Phase II Clinical Trial of Combination Chemotherapy With Dexamethasone for Lymphoma in Dogs
Dogs with histologically confirmed lymphoma were treated with a 14-week induction chemotherapy protocol that included dexamethasone. A phase II clinical trial was done using a standard two-stage design. Complete remission occurred in 21 (88%) dogs, with a median initial progression-free interval of 186 days. Toxicity was mild and self-limiting in the majority of dogs.
Introduction
Lymphoma is a commonly diagnosed malignancy in dogs.1 Canine lymphoma is typically responsive to chemotherapy, and many protocols have been used to treat the disease over the past two decades. The most successful protocols are composed of multiple drugs given sequentially and have historically included a protracted phase of maintenance chemotherapy.2–15 High-intensity induction chemotherapy in humans with lymphoma provides improved survival without requiring the use of maintenance chemotherapy.16–19 Induction chemotherapy without maintenance chemotherapy in dogs with lymphoma has provided comparable complete remission rates and initial progression-free intervals (PFIs) when compared to other conventional protocols.4,10,12,20–23 Combination chemotherapy protocols including dexamethasone have been successful at inducing durable remissions in humans with lymphoid malignancies.19,24–32 The purpose of this study was to evaluate the remission rate and length of the initial PFI of an induction cyclophosphamide, vincristine, doxorubicin, and prednisone (COAP)-like chemotherapy protocol, with dexamethasone replacing the prednisone, for the treatment of canine lymphoma.
Materials and Methods
Study Subjects
Twenty-four client-owned dogs were included in a prospective study from June 2003 to June 2004. Dogs were eligible for the study if they had histologically confirmed lymphoma on lymph node biopsy and had no cutaneous, alimentary, mediastinal, or nervous system involvement. Other inclusion criteria included informed owner consent and no previous treatment with cytotoxic drugs. All dogs were required to undergo a complete clinical staging evaluation, including a complete blood count (CBC), serum biochemical profile, urinalysis, thoracic radiography, abdominal ultrasonography, measurement of peripheral lymph nodes, and bone marrow aspiration and cytology. A clinical stage was assigned to each dog based on the World Health Organization (WHO) system for canine lymphoma.33
Study Design
This prospective study was designed as a standard, two-stage, phase II clinical trial to evaluate efficacy.34 All dogs were treated with a 14-week chemotherapy protocol [Table 1] that included cyclophosphamide,a vincristine,b doxorubicin,c and dexamethasone.d Dogs were evaluated by physical examination and CBC before administration of each dose of injectable chemotherapy. The study was initially designed for 14 dogs, and if more than two complete responses were observed (i.e., >20% response rate), an additional 10 dogs would be added to the study to better estimate the response rate.34 If the first five dogs had <20% complete response, the study would be abandoned.
Complete remission was defined as the resolution of all evidence of lymphoma. Partial remission was defined as ≥ 50% decrease in lymph node volume. Stable disease referred to <50% change in lymph node volume. Progressive disease indicated a ≥ 50% increase in lymph node volume. The PFI was defined as the interval between the date of the first dose of chemotherapy and the date of first relapse. Survival time was defined as the interval between the date of first dose of chemotherapy and date of death.
Immunophenotyping was performed on lymph node cytology and histopathological samples when available. If dogs had prior steroid usage, the cumulative dose and length of steroid treatment prior to chemotherapy were recorded.
Toxicity associated with this protocol was assessed using the National Cancer Institute criteria for chemotherapy toxicity, and levels ranged from grade 0 (i.e., no toxicity) to grade 4 (i.e., severe neutropenia or intractable vomiting or diarrhea).3 If more than a grade 1 (mild) toxicity occurred, the dose of the offending drug was reduced by 20% for all subsequent treatments. Potential side effects of dexamethasone (e.g., polyuria, polydipsia, polyphagia, anorexia, vomiting) were also recorded.
After completion of the protocol, the dogs were examined monthly until relapse. Relapse was defined as enlarged lymph nodes with cytological or histopathological evidence of lymphoma. Response to rescue chemotherapy and type of rescue chemotherapy were noted, along with the cause of death for each dog.
Statistical Analysis
All dogs were included in all statistical analyses once they began the chemotherapy protocol. Prognostic factors examined in this study included gender, prednisone administration prior to the chemotherapy protocol, clinical stage and substage, and immunophenotype of the lymphoma. The product-limit method was used to generate estimated response curves, and these curves were compared between groups using the log-rank test.35,36 Cases were censored if the dogs developed another disease that required additional chemotherapy, were lost to follow-up, or were still alive at the time of data analysis. Data analysis occurred when the PFI was reached in all dogs. All statistical tests were done using commercial software.e Values of P <0.05 were considered significant. All data are expressed as means ± standard deviation unless otherwise noted.
