participating institutions. Written informed consent was obtained from all owners before treatment started.

Treatment Protocol

All dogs received prednisone (1 mg/kg) PO q48h, omeprazole (0.7 mg/kg) PO q24h, and diphenhydramine (2–4 mg/kg) PO q8h for 72 hours before initiation of toceranib treatment. Toceranib was given at a target dosage of 2.75 mg/kg PO on a Monday, Wednesday, Friday schedule on days alternating with prednisone administration. The dosage of toceranib used (2.75 mg/kg) was based on previous information indicating fewer adverse effects with roughly equivalent antitumor activity with this decreased dose.[10] The study protocol duration was 16 weeks, and at the end of the study, dogs could continue to receive toceranib until the development of progressive disease (PD) or 1 year after treatment initiation.

Radiation treatment was begun 7 days after the start of toceranib. This delay was to ensure that toceranib was tolerated and to maximize the likelihood of therapeutic toceranib plasma concentrations before starting RT. Dogs were treated with a total of 24 Gy that was delivered in either 3 fractions of 8 Gy on days 0, 7, 21, or 4 fractions of 6 Gy on days 0, 7, 14, 21. The radiation schedule and treatment plan were left to the discretion of the radiation oncologist. Radiation was delivered at a target time of 6–8 hours after toceranib dosing to correspond with peak plasma concentrations of the drug.[19]

Response Criteria

Tumor response was assessed by physical examination and tumor measurements at enrollment, weeks 1, 2, 3, 4, 6, 10, 16, and every 6 weeks thereafter. Response was determined using RECIST criteria and categorized as complete response (CR): complete resolution of all disease; partial response (PR): ≥30% decrease in maximal diameter; stable disease (SD): percent difference in maximal diameter between −30 and 20%; and PD ≥ 20% increase in maximal diameter or the appearance of new lesions.[20]

Toxicity Criteria

At each visit, a detailed owner history was obtained. If dogs experienced any grade of known gastrointestinal adverse effects of toceranib, the drug was discontinued until the signs resolved. If the problem was severe or persistent, a dose reduction was performed. CBCs were performed at each visit and biochemical profiles at weeks 1, 4, 10, 16, and every 6 weeks thereafter. Toxicities were graded using the Veterinary Cooperative Oncology Group Common Terminology Criteria for Adverse Events (VCOG CTCAE v1.0).[21]

Acute radiation adverse events were graded according to the Veterinary Radiation Therapy Oncology Group (VRTOG) acute radiation morbidity scoring scheme and categorized as grade 0, no change over baseline; grade 1, erythema, dry desquamation, alopecia/epilation; grade 2, patchy moist desquamation; grade 3, confluent moist desquamation with edema, ulceration, necrosis, or hemorrhage.[22]

C-Kit Mutation Status

Nucleic Acid Extraction

Genomic DNA samples were prepared from Wright-Giemsa stained fine needle aspirates or formalin-fixed paraffin-embedded samples. The slides were confirmed to have ≥10% of the cell population as mast cells. Qiagen AL buffer was applied to the slides, which were then scraped with a straight-edged razor into microcentrifuge tubes. DNA extraction then was performed with a commercial kit according to the manufacturer's instructions.1 Shavings from wax blocks were collected into microcentrifuge tubes and samples were processed with a commercially available kit.2

PCR Amplification of C-Kit Exons 8 and 11

Regions of c-kit exons 8 and 11 were amplified with primers directed against sequences flanking the characterized ITDs (internal tandem duplications). The primer pair for exon 8 consisted of (TGACCTATGGCCATTTCTCT) coupled with (56FAM-AATCCTGCAACCACACACTG). The pair for exon 11 consisted of (CAGTGGAAGGTTGTTGAGGAG) coupled with (VIC-CATGGAAAGCCCCTATTTCA). Amplifications were performed with a commercially available PCR kit.3 Primer concentrations used were 400 nM each.

