Masitinib is a TKI that targets canine c-Kit, with an IC50 of approximately 100nM for the WT c-Kit and 5 to 100nM for isoforms that have mutations in the extracellular (exon 8 or 9) and juxtamembrane (exon 11) domains of c-Kit and are commonly found in MCTs. Several reports1–4 support the involvement of c-Kit, the receptor for stem cell factor, in the regulation of mast cell proliferation and differentiation for physiologically normal conditions, as well as in processes involved in the development of MCTs. Preclinical in vivo experiments have revealed antitumor properties of masitinib in nude mice that received a subcutaneous graft of tumor cells that expressed c-Kit.5 Therefore, masitinib may also exert antitumor activities in MCTs of dogs.
Results from a 6-month, phase 3, double-blind, placebo-controlled clinical trial revealed the safety and efficacy of masitinib in the treatment of dogs with nonmetastatic grade 2 or grade 3 cutaneous MCTs.6 Analysis of data indicated that masitinib significantly improved TTP, although the response rate at 6 months for masitinib-treated dogs was not significantly different from that for placebo-treated dogs. To further assess the long-term impact of masitinib on the survival of dogs with MCTs, dogs treated with masitinib and benefiting from treatment were enrolled in a compassionate program. After treatments were revealed, 24 dogs with nonresectable tumors continued to receive masitinib and 56 dogs with nonresectable tumors (49 from the masitinib group and 7 from the placebo group) were monitored to determine survival analysis. Thus, the objective of the study reported here was to provide follow-up information and analyses on the same dogs approximately 2.5 years after the completion of the original study's pivotal phase6 and to determine the benefits of long-term use of masitinib in dogs with cutaneous MCTs and the predictive value of short-term tumor responses for estimating long-term survival.
Materials and Methods
Sample population—The pivotal clinical study6 was a multicenter, randomized, double-blind, placebo-controlled trial that was conducted at 25 veterinary centers in the United States and France. Each participating veterinary hospital adhered to guidelines established for Good Clinical Practice, and all dogs were cared for in accordance with each institution's standard of animal care. Consent was obtained from each owner for participation of dogs in the study.
Eligible dogs were client-owned pets with measurable nonresectable or recurrent postsurgical grade 2 or 3 MCTs without lymph node or visceral metastasis. Dogs were classified as having nonresectable tumors when they had at least 1 tumor that could not be removed surgically (eg, resection was unlikely to be complete if surgery was performed or complete resection would lead to amputation, which would not be acceptable to the owner).
Classification of MCTs—Tumors were staged on the basis of results of a CBC, serum biochemical analysis, thoracic radiography (2 views), and abdominal ultrasonography and evaluation of a fine-needle aspirate obtained from a regional lymph node (if palpable) and graded in accordance with grading criteria described in another study.7 The presence of mutations in the sequence of the receptor c-Kit expressed by tumors was evaluated by use of reverse transcriptase—PCR assay performed on biopsy specimens obtained from 124 cutaneous MCTs; biopsy specimens for 8 dogs could not be evaluated.
Treatment—Participating dogs received daily oral administration of a placebo or masitinib (12.5 mg/kg/d in the form of tablets that contained 25, 50, or 100 mg of masitinib mesilate) until tumor progression was detected, as determined on the basis of World Health Organization criteria.8 Dogs were randomly assigned to one of the treatment arms with a placebo-to-masitinib ratio of 1:4, with stratification on the basis of tumor grade (grade 2 vs grade 3) and tumor status (nonresectable vs recurrent after surgery).
Reasons for discontinuation in the study were adverse events, lack of efficacy, major deviation from the protocol, lack of treatment compliance, dogs lost to follow-up monitoring, and withdrawal of owner's consent. In the case of adverse events, specific rules were applied for treatment interruptions, dose reductions, or treatment discontinuations.6
Pivotal 6-month study—In the original pivotal study phase,6 dogs were evaluated regularly during the first 6 months (weeks 1, 2, 4, 6, 8, 12, 16, 20, and 24) after onset of treatment. Dogs that had a complete or partial response or that had stable disease following 6 months of treatment were eligible to continue receiving the treatment (masitinib or placebo) by entering a compassionate program, wherein they were evaluated at 4-week intervals until progression of disease was detected. After treatments administered to the dogs were unblinded, only dogs receiving masitinib could continue in the compassionate program; these dogs were evaluated at 12-week intervals. At each examination during the compassionate-use period, tumors were measured and response to treatment was evaluated in accordance with World Health Organization criteria.8 After treatment was discontinued, investigators regularly assessed the survival of dogs by contacting clients via telephone at monthly intervals.
