Evaluation of the Tolerability of Combination Chemotherapy with Mitoxantrone and Dacarbazine in Dogs with Lymphoma
ABSTRACT
Combination chemotherapy can be an effective option for treating resistant lymphoma in dogs. This retrospective study examined the tolerability and efficacy of the combination of 5-(3,3-dimethyl-1-triazeno)-imidazole-4-carboxamide (dacarbazine) (DTIC) in a population of dogs with lymphoma resistant to a doxorubicin-containing chemotherapy protocol. Mitoxantrone was administered at 5 mg/m 2 IV over 10 min followed by DTIC at 600 mg/m 2 IV over 5 hr, every 3 wk. All dogs were treated with prophylactic trimethoprim–sulfadiazine and metoclopramide. The frequency of grade 4 neutropenia was 18%, and 5% of dogs were hospitalized from sepsis. Gastrointestinal toxicity was uncommon. The overall response rate was 34% (15 of 44; 95% confidence interval 20–48%) for a median duration of 97 days (range 24–636 days, 95% confidence interval 44–150 days). Fourteen of 15 dogs who received mitoxantrone and DTIC as first rescue responded to treatment. Dogs who achieved complete remission to their initial L-asparaginase, cyclophosphamide, doxorubicin, vincristine, and prednisone chemotherapy protocol were more likely to respond to mitoxantrone and DTIC (23 versus 11%, P = .035). The combination of mitoxantrone and DTIC is a safe treatment option for resistant lymphoma in dogs.
Introduction
Resistance to chemotherapy is a common cause of treatment failure in dogs with lymphoma and is the major factor limiting successful management of this disease. Rescue therapies are used with the goal of attempting to induce remission in a patient with lymphoma that has failed first-line treatment or to re-establish remission in a patient who relapsed after previous treatment. Response rates for the most common rescue protocols used to treat canine lymphoma range between 30 and 72%, with durations of response lasting between 1 and 5 mo.1–8
One strategy to overcome the challenges inherent to treating dogs with resistant lymphoma is to use combination chemotherapy. The advantages of combination chemotherapy over single-agent chemotherapy include additive or synergistic interactions between drugs, the ability to overcome or delay multidrug resistance (MDR), and achieving higher dose intensity (milligram per meter squared body surface area [BSA] per week).9–11 When selecting agents to be used in combination, recommendations are to (1) choose drugs with nonoverlapping toxicities so each drug can be administered at near maximal dosages, (2) combine agents with different mechanisms of action and minimal to no cross-resistance, (3) use drugs that are proven as single-agent treatments, and (4) administer combination treatments at an early stage of disease and at a schedule with minimal treatment-free period between cycles to maximize cell kill.9–11
5-(3,3-dimethyl-1-triazeno)-imidazole-4-carboxamide (dacarbazine) (DTIC) is a nonclassical alkylating chemotherapy agent that acts via methylation of DNA at the O6 methyl group of guanine.12 Other potential mechanisms of action include inhibition of purine synthesis and direct DNA damage.12 Greissmayr et al. reported an overall response rate of 35% in dogs with relapsed lymphoma treated with single-agent DTIC.13 In vitro and in vivo studies indicate synergism between dacarbazine and anthracycline agents.14–16 In dogs with resistant lymphoma, dacarbazine combined with anthracyline agents, either doxorubicin or dactinomycin, resulted in overall response rates of 53–71%, numerically robust responses for this population of patients.2,17 The major dose-limiting toxicity (DLT) of DTIC in dogs is gastrointestinal toxicity.18
Mitoxantrone is a synthetic dihydroxyanthracenedione antineoplastic drug that shares similar mechanisms of action with anthracycline chemotherapy agents, including the ability to bind to nucleic acids to inhibit DNA and RNA synthesis.19 This intercalation between DNA strands creates single- and double-strand breaks and induces apopotosis.19 Compared with the anthracyline agent doxorubicin, mitoxantrone has less potential for free radical formation, which is a possible explanation for the reduced cardiotoxicity seen with this drug.19 These features make mitoxantrone an attractive treatment option for dogs previously treated with maximally cumulative dosages of doxorubicin, a common clinical scenario for patients with relapsed lymphoma. Single-agent mitoxantrone is an effective treatment for dogs with both chemotherapy-naïve and relapsed lymphoma, with complete response rates ranging between 26 and 47%.4,20 Combination treatment with mitoxantrone has not yet been evaluated in dogs. The DLT of mitoxantrone in dogs is myelosuppression.21,22
The purpose of this retrospective study was to describe toxicity associated with the combination of mitoxantrone and DTIC in dogs. A secondary goal was to report the overall response rate to treatment with mitoxantrone and DTIC in dogs with resistant peripheral nodal lymphoma. The hypothesis was that the combination of mitoxantrone and DTIC would be well tolerated and an effective treatment for dogs with resistant lymphoma.
