Evaluation of toceranib for treatment of apocrine gland anal sac adenocarcinoma in dogs

Abstract Background There is no widely accepted standard medical treatment for apocrine gland anal sac adenocarcinoma (AGASACA) in dogs. Targeted agents such as toceranib may be effective in treatment of AGASACA, but the number of clinical reports investigating its efficacy is limited. Hypothesis/Aim To evaluate the efficacy of toceranib treatment of AGASACA in dogs, and to assess prognostic factors in the study population. Our hypothesis was that toceranib would provide a clinical benefit in the treatment of dogs with AGASACA. Animals Thirty‐six client‐owned dogs with either a cytologic or histologic diagnosis of AGASACA that were treated with toceranib alone or in combination with surgery, nonconcurrent chemotherapy or both. Methods Retrospective study. Result The median progression‐free survival (PFS) and overall survival time (OST) for the study population was 313 days and 827 days, respectively. A clinical benefit from toceranib treatment was observed in 69% of dogs, with 20.7% of dogs experiencing partial response and 48.3% of dogs experiencing stable disease. Dogs that responded to toceranib treatment had significantly prolonged PFS and OST. Hypercalcemia was a negative prognostic factor for clinical outcomes. Conclusions Toceranib is effective in the treatment of AGASACA in dogs. Prospective, controlled clinical trials are needed to determine the efficacy of toceranib in comparison to other treatment protocols for dogs with AGASACA.

occurs before distant metastasis to sites such as the lungs, liver, and spleen. 3 Polyuria and polydipsia may be present as a consequence of paraneoplastic hypercalcemia, which is reported in approximately 25% to 50% of cases. 3,5,6 Despite its high propensity to metastasize, this tumor tends to have an overall indolent disease course with median survival times of 1-2 years with single or multimodal therapeutic modalities, including surgery, 6,7 chemotherapy, 8,9 and radiation. 5 Although numerous chemotherapeutic agents (carboplatin, 8 actinomycin D, 9 mioxantrone, 5 and melphalan 6 ) have been investigated, the definitive role of adjuvant chemotherapy in the treatment of AGASACA remains unclear. Alternative treatment options such as targeted therapy also require further assessment to define therapeutic benefit.
Toceranib phosphate (Palladia) is a small molecule tyrosine kinase inhibitor approved by the United States Food and Drug Administration (FDA) for the treatment of high-grade cutaneous mast cell tumors in dogs. 10 Toceranib has activity against several members of the splitkinase family including vascular endothelial growth factor receptor (VEGFR), platelet-derived growth factor receptor (PDGFR), Kit and Flt-3. 11,12 Toceranib's mechanism of action results from competitive blockade of the ATP-binding site of tyrosine kinase receptors, impairing phosphorylation and downstream signaling. 13 With both antiangiogenic as well as antitumor effects, toceranib has potential therapeutic activity in a wide range of tumor types in dogs. [14][15][16][17][18][19][20][21][22][23] A phase I clinical trial and retrospective case series suggest that toceranib exhibits potential clinical antitumor activity against mixed mammary carcinoma, soft tissue sarcoma, multiple myeloma, AGASACA, osteosarcoma, thyroid carcinoma, head and neck carcinoma, and nasal carcinoma. 12,24 Use of toceranib for treatment of AGASACA has been evaluated in a limited number of studies. In 1 study of 32 dogs with AGASACA, a clinical benefit from toceranib was reported in 87.5% of dogs, with 25% achieving partial response (PR) and 75% experiencing stable disease (SD). 24 The median duration for PR was 22 weeks and the median duration for SD was 30.5 weeks. A more recent retrospective study evaluated dogs diagnosed with stage 4 AGASACA, and 13 of 15 dogs experienced clinical benefit from toceranib treatment by stabilization of their disease. 25 Thus, toceranib may provide an attractive adjunctive treatment option for AGASACA in dogs, and further investigation of its effectiveness is needed.
Our primary aim was to retrospectively evaluate the efficacy of adjuvant and primary toceranib treatment in dogs with AGASACA. A secondary aim was to identify prognostic factors in the study population. A clinical benefit was defined as complete response (CR) or PR at any point in time or SD for at least 10 weeks. We hypothesized that toceranib would provide clinical benefit in dogs treated for AGASACA.

| Case selection and data collection
Dogs with a definitive ante-mortem diagnosis of AGASACA were identified by reviewing medical records from University of Wisconsin

| Antitumor response assessment
Tumor response was assessed in 29 patients with macroscopic disease by using the Response Evaluation Criteria in Solid Tumors (RECIST) 1.1. 26 Tumor assessment was performed using caliper measurement, thoracic radiographs, ultrasonography, or CT as indicated. A CR was defined as complete regression of disease; PR was defined as

| Assessment of adverse events (AEs)
All patients were evaluated at the UWVC before toceranib treatment. Physical examination, CBC, serum biochemistry profile, urinalysis, and blood pressure measurement were recommended before treatment, 1 month after initiation of toceranib treatment and every 2 to 3 months while receiving toceranib. Urine protein-to-creatinine ratio was performed if results of a patient's urinalysis or blood pressure measurement indicated proteinuria or systemic hypertension. All AEs were graded according to the Veterinary Co-operative Oncology Group's common terminology criteria for AEs (VCOG-CTCAE v1.1). 27 For gastrointestinal AEs, supportive care included, but was not limited to, a 7-day toceranib treatment holiday, antidiarrheal agents (eg, metronidazole, tylosin), antiemetics (eg, maropitant, metoclopramide, ondansetron) and gastric protectants (eg, omeprazole, famotidine).