Endpoints for follow-up monitoring for up to 24 months—Endpoints were TTP and tumor response rates at 12 and 24 months after onset of treatment for masitinib-treated dogs and overall survival and survival rate at 12 and 24 months for both masitinib-and placebo-treated dogs. The TTP was defined as the interval between the first day of treatment and the time of tumor progression. If tumor progression was not observed at the time of the last tumor assessment, data were censored at the time of last tumor assessment. Overall survival was defined as the interval between the first day of treatment and the time of death. If the dog was still alive at the time of last contact, data were censored at the time of last contact.
The majority (42/52 [80.8%]) of the dogs that entered the compassionate program had nonresectable tumors. Therefore, endpoints for follow-up monitoring were analyzed in the subgroup of dogs with nonresect-able tumors.
Statistical analysis—Statistical analyses were performed by use of commercial software.a Values for all statistical tests were considered significant at P < 0.05. In addition, commercial graphing softwareb was used to plot Kaplan-Meier curves and calculate median ratios with 95% CIs.9
For time-to-event analyses (ie, TTP and overall survival), survival curves were generated by use of the Kaplan-Meier method. The effect of masitinib was evaluated by determining medians, the median ratio, and the Cox proportional hazard ratio with 95% CI; data were analyzed by use of a log-rank test.
Survival rates at 12 and 24 months were calculated as Kaplan-Meier estimates with 95% CIs and as the percentages of observed cases. For observed cases, only dogs with confirmed survival at the time point or that died before reaching the time point were included in the analysis. The effect of masitinib (vs the placebo) was analyzed by use of a Fisher exact test.
Tumor response rates (defined as the sum of complete and partial responses) at 12 and 24 months were calculated as the percentages of observed cases. Observed cases included all dogs, except for dogs that were removed from the study without disease progression before reaching the time point. Disease control rate was defined as the sum of tumor response and stable disease.
The predictive value of tumor response and disease control at 6 weeks or 6 months after onset of treatment for long-term survival (12 or 24 months) was determined. Several variables were calculated, including positive predictive value (percentage of dogs that were alive at 12 or 24 months among dogs with a tumor response at 6 months), negative predictive value (percentage of dogs that died by 12 or 24 months among dogs that did not have a tumor response at 6 months), sensitivity (percentage of dogs that had a tumor response at 6 months among dogs that were alive at 12 or 24 months), and specificity (percentage of dogs that did not have a tumor response among dogs that died by 12 or 24 months). Significant differences were tested by use of the Mantel-Haenszel χ2 test and Yule coefficient.
Results
Animals—A total of 132 dogs with nonresectable grade 2 or 3 MCTs were initially recruited into the study's pivotal phase; this included 106 dogs treated with masitinib and 26 dogs treated with a placebo. At the time the treatments for the pivotal study6 were unblinded, 22 masitinib-treated dogs and 1 placebo-treated dog remained in the compassionate program; the placebo-treated dog was removed.
Among the 22 dogs treated after the pivotal treatments were revealed, 11 were removed from the study, 5 were lost to follow-up monitoring, and 6 continued to receive treatment. Of those 11 dogs, 6 were removed because of lack of efficacy, 4 were removed because of adverse events, and 1 was removed because the owner withdrew consent.
Disease characteristics—Baseline characteristics of dogs with nonresectable tumors at the time of inclusion into the pivotal study were evaluated (Table 1). Disease characteristics were similar between the placebo-treated and masitinib-treated groups.
Baseline disease characteristics (at time of inclusion into the pivotal study) of dogs with nonresectable MCTs that received masitinib at 12.5 mg/kg/d or a placebo.
| Variable | Masitinib (n = 106 dogs) | Placebo (n = 26 dogs) |
|---|---|---|
| Time from diagnosis (mo)* | ||
| Mean ± SD | 8.0 ± 18.4 | 6.1 ± 14.1 |
| Median (range) | 1.4 (0–102.2) | 0.6 (0–67.7) |
| Tumor gradet | ||
| 2 | 97 (91.5) | 24 (92.3) |
| 3 | 9 (8.5) | 2 (7.7) |
| c-Kit status† | ||
| WT | 69 (65.1) | 17 (65.4) |
| Mutated | 31 (29.2) | 7 (26.9) |
| Missing | 6 (5.7) | 2 (7.7) |
Values did not differ significantly (P ≥ 0.05) between groups.