Materials and Methods
Study Population
Data was obtained retrospectively from medical records from three institutions (Cornell University Hospital for Animals, Animal Emergency and Referral Associates, VCA West Los Angeles Animal Hospital) of client-owned dogs who received combination treatment with IV mitoxantrone and DTIC (Mitox/DTIC) between November 2006 and May 2009. During that time, Mitox/DTIC was used for rescue treatment in dogs with confirmed peripheral nodal lymphoma who had developed resistance or failed to respond to previously administered doxorubicin-based chemotherapy. This chemotherapy combination was not used in dogs who had received any myelosuppressive chemotherapy within 2 wk or had pretreatment inappetance, vomiting, diarrhea, or hematological abnormalities including clinically significant cytopenias or evidence of hepatic or renal dysfunction. Consent from clients was obtained before any chemotherapy treatment.
Treatment Protocol
The dosages of mitoxantrone and DTIC used in combination in this study were based on analysis of a previously constructed tolerable dose diagram. As was customary at our institutions to the general approach of exploring new chemotherapy combinations, a tolerable-dose diagram illustrating the organ-specific DLTs for various combinations of mitoxantrone and DTIC was prepared (Figure 1). DLTs of mitoxantrone was based on previously published reports.21–23 DLTs of DTIC was based on retrospective analysis of records of dogs treated at our institutions prior to 2006. At that time, information regarding the toxicity of single-agent DTIC was not yet published.13 Starting dosages were selected by examining the relationship of the DLTs of both drugs. Based on this analysis, the starting dosages of mitoxantrone and DTIC of 5 and 600 mg/m2 BSA, respectively, seemed appropriate.
Citation: Journal of the American Animal Hospital Association 55, 2; 10.5326/JAAHA-MS-6878
Mitoxantronea was administered IV at a dosage of 5 mg/m2 BSA diluted in 30 mL of 0.9% NaCl over 10–15 min. DTICb was administered IV at a dosage of 600 mg/m2 BSA, reconstituted in sterile water to achieve a concentration of 10 mg/mL, and the prescribed dose was further diluted in saline solution. The volume of saline used for dilution was based on BSA as follows: 1000 mL saline for dogs >1 m2 BSA, 250 mL of saline for dogs 0.4–1 m2 BSA, and 100 mL saline for dogs <0.4 m2 BSA. Dolasetronc was administered as an antiemetic at a dosage of 0.6 mg/kg. DTIC and dolasetron were administered IV through an indwelling catheter. Specifically, dogs received mitoxantrone followed by a bolus of dolasetron IV. The calculated dose of DTIC was then infused during a 5 hr period. Starting 1 day after treatment, dogs received a prophylactic antibiotic (trimethoprim–sulfadiazined, 15 mg/kg per os [PO] q 12 hr for 14 days) and a prophylactic antiemetic (metoclopramidee, 0.5 mg/kg PO q 8 hr for 7 days). The Mitox/DTIC protocol was administered every 21 days as long as patients sustained a response and had recovered from any toxic effects associated with a preceding treatment (Table 1). For dogs with grade 4 neutropenia (<500 μL) and/or grade 4 thrombocytopenia (<15,000 μL) at the nadir following the first Mitox/DTIC treatment, subsequent mitoxantrone dosages were reduced to 4 mg/m2. If greater than or equal to grade 3 gastrointestinal toxicity developed, the dosage of DTIC was reduced to 500 mg/m2.
Assessment of Response and Toxicity
All dogs had baseline complete blood count (CBC) and serum biochemistry panels performed immediately prior to starting treatment with Mitox/DTIC. Dogs were evaluated by a physical examination and CBC 7, 14, and 21 days after the first treatment. Serum biochemistry was performed every other treatment, or if clinically indicated.
