• Open Access

Phase I Study to Determine the Maximal Tolerated Dose and Dose-Limiting Toxicities of Orally Administered Idarubicin in Dogs with Lymphoma


  • DM Vail and BD Husbands contributed equally to this work.

Corresponding author: Aoibhinn McDonnell, Pfizer Ltd, Ramsgate Road, Sandwich, Kent, CT13 9NJ, UK; e-mail: Aoibhinn.McDonnell@pfizer.com.



Idarubicin, a PO bioavailable anthracycline antibiotic-class chemotherapeutic, could have substantial convenience advantages over currently available similar class agents in use that require IV delivery.


The primary objective of this study was to determine the maximally tolerated dose (MTD), dose-limiting toxicities (DLTs), and basic pharmacokinetic parameters of oral idarubicin exposure in dogs with lymphoma after a single oral dose. A secondary objective was to document preliminary antitumor efficacy in an expanded treatment cohort using the established MTD.


Client-owned dogs with measurable lymphoma.


Dogs (n = 31) were enrolled in a prospective open label phase I study of oral idarubicin. By means of a 3 + 3 cohort design, dose escalations were made with 3 dogs per dose level, and the MTD was established based on the number of patients experiencing a DLT. Plasma concentrations of idarubicin and idarubicinol were determined by postdose sampling. Assessment of antitumor efficacy focused on evaluation of accessible, measurable lymph nodes and skin lesions by modified RECIST guidelines.


The MTD in dogs > 15 kg body weight was 22 mg/m2. Adverse hematologic events (neutropenia and thrombocytopenia) were the predominant DLT and generally correlated with higher plasma concentrations of idarubicin and idarubicinol.

Conclusions and Clinical Importance

PO administered idarubicin was generally well-tolerated and had preliminary antitumor activity in dogs with lymphoma. Furthermore, the potential clinical advantage of a safe and efficacious oral anthracycline alternative supports further investigations of this agent in repeated-dose, randomized clinical trials.


maximal tolerated dose


dose-limiting toxicity


multidrug resistance


upper limit of normal


Veterinary Co-operative Oncology Group Common Terminology Criteria for Adverse Events


adverse event


alkaline phosphatase


alanine aminotransferase


progressive disease


complete response


partial response


stable disease

Anthracycline antibiotic-class cytotoxic chemotherapeutics, in general, and doxorubicin, in particular, are the mainstay of treatment for dogs with multicentric lymphoma, either alone or in combination with other agents.[1] Idarubicin is a 4-demethoxy-anthracycline analogue of daunorubicin that is approved for oral use in humans for the treatment of acute nonlymphoblastic leukemia (when IV administration cannot be employed) and as a single agent in the treatment of advanced breast cancer after failure of first-line chemotherapy.[2] Several studies have compared idarubicin with doxorubicin-based treatment protocols in people with various tumor types; some report equivalent antitumor activity whereas others report inferior activity for idarubicin.[3-5]

Idarubicin has several theoretical advantages when compared with doxorubicin in patients who may benefit from an anthracycline-class chemotherapeutic. Although these advantages only have been explored in trials in human patients and not in veterinary species, they include less clinical and echocardiographically identified cardiotoxicity,[6] high lipophilicia which allows blood brain barrier passage,[7] superior bioavailability in the gastrointestinal tract,[8] and less cellular efflux by the multidrug resistance (MDR) efflux pump (P-glycoprotein) compared with doxorubicin or daunorubicin,[9] which might improve antitumor efficacy and allow use in some patients with doxorubicin-resistant tumors.[10]

As a prelude to investigations of comparative efficacy and safety of oral idarubicin, the phase I trial reported here was initiated to determine the MTD, DLT, and adverse event profile in companion dogs. A secondary objective of this study was to document preliminary antitumor efficacy in an expanded treatment cohort using the MTD established after completion of the 1st objective.