*Represents results for only 105 masitinib-treated dogs.
†Values reported are number (percentage) of dogs.
Impact of masitinib on TTP—Kaplan-Meier analysis of TTP at the time at which treatments were unblinded in the pivotal study6 indicated that masitinib treatment significantly (hazard ratio, 2.17; P = 0.001) improved the TTP in dogs with nonresectable tumors (median, 173 days), compared with the TTP in dogs receiving the placebo treatment (75 days). Significant improvement was detected regardless of c-Kit mutational status. For dogs having nonresectable tumors with WT c-Kit, masitinib treatment significantly (hazard ratio, 1.90; P = 0.027) improved the TTP (median, 140 days), compared with the TTP for the placebo treatment (75 days). For dogs having nonresectable tumors with mutated c-Kit, masitinib treatment significantly (hazard ratio, 4.42; P = 0.002) improved the TTP (median, 241 days), compared with the TTP for the placebo treatment (83 days).
Analysis of TTP was performed for data obtained after treatments were unblinded in the pivotal study. The curve for masitinib-treated dogs reached a plateau, which was estimated at 28.4% of dogs with nonresectable tumors (Figure 1). The plateau was evident regardless of c-Kit mutational status.
Kaplan-Meier curves for TTP in 132 dogs with nonresectable MCTs (A), 86 dogs with nonresectable tumors that had WT c-Kit (B), and 38 dogs with nonresectable tumors that had mutated c-Kit (C). In each graph, curves represent placebo-treated dogs (black diamonds; determined at the time treatments were unblinded in the pivotal study6), masitinib-treated dogs (black circles; determined at the time treatments were unblinded in the pivotal study6), and dogs in the follow-up study reported here (white circles).
Citation: American Journal of Veterinary Research 71, 11; 10.2460/ajvr.71.11.1354
Overall survival in dogs receiving masitinib—Survival at 12 and 24 months was confirmed for 120 of 132 (90.9%) and 103 of 132 (78.0%) dogs with nonresectable tumors, respectively (Table 2). Masitinib treatment significantly improved survival rates at both 12 and 24 months, compared with results for the placebo treatment. At 12 months, a significantly (P = 0.024) higher proportion of dogs treated with masitinib were alive (59/95 [62.1%]), compared with the proportion of placebo-treated dogs (9/25 [36.0%]) that were alive. At 24 months, a significantly (P = 0.04) higher proportion of dogs treated with masitinib were alive (33/83 [39.8%]), compared with the proportion of placebo-treated dogs (3/20 [15.0%]) that were alive.
Survival rates at 12 and 24 months in dogs with nonresectable MCTs that were treated with masitinib (n = 106 dogs) or a placebo (26).
| 12 months | 24 months | |||
|---|---|---|---|---|
| Variable | Masitinib | Placebo | Masitinib | Placebo |
| Nonresectable MCTs | ||||
| No. of dogs | 106 | 26 | 106 | 26 |
| No. of observed cases | 95 | 25 | 83 | 20 |
| No. (%) alive | 59 (62.1)a | 9 (36.0)b | 33 (39.8)c | 3 (15.0)d |
| No. (%) that died | 36 (37.9) | 16 (64.0) | 50 (60.2) | 17 (85.0) |
| Kaplan-Meier estimate (%)* | 65.1 (55.1–73.5) | 36.9 (18.9–55.1) | 48.9 (38.4–58.5) | 29.5 (12.0–49.6) |
| Nonresectable MCTs with WT c-Kit | ||||
| No. of dogs | 69 | 17 | 69 | 17 |
| No. of observed cases | 62 | 16 | 54 | 12 |
| No. (%) alive | 38 (61.3) | 7 (43.8) | 23 (42.6) | 2 (16.7) |
| No. (%) that died | 24 (38.7) | 9 (56.2) | 31 (57.4) | 10 (83.3) |
| Kaplan-Meier estimate (%)* | 64.6 (52.0–74.7) | 45.3 (21.0–66.9) | 51.8 (38.7–63.4) | 34.0 (10.8–59.2) |
| Nonresectable MCTs with WT c-Kit | ||||
| No. of dogs | 31 | 7 | 31 | 7 |
| No. of observed cases | 27 | 7 | 23 | 6 |
| No. (%) alive | 17(63.0)e | 1(14.3)f | 7 (30.4) | 0 (0) |
| No. (%) that died | 10 (37.0) | 6 (85.7) | 16 (69.6) | 6 (100) |
| Kaplan-Meier estimate (%)* | 66.0(46.0–80.1) | 14.3(0.7–46.5) | 41.4 (22.5–59.4) | NR |
*Results reported represent the value (95% CI).