Gene scanning analysis was performed using a capillary electrophoresis machine.4 Amplified fragments were run with Gene Scan-600 LIZ size standards. Raw data were analyzed with a commercially available genotype analysis software.5

Statistical Analysis

Progression-free interval (PFI) was defined as the time from starting medications (diphenhydramine, omeprazole, and prednisone) to the time of PD. Similarly, survival time (ST) was defined as the time from the start of medications to the time of death. Animals were censored from PFI/survival analysis if they were progression-free/alive at the time of last follow-up, lost to follow-up, or died of an unrelated cause. PFI and ST curves were generated using the Kaplan–Meier product limit method and groups compared using log rank analysis. All statistical analyses were performed using a commercial software package.6 A P value of .05 was considered significant for all analyses.


Seventeen client-owned dogs were included in this study and were treated at Colorado State University, The Ohio State University, and Red Bank Veterinary Hospital. Animal and tumor characteristics are reported in Table 1. Histopathology was available for 5 dogs: 4 tumors were grade III and 1 was grade I. The remaining 12 dogs had cytologic diagnoses only. Two dogs received prior surgical treatment, but all others were naïve to any treatment. Median tumor diameter before treatment was 4.3 cm, with a range of 1.5–20 cm. Eight dogs had lymph node involvement at the time of enrollment and 2 had multiple tumors.

Table 1. Characteristics of dogs at time of enrollment.
Median age, years (range)7 (2–12)
Median weight, kg (range)28.1 (3.6–50.5)
Labrador Retriever5
Mixed breed4
Others (1 each)4

Treatment and Response

Fifteen of the 17 dogs received hypofractionated RT and toceranib as intended. Fourteen dogs received 3 fractions of 8 Gy and 1 dog received 4 fractions of 6 Gy. Two dogs were withdrawn from the study before starting toceranib or RT: 1 after tumor rupture and 1 for inability to tolerate adverse effects associated with prednisone. However, these dogs were included in outcome analysis. On an intent-to-treat basis, the overall response rate (ORR) was 76.4%, with a median time to maximal response of 32 days (range, 2–156). Ten dogs (58.8%) experienced CR, 3 (17.6%) dogs experienced PR, 3 (17.6%) dogs had SD, and 1 (5.9%) dog had PD (Fig 1). When the 2 dogs that did not receive toceranib or RT were excluded, the ORR was 86.7% (66.7% CR, 20% PR).

Figure 1.

Waterfall plot depicting maximal tumor response for each dog. * = dogs not receiving radiation treatment.

Thirteen dogs completed the 16-week treatment protocol. Of the 13 dogs, 1 did not continue toceranib because of continued intermittent gastrointestinal adverse effects. Of the 4 dogs that were withdrawn before the 16-week examination, 2 were withdrawn before starting treatment as previously mentioned, and 2 were withdrawn for PD. The median starting dosage of toceranib was 2.75 mg/kg (mean, 2.73 mg/kg; range, 2.35–2.89 mg/kg) and the median ending dosage was 2.53 mg/kg (mean, 2.52 mg/kg; range, 1.76–2.96 mg/kg).

The overall median PFI was 316 days and the median ST was not reached (Fig 2). The mean follow-up time in the censored dogs was 374 days. At 1 year, 70% of dogs were alive and 44.8% were disease-free. Nine dogs experienced PD. The times and patterns of treatment failure are depicted in Figure 3. Four dogs received rescue treatment including 2 masitinib, 1 surgery and RT, and 1 vinblastine and RT.

Figure 2.

Kaplan–Meier plot depicting progression-free interval and survival time.

Figure 3.

Progression-free interval and reason for progression for each dog. Shaded bars represent dogs without disease progression at the time of last follow-up. * = dogs with activating c-kit gene mutations. ** = dogs that did not receive radiation treatment.


Gastrointestinal toxicity was observed in all dogs receiving toceranib. Ten dogs developed increased serum liver enzyme activity. Three dogs experienced grade I toxicities as their most severe adverse event, 6 dogs had grade II, 7 grade III, and 1 grade V as the most severe event. The single grade V toxicity was observed in the dog that was euthanized after tumor rupture, before initiating treatment. Information regarding grade and description of all reported toxicities is listed in Table 2. Of the 15 dogs that received treatment as intended, toceranib was temporarily discontinued for drug-related adverse effects in 13 (87%) dogs and dose reductions were performed in 8 (53%) dogs. Two dogs had dose delays during the 1st month of treatment, but all patients received toceranib before RT.