Materials and Methods

Patient Selection

Client-owned dogs presented to the teaching hospitals at the University of Wisconsin-Madison and the University of Minnesota with previously untreated or relapsed lymphoma underwent initial screening followed by procurement of signed owner informed consent. All enrolled dogs had histologic confirmation of lymphoma; untreated cases by histologic assessment of surgical biopsy samples (excised lymph node) or, in the case of relapsed patients, fine needle aspirate cytology at the time of study entry to confirm recurrence of previously histologically confirmed lymphoma. Inclusion criteria included dogs ≥ 1 year of age, ≥ 15 kg body weight, with measurable Stage II–V, substage a lymphoma.[11] Immunophenotyping was not a requirement for entry and the minimal requirements for staging were physical examination, CBC, serum biochemistry, and urinalysis. Additional tests such as thoracic radiographs, abdominal ultrasound examination, and bone marrow biopsy were not a requirement of this protocol and were performed at the discretion of the investigator, depending on the individual case requirement. Dogs were required to have adequate organ function as indicated by standard laboratory tests (CBC, serum biochemistry, and urinalysis) on Day 0. Specifically, dogs were required to have an absolute neutrophil count ≥ 1,500 cells/μL, hematocrit > 25%, platelet count ≥ 100,000/μL, alanine transaminase activity ≤ 4 times the upper limit of normal (ULN), and serum creatinine concentration ≤ 2 times the ULN; lethargy/fatigue status (Veterinary Co-operative Oncology Group Common Terminology Criteria for Adverse Events [VCOG-CTCAE] version 1.0)[12] of either 0 or 1; and life expectancy of at least 1 month. Patients were excluded from trial participation if they had received prior radiation treatment, immunotherapy at any point, or chemotherapy with or without glucocorticoid treatment less than 2 weeks before trial entry; showed evidence of overt cardiac abnormalities at screening; had serious systemic medical conditions; or were pregnant or lactating females. Dogs that met all inclusion criteria and for which none of the exclusion criteria applied were admitted to the study.

Trial Design

The clinical protocol was approved by the contributing institutions’ Animal Care and Use Committees. This study was conducted as a prospective open label phase I dose-cohort (3 + 3) escalation design[13] that investigated the MTD and DLTs occurring over a 21-day period after a single PO dose of idarubicin. Dogs enrolled in the study received 1 PO dose of idarubicin on Day 0 with a starting dosage of 12.5 mg/m2 (cohort 1; Table 1), which was based on normal laboratory dog data (Pfizer internal data, not shown). Dose escalation was carried out wherein dose escalations were made with 3 dogs per dose level, and the MTD was established based on the number of patients experiencing a DLT. A DLT was defined as grade ≥ 3 for any adverse event (VCOG-CTCAE version 1.0)[12] with the exception of hematologic toxicity, where grade 3 neutropenia or thrombocytopenia that resolved before the next scheduled treatment was not deemed dose-limiting. Any grade 4 neutropenia or thrombocytopenia was deemed a DLT. If no DLTs were observed in the first cohort of dogs after the 21 day assessment, a second cohort of 3 dogs was treated at the next higher dose cohort. If a DLT was observed in 1 dog, the cohort was expanded to 6 dogs. If ≥ 2 DLTs were noted in any initial or expanded cohort, no further dose escalations were performed and the maximally tolerated dose (MTD) was considered to have been exceeded. The interval between dose escalations was 21 days (ie, a new cohort of patients was not started until 21 days after the last dog in the previous lower dose cohort was dosed). There was no intrapatient dose escalation in this study. Once the MTD was established, this cohort was expanded to include an additional 7 dogs for a total of 13 dogs treated at MTD. Table 1 summarizes dosing cohort information.

Table 1. Scheduled idarubicin treatment cohorts.
CohortDose (mg/m2) Dogs TreatedNumber of Dogs Experiencing Dose-Limiting Adverse EventsResponse Observed (Day 21)
  1. a

    Established as the maximum tolerated dose.

  2. b

    3 of 6 dogs in cohort 2 completed the 21-day visit.

  3. c

    10 of 13 dogs in cohort 4 completed the 21-day visit.