abValue differs significantly (P = 0.024) between groups with different superscript letters. cdValue differs significantly (P = 0.040) between groups with different superscript letters. e,fValue differs significantly (P = 0.035) between groups with different superscript letters.
NR = Not reached.
Median overall survival time was 617 days (95% CI, 433 to 937 days) for masitinib-treated dogs (Table 3). However, this was not significantly (P = 0.121) different from the median survival time of 322 days (95% CI, 224 to 721 days) for the placebo-treated dogs (median ratio, 1.92 [95% CI, 1.33 to 2.50]; hazard ratio, 1.54 [95% CI, 0.89 to 2.66]).
Overall survival time (days) in dogs with nonresectable MCTs that were treated with masitinib (n = 106 dogs) or a placebo
| Variable | Masitinib* | Placebo* | Hazard ratio† | Pvalue‡ |
|---|---|---|---|---|
| Nonresectable MCTs (n = 132 dogs) | 617 (433–937) | 322 (224–721) | 1.54 (0.89–2.66) | 0.121 |
| Nonresectable MCTs with mutated c-Kit (n = 38 dogs) | 533 (317-infinity) | 162 (23–340) | 3.07 (1.18–8.02) | 0.016 |
| Nonresectable MCTs with WT c-Kit (n = 86 dogs) | 779 (434–938) | 361 (224-infinity) | 1.26 (0.62–2.56) | 0.516 |
*Results reported are median (95% CI).
†Results reported are value (95% CI).
‡Results were considered significant at P < 0.05 (log-rank test).
Median survival time for masitinib-treated dogs having nonresectable tumors with WT c-Kit (n = 69) was 779 days (95% CI, 434 to 938 days). This did not differ significantly (hazard ratio, 1.26 [95% CI, 0.62 to 2.56]; P = 0.516) from the median survival time for 17 placebo-treated dogs (361 days [95% CI, 224 to infinity]; median ratio, 2.16 [95% CI, 1.66 to 2.66]). Masitinib treatment did not have a significant impact, except for the median ratio (Table 3). At 12 months, 38 of 62 (61.3%) masitinib-treated dogs were alive, which did not differ significantly (P = 0.260) from the proportion of placebo-treated dogs (7/16 [43.8%]) that were alive (Table 2). At 24 months, 23 of 54 (42.6%) masitinib-treated dogs were alive, which did not differ significantly (P = 0.113) from the proportion of placebo-treated dogs (2/12 [16.7%]) that were alive.
Survival time of masitinib-treated dogs that had nonresectable tumors with mutated c-Kit was significantly (hazard ratio, 3.07 [95% CI, 1.18 to 8.02]; P = 0.016) longer (median, 533 days [95% CI, 317 to infinity]), compared with the survival time of placebo-treated dogs (median, 162 days [95% CI, 23 to 340 days]; median ratio, 3.29 [95% CI, 2.90 to 3.68]; Table 3). At 12 months, 17 of 27 (63.0%) masitinib-treated dogs were alive, which differed significantly (P = 0.035) from the proportion of placebo-treated dogs (1/7 [14.3%]) that were alive (Table 2). At 24 months, 7 of 23 (30.4%) masitinib-treated dogs were alive; however, this did not differ significantly (P = 0.289) from the proportion of placebo-treated dogs (0/6 [0%]) that were alive.