Results
Twenty-four dogs were enrolled in the clinical trial. They included castrated males (n=6), spayed females (n=17), and one intact female dog. The male-to-female ratio was 0.33. The mean age of the dogs was 8.5±2.38 years (range 3.8 to 13.1 years), and mean weight was 29.4±13.09 kg (range 3.0 to 52.9 kg). The study population included mixed-breed dogs (n=4), basset hounds (n=3), golden retrievers (n=3), German shepherd dogs (n=2), and 12 other pure breeds (n=1 each). The majority of dogs had advanced disease. Seven dogs had stage 5 lymphoma (one had substage a; six had substage b). Thirteen dogs had stage 4 lymphoma (four had substage a; nine had substage b). Two dogs had stage 3a disease; one dog had stage 2a; and one dog had stage 1a lymphoma. Clinical stage (P=0.504), clinical substage (P=0.992), and gender (P=0.719) were not significantly associated with initial PFIs.
Complete remission occurred in 21 of the dogs, and the remaining three dogs were classified as partial remissions. Of the three dogs that had partial remission, one was euthanized 2 weeks into the study based on the owner’s request (personal reasons). One of the other dogs was considered a partial remission after 6 weeks of the protocol, and this dog was switched to another chemotherapeutic protocol in an attempt to induce a complete remission. The third dog completed the protocol and was treated with maintenance chemotherapy. The latter two dogs were considered failures at the time of chemotherapy protocol deviation. The overall median PFI was 186 days [Figure 1], with a median estimated survival time of 294 days for all dogs [Figure 2]. One dog in complete remission was lost to follow-up at 219 days. One dog developed tonsillar squamous cell carcinoma at 126 days and was censored when started on carboplatin. Mean follow-up time for all dogs was 289 days (range 13 to 542 days). Of those dogs lost to follow-up (n=1) or still alive at data analysis (n=2), 19 were euthanized because of progression of the lymphoma, one dog was euthanized because of renal failure, and one dog died from tonsillar squamous cell carcinoma.
Immunophenotyping was done on 19 cases. B-cell lymphoma was identified in 16 of the dogs, and T-cell lymphoma was seen in three dogs. Five dogs had no tissue specimens available for immunophenotyping. Comparison of immunophenotyping was not possible because of the low numbers of T-cell lymphomas. Seven dogs had received prednisone prior to beginning the chemotherapy protocol. Cumulative pretreatment prednisone exposure in those dogs ranged from 0.46 to 212.80 mg/kg. Median duration of the prednisone was 7 days (range 2 to 1460 days). Prednisone administration (P=0.680) before chemotherapy was not significantly associated with the initial PFI.
Toxicity was observed after 38 of 288 doses of chemotherapy [Table 2]. The majority of adverse events were mild and self-limiting and did not result in treatment delays, dosage reductions, or hospitalizations. Dogs with mild gastrointestinal toxicity (grade 1) were easily managed by dietary modification (increased fiber in the diet). Gastrointestinal or hematological toxicity was responsible for 36 of the 38 events, and sterile hemorrhagic cystitis was responsible for the remaining two events. Ten cyclophosphamide-related events resulted in a dosage reduction and treatment delay. One dog required hospitalization after cyclophosphamide because of neutropenia, and two dogs developed sterile hemorrhagic cystitis. Of the latter two dogs, one developed sterile hemorrhagic cystitis at the end of the study and required no protocol modification, and the other developed cystitis after the first dose of cyclophosphamide and was given chlorambucilf instead for the remainder of the study. Only four vincristine treatments resulted in dosage reductions or drug substitutions. Two dogs required a 20% dose reduction because of a grade 3 toxicity, and they successfully completed the protocol on the reduced dose. One of the dogs required hospitalization following vincristine treatment because of paralytic ileus. In three dogs with gastrointestinal toxicity, vinblastineg (2 mg/m2 intravenously) was substituted for vincristine. Five dogs had gastrointestinal side effects during the time of both vincristine and dexamethasone administration. Mild, self-limiting vomiting or diarrhea occurred in six dogs after doxorubicin administration, and one dog required a dosage reduction because of neutropenia. None of the dogs required hospitalization for adverse events from doxorubicin.
