Individualized chemotherapy drug dose escalation in dogs with multicentric lymphoma

2.1 Dog selection

Colorado State University Institutional Animal Care and Use Committee and Clinical Review Board approval was acquired and signed informed consent was obtained from owners before enrollment. Dogs with newly diagnosed, high-grade, multicentric lymphoma confirmed via cytology or histopathology were prospectively enrolled into a cohort study from January 2018 to September 2020 at the Colorado State University Veterinary Teaching Hospital (CSU-VTH). All dogs were required to weigh greater than or equal to 9 kg, which was the minimal weight necessary for the volume of blood collected for pharmacokinetic analysis. All immunophenotypes were considered eligible. In addition, the owners had to plan to pursue chemotherapy with the CHOP protocol at CSU. Exclusion criteria included dogs that received any prior chemotherapy for multicentric lymphoma, including corticosteroids for greater than 7 days before starting the CHOP protocol. Dogs were also excluded if they had cardiac insufficiency that precluded them from receiving DOX, if they had a concurrent malignancy or other serious systemic disorder, or if they received homeopathic or alternative therapies within 1 day of the start of the study. Dogs receiving supplements, including chondroitin sulfate, vitamins, essential fatty acids, and glucosamine were considered eligible. All dogs had a complete blood count (CBC), chemistry panel, and urinalysis within 7 days of enrollment in the trial to screen for systemic disorders. Thoracic radiographs and abdominal ultrasound were encouraged if clinically indicated but not required. Dogs were assigned an approximate stage based on the World Health Organization clinical staging system.⁷

2.2 Treatment protocol

All chemotherapy agents were purchased from commercial vendors. CHOP chemotherapy was administered according to the 15-week protocol used at CSU (Table S1).⁸ The first cycle of CHOP was administered at standard doses (VCR 0.7 mg/m², CTX 250 mg/m², DOX 30 mg/m²). VCR and DOX were given IV, and CTX was given PO. Prednisone or prednisolone was prescribed per the protocol pending pharmacy availability. Oral furosemide was administered concurrently with CTX at 1-2 mg/kg. For dogs weighing less than 15 kg, DOX was dosed at 1 mg/kg. Blood samples for pharmacokinetic analysis were collected before treatment and at varying timepoints after treatment based on known pharmacokinetic parameters of each drug. VCR samples were obtained 5 minutes, 15 minutes, 30 minutes, 1 hour, 2 hours, and 4 hours after dosing. CTX samples were collected 15 minutes, 30 minutes, 1 hour, 2 hours, and 4 hours after dosing. DOX samples were collected 5 minutes, 45 minutes, and 1 hour after dosing. (Table 1).

Doses of drugs that did not cause dose-limiting adverse effects (AEs) were increased for subsequent treatments according to a standardized escalation protocol, and dose reductions were performed in accordance with a specified protocol (Tables 2 and 3). Once a dog required a dose reduction for a drug, no further dose escalation was implemented for subsequent doses of that drug. Supportive medications for treatment of AEs were given at the discretion of the prescribing clinician and were not limited by the study. Lymph node biopsies were collected before starting chemotherapy and at the time of relapse for biobanking. Blood samples (whole blood and serum) were also collected at the time of each VCR treatment and monthly after completion of the CHOP protocol for biobanking. Complete blood counts were performed before each chemotherapy treatment, as well as at the expected drug nadir after any dose adjustment. In addition, an owner quality of life survey and measurement of target and nontarget lesions were performed at every visit to assess response to therapy.

2.3 Response and toxicosis evaluation

Response to chemotherapy was assessed using the VCOG response evaluation criteria for peripheral nodal lymphoma in dogs.⁹ Dogs were considered off-study if there was evidence of disease progression, if AEs occurred that warranted discontinuation of chemotherapy, or in the instance of death by any cause. If a dog died while enrolled in the study, a necropsy exam was requested but not required.

All AEs were recorded and graded according to the VCOG CTCAE v1.1.¹⁰ This version of VCOG was utilized because of available published material at the start of enrollment. AEs were determined via owner reported history, clinical abnormalities at time of evaluations, and clinicopathologic assessment.

2.4 Pharmacokinetic analysis

DOX, CTX, and 4-OH-cyclophosphamide were analyzed by LC/MS/MS as previously described.^{11, 12} A detailed description of the methods for pharmacokinetic analysis can be found in the supplemental material.