Clinical observations and sample collections were made on 3 visits during the study period: Day 0, Day 7 (±1 day), and Day 21 (±2 days). Blood samples were collected for serum biochemistry (pretreatment, Day 7, and Day 21), hematology (pretreatment, Day 7, and Day 21) and pharmacokinetic (PK) plasma analysis (predose, 2 and 4 hour, postdose). These plasma sampling times were chosen because they were near Tmax,obs for idarubicin (2 hour), and the equally active metabolite idarubicinol (4 hour), in normal laboratory dogs (Pfizer internal data, not shown). Urine was collected for urinalysis twice during the study period (pretreatment and Day 21). Body weight was recorded on Day 0 and on Day 21. Clinical observations involved physical examination and evaluation of clinical response to treatment. At Day 21, postdosing, dogs were removed from protocol and offered continued antineoplastic treatment (available standard of care) at trial-subsidized expense. Clients were allowed to withdraw from the study before 21 days for disease progression or other signs of ill health, at the discretion of the investigator, owner, or both.

Drug Dosing

Dogs were to be given a small meal and dosed immediately after or within 1 hour after feeding. If the animal refused to eat, treatment still was administered as prescribed. Three oral capsule dose-sizes were employed: 1, 5, and 10 mg capsules (5 and 10 mg capsules were commercially available whereas the 1 mg capsule was prepared by Pfizer Animal Health using the 5 mg capsule blend). Animals were dosed according to protocol dose charts provided to ensure dose within ± 20% accuracy of the targeted mg/m2 dose range.

Pharmacokinetic Plasma Analysis Methods

Idarubicin and idarubicinol (the main active metabolite of idarubicin)[14] were quantified in plasma samples by means of an LC-MS/MS method. Standards and QC samples were prepared by diluting idarubicin and idarubicinol stock solutions in pooled control beagle dog plasma.a Sample preparation was accomplished by the Hamilton Microlab (STAR robotic pipetting workstation (Hamilton Company, Reno, NV)) and quantitation of analyte in plasma samples was accomplished with an LC-MS/MS assay. Plasma standards were prepared over a concentration range of 0.1–2,000 ng/mL in control dog plasma from working standard solutions prepared in 90 : 10 (v : v) methanol : water. A 50 μL aliquot of standard or sample plasma was added to 96-well polypropylene cluster tubesb containing 200 μL of internal standard solution (epirubicin at 200 ng/mL in acetonitrile) and 50 μL of control dog plasma or 90 : 10 methanol : water. Samples were vortexed on a platform shaker for 1 minute and centrifuged for 10 minutes at approximately 3000 × g. Approximately 250 μL of supernatant was transferred to an evaporation plate and evaporated to dryness. Samples were reconstituted in 100 μL of LC mobile phase (50/50 mobile phase A/B, v/v), sealed, mixed, and placed in an autosampler for analysis by LC–MS/MS.

A 10 μL volume of the sample in each well was injected onto a Acquity UPLC BEH C18, 1.7 μm, 2.1 × 50 mm analytical columnc and eluted at 0.8 mL/min with gradient elution from 50% : 50% mobile phase A : mobile phase B to 90% mobile phase B in 0.4 minutes, where mobile phase A was 90/5/5 water/acetonitrile/methanol with 10 mM ammonium acetate and mobile phase B was 10/45/45 water/acetonitrile/methanol. Detection was performed by means of a Sciex API-4000 mass spectrometer with positive electrospray ionization (ESI) with multiple reaction monitoring of parent/product ion transitions of m/z 498.3/291.1, 500.3/291.1, and 544.3/321.0 for idarubicin, idarubicinol, and epirubicin (used as an internal standard), respectively. Total run time was 1.0 minute. The retention times were 0.40, 0.38, and 0.34 minutes for idarubicin, idarubicinol, and epirubicin, respectively. Calibration curves were generated by linear regression of the peak area ratios (analyte/internal standard) with 1/×2 weighting. Independently prepared quality control (QC) samples in dog plasma were assayed in each run. The lower limit of quantitation for the assay was 0.5 ng/mL for idarubicin and 0.1 ng/mL for idarubicinol. Bioanalytical data were stored in Watson LIMS (v7.2.0.03).d Estimated Cmax for idarubicin and idarubicinol was determined from the mean plasma concentrations at 2 and 4 hours, respectively.