Of 288 total doses of chemotherapy administered during the study period, only two adverse events resulted in hospitalization. No protocol-related deaths occurred. Only two dogs required dosage reduction or substitution of more than one cytotoxic drug. The dexamethasone was well tolerated by all dogs. Mild to moderate polyuria (n=20), polydipsia (n=18), or polyphagia (n=5) occurred from the dexamethasone.
Of the 22 dogs that relapsed, treatment was continued in 11 dogs using various chemotherapeutic protocols, and 11 were not treated based on the owners’ wishes. All 11 of the dogs that received subsequent chemotherapy had finished the initial induction protocol. Eight of the dogs were retreated with this same protocol, which resulted in a second complete remission. Two other dogs were treated with a similar protocol, which resulted in one partial remission and one complete remission.22 The remaining dog was treated with only lomustine,h which resulted in a partial remission.
Discussion
The median PFI in this study approached that of some previous reports but fell short of others.2–4,7,9–12,15,21–23,37–41 The median PFI may have been shorter because most dogs in the study population had advanced (stage 4 and 5) disease, which may have carried a poor prognosis.2,15,21,23,38,40,42 It was also possible that dexamethasone may have shortened remission times, owing to upregulation of protective pathways. Because the PFIs and survival times observed in the present study were shorter than in other published protocols, substituting dexamethasone for prednisone may have a detrimental effect on PFI and survival.43 Ideally, efficacy comparisons among various lymphoma protocols should be investigated in a randomized, phase III clinical trial.
The toxicities observed in this study did not differ from those in previous studies on canine lymphoma.2,3,5–7,10,12–14,21–23,37,38,41,44–46 The chemotherapy-related adverse effects encountered with this protocol were easily addressed by dosage reductions, drug substitutions, and dietary management. Clinical effects of the dexamethasone may have been tolerable by the owners because of the short and discontinuous schedule of administration. Dexamethasone was not discontinued in any dogs because of side effects. Gastrointestinal side effects did occur in a few dogs during the weeks of vincristine and dexamethasone administration; however, paralytic ileus was diagnosed in those dogs, which suggested that vincristine—not dexamethasone—was the cause of the clinical signs.
All dogs that were re-treated with this protocol at signs of first relapse achieved a second complete remission, which was comparable to previously published rescue protocols.23,38,47–54 Initial administration of dexamethasone did not seem to decrease the response to rescue chemotherapy; however, the observed success of this second induction may have been related to the small number of dogs treated.
Conclusion
Dogs with lymphoma were entered into a phase II clinical trial using a short-duration, COAP-like protocol, with dexamethasone substituted for prednisone. The complete remission rate was 88%; however, the median PFI was only 186 days. Toxicity was mild and self-limiting. All dogs reinduced with the same protocol at relapse achieved a second complete remission. Based on these results, this protocol was effective for inducing first and second remissions in dogs with lymphoma and may represent another viable treatment option. The effect of dexamethasone on PFI and survival time is currently undefined and requires further study.
Cytoxan; Bristol-Myers Squibb, Princeton, NJ 08543
Vincristine; Faulding Pharmaceutical, Paramus, NY 07652
Doxorubicin; GensiaSicor Pharm, Irvine, CA 92618
Dexamethasone; Roxane Laboratories, Columbus, OH 43216
SigmaSTAT 3.0; SPSS Inc., Chicago, IL 60606
Leukeran; GlaxoSmithKline, Research Triangle Park, NC 27709
Vinblastine; Bedford Laboratories, Bedford, OH 44146
CeeNU; Bristol-Myers Squibb, Princeton, NJ 08543
Acknowledgments
The authors thank Amalia deGortari, MVZ, and Rosanne Thomas, RVT, for their technical support.
Citation: Journal of the American Animal Hospital Association 43, 1; 10.5326/0430027
Citation: Journal of the American Animal Hospital Association 43, 1; 10.5326/0430027
Initial progression-free interval for dogs (n=24) with lymphoma treated with a short induction chemotherapeutic protocol. The median progression-free interval was 186 days. The dots represent dogs that were censored (n=2). One dog in complete remission was lost to follow-up at 219 days, and one dog was censored for starting carboplatin for tonsillar squamous cell carcinoma.
Kaplan-Meier survival curve for dogs (n=24) with lymphoma treated with a short induction chemotherapeutic protocol. The median survival time was 294 days. The dots represent dogs that were censored (n=3). One dog in complete remission was lost to follow-up at 219 days, and two dogs were still alive at the time of data analysis.
Contributor Notes