2.5 Pharmacokinetic calculations and analysis

2.6 Statistical analysis

Assuming a 50% hypothetical escalation percentage, 30 dogs were enrolled to allow for calculation of the observed escalation percentage with 37.5% variability, equating to 95% confidence intervals of 31.3%-68.7%. The number of dogs that were able to be successfully dose escalated for at least 1 drug were expressed as a percent of total dogs enrolled. Continuous data were reported as median and ranges, and categorical data were expressed as frequencies and percentages. Progression-free interval (PFI) and overall survival (OS) were calculated from date of treatment initiation to the date of PD or death, respectively. Overall response rate (ORR) was calculated as a percentage of dogs that had a partial response (PR) or complete response (CR). Drug C_max and area under the curve (AUC) were compared in dogs escalated or not escalated for each drug via a 2-tailed, unpaired T test or Mann-Whitney test depending on data normality. Body weight and ability to escalate each drug was evaluated using a 2-tailed, unpaired T test or Mann-Whitney test depending on data normality. Change in neutrophil count post treatment was correlated with AUC and C_max via simple linear regression. Dogs were censored if they were withdrawn or lost to follow-up before PD occurred. The Kaplan-Meier methodology was used to estimate and display the distribution of PFS and OS, and cohorts were compared using logrank analysis. Variables with values of P ≤ .05 were considered significant. All statistical analysis was performed with a commercial software package (Prism v9, GraphPad Software, La Jolla, California).

3.1 Study sample

Thirty dogs were prospectively enrolled in this study. Numerous breeds were represented, including Mixed Breed (14), Labrador retriever (4), Boxer (3), Tibetan terrier (1), American Pit Bull Terrier (1), Bernese Mountain Dog (1), Boykin spaniel (1), Goldendoodle (1), Greyhound (1), Labradoodle (1), Rottweiler (1), and Yorkshire terrier (1). Information regarding age, weight, sex, approximate stage, substage, and immunophenotype is shown in Table 4.

3.2 Dose escalation

In total, 23 dogs were eligible to dose escalate. Five of the 7 dogs not escalated were withdrawn before escalation because of disease progression within the first cycle at standard chemotherapy doses. One dog was euthanized 1 week after the first dose of VCR after development of a fever and swollen joint suspected to be secondary to lymphoma-induced systemic vasculitis. This was not considered to be secondary to the dose of VCR. One dog was not escalated because of persistent thrombocytopenia and panleukopenia. Of those with the opportunity to escalate, 78% (18/23) were successfully escalated in at least 1 drug during the CHOP protocol. A total of 63 dose escalation events occurred: 21 for VCR, 27 for CTX, and 15 for DOX. Vincristine was successfully escalated to 0.8 mg/m² in 3 dogs, 0.9 mg/m² in 6 dogs, and 1.0 mg/m² in 2 dogs. Cyclophosphamide was successfully escalated to 300 mg/m² in 8 dogs, 350 mg/m² in 5 dogs, and 400 mg/m² in 3 dogs. Doxorubicin was successfully escalated to 35 mg/m² in 5 dogs, 40 mg/m² in 1 dog, 45 mg/m² in 1 dog, 1.4 mg/kg in 1 dog, and 1.6 mg/kg in 1 dog. No further dose escalation done in any dog beyond 1.0 mg/m² (VCR), 400 mg/m² (CTX), and 45 mg/m²/1.6 mg/kg (DOX) as the dogs had reached the end of their treatment protocols.

3.3 Adverse events

All 30 dogs were available for AE assessment. Neutropenia was the most common dose-limiting toxicosis for all drugs. The most frequently observed AEs and severe AEs are available in Table 5 and Table 6, respectively; a complete list of AEs is in Table S2. The highest reported AEs were grade 4 in 10 dogs and were associated with neutropenia and thrombocytopenia. One of these occurred after standard dosing of VCR. Grade 4 neutropenia occurred after VCR in 2 dogs (0.7 mg/m² [1]) and 0.9 mg/m² [1] dose), for CTX in 4 dogs (300 mg/m² [3] and 400 mg/m² [1] dose), and for DOX in 3 dogs (31.5 mg/m² [1], 35 mg/m² [1], and 40 mg/m² [1] dose). One dog had a documented fever associated with the grade 4 neutropenia. One dog also experienced grade 4 thrombocytopenia after 35 mg/m² of DOX. Two dogs experienced grade 3 diarrhea after 30 mg/m² and 35 mg/m² of DOX, resulting in subsequent dose reductions. One episode of grade 3 vomiting, 2 episodes of grade 3 anorexia, 1 episode of grade 3 ileus, and 1 episode of grade 3 hematochezia were also documented. All other GI toxicosis was grade 1 or 2. In addition, 1 dog had grade 3 sterile hemorrhagic cystitis after 400 mg/m² of CTX, despite cotreatment with furosemide. One dog developed a fever and swollen joint after his first dose of VCR. No dogs died as a result of drug toxicosis, and no dogs were withdrawn from the study secondary to AEs.

