Therapeutic effect of low-dose imatinib on pulmonary arterial hypertension in dogs

Abstract

This was a pilot study to determine the effectiveness of low-dose imatinib therapy for hemodynamic disturbances, including pulmonary arterial hypertension (PAH), and clinical manifestations caused by chronic heart failure in dogs. Six client-owned dogs with PAH were administered imatinib mesylate orally, 3 mg/kg body weight q24h, for 30 d. Physical examination, blood biochemical tests, radiography, and Doppler echocardiography were performed prior to imatinib administration and again 30 days after administration. Clinical scores were significantly reduced after imatinib treatment. Systolic pulmonary arterial pressure, heart rate, maximum tricuspid regurgitation velocity, left atrium/aorta ratio, right and left ventricular Tei indexes, early diastolic transmitral flow wave/mitral annulus velocity ratio, and plasma atrial natriuretic peptide concentration decreased significantly after therapy. Diastolic blood pressure, stroke volume, cardiac output, and left ventricular fractional shortening increased significantly after therapy. These results indicate that low-dose imatinib therapy was effective for heart failure in dogs with PAH.

Introduction

Pulmonary arterial hypertension (PAH) is characterized by systolic pulmonary arterial pressure (sPA) of more than 30 mmHg or mean pulmonary arterial pressure of more than 20 mmHg (1). In human medicine, PAH is 1 of the prognostic factors in intractable cases of heart failure (2,3), progressive heart failure with poor prognosis, and the risk of death (4). The prognosis in dogs with PAH has also been reported to be poor with median survival times of 3 to 91 d from diagnosis (5,6). In dogs with asymptomatic mitral regurgitation (MR), those with more advanced MR will develop PAH (7). Therefore, treatments to manage PAH in dogs with chronic heart failure may alleviate clinical signs and improve survival times.

Pulmonary arterial hypertension is due to increased pulmonary vascular resistance that is caused by pulmonary artery vasoconstriction and pulmonary vascular remodeling (8,9). Several extracellular and intracellular signaling abnormalities have been implicated in such remodeling (10). The expression of platelet-derived growth factor (PDGF) and PDGF receptor mRNAs was markedly increased in pulmonary vascular lesions of humans with PAH (11). In addition, there were more pulmonary artery smooth muscle cells in patients with PAH compared with individuals without PAH (12), and these cells proliferated significantly in the presence of PDGF in vitro(12). The prevention of excessive PDGF-induced proliferation of pulmonary artery smooth muscle cells (PASMCs) may result in the suppression of pulmonary blood vessel remodeling (11). It has been suggested that PDGF may play an important role in human PAH and that novel therapeutic strategies targeting the PDGF pathway should be tested in clinical trials (11).

Imatinib is a therapeutic drug that inhibits the activation of PDGF by impeding phosphorylation of the PDGF receptor tyrosine kinase (13). This agent has been used for treatment of chronic myeloid leukemia, gastrointestinal stromal tumor, and scleroderma in humans, and for mastocytoma and gastrointestinal stromal tumor in dogs (14,15). Imatinib also inhibits the PDGF receptor, a factor in the molecular pathogenesis of PAH (16). In in vitro culture of PASMCs from human patients with idiopathic PAH, PDGF-induced proliferation and migration of PASMCs are inhibited by imatinib (11). It has been reported that imatinib reversed vascular remodeling and cor pulmonale in monocrotaline-induced PAH in rats (17) and was effective for the treatment of a human patient with a rapidly progressing form of familial idiopathic PAH (18). Recently, it has also been reported that tyrosine kinase inhibitors including imatinib have potent pulmonary vasodilatory activity in pulmonary hypertensive rats, which could contribute to their long-term beneficial effect against pulmonary hypertension (19). However, to the best of our knowledge, there are no reports on the effectiveness of PDGF receptor antagonists, such as imatinib, for PAH in dogs. The purpose of the present pilot study was to determine the effect of low-dose imatinib therapy on both hemodynamic parameters and clinical manifestations in dogs with PAH due to advanced mitral valve disease and heartworm disease.

Materials and methods

Animals

Six client-owned dogs with PAH were prospectively recruited at the Arita Sougo Animal Hospital and the Veterinary Medical Center of Tottori University. The data for all dogs were obtained from July 2010 to July 2011. Informed consent was obtained from every dog owner. The diagnosis of PAH was defined as sPA of more than 30 mmHg and was calculated using the modified Bernoulli equation and estimated right atrial pressure. The breed, age, gender, body weight (BW), diagnosis of cardiac failure, main etiology of pulmonary hypertension, main clinical manifestations, the International Small Animal Cardiac Health Council (ISACHC) severity classification, and history of medications of the 6 dogs investigated are summarized in Table 1. The mean ± standard deviation (SD) of sPA immediately before imatinib administration was 63.3 ± 24.9 mmHg. The ISACHC severity classification ranged from II to IIIa, and syncope and cough were observed in all cases. The cough had been present for 1 to 3 y in all dogs. In dogs with heartworm (Cases 5 and 6), eosinophilic bronchitis by the filaria may be involved in causing the cough. The dogs with heartworm were not given adulticide before imatinib therapy and had a few heartworms detected by echocardiography during imatinib therapy.

Table 1.

