4.1 Demography
4.1.1 Patient characteristics
A total of 41 dogs fulfilled the inclusion criteria. Twenty-one dogs received VBL-based chemotherapy, including eleven dogs treated with VCP protocol and ten dogs treated with VP protocol, and 20 dogs received imatinib target therapy. There were 24 (58.5%) pure breeds and 17 (41.5%) mixed breeds. The median age was eleven years (range 6-15 years) in the VBL group and ten years (range 4-17 years) in the imatinib group. Twenty-one (51.2%) dogs were female (1 intact, 20 spayed), and twenty (48.8%) dogs were male (11 intact, 9 castrated). Median body weight was significantly higher in VBL group (17.6 kg, range 4.9-36.8 kg) compared to imatinib group (10.9 kg, range 2.6-34.5 kg; P = 0.003). No significant differences observed in the breed, age, and, sex (Table 4).
4.1.2 Tumor features and previous treatments
In VBL group, one dog (4.8%) was stage 1, 4 dogs (19.0%) were stage 2, 7 dogs (33.3%) were stage 3, and 9 dogs (42.9%) were stage 4; however, abdominal ultrasound was not performed for staging in 3 dogs (1 classified as stage 2, 2 classified as stage 3).
In imatinib group, 6 dogs (30.0%) were stage 2, 8 dogs (40.0%) were stage 3, and 6 dogs (30.0%) were stage 4; abdominal ultrasound was not performed in 5 dogs (1 classified as stage 2, 4 classified as stage 3). In the VBL group, ten dogs (47.6%) were substage a, and 11 dogs (52.4%) were substage b. In imatinib group, 11 dogs (55.0%) were substage a, and 9 dogs (45.0%) were substage b. The median sum of the diameter of target lesions
the imatinib group. In the VBL group, tumors were located on the trunk in six dogs (28.6%), on limbs in three dogs (14.3%), and ten dogs (47.6%) had multiple cutaneous tumors, two dogs (9.5%) had only metastasized lymph nodes as gross lesions. In imatinib group, tumors were located on the trunk in five dogs (25.0%), on limbs in two dogs (10.0%), on the head in two dogs (10.0%), and nine dogs (45.0%) had multiple cutaneous tumors, two dogs (10.0%) had only metastasized lymph nodes as gross lesions. The tumors were local recurrent in eleven dogs (52.4%) in the VBL group and ten dogs (50.0%) in the imatinib group. No significant difference observed in the clinical stage, substage, sum of the diameter of target lesions, tumor location, and local recurrent tumor between the VBL group and the imatinib group (Table 5). However, in VBL group, there were more dogs classified as substage a in dogs receiving VCP protocol (n = 8; 72.7%) than in dogs receiving VP protocol (n = 2; 20.0%; P = 0.030).
Seven dogs (33.3%) in the VBL group were diagnosed with cMCTs by histopathology, including 2 Patnaik grade II, 1 Patnaik grade II-III, 1 Patnaik grade III, and 3 Kiupel high-grade cMCTs. Ten dogs (50.0%) dogs in the imatinib group were diagnosed with cMCT by histopathology, including 1 Patnaik grade I-II, 2 Patnaik grade II-III, 2 Patnaik grade III, and 5 Kiupel high-grade cMCTs. All dogs with Patnaik grade I-II or Patnaik grade II cMCTs were presented with LN metastasis or distant metastasis.
Four dogs (19.0%) had received other systemic treatments previously in VBL group, including imatinib (n = 1), palladia (n = 2), and CTX (n = 1). Nine dogs (45.0%) had received other systemic treatments previously in imatinib group, including VBL (n = 4), CTX (n = 4), chlorambucil (n = 2), palladia (n = 5). Ten dogs (47.6%) had received steroids previously in the VBL group, and 13 dogs (65.0%) had received steroids previously in the imatinib group. No significant difference observed in the distribution of dogs that had received previous systemic treatment or steroids between two groups (Table
5). Ancillary medications were concurrently given for prevention or alleviation of paraneoplastic syndromes in all cases based on the clinicians’ judgment, including diphenhydramine, famotidine, and sucralfate.
4.2 Treatment of VBL group
A total of 21 dogs received a median of 3 VBL doses (range 1-8 doses), with median dosage at 2.25 mg/m2 (range 2-2.55 mg/m2) and a median dose interval of 2 weeks (range 1-6 weeks). For dogs receiving VCP protocol, the median dosage of VBL was 2.25 mg/m2 (range 2-2.5 mg/m2), with median dose interval of 3 weeks (range 1-6 weeks); the median dosage of CTX was 250 mg/m2 (range 200-250 mg/m2), with median dose interval of 3 weeks (range 1-6 weeks). For dogs receiving VP protocol, the median dosage of VBL was 2.39 mg/m2 (range 2-2.55 mg/m2), with a median dose interval of 2 weeks (range 1-4 weeks). Six dogs experienced dose escalation of VBL (3 in VCP subgroup, 3 in VP subgroup), and two dogs experienced dose reduction of VBL due to unfavorable adverse effects (all in VCP subgroup). All dogs received concurrent prednisolone, and the median dosage of prednisolone was 0.85 mg/kg/day (range 0.25-2.25 mg/kg/day). For dogs receiving VCP protocol, the median dosage of prednisolone was 0.85 mg/kg/day (range 0.25-2 mg/kg/day); for dogs receiving VP protocol, the median dosage of prednisolone was 0.55 mg/kg/day (range 0.25-2.25 mg/kg/day). There was no significant difference in the dosage of prednisolone (P = 0.695) between dogs treated with VCP or VP protocol.