Treatment Response Evaluation

Although tumor response is not a primary objective of phase I trial designs, response (lymph node or lesion) was determined by criteria consistent with the Response Evaluation Criteria for Peripheral Nodal Lymphoma in Dogs (v1.0)[15] based on Day 0 (pretreatment) and Day 21 (final study visit) tumor/lesion caliper measurement evaluation.

Statistical Analysis

In keeping with acute phase I trial design, no analyses were performed beyond simple descriptive statistics (eg, mean, median) for the population under study. Pharmacokinetic analysis is discussed previously under plasma analysis methods.



Patient demographics, tumor stage and immunophenotype, and prior treatment exposure are described in Table 2. There were 17 different breeds in study; the predominant breeds were Golden Retriever (n = 8), Labrador Retriever (n = 6), Boxer (n = 2), and Vizsla (n = 2). Twelve dogs (35%) had relapsed after prior chemotherapy protocols, the majority (9; 29%) having received a prior anthracycline (doxorubicin) agent, usually as part of a combination protocol. Twenty-four dogs (77%) completed the 21-day study; 5 progressed before Day 21 and were removed to seek alternative treatment, and 2 died before Day 21 (1 because of acute liver failure with jaundice determined at necropsy to be secondary to diffuse infiltrative carcinoma undiagnosed at study entry, the other euthanized because of intractable seizures on day 16 posttreatment).

Table 2. Patient demographics.
CharacteristicsValue (Percent)
  1. a

    Stage V proportion may be underestimated as not all patients had bone marrow aspiration performed.

Age (years)
Body weight (kg)
Male intact2 (6)
Male neutered15 (48)
Female intact0
Female neutered14 (45)
Tumor stagea
II2 (6)
III23 (74)
IV3 (10)
V3 (10)
Tumor immunophenotype
B cell16 (52)
T cell8 (26)
Null cell1
Not determined6 (19)
Prior treatment
Naïve19 (61)
Prior chemotherapy12 (35)
anthracycline containing9 (29)
anthracycline naïve3 (10)

Dose and Safety Evaluation

Adverse events (AEs) are summarized in Table 3. The MTD was established as 22 mg/m2. Hematologic AEs (neutropenia and thrombocytopenia) were the predominant DLT in this study, occurring between 7 and 10 days posttreatment. Three of 13 dogs in the expanded cohort experienced DLTs. At the maximal administered dosage of 25 mg/m2, 2 of 6 dogs had DLTs. Twenty-one dogs (68%) enrolled presented with AEs on at least one occasion during the study. The most common clinical chemistry-related AE was increased bilirubin concentration, which occurred in 5 dogs; only 1 case was grade 3 (2.4 mg/dL), was not associated with increases in serum alkaline phosphatase (ALP) or alanine aminotransferase (ALT), and resolved on subsequent evaluation.

Table 3. Adverse events by grade and cohort.
AE ClassificationGradeDose Cohort (Dogs)
1 (n = 3)2 (n = 6)3 (n = 3)4 (n = 13)5 (n = 6)
Neutropenia1   1 
3   11
4   22
Thrombocytopenia1 112 
211 1 
3 1   
4   12
3 1   
Anorexia124 1 
3 1   
Vomiting1 1 12
3 1   
Diarrhea1   32
2 1   
Weight loss1   1 
Elevated bilirubin1    1
2   21
3   1 
Elevated ALP1     
2 1   

One dog in the 22 mg/m2 group developed hyphema and uveitis concurrent with grade 4 neutropenia, thrombocytopenia, and progressive disease (PD). Although this could be attributed to idarubicin-caused hematologic AEs, the ocular AEs resolved upon achievement of complete response (CR) after a subsequent rescue protocol, and again returned months later concomitant with relapse of lymphoma implying disease causality.