Tumor response at 6 months for prediction of long-term survival—Tumor response at 6 months was the primary endpoint for the present study, and the predictive value of this endpoint for estimation of long-term survival was analyzed (Table 4). Tumor response at 6 months was predictive of long-term survival at 12 months (Yule coefficient, 0.89; P < 0.001) and 24 months (Yule coefficient, 0.79; P < 0.001). Indeed, 19 of 20 (95.0%) dogs that had a tumor response at 6 months were alive at 12 months, and 14 of 18 (77.8%) dogs that had a tumor response at 6 months were alive at 24 months. However, sensitivity was only 32% for survival at 12 months and 42% for survival at 24 months, which suggested that a lack of tumor response was not necessarily predictive of a poor prognosis and that stabilization of the disease may also have provided long-term clinical benefits.
Predictive value of tumor response or tumor control at 6 weeks or 6 months for survival at 12 or 24 months in dogs with nonresectable MCTs that were treated with masitinib or a placebo.
| Variable | 12 months | 24 months |
|---|---|---|
| Tumor response at 6 months* | ||
| Dogs alive (No. [%]) | ||
| Among dogs with tumor response at 6 months | 19/20 (95.0) | 14/18 (77.8) |
| Among dogs without tumor response at 6 months | 40/75 (53.3) | 19/65 (29.2) |
| Positive predictive value (%) | 95 | 78 |
| Negative predictive value (%) | 47 | 71 |
| Sensitivity (%) | 32 | 42 |
| Specificity (%) | 97 | 92 |
| Yule coefficient | 0.89 | 0.79 |
| P value (Mantel-Haenszel χ2 test)† | < 0.001 | < 0.001 |
| Tumor control at 6 months‡ | ||
| Dogs alive (No. [%]) | ||
| Among dogs with tumor control at 6 months | 33/35 (94.3) | 25/31 (80.7) |
| Among dogs without tumor control at 6 months | 26/60 (43.3) | 8/52 (15.4) |
| Positive predictive value (%) | 94 | 81 |
| Negative predictive value (%) | 57 | 85 |
| Sensitivity (%) | 56 | 76 |
| Specificity (%) | 94 | 88 |
| Yule coefficient | 0.91 | 0.92 |
| P value (Mantel-Haenszel χ2test)† | < 0.001 | |
| Tumor response at 6 weeks* | ||
| Dogs alive (No. [%]) | ||
| Among dogs with tumor response at 6 weeks | 21/33 (63.6) | 12/28 (42.9) |
| Among dogs without tumor response at 6 weeks | 38/62 (61.3) | 21/55 (38.2) |
| Positive predictive value (%) | 64 | 43 |
| Negative predictive value (%) | 39 | 62 |
| Sensitivity (%) | 36 | 36 |
| Specificity (%) | 67 | 68 |
| Yule coefficient | 0.05 | 0.10 |
| P value (Mantel-Haenszel χ2test)† | 0.822 | 0.681 |
| Tumor control at 6 weeks‡ | ||
| Dogs alive (No. [%]) | ||
| Among dogs with tumor control at 6 weeks | 48/69 (69.6) | 28/59 (47.5) |
| Among dogs without tumor control at 6 weeks | 11/26 (42.3) | 5/24 (20.8) |
| Positive predictive value (%) | 70 | 47 |
| Negative predictive value (%) | 58 | 79 |
| Sensitivity (%) | 81 | 85 |
| Specificity (%) | 58 | 38 |
| Yule coefficient | 0.51 | 0.55 |
| P value (Mantel-Haenszel χ2test)† | 0.015 | 0.026 |
*Tumor response includes complete or partial responses.
†Values were considered significant at P < 0.05.
‡Tumor control includes complete or partial responses and stable disease.
Overall survival on the basis of tumor status at 6 months—We further analyzed the overall survival by subgroups defined on the basis of tumor status at 6 months (ie, dogs with a tumor response, dogs with stable disease, and dogs with tumor progression; Figure 2). Dogs treated with masitinib that had tumor progression at 6 months had an overall survival that did not differ significantly (hazard ratio, 0.87; P = 0.612) from that of placebo-treated dogs. Moreover, dogs that responded to treatment with masitinib at 6 months had a significantly (hazard ratio, 5.18; P < 0.001) better overall survival than did dogs treated with the placebo.
Kaplan-Meier curves for overall survival in dogs with nonresectable MCTs on the basis of tumor response at 6 months. Curves represent 41 placebo-treated dogs (black diamonds), 26 masitinib-treated dogs that had a complete or partial tumor response at 6 months (white circles), 20 masitinib-treated dogs that had stable disease at 6 months (black circles), and 115 masitinib-treated dogs that had progression of disease at 6 months (black squares).