Table 2. Adverse events reported.
 Grade IGrade IIGrade IIIGrade IVGrade V
Bone marrow14    
Weight loss11   
Otitis externa 1   
MCT rupture    1
Corneal ulcer 2   
Ocular discharge11   
Urinary incontinence1    
Hearing loss 1   
Nasal discharge1    
Radiation effects24    
Acute effects7    
Dry desquamation7    
Late effects17    

Toxicity from RT was minimal. The most common acute effects were dry desquamation and alopecia. Eight dogs experienced grade I acute effects and no grade II or III acute effects were seen. The most common late radiation effects were alopecia and leukotrichia, with 10 dogs experiencing grade I adverse events. No grade II or III events were observed.

C-Kit Mutation Status

Samples for evaluation of c-kit mutation status were available for 14 dogs. Eight dogs had no mutation identified, 1 had an exon 8 mutation, and 5 had exon 11 mutations. The median PFI of dogs with mutant c-kit was 188.5 days (27 weeks) and was not reached in dogs with wild-type c-kit (= .0017; Fig 4). No dogs with c-kit mutant MCT and 87.5% of dogs with c-kit wild-type MCT were disease-free at 1 year. There was no statistical difference in overall ST based on c-kit mutation status (= .25). At 1 year, 66.7% of dogs with c-kit mutant MCT and 100% of dogs with c-kit wild-type MCT were alive.

Figure 4.

Progression-free interval and overall survival time based on c-kit gene mutation status.


The goal of this study was to determine the efficacy and tolerability of the combination of hypofractionated RT, prednisone, and toceranib for the treatment of canine MCT. The results of this study indicate that this treatment protocol is efficacious and tolerable in dogs with measurable MCT. No potentiation of RT adverse effects was observed, but the majority of dogs required a toceranib holiday and over half of the dogs had a dose reduction. The ORR for the 15 dogs that received treatment as intended was 86.7%, with 10 CR, 3 PR, and 2 SD. The median PFI was 316 days and time to best response was 32 days. The response rate and PFI in this study were greater than those reported for single-agent toceranib in dogs with measurable MCT, in which an ORR of 48.2% and median time to progression of 18.1 weeks (126 days) were reported.[11] A difference between these 2 studies was the use of prednisone. Dogs in the current study received prednisone and those in the previous study evaluating single-agent toceranib did not. However, the low and infrequent dosage of prednisone used in this study (1 mg/kg every other day) is unlikely to have been associated with clinically relevant single-agent antitumor activity.[23] The authors are not aware of any studies evaluating the use of toceranib and prednisone for the treatment of measurable MCT.

Previous studies evaluating RT as a sole treatment using total dosages between 40 and 45 Gy resulted in 1-year control rates of approximately 50%.[5] A retrospective multi-institutional study evaluating dogs with measurable MCT receiving either hypofractionated RT and prednisone or hypofractionated RT and chemotherapy reported an ORR of 41.5% and PFI of 68.5 days.7 In contrast, a study evaluating the combination of hypofractionated RT and prednisolone reported an ORR of 88.5%, which is equivalent to that of the current study. However, the PFI reported was 1,031 days, which is higher than that found in the current study (PFI = 316 days) and earlier literature.[9] It is difficult to compare our results with the previous study, because the previous study was retrospective. We can speculate, however, that specific differences between the 2 studies may account for some discrepancy in PFI. In the current study, only 15 dogs received treatment whereas the previous study included 35 dogs. It is possible that if a larger population were studied, the results of the current study would be different. In addition, only 1 dog in the previous study had a grade III MCT, and although grade was available for only 5 dogs in our study, 4 were grade III MCT. It is possible that the PFI was shorter in our study attributable to a large percentage of biologically aggressive MCT. Furthermore, c-kit mutation status was not evaluated in the previous study for its effect on outcome.