One dog in the 22 mg/m2 dosing group developed seizure activity and tetraparesis beginning on day 6 postidarubicin treatment concomitantly with grade 4 neutropenia. A CSF tap and MRI exam did not identify the cause, and the dog was euthanized on day 16 because of poorly controlled seizures, anorexia and lateral recumbency. Bilateral adrenocortical hyperplasia was evident on gross necropsy examination and a mass in the right adrenal gland presumed to be a functional adenoma. On histopathology, a cortical adenoma was confirmed in the right adrenal gland and the left adrenal gland had diffuse cortical hyperplasia, which was moderate and chronic. Histiocytic and lymphoplasmocytic pituitary adenitis, which was focally extensive, marked, and subacute, was detected on histology. The cause of the seizures was unclear, and the pathologist reported the seizures may have been caused by inflammation of the pituitary gland, possibly causing edema and compression of the adjacent brainstem.

Pharmacokinetic Analysis

Plasma concentrations versus the mean dose for each treatment group are shown for idarubicin plus idarubicinol in Figure 1. Data were fitted by linear regression to show the trend in plasma exposure, but no formal analysis of dose proportionality was conducted because the sampling times only represent estimates of Cmax. All grade 4 neutrophil and thrombocyte adverse events and the majority of tumor responses were observed in conjunction with higher estimated plasma Cmax concentrations (Fig 2).

Figure 1.

Plasma drug concentrations of idarubicin (IDA), idarubicinol (IDOL), and idarubicin plus idarubicinol (IDA + IDOL) at the sampling times indicated. Data are shown as mean ± SD.

Figure 2.

Estimated combined Cmax (idarubicin + idarubicinol) plotted against individual subjects identified by treatment group. Four subjects from the 2 highest dose groups with Grade IV neutropenia and thrombocytopenia are denoted by arrows. Subjects demonstrating complete or partial tumor response the final response evaluation after treatment (day 21) are denoted by asterisks.

Treatment Response

Observed tumor responses at Day 21 study termination are presented in Table 1. Of 19 dogs receiving at least the MTD established dose of idarubicin (ie, cohorts 4 and 5), 11 demonstrated a treatment response, and 8 either did not achieve a response (n = 5) or did not complete the Day 21 evaluation (n = 3). Of the 11 responders, 9 had B cell, 1 null cell, and 1 undetermined immunophenotypic tumors; 9 were treatment naïve and 2 had relapsed after prior chemotherapy (1 including doxorubicin) before study entry. Of the 8 unresponsive dogs, 4 had B cell, 3 T cell, 1 null cell, and 1 undetermined immunophenotypic tumors; 6 were treatment naïve, and 2 had relapsed after prior chemotherapy (1 including doxorubicin) before study entry.


With the exception of substage b exclusion required for phase I evaluation, patient demographics, tumor stage, and immunophenotypic breakdown of subjects in this study population were similar to previous reports of pet dogs with multicentric non-Hodgkin's lymphoma.[1] A substantial proportion (35%) of dogs in study were heavily pretreated with chemotherapy protocols commonly in use for lymphoma and, as such, those cases probably represent similar drug-resistant populations enrolled in phase I trials of novel agents.