Citation: American Journal of Veterinary Research 71, 11; 10.2460/ajvr.71.11.1354
For dogs with stable disease at 6 months, overall survival was significantly (hazard ratio, 5.10; P = 0.001) better than the overall survival of dogs treated with the placebo but did not differ significantly (hazard ratio, 0.68; P = 0.498) from the overall survival of dogs that had a tumor response at 6 months. These results confirmed that stabilization of disease for 6 months provided long-term clinical benefits similar to those for a tumor response.
Tumor control at 6 months for prediction of long-term survival—Controlled disease included both tumor response and tumor stabilization. We analyzed the use of tumor control at 6 months for estimation of long-term survival (Table 4). The analysis revealed that tumor control at 6 months was a good predictor of survival at 12 months (Yule coefficient, 0.91; P < 0.001) and at 24 months (Yule coefficient, 0.92; P < 0.001). Sensitivity was 56% and 76% at 12 and 24 months, respectively.
Tumor response at 6 weeks for prediction of long-term survival—Other investigators have proposed tumor response at 6 weeks as the primary endpoint.10 Therefore, we also analyzed the use of tumor response at 6 weeks for estimation of long-term survival (Table 4). Tumor response at 6 weeks was not predictive of long-term survival at 12 months (Yule coefficient, 0.05; P = 0.882) or 24 months (Yule coefficient, 0.10; P = 0.681). Indeed, 21 of 33 (63.6%) dogs that had a tumor response at 6 weeks were alive at 12 months (compared with 38/62 [61.3%] dogs that did not have a tumor response at 6 months), and 12 of 28 (42.9%) dogs that had a tumor response at 6 weeks were alive at 24 months (compared with 21/55 [38.2%] dogs that did not have a tumor response at 6 weeks).
At 6 weeks, tumor control was predictive for long-term survival at 12 months (Yule coefficient, 0.51; P = 0.015) and 24 months (Yule coefficient, 0.55; P = 0.026). However, the predictive value of tumor control at 6 weeks (especially the specificity) was lower than those of tumor response at 6 months or tumor control at 6 months.
Long-term tumor response at 12 and 24 months resulting from masitinib treatment—Complete response rates were assessed on observed cases. At 12 and 24 months, the number of evaluable dogs with masitinib-treated nonresectable tumors was 74 of 106 (69.8%) dogs and 67 of 106 (63.2%) dogs, respectively. Among dogs with nonresectable tumors, 11 of 74 (14.9%) dogs treated with masitinib had complete tumor responses at 12 months and 6 of 67 (9.0%) dogs had complete tumor responses at 24 months, which suggested that this subset of patients had long-term tumor control or a potential cure. Thus, masitinib induced long-term complete responses in this subset of patients.
Among dogs that had nonresectable tumors with WT c-Kit, 5 of 50 (10.0%) treated with masitinib had complete tumor responses at 12 months and 3 of 46 (6.5%) had complete tumor responses at 24 months. Among dogs that had nonresectable tumors with mutated c-Kit, 6 of 18 (33.3%) treated with masitinib had complete tumor responses at 12 months and 3 of 15 (20.0%) had complete tumor responses at 24 months. Thus, long-term complete responses were evident regardless of c-Kit mutational status, although complete responses were more frequent among dogs that had tumors with mutated c-Kit.
Discussion
In the study reported here, the long-term outcome of dogs that had nonresectable grade 2 or 3 MCTs treated with masitinib was evaluated. Three major conclusions can be drawn.
First, the long-term follow-up data reported here provided further evidence that masitinib is effective in the treatment of nonresectable MCTs (similar to the results reported for the pivotal study6) and provided benefits in terms of long-term survival. Second, analysis revealed that the best short-term response at 6 weeks was not predictive of long-term survival in dogs treated with masitinib. Third, analysis revealed that control of disease at 6 months was highly predictive of long-term survival. Indeed, dogs with a tumor response at 6 months and dogs with stable disease at 6 months had similar long-term survival, which confirmed that tumor stabilization also provided important clinical benefits in terms of long-term survival.