In the current study, dogs with MCT possessing activating mutations in the c-kit gene had a median PFI of 27 weeks, whereas in dogs with wild-type c-kit gene the PFI was not reached. This finding is not surprising, because it has been shown previously that MCT with activating c-kit gene mutations are more biologically aggressive than those with wild-type c-kit genes.[24-26] In addition, in a clinical field trial evaluating toceranib in dogs with MCT, those with c-kit gene mutations had shorter PFI than those with wild-type c-kit (C. London, unpublished data).

The treatment protocol reported in this article overall was well tolerated with no grade IV or V hematologic toxicities. No clinically relevant degranulation episodes were seen and all radiation-associated adverse effects were mild. Gastrointestinal toxicity associated with toceranib has been well documented in previous studies.[10, 11, 27] All dogs in this study that received toceranib experienced at least 1 mild to moderate gastrointestinal adverse event. The number and grade of gastrointestinal events are listed in Table 2. All instances of gastrointestinal toxicity were documented, but in 2 dogs, the owner reported the events after the signs had resolved without any change in treatment schedule or dose. In the 13 dogs for which the owners reported the event sufficiently early, gastrointestinal toxicities resolved after temporary discontinuation of toceranib and use of supportive medications (eg, antidiarrheals, antiemetics, antibiotics) as needed. However, breaks in treatment may have put animals at higher risk for disease progression. In 2 cases, disease progression occurred after a prolonged treatment delay (>3 weeks). In cases of mild gastrointestinal toxicity, it may not be necessary to discontinue treatment, but supportive medications should be started instead.

Hepatotoxicosis was observed in 80% (12/15) of dogs that received toceranib in this study. The authors are not aware of any other studies that have reported hepatoxicity associated with toceranib use. This is the first study to evaluate the use of toceranib and prednisone in combination, and it is not uncommon to see increases in ALP activity in dogs receiving corticosteroids. Several dogs also had increases in other liver enzyme activities (AST and ALT), which are not commonly seen with prednisone administration alone. In addition, several dogs experienced increased liver enzyme activity only 1 month after starting treatment. Several animals were given a hepatoprotectant (S-adenosylmethionine and silybin) and liver function tests improved in all of these dogs. The combination of prednisone and toceranib may have additive effects on the liver.

A limitation of this study is the lack of single-agent control arms to evaluate toceranib and prednisone; RT and prednisone; and toceranib, prednisone, and RT. In addition, the number of dogs in this study is small, making it challenging to draw definitive conclusions. Lastly, tumor measurements were not performed by a single individual, which could have resulted in variation in measurements.

In conclusion, this study indicates that treatment with the combination of hypofractionated RT, prednisone and toceranib is well tolerated and effective for unresectable canine MCT. This protocol should be used for comparison in a multiple arm study. Mild gastrointestinal and hepatic toxicities were commonly observed with this protocol, and resolved with supportive treatment and dose modification. It may not be necessary to discontinue treatment for grade 1 gastrointestinal toxicities, but provide supportive medications if needed and consider a drug holiday or dose reduction for higher-grade toxicities or if resolution is not seen after supportive treatment. The authors recommend starting hepatoprotectants at the first sign of hepatoxicity when using this protocol. Future randomized prospective studies should be performed to compare the described treatments alone and in combination.


The authors thank the clinical trials teams at the participating institutions. Pfizer Animal Health provided funding for this study.


  1. 1

    Qiagen DNEasy Blood and Tissue Kit, Qiagen, Valencia, CA

  2. 2

    QIAampR DNA FFPE Tissue Kit, Qiagen

  3. 3

    Phusion Blood Direct PCR Kit, Thermo Scientific, Waltham, MA

  4. 4

    ABI Prism 3730xl genetic analyzer, Applied Biosystems, Carlsbad, CA

  5. 5

    GeneMarker Vers 1.85, SoftGenetics, State College, PA

  6. 6

    Prism 5.0b for Macintosh, GraphPad Software, San Diego, CA

  7. 7

    Rau SE, Clifford C, Lang V, et al. Treatment of bulky MCT with palliative radiation treatment ± chemotherapy: A multi-institutional retrospective study. Proceedings of the 27th Veterinary Cancer Society Conference, Fort Lauderdale, FL, November 1–4, 2007 (abstract)