The established MTD of 22 mg/m2 for a single PO dose of idarubicin was determined after completion of 5 dosing cohorts, a dose de-escalation from the maximally delivered dose of 25 mg/m2 (cohort 5), and subsequent conformation by expansion of the 22 mg/m2 group to 13 total dogs treated. This MTD is comparable with that generated in healthy laboratory Beagle dogs (Pfizer internal data, not shown) that demonstrated an average dose of 0.7 mg/kg (approximately equivalent to 22 mg/m2) was tolerated whereas 0.9 mg/kg (approximately equivalent to 28 mg/m2) was not tolerated. We chose a starting dosage that was approximately 50 to 60% of the MTD in Beagle dogs for this study per current recommendation for phase I trials in tumor-bearing client-owned dogs.[13] The AE profiles observed in this study are consistent with the anthracycline-class chemotherapeutics and most AEs were transient, manageable, and self-limiting. DLTs were predominately hematologic (neutropenia and thrombocytopenia). Modest increases (grade 1–2) in serum bilirubin concentrations were observed in 4 dogs in the study, and 1 dog experienced a grade 3 increase (2.4 mg/dL) that was not associated with increases in ALP or ALT activity and normalized in subsequent evaluations. Transient changes in hepatic enzyme activities are reported in < 5% of humans treated with idarubicin-containing protocols.[16] It would be prudent to include assessment of hepatic function in any subsequent repeat-dose idarubicin investigations. Because this was a single dose study and cardiac function was not assessed, no conclusions regarding the superior cardiac AE profile of idarubicin reported in other species[6] can be made.

In general, plasma concentrations of idarubicin and idarubicinol increased with increasing idarubicin doses (Fig 1), although the dose range was narrow and the number of subjects in each treatment group was different. Cmax is thought to be related to the most frequently observed toxicities of anthracyclines, namely, cardiotoxicity, and neutropenia.[17, 18] As expected, the hematologic DLTs in this study were observed in conjunction with higher estimated plasma Cmax concentrations (Fig 2).

Although antitumor activity is not an end-point of phase I studies, especially those employing a single dose of an investigational agent, 11 responses (58%) were noted in the 19 dogs receiving at least the MTD (cohorts 4 and 5), suggesting activity for dogs with lymphoma. This was not unexpected because idarubicin previously has shown activity in cats and people with lymphoma.[14, 19] Although cohort numbers are too low and the population too heterogeneous (ie, immunophenotype, naïve versus relapse) to determine a true dose-response, activity was not observed in lower dosing cohorts and the majority occurred in conjunction with higher estimated plasma Cmax concentrations (Fig 2). Two of the 19 dogs in the higher dosing cohorts had relapsed after prior anthracycline treatment (doxorubicin), and although one achieved a CR to idarubicin, numbers are insufficient to draw conclusions regarding the theorized lack of cross-resistance between these 2 anthracyclines, and the reported lower cellular efflux of idarubicin by the multidrug resistance (MDR) efflux pump (P-glycoprotein). These potential advantages of idarubicin over doxorubicin remain theoretical in dogs, and this study was not designed to evaluate them.

Limitations of the current study are similar to those inherent in phase I dose-finding trials in general. Because of small patient numbers, varied dose-intensity cohorts, and the single dosing design, an accurate assessment of response rates, response durability, and cumulative adverse event profiles must await larger repeated-dosing phase II trials. The lack of consistent clinical staging was an additional minor limitation; but this limitation would only lead to understaging, and because adverse event profiles generally are higher and response rates generally lower in more advanced disease this would not have resulted in falsely optimistic results.

In summary, PO administered idarubicin generally was well-tolerated and showed preliminary antitumor activity in dogs with lymphoma. Furthermore, the potential clinical advantage of a safe and efficacious oral anthracycline alternative supports further investigations of this agent in repeated-dose, randomized clinical trials.


The authors thank the Clinical Trials clinicians at the Universities of Wisconsin and Minnesota for their expert patient care and data accrual.

Location where work performed: Clinical cases were treated at the School of Veterinary Medicine, University of Wisconsin-Madison, and the University of Minnesota College of Veterinary Medicine. Pharmacokinetic analysis performed at Pfizer Animal Health, Kalamazoo, MI.

Conflict of Interest: Dr Vail periodically receives honoraria from Pfizer Animal Health for his consultancy work in oncology. Drs Kamerling, Simpson, and McDonnell are employees of Pfizer Animal Health. This study was supported by Pfizer Animal Health, Kalamazoo, MI.


  1. 1

    EDTA anticoagulant, Bioreclamation, Inc, Westbury, NY

  2. 2

    Micronic Systems, Lelystad, the Netherlands

  3. 3

    Waters Corporation, Milford, MA

  4. 4

    Thermo Electron Corporation, Waltham, MA