Tumor response was determined at each examination by measuring lymph nodes directly with calipers. Response to therapy was categorized using previously established guidelines for defining complete response (CR; the disappearance of all clinical evidence of disease and no new sites of disease identified), partial response (PR; at least 30% reduction in mean sum of longest diameter of target node[s]), progressive disease (PD; at least 20% increase in mean sum of longest diameter of target node[s] or development of new lesion[s]), or stable disease (<30% reduction in mean sum of longest diameter of target node[s]) or <20% increase in mean sum of longest diameter of target node[s]).24 Response categories were required to persist for 21 days or more. Any response less than PR or a duration of <21 days was defined as no response.
Toxicity of the Mitox/DTIC protocol was determined based on the owner’s descriptions, results of physical examination findings, and clinicopathologic data. Toxicity was graded in accordance with the Veterinary Cooperative Oncology Group Common Terminology Criteria for Adverse Events.25 Hematologic nadirs were recorded based on the lowest reported value for each dog and each treatment. Nonhematologic toxicities were recorded as the maximum grade for a specific event for each treatment.
Statistical Analysis
Hematologic toxic effects were summarized by summary statistics, and hematologic nadirs were reported as a minimum value for each dog and each treatment. Nonhematologic toxic effects were summarized as a maximum grade for a specific type of event for each treatment. Factors examined for their potential influence on risk of developing grade 4 neutropenia post Mitox/DTIC included body weight and overall CR duration to standard chemotherapy with L-asparaginase, cyclophosphamide, doxorubicin, vincristine, and prednisone (L-CHOP).
Overall response rate was defined as the number of dogs achieving CR or PR, compared with the total number of dogs treated. CR and PR rates were defined as the number of dogs achieving CR or PR, compared with the total number of dogs treated. Response duration was calculated using the Kaplan-Meier method and was defined as the number of days from the first day of the Mitox/DTIC protocol until relapse for dogs who achieved CR or progression of disease for dogs who achieved PR. Dogs still in remission and dogs lost to follow-up were included in Kaplan-Meir analyses until the last day follow-up information was collected and then were censored. Overall survival was not evaluated because of confounding influences of euthanasia and the owner’s willingness to pursue other treatments. Responders (CR and PR) were compared with nonresponders with respect to weight, response (CR versus PR or no response) to L-CHOP, and overall L-CHOP CR duration (from start of L-CHOP chemotherapy to initiation of first rescue protocol).
Student t tests were used for analysis of continuous Gaussian data and Mann-Whitney U tests were used for continuous non-Gaussian data. The χ2 test of independence and Fisher exact test (when values were <5) were used for analysis of categorical variables. All analyses were two-sided, and P ≤ .05 was considered significant. The 95% confidence intervals (CI) were determined for response proportion and response duration. All statistical calculations were performed by use of a computer software programf.
Results
Study Subject Characteristics
Forty-four dogs with resistant peripheral nodal lymphoma treated with a combination of Mitox/DTIC were identified in the retrospective search of medical records. Thirty were males (24 neutered, 6 intact) and 14 were females (13 spayed, 1 intact). The median body weight was 35 kg (range 4–61.9 kg) and the median age was 7.8 yr (range 2.2–13.9 yr). Complete staging at the time of initial lymphoma diagnosis was available for 40 (91%) dogs. One dog (3%) was classified as stage 2, 10 dogs (25%) were classified as stage 3, 15 dogs (38%) were classified as stage 4, and 14 dogs (35%) classified as stage 5. Seventeen dogs (43%) were considered substage a, and 23 dogs (58%) were considered substage b. Immunophenotyping was available for 15 dogs (34%; (11 B-cell, 4 T-cell).
Previous Treatments
All dogs were treated with at least one chemotherapy protocol prior to receiving Mitox/DTIC. The median number of total drugs received prior to Mitox/DTIC was 6 (range 1–12). Forty dogs (91%) received an L-CHOP based protocol similar to that described by Garrett et al. as their initial treatment.26 In addition to L-CHOP drugs, some dogs received other agents in their initial protocol. Twelve dogs received methotrexate; five dogs received lomustine, mechlorethamine, and/or procarbazine; and two dogs received mitoxantrone substituted for doxorubicin. Both dogs relapsed before completing their initial L-CHOP protocol and were therefore considered resistant to mitoxantrone. The dose and number of mitoxantrone treatments each dog received was not available from the medical records. Four dogs were treated with single-agent doxorubicin as their initial protocol, two of whom received a single dose of vincristine (one with L-asparaginase) prior to doxorubicin.