To our knowledge, the study reported here is the first in which significant drug-induced improvement of long-term survival in dogs with MCTs has been described. Investigators in most clinical studies have evaluated the efficacy of cytotoxic chemotherapeutics or other TKIs on short-term tumor response.11,12 Although tumor response or improved TTP does not necessarily correlate with long-term survival,13–16 the follow-up data of the present report provide direct evidence of significant improvement in survival as a result of masitinib treatment of dogs with nonresectable tumors, which is the ultimate objective of anticancer treatments. Moreover, complete responses at 24 months were detected in 6 of 67 (9.0%) dogs with nonresectable MCTs treated with masitinib, which suggested that these patients might have been cured.
For dogs with nonresectable tumors, masitinib treatment improved long-term survival, regardless of c-Kit mutational status. Masitinib treatment induced long-term complete responses, regardless of c-Kit mutational status. However, the improvement in survival was more evident in dogs that had tumors with mutated c-Kit.
Mutations in the extracellular or juxtamembrane domain of c-Kit result in ligand-independent cell proliferation. They are detected in up to 30% of high-grade MCTs and are believed to play a prominent role in the tumorigenesis of malignant mast cells.1,2,4 Considering that masitinib has high in vitro antiproliferative properties for ligand-independent proliferation of cells expressing the mutated c-Kit (IC, 5 to 100nM), it was expected to be particularly effective for tumors that had a mutated c-Kit. Analysis of the data confirmed that the efficacy of masitinib treatment was particularly evident in dogs with tumors that had a mutated c-Kit.
The majority (approx 75%) of dogs with MCTs have tumors that express WT c-Kit.1 The exact cellular mechanisms involved in the tumorigenesis of malignant mast cells expressing WT c-Kit remain to be elucidated. Considering that masitinib also has high in vitro antiproliferative properties for ligand-dependent proliferation of cells expressing the WT c-Kit (IC50, 100nM), it was expected to be effective in dogs with tumors expressing WT c-Kit. In this subpopulation, masitinib significantly improved TTP.
Tumor response at 6 weeks was not a good predictor of long-term survival. The poor predictive value for the short-term (ie, 6 weeks) tumor response may have been a result of the fact that the size of MCTs typically fluctuates over time, depending on the degree of inflammation of the tumor. In the study reported here, placebo-treated dogs sometimes had short-lasting remission of the MCTs.
In addition, the lack of correlation between short-term response and long-term survival is in agreement with data from human clinical oncological studies.14–16 The US FDA determined that response rate generally should not be the sole basis for approval of a drug because the potential benefit associated with a response did not necessarily outweigh the substantial toxic effects of oncology drugs and the correlation between response rate and survival or between response rate and clinical benefit was not clearly established.16 A review of results for clinical trials suggests that tumor response is not a reliable replacement outcome for survival. Trials14,15 in which investigators have established treatment effects on the basis of a response rate endpoint have not revealed any change in mortality rates.
Finally, studies17,18 with other TKIs, such as inhibitors of vascular endothelial growth factor, have revealed that these drugs are able to generate a short-term tumor response but that they increase the invasiveness of tumor cells and ultimately reduce survival. Therefore, efficacy of anticancer drugs in dogs with MCTs, as evaluated on the basis of short-term tumor response, should be interpreted with caution and confirmed by further evaluations of survival, when possible.
Disease control at 6 months had a high predictive value for long-term survival. Tumor response and tumor stabilization at 6 months translated into similar long-term survival, which confirmed that stable disease at 6 months was predictive of a good prognosis in dogs with MCTs treated with masitinib. In fact, tumor control at 6 months was highly predictive of long-term survival. Considered together, these data indicated that sustained tumor stabilization for at least 6 months provided substantial clinical benefits in terms of long-term survival and suggested that the evaluation of treatment efficacy in veterinary practice or during clinical trials should not be based on initial tumor response alone. This introduces a new paradigm for TKIs. In contrast to chemotherapeutics, the success of which depends on short-term response, the success of TKIs depends on achieving long-term (6 months) control of tumors and ultimately improvement in survival.
The follow-up analysis of the pivotal study phase6 of masitinib for the treatment of MCTs revealed that masitinib significantly increased survival rates at 12 and 24 months in dogs with nonresectable tumors. Control of disease at 6 months, but not best response at 6 weeks, was predictive of long-term survival in dogs treated with masitinib, which suggested that short-term response may be irrelevant for assessing clinical efficacy of TKIs for use in treatment of MCTs.