4.3 Treatment of imatinib group
The median dosage of imatinib was 9.3 mg/kg/day (range 2.3-12.7 mg/kg/day). Two
for dose reduction was CR in 2 dogs and SD in 1 dog. Eighteen dogs received concurrent prednisolone, with the median dosage at 0.5 mg/kg/day (range 0.15-2 mg/kg/day).
The median dosage of prednisolone was not statistically different between the VBL group and the imatinib group (P = 0.212).
4.4 Outcome
4.4.1 Response
In the VBL group, one dog achieved CR, eight dogs achieved PR, and 12 dogs achieved SD. The ORR in the VBL group was 42.9%, and the CBR was 100.0%. The ORR in VCP subgroup was 54.5% (1 CR and 5 PR), and was 30.0% (3 PR) in VP group (P = 0.387). The median time to response was 14 days (range 7-49 days) for responders.
In the imatinib group, three dogs achieved CR, four dogs achieved PR, and nine dogs achieved SD. The ORR in the imatinib group was 35.0%, and the CBR was 80.0%. The median time to response was 17 days (range 7-35 days) for responders.
There was no statistical difference in ORR between 2 groups (P = 0.606); however, the CBR was significantly higher in the VBL group than in the imatinib group (P = 0.048) (Table 6).
4.4.2 PFI
The median PFI was 83 days (range 7-239 days) in VBL group, with a median PFI of 121 days (range 7-239 days) in VCP subgroup and 55 days (range 14-83 days) in VP subgroup (P = 0.290). Eleven dogs were censored during follow-up, and the reason for drug withdrawal included the clients’ request (n = 4), unrelated death (n = 2), unfavorable side effects (n = 2), remaining CR (n = 1), remaining SD (n = 1), and undergoing surgical
excision (n = 1).
The median PFI was 51 days (range 7-415 days) in the imatinib group. Nine dogs were censored during follow-up, and the reason for drug withdrawal included the clients’
request (n = 4), unrelated death (n = 1), remaining CR (n =1), remaining SD (n = 1), receiving RT (n = 1), and lost to follow-up (n = 1).
The median PFI between 2 groups was not significantly different (P = 0.885), and was neither statistically different for dogs that achieved objective response or clinical benefit (P = 0.080 and P = 0.387, respectively) (Table 7) (Figure 1-3).
4.5 Toxicity
Adverse effects were noticed in 15 (71.4%) dogs in the VBL group and 7 (35.0%) dogs in the imatinib group (P = 0.019) (Table 9) and were considered mild in most instances (Table 8).
In VBL group, neutropenia was noticed for five episodes (all were grade 1) in 5 (23.8%) dogs, lethargy was noticed for eight episodes (grade 1) in six (28.6%) dogs, anorexia was noticed for five episodes (2 grade 1, 3 grade 2) in five (23.8%) dogs, GI signs were noticed for eleven episodes (grade 1) in six (28.6%) dogs, and elevation of alanine transaminase (ALT) was noticed for twelve episodes (6 grade 1, 5 grade 2, 1 grade 3) in nine (42.9%) of dogs. There was no significant difference in the incidence of adverse effects between VCP and VP subgroups (Table 10).
In imatinib group, neutropenia and lethargy were not reported in any dog, anorexia was noticed for three episodes (grade 1) in two (10.0%) dogs, GI signs were noticed for two episodes (grade 1) in two (10.0%) dogs, and elevation of ALT was noticed for six episodes (2 grade 1, 4 grade 2) in four (20.0%) dogs (Table 9).
The incidence of neutropenia and lethargy was significantly higher for dogs in the VBL group than the imatinib group (P = 0.048 and P = 0.021, respectively) (Table 9).
4.6 Prognostic factors for PFI
Prognostic factors that were subjected to the univariate analysis included age (< or
≥ median age), breed, sex, body weight (< or ≥ median body weight), the sum of the diameter of the target lesions (< or ≥ median sum), local recurrence or not, tumor location, the median dosage of prednisolone (< or ≥ median dosage of prednisolone), stage, substage, had received previous steroid treatment or not, previous systemic treatment, and objective response to the treatment.
Variables that had a significant influence on PFI identified on the univariate analysis included age, stage, and objective response (Table 11 and Table 12). Median PFI was significantly shorter for dogs older than 11 years old (51 days versus 189 days, P = 0.048) (Figure 4). Dogs classified as stage 4 had significantly shorter median PFI (35 days) compared to dogs classified as stage 2 (not reached, P = 0.005) and stage 3 (189 days, P
= 0.017) (Figure 5 and Figure 6). The median PFI for non-responders was also significantly shorter (40 days versus 160 days, P = 0.025) (Figure 7).
Upon multivariate analysis, age ≥ 11 years-old remained to be a negative prognostic factor for PFI (P = 0.033), with a hazard ratio of 3.710 (95% CI, 1.110-12.408). Dogs classified as stage 2 had significantly longer PFI compared to stage 4 (HR, 0.223; 95%
CI, 0.050-0.996; P = 0.049). There was no significant difference between the PFI of responders and non-responders (P = 0.231) (Table 13).