Of the dogs that had the opportunity to dose escalate, 4 dogs (17%) were hospitalized. Of these, 1 dog was hospitalized because of a grade 4 neutropenia after DOX at 31.5 mg/m², 1 because of a grade 3 neutropenia after DOX at 35 mg/m², and 1 because of a grade 3 neutropenia after VCR at 1.0 mg/m². When the VCR dog was hospitalized, his neutropenia was resolving; however, he developed nausea and hyporexia and was hospitalized for supportive care. The remaining dog was hospitalized for acute vomiting, lethargy, hyporexia, and defecating foreign material. This was suspected to be secondary to foreign body ingestion rather than a consequence of chemotherapy, as it occurred 13 days after DOX administration. Three additional dogs were hospitalized after standard dosages of chemotherapy. Of these, 1 was hospitalized and eventually euthanized after the first dose of vincristine after development of a fever and swollen joint suspected to be secondary to lymphoma-induced systemic vasculitis, as previously mentioned. Findings on necropsy examination included profound fibrinoid and leukocytoclastic vasculitis affecting numerous small and medium caliber vessels in the subcutis of the right forelimb. While the exact etiology of this cutaneous vasculitis was unclear, it was considered possible that vascular infiltration and destruction by neoplastic cells incited a systemic vasculitis in this dog based on the evidence of similar fibrinoid vascular necrosis in sections of the mandibular lymph node associated with cuffing of those vessels by neoplastic lymphocytes.

3.4 Dose reductions/delays

Thirty-nine dose reductions occurred across all dogs. Vincristine was dose-reduced 14 times, CTX was dose-reduced 12 times, and DOX was dose-reduced 13 times. Vincristine was dose-reduced in 12 dogs (40%), CTX in 11 dogs (36%), and DOX in 13 dogs (43%). The average dose reduction was 15% for VCR, 15% for CTX, and 10% for DOX. The most frequent reason for dose reduction across all agents was neutropenia, occurring 31 times (80% of dose-reductions). The other catalyst for reduction was GI toxicosis in 8 instances (20% of dose-reductions). Dose delays occurred in 16 dogs (53%). The average duration of dose-delay was 7.3 days (range, 4-14 days). Dose delay was most commonly attributed to neutropenia.

3.5 Pharmacokinetics

DOX, CTX, and VCR plasma concentrations were assessed at designated timepoints after administration. DOX C_5min was significantly lower in dogs that could be successfully escalated (P = .03). The VCR C_5min was significantly correlated with the magnitude of neutrophil change (P = .01). Drug C_max was not significantly correlated with the ability to escalate for CTX or VCR nor with magnitude of neutrophil change for CTX or DOX. AUC was not significantly correlated with magnitude of neutrophil change. Body weight was not correlated with the ability to escalate each drug.

3.6 Outcomes

The ORR was 100%, with 19 (63%) dogs achieving a CR, and 11 (37%) dogs experiencing a PR. The median PFI was 171 days (range, 21-332 days; 203.5 for B cell, 83 for T cell). The median OST was 254 days (range, 6-629 days; 260 for B cell, 222 for T cell). One dog was censored from survival analysis at 213 days, as the dog was lost to follow-up after a single rescue dose of CCNU/L-asparaginase. Three other dogs were censored, as they were alive at the time of manuscript submission, with follow-up times of 866, 1025, and 1155 days. Twenty-five dogs were withdrawn from study because of disease progression. Two were withdrawn because of death or euthanasia. The dog that died was suspected to have a fatal arrhythmia, which was suspected to be secondary to chronic active interstitial pneumonia that spread to the left atrium based on necropsy examination. The other dog was withdrawn after euthanasia was elected because of systemic vasculitis after his first dose of VCR. One dog was withdrawn because of excessive myelosuppression. Two dogs are still in remission after completing their initial CHOP protocol.