Forty-one dogs (93%) responded to their initial protocol (CR 34, PR 7) for a median duration of 137 days (range 21–781 days). Median duration of response for dogs achieving a CR was 172 days (range 22–781 days).
At the time of first relapse, 18 dogs (41%) were reinduced with the same agent(s) as their initial treatment protocol. Five dogs did not respond to reinduction. The median duration of second CR/PR for the remaining 13 dogs was 34 days (range 22–181 days).
Twenty-eight dogs (64%) received Mitox/DTIC as their first rescue protocol. Sixteen dogs (57%) received at least one rescue protocol before Mitox/DTIC (lomustine [6]; mechlorethamine, vinblastine, procarbazine, and prednisone [4]; mechlorethamine, vincristine, procarbazine, and prednisone [3]; methotrexate and lomustine/DTIC [1]; doxorubicin [1]; vincristine [1]; and l-asparaginase [2]). The median number of rescue protocols was 1 (range 1–3). The median number of rescue drugs was 1 (range 1–5). Three dogs responded to their first rescue treatment (other than Mitox/DTIC) for 49, 93, and 368 days. Thirteen dogs did not respond to any rescue treatment before Mitox/DTIC. Three dogs did not respond to either their initial protocol or any rescue protocol before Mitox/DTIC. The median duration of time for all 44 dogs from first diagnosis of lymphoma to treatment with Mitox/DTIC was 166 days (range 27–1010 days).
Mitox/DTIC Treatment and Toxicoses
Eighty-seven Mitox/DTIC treatments were given to the 44 dogs. The median number of treatments was 1 (range 1–6 treatments). The number of treatments given to the dogs was as follows: 1 (n = 29), 2 (n = 4), 3 (n = 2), 4 (n = 3), 5 (n = 4), and 6 (n = 2). All dogs were initially treated at the planned Mitox/DTIC dosages. Dose reductions occurred in 7 dogs following the first treatment. Five dogs had mitoxantrone reduced to 4 mg/m2 because of neutropenia (grade 4 with fever, n = 1; grade 4 no fever, n = 1; grade 3 with fever, n = 1; and grade 3 no fever, n = 2). CBCs were available for 4 of these 5 dogs 7 days following their second, dose-reduced treatment. One dog had asymptomatic grade 2 neutropenia, and the remainder had normal neutrophil counts. Two dogs had mitoxantrone reduced to 4 mg/m2 and DTIC reduced to 500 mg/m2. Both of those dogs experienced grade 4 neutropenia, fever, and grade 3 anorexia and were hospitalized and recovered with supportive care. No hematologic or gastrointestinal toxicity occurred following the subsequent dose reduction in either dog.
Data on the hematologic toxic effects for dogs in the study is summarized in Table 2. The toxicoses represent the maximum grade observed for a specific dog after the first treatment. Neutropenia was the most frequent effect. Twenty-two dogs had CBCs performed at all planned time points (days 7, 14, and 21 following treatment). The median neutrophil nadir in those dogs was 901 cells/μL (range 48–6000 cells/μL). Neutrophil counts were available for 38 dogs 7 days posttreatment (median count 1310 cells/μL; range 40–28,593 cells/μL), 32 dogs 14 days posttreatment (median count 5491 cells/μL; range 1030–28,380 cells/μL), and 28 dogs 21 days posttreatment (median count 7507 cells/μL; range 2800–22,960 cells/μL).
The neutrophil nadir occurred on day 7 posttreatment in all but one case (grade 4 neutropenia noted day 11 posttreatment.) The neutrophil count in this dog on day 7 posttreatment showed grade 2 neutropenia (1100 cells/μL). Overall, 10 dogs (26%) developed grade 3 neutropenia, and 7 dogs (18%) developed grade 4 neutropenia. Four dogs developed fever (>103.0°F). Three dogs with fever had grade 4 neutropenia, and 1 dog with fever had grade 3 neutropenia. Each recovered with supportive treatments. Both dogs weighing less than 10 kg developed asymptomatic grade 3 neutropenia. There was no difference between dogs who developed grade 4 neutropenia and dogs who did not with regard to body weight (P = .51) or initial L-CHOP CR duration (P = .53; Table 3).