| Statistical analysis
Descriptive statistics were used to summarize the variables collected on the group of patients treated and to define toceranib dosing strata.
The Kaplan-Meier (KM) product of survival probabilities was used to assess PFS and OST curves for the population and for different cohorts of interest. Dogs lost to follow-up were censored in the survival analysis at the last known follow-up event. Survival curves were compared using the log rank test. A multivariable Cox proportional hazard model was used to assess the rate of possible death events given the prognostic factors. A P-value for the log rank test of <.05 was considered statistically significant. All analyses were performed in R, 28 and the KM curves and Cox proportional hazard models were performed using the survival package (GraphPad Software, La Jolla, California). 29 3 | RESULTS

| Study population
Thirty-six dogs diagnosed with AGASACA and treated with toceranib alone or in combination with surgery, nonconcurrent chemotherapy or both during the study's time period fulfilled the study's inclusion criteria. The population consisted of 11 female dogs and 25 male dogs, and all dogs were either spayed or neutered. Nineteen different breeds were represented (Table 1) Table 2.

| Treatments
The majority of dogs (23/36, 63.9%) underwent surgery for removal of the primary tumor before treatment with toceranib. Regional

| Treatment outcomes
Toceranib-associated median OST and the median OST from the time of diagnosis (36 dogs) were 434 days ( Figure 1A) and 827 days ( Figure 1B Response to treatment was significantly associated with both PFS (P < .001) and toceranib-associated OST (P = .02; Figure 3 Figures S1 and S2). Median PFS and OST for each cohort are listed in Table 4.  (Table 4).

| DISCUSSION
Our primary objective was to assess the efficacy of toceranib in dogs with AGASACA. The demographics of our study population were consistent with previous reports, providing support for an accurate representation of dogs with AGASACA. [1][2][3][4][5]30  not directly comparable to other studies that calculated OST from the initial treatment or disease diagnosis. Thus, we also calculated OST from the time of diagnosis, which was 827 days ( Figure 1B). This outcome is more consistent with the previously reported survival times, ranging from 16 to 31 months. [4][5][6]30 The rate of metastatic disease in our study population (86.1%) was consistent with that of previous reports, but skewed toward the higher end of the range. Again, the majority of dogs in our study were evaluated later in the course of their disease, because toceranib often was used as adjuvant treatment after surgery or salvage treatment after failure of cytotoxic chemotherapy. Therefore, this higher rate of metastasis is likely a result of selection bias. It is also possible that the toceranib-associated survival time (434 days) in our study was falsely decreased because of selection bias and the advanced extent of disease in most patients.
A secondary aim of our study was to identify prognostic factors in the study population. Multivariate subgroup analysis, however, can be subject to bias and statistical concerns when used in small cohorts and in retrospective analysis. 31 Our multivariate Cox hazard analysis suggests that clinical benefit from toceranib treatment was a positive prognostic factor for clinical outcome, but toceranib dosage was not found to be prognostic ( Hypercalcemia also was found to be a negative prognostic factor with regard to clinical outcome ( Table 4). The prevalence of hypercalcemia (38.9%) in our study was similar to findings in previous reports, with hypercalcemia noted in 25%-50% of dogs with AGASACA. 3,5,6 The prognostic relevance of hypercalcemia in AGASACA is discordantly reported in the veterinary literature. 3,5,30 Multiple studies have found no statistical significance associated with hypercalcemia. 3,5,30 However, other studies have found hypercalcemia to be a negative prognostic factor associated with shorter survival. 4 It is possible that interactions exist between hypercalcemia and other prognostic variables, which may explain the presence of prognostic significance in the multivariate Cox analysis but not in the KM analysis.
Our study was a retrospective analysis of observational clinical data. Therefore, imbalance across the variables studied and the presence of missing data are to be expected because patient restaging was inconsistent and several patients were lost to follow-up. Thirteen dogs were censored in the PFS analysis (including both macroscopic and microscopic disease settings) and 7 dogs were censored in the OST analysis (including both macroscopic and microscopic disease settings) because of lack of follow-up data. Also, few dogs (n = 7) were included in PFS and OST analysis for dogs treated in the microscopic disease setting. Furthermore, dogs in our study often received multiple treatment modalities including surgery and chemotherapy, in addition to toceranib. Among patients treated by chemotherapy, a variety of drugs and protocols were used. Thus, it is difficult to compare differences and interactions among these various treatment protocols. A larger database may be required for more thorough analysis, which would demand a collaborative effort with other veterinary hospitals.

| CONCLUSIONS
A clinical benefit from toceranib treatment was observed in the majority of dogs in our study (69%). Of dogs that experienced clinical benefit, most experienced stabilization rather than regression of their disease. Toceranib treatment-associated OST and PFS were 434 and 313 days, respectively. Response to toceranib treatment was associated with both improved OST and PFS. These results support the clinical efficacy of toceranib in the treatment of dogs with AGASACA.
Prospective, controlled clinical trials are needed to further evaluate the efficacy of toceranib in comparison to other treatments for dogs with AGASACA.

CONFLICT OF INTEREST DECLARATION
Authors declare no conflict of interest.

OFF-LABEL ANTIMICROBIAL DECLARATION
Authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION
Authors declare no IACUC or other approval was needed.

HUMAN ETHICS APPROVAL DECLARATION
Authors declare human ethics approval was not needed for this study.