ABBREVIATIONS
| CI | Confidence interval |
| IC50 | Median inhibitory concentration |
| MCT | Mast cell tumor |
| TKI | Tyrosine kinase inhibitor |
| TTP | Time to progression |
| WT | Wild type |
SAS, version 9.1, SAS Institute Inc, Cary, NC.
GraphPad Prism, GraphPad Softwares Inc, La Jolla, Calif.
References
- 1.↑
Letard S, Yang Y & Hanssens K, et al. Gain-of-function mutations in the extracellular domain of KIT are common in canine mast cell tumors. Mol Cancer Res 2008; 6:1137–1145.
- 2.↑
London CA, Kisseberth WC & Galli SJ, et al. Expression of stem cell factor receptor (c-kit) by the malignant mast cells from spontaneous canine mast cell tumours. J Comp Pathol 1996; 115:399–414.
- 3.
Qiu FH, Ray P & Brown K, et al. Primary structure of c-kit: relationship with the CSF-1/PDGF receptor kinase family-oncogenic activation of v-kit involves deletion of extracellular domain and C terminus. EMBO J 1988; 7:1003–1011.
- 4.
Zemke D, Yamini B, Yuzbasiyan-Gurkan V. Mutations in the juxtamembrane domain of c-KIT are associated with higher grade mast cell tumors in dogs. Vet Pathol 2002; 39:529–535.
- 5.↑
Dubreuil P, Letard S & Ciufolini M, et al. Masitinib (AB1010), a potent and selective tyrosine kinase inhibitor targeting KIT. PLoS One 2009; 4:e7258.
- 6.↑
Hahn KA, Ogilvie G & Rusk T, et al. Masitinib is safe and effective for the treatment of canine mast cell tumors. J Vet Intern Med 2008; 22:1301–1309.
- 7.↑
Patnaik AK, Ehler WJ, MacEwen EG. Canine cutaneous mast cell tumor: morphologic grading and survival time in 83 dogs. Vet Pathol 1984; 21:469–474.
- 8.↑
Miller AB, Hoogstraten B & Staquet M, et al. Reporting results of cancer treatment. Cancer 1981; 47:207–214.
- 9.↑
Machin D, Cheung YB, Parmar M. Survival analysis: a practical approach. 2nd ed. Chichester, England: John Wiley & Sons Inc, 2006.
- 10.↑
London CA. A tyrosine kinase inhibitors in veterinary medicine. Top Companion Anim Med 2009; 24:106–112.
- 11.
London CA, Malpas PB & Wood-Follis SL, et al. Multi-center, placebo-controlled, double-blind, randomized study of oral toceranib phosphate (SU11654), a receptor tyrosine kinase inhibitor, for the treatment of dogs with recurrent (either local or distant) mast cell tumor following surgical excision. Clin Cancer Res 2009; 15:3856–3865.
- 12.
London CA, Seguin B. Mast cell tumors in the dog. Vet Clin North Am Small Anim Pract 2003; 33:473–489.
- 13.
Benjamin RS, Choi H & Macapinlac HA, et al. We should desist using RECIST, at least in GIST. J Clin Oncol 2007; 25:1760–1764.
- 14.
Buyse M, Thirion P & Carlson RW, et al. Relation between tumour response to first-line chemotherapy and survival in advanced colorectal cancer: a meta-analysis. Meta-Analysis Group in Cancer. Lancet 2000; 356:373–378.
- 15.
Fleming TR, DeMets DL. Surrogate end points in clinical trials: are we being misled? Ann Intern Med 1996; 125:605–613.
- 16.↑
Johnson JR, Williams G, Pazdur R. End points and United States Food and Drug Administration approval of oncology drugs. J Clin Oncol 2003; 21:1404–1411.
- 17.
Ebos JM, Lee CR & Cruz-Munoz W, et al. Accelerated metastasis after short-term treatment with a potent inhibitor of tumor an-giogenesis. Cancer Cell 2009; 15:232–239.
- 18.
Paez-Ribes M, Allen E & Hudock J, et al. Antiangiogenic therapy elicits malignant progression of tumors to increased local invasion and distant metastasis. Cancer Cell 2009; 15:220–231.