Platelet counts were difficult to assess because of the frequency of reported platelet clumps. Excluding cases in which platelet counts were <200,000 μL and clumps were observed (e.g., in which true grade of toxicity could not be determined), platelet counts were available for 12 dogs at all planned time points (days 7, 14, and 21). The median platelet nadir for these dogs was 201,000 μL (range 21,000–401,000 μL) and occurred at a median of 7 days after treatment (range 7–21 days). The median platelet count for all dogs (excluding cases in which platelet counts were <200,000 μL and clumps were observed [e.g., in which true grade of toxicity could not be determined]) with available CBCs at day 7 (n = 28) was 177,500 μL (range 34,000–753,000 μL), day 14 (n = 26) was 250,500 μL (range 21,000–570,000 μL), and day 21 (n = 24) was 300,500 μL (range 25,000–1,089,000 μL).
Adverse gastrointestinal effects occurred in 3 dogs. One dog experienced self-limiting vomiting during the DTIC infusion. Two dogs experienced grade 3 anorexia, and 1 also experienced grade 1 diarrhea. Both dogs simultaneously experienced grade 4 febrile neutropenia and were concurrently hospitalized for their signs and recovered with supportive care. The overall hospitalization rate was 5% (2/44 dogs).
Response to Treatment
Response to treatment with Mitox/DTIC was available for 42 dogs; 2 dogs were lost to follow-up after treatment one and were therefore considered nonresponders. The overall response rate was 34% (15/44; 95% CI 20–48%) for a median duration of 97 days (range 24–636 days, 95% CI 44–150 days). Eight dogs (18%, 95% CI 7–30%) achieved a CR for a median duration of 123 days (range 37–636 days, 95% CI 35–211 days). Five dogs were in CR after treatment one. Two dogs were in CR after treatment three. One dog was in CR based on palpation of peripheral lymph nodes after treatment one but had persistent lymphocytosis until treatment five when this resolved. This dog remained in a clinical remission (nodes and lymphocyte count) until being lost to follow-up at 636 days. Seven dogs (16%, 95% CI 5–27%) achieved a PR for a median duration of 56 days (range 24–112 days, 95% CI 20–92 days). All dogs achieved a PR following treatment one. Five dogs (4 CR, 1 PR) were censored from analysis for response duration after being lost to follow-up and lack of confirmation of relapse or disease progression. The median time to follow-up in these dogs was 112 days (range 63–636 days).
Phenotype was available for 4 of the 15 dogs who responded to Mitox/DTIC; 2 had B-cell lymphoma (1 CR, 1 PR), and 2 had T-cell lymphoma (both CR). Fourteen of the 15 (93%) responding dogs were resistant to L-CHOP chemotherapy and received Mitox/DTIC as their first rescue (5/6 CR, 9/9 PR). Five dogs experienced a longer remission duration to Mitox/DTIC than to their initial L-CHOP protocol. For those dogs, the median response duration to Mitox/DTIC was 84 days (range 42–636 days) versus 30 days (range 14–84 days) to L-CHOP. One of these dogs previously received mitoxantrone (initial remission duration 84 days, remission duration for Mitox/DTIC = 97 days.) The other dog who previously received mitoxantrone experienced a numerically similar remission duration to L-CHOP as it did to Mitox/DTIC (113 versus 112 days). Five of 7 dogs who developed grade 4 neutropenia responded to treatment (3 CR, 161, 123, and 37 days; 2 PR, 56 and 24 days). Two of these dogs also experienced gastrointestinal toxicity and neutropenia as described above.
There was no significant difference between responders and nonresponders with respect to weight (P = .075) or overall L-CHOP CR duration (P = .98). Dogs who achieved CR with a previous L-CHOP protocol were significantly more likely to respond to Mitox/DTIC than dogs achieving a PR or those with no response (23 versus 11%, P = .035; Table 4).
Discussion
The objective of this retrospective study was to determine the tolerability of the combination of mitoxantrone and DTIC in dogs with resistant peripheral nodal lymphoma. Results showed a combination of mitoxantrone at 5 mg/m2 and DTIC at 600 mg/m2 given intravenously every 3 wk with prophylactic antibiotics and antiemetics was well tolerated. Hematological toxicity, specifically neutropenia, was the most common adverse event.
The median neutrophil count at the nadir (1.310 × 109 cells/L) and frequency of grade 4 neutropenia (at least 18%) and febrile neutropenia (at least 9%) in this study is comparable to those of other DTIC containing protocols previously used to treat dogs with lymphoma. Nine of 71 dogs (12.6%) treated with doxorubicin or dactinomycin and DTIC were hospitalized because of neutropenic sepsis, including 1 death from treatment.17 Flory et al. evaluated the combination of lomustine and DTIC in dogs with resistant lymphoma.8 The median neutrophil count at the nadir was 1.275 × 109 cells/L, and the frequency of grade 4 neutropenia was 26%.8 In an earlier study by Van Vechten et al., 20% of dogs treated with doxorubicin and DTIC experienced severe neutropenia.2 Although the overall frequency of grade 3–4 neutropenia in this study was at least 45%, the risk of significant complications related to myelosuppression was low, with only 2 of 44 dogs (5%) requiring hospitalization and no dogs dying from treatment. Importantly, because of the retrospective design of the current study, it is possible that the frequency of myelosuppression is higher because follow-up laboratory data was missing in some dogs, further supporting a lack of substantial risk from treatment. Additionally, the prophylactic use of trimethoprim–sulfadiazine could have resulted in reduced complications secondary to hematological toxicity, as was observed in dogs with lymphoma or osteosarcoma treated with doxorubicin.27
Several studies have associated body weight with frequency of neutropenia.28–31 Specifically, smaller dogs are more likely to experience neutropenia than larger dogs when treatment is administered based on BSA dosing. In this study, grade 4 neutropenia was not associated with body weight. This may have occurred because the median body weight of dogs who developed grade 4 neutropenia in this study (18 kg) was increased compared with the cutoffs used in other studies (≤10–15 kg).28–31 Alternatively, this may represent variability in susceptibility to neutropenia related to undetermined characteristics of the population examined in this study. The degree of myelosuppression observed in this study could also be confounded by undocumented bone marrow involvement, degree of pretreatment with multiple drugs/protocols, or both. The duration of treatment to L-CHOP chemotherapy was considered a possible factor that could influence cytopenias in that it could be a surrogate for the quantity of chemotherapy received by the dog in question as well as the duration of time between treatment protocols. Both factors could theoretically influence long-term hematopoietic function, leading to increased risk for cytopenias following rescue treatment. As the duration of CR to overall L-CHOP protocol was not associated with development of grade 4 neutropenia in dogs treated with Mitox/DTIC, chronic, cumulative myelosuppression seems less likely. The lack of relationship between body weight and grade 4 neutropenia may also represent a type 2 error as a result of small sample size.
Previous studies report frequent adverse gastrointestinal effects with combination treatment with doxorubicin and DTIC. In one study of dogs with relapsed lymphoma, mild vomiting occurred following 50% of the treatment cycles, as did mild diarrhea (∼30%) and severe anorexia (8%).2 Finotello et al. reported one-third of dogs with hemangiosarcoma experienced gastrointestinal signs after treatment with doxorubicin and DTIC, although all instances were either grade 1 or 2.32 In both studies, a strategy to reduce toxicity was implemented, in which the planned dosage of DTIC was given over the course of several days, rather than during a single infusion.2,32 In two separate studies in which DTIC was given at a total dose of 800 mg/m2 as a single infusion along with doxorubicin, gastrointestinal toxicity was more pronounced.17,33 Grade 3 or 4 gastrointestinal toxicity occurred in 9 of 17 dogs with relapsed lymphoma, and 5 of those dogs were concurrently hospitalized for neutropenic sepsis.17 Twenty percent of dogs with hemangiosarcoma treated with doxorubicin, DTIC, and vincristine discontinued treatment because of chemotherapy-related toxicities.33 Although direct comparisons between protocols and among different tumor types cannot be made, in the current study, adverse gastrointestinal effects occurred in only 3 dogs (7%), with 2 of the affected dogs showing concurrent sepsis. This could be a result of a comparably lower total dose of DTIC given in combination with mitoxantrone, the use of prophylactic antibiotics, or both.27
Although response to treatment was not the primary endpoint of this study, the overall response rate of 34% is comparable to other rescue protocols for dogs with relapsed or resistant lymphoma.1–8 Many factors influence the choice of and sequence of rescue protocols, including owner-imposed restrictions relating to cost and/or time commitment and concern for side effects. As such, there is no universally agreed-upon recommended approach to rescue therapy in canine lymphoma. Acquired models of drug resistance indicate that although treatment can be successful in killing off a population of susceptible tumor cells, it also can promote unopposed growth of a subpopulation of resistant cells present in the original tumor.34 With time and subsequent drug therapy, resistance prevails. It is not surprising, therefore, that the majority (14 of 15) of dogs who responded to Mitox/DTIC received this protocol as first rescue. These dogs were classified as having resistant, but not yet refractory, lymphoma and perhaps more likely to respond to earlier protocols.35 Likewise, dogs who achieved a CR to their initial L-CHOP protocol were significantly more likely to respond to Mitox/DTIC than dogs achieving only a PR or NR (23 versus 11%, P = .035). This may also explain why response to Mitox/DTIC was not associated with overall CR duration to L-CHOP, as regardless of duration of time to relapse, the majority of responders were not heavily pretreated with chemotherapy, and therefore potentially less likely to express MDR.36
The inaccuracy of BSA dosing could lead to increased relative dose-intensity of drugs in smaller dogs, lending to more favorable response rates and durations to treatment.37 However, this hypothesis was not supported in this study, in which response to Mitox/DTIC was not associated with body weight. Further studies are necessary to support or refute the relationship between body weight and response to treatment in veterinary cancer patients.
It is interesting that both dogs who received mitoxantrone as part of their initial treatment experienced durable complete remissions to Mitox/DTIC (97 and 112 days) even though they were clinically considered to be resistant to single-agent mitoxantrone. Similar results were seen in 8 of 15 dogs who responded to doxorubicin/DTIC who had previously failed treatment with doxorubicin alone.2 Overexpression of permeability glycoprotein (P-gp) is a major contributing factor to the MDR phenotype.38–41 Lymphoma cells with the MDR phenotype are resistant to drugs that are substrates of P-gp including vincristine, doxorubicin, mitoxantrone, and prednisone.6 Alkylating agents are not substrates of the P-gp pump and rarely show cross-resistance, making them attractive options for treating dogs with resistant lymphoma.12 The combination of doxorubicin and dacarbazine was previously shown to have synergism against sarcomas and Hodgkin lymphoma in humans.42–47 One proposed mechanism of action is the drugs may affect cells at different times in the cell cycle. Although both drugs are considered cell cycle nonspecific, doxorubicin affects cell cycle metabolism primarily during S phase, whereas dacarbazine may affect cells in the G2 phase.48,49 Alternatively, dacarbazine, an alkylating agent, is not susceptible to P-gp-mediated cellular resistance, and therefore may eradicate those cells with a multiple drug-resistant phenotype. This could lead to increased responsiveness of remaining cells to doxorubicin.2 Whether such relationships exist between mitoxantrone and DTIC is unknown. Also, dogs in the present study may have responded only to DTIC in the Mitox/DTIC protocol, as the response rate to single-agent DTIC was numerically similar (35%) to that of combination Mitox/DTIC (34%).13 However, 93% of dogs treated with DTIC alone were partial responders, and the dose of DTIC in that study (800–1000 mg/m2) was higher than what was used in combination in this study.13 A randomized prospective study is necessary to accurately compare response rates and efficacy of single-agent DTIC with the combination of Mitox/DTIC.
There are several limitations to the current study, including those inherent to a retrospective analysis such as absent data and incomplete records. As previously mentioned, CBCs were not available for each dog at all planned time points; therefore, it is possible to have missed hematological nadirs and/or underestimated overall toxicity. Because of frequent clumping, it was difficult to accurately report platelet counts, although in most cases, platelet estimates indicated adequate counts were present. Response evaluation was not available for two dogs as they were lost to follow-up prior to day 21 post initial treatment. Patient population heterogeneity represents another limitation, including differences in initial treatment protocol and order of rescue protocols. Patients were not fully staged, either at the onset of diagnosis or at relapse, and phenotype was lacking for most patients, lending to possibly underestimated influences of either of these parameters on outcome.
Conclusion
The combination of mitoxantrone and DTIC with concomitant antibiotics and antiemetics is a safe and effective regimen to rescue dogs with resistant lymphoma. Phase 2/3 studies are necessary to determine if the combination Mitox/DTIC is superior to single-agent mitoxantrone or single-agent DTIC in the rescue setting for dogs with resistant lymphoma.
Tolerable-dose diagram illustrating the organ-specific DLT curves for various combinations of mitoxantrone and DTIC. DLTs of mitoxantrone were based on previously published results. DLTs of DTIC were based on retrospective review of cases treated at the Cornell University Hospital for Animals. The dose combination for the protocol used in this study is indicated by the asterisk (*). DLT, dose-limiting toxicity; DTIC, 5-(3,3-dimethyl-1-triazeno)-imidazole-4-carboxamide (dacarbazine).
Contributor Notes