Clinical characteristics and medication history of 6 dogs with PAH^a

Case number	Breed	Age (y)	Gender	BWd (kg)	Cardiac failure	Main etiology of PAH	Clinical signs	ISACHCh	History of medications and duration
1	Shih tzu	14	Mb	3.6	MRe, TRf	Pulmonary venous PAH due to MR	Syncope, cough, exercise intolerance, ascites	IIIa	Alacepril (3.5 mg/kg BW, PO, q12h), pimobendan (0.2 mg/kg BW, PO, q12h), furosemide (1.5 mg/kg BW, PO, q12h), spironolactone (2 mg/kg BW, PO, q12h), hydrochlorothiazide (2 mg/kg BW, PO, q12h), beraprost sodium (0.8 μg/kg BW, PO, q12h) for 5 mo
2	Maltese	14	Fc	2.5	MR, TR	Pulmonary venous PAH due to MR	Syncope, cough, exercise intolerance	IIIa	Benazepril (0.4 mg/kg BW, PO, q12h), pimobendan (0.2 mg/kg BW, PO, q12h) for 4 mo
3	Shih tzu	16	M	4.6	MR, TR	Pulmonary venous PAH due to MR	Syncope, cough, exercise intolerance	II	Alacepril (1.4 mg/kg BW, PO, q12h), pimobendan (0.12 mg/kg BW, PO, q12h) for 4 mo
4	Poodle	7	M	4.3	MR, TR	Pulmonary venous PAH due to MR	Syncope, cough	II	Benazepril (0.5 mg/kg BW, PO, q12h), pimobendan (0.15 mg/kg BW, PO, q12h) for 1 mo
5	Mongrel	9	F	8.0	MR, TR, PRg	Pulmonary arterial PAH due to Dirofilaria immitis	Syncope, cough, exercise intolerance, pleural effusion	IIIa	Alacepril (2 mg/kg BW, PO, q12h), pimobendan (0.2 mg/kg BW, PO, q12h), furosemide (1.4 mg/kg BW, PO, q12h), spironolactone (2 mg/kg BW, PO, q12h) for 1 mo
6	Mongrel	14	F	9.0	MR, TR, PR	Pulmonary arterial PAH due to Dirofilaria immitis	Syncope, cough, exercise intolerance, ascites	IIIa	Alacepril (2 mg/kg BW, PO, q12h), pimobendan (0.2 mg/kg BW, PO, q12h), furosemide (1.4 mg/kg BW, PO, q12h), spironolactone (2 mg/kg BW, PO, q12h) for 1 mo

^aPAH — Pulmonary arterial hypertension.

^bM — Male.

^cF — Female.

^dBW — body weight.

^eMR — Mitral regurgitation.

^fTR — Tricuspid regurgitation.

^gPR — Pulmonary regurgitation.

^hISACHC — Severity classification by International Small Animal Cardiac Health Council.

Medications

At the time of this study, all the dogs had been treated by a polypharmacy approach with therapeutic drugs for PAH, such as angiotensin-converting enzyme inhibitors, calcium-channel antagonists, prostacyclin pharmaceuticals, phosphodiesterase inhibitors, and diuretics for 1 to 5 mo (Table 1). In all cases, the lower dose of imatinib mesylate (Glivec; Novartis Pharma, Tokyo, Japan), 3 mg/kg BW, PO, q24h was administered for 30 d. Before and after imatinib administration, all dogs continued to receive these previous medications as described without any change. Physical examination, hematological and blood biochemical examinations, heart rate (HR) and blood pressure measurements, radiography, and Doppler echocardiography were performed at least twice, before imatinib administration (Pre) and again 30 d after imatinib administration (Post).

Clinical evaluations

Clinical signs, such as cough, exercise intolerance, syncope, ascitic fluid, and peripheral edema were examined at both pre- and post-investigations in all dogs. Ascites and edema were determined by physical examination, radiography, and echography, and other findings were determined by questioning the owner. The degree of clinical manifestation was assessed using the scoring method that was a modification of methods reported previously (20), (Table 2), including cough, exercise intolerance, syncope, and ascites-edema scores. Total score was calculated as the sum of the 4 scores.

Table 2.

Clinical manifestation score before and after imatinib administration in 6 dogs

Variable	Prea	Postb
Cough scorec	2.0 ± 0.6g	1.2 ± 0.4*
Exercise intolerance scored	1.3 ± 0.8	0.8 ± 0.4
Syncope scoree	1.2 ± 0.4	0.5 ± 0.5
Ascites and edema scoref	1.0 ± 1.1	0.3 ± 0.5
Total score	5.5 ± 2.6	2.7 ± 1.5*

^aPre — immediately before imatinib administration.

^bPost — 30 days after imatinib administration.

^cCough score — 0 = none, 1 = mild (coughing stops after 2 or 3 coughs per episode, and the frequency of coughing episodes is less than 9 times per day), 2 = moderate (coughing stops after 2 or 3 coughs per episode, and the frequency of coughing episodes is more than 9 times per day), and 3 = severe (coughing continues for more than 4 coughs, and the frequency of coughing episodes is more than 9 times in a day).

^dExercise intolerance score — 0 = none, 1 = mild (the dog can bark with pleasure, but tends to sleep), and 2 = severe (the dog does not bark at all, and hardly moves).

^eSyncope score — 0 = none, 1 = mild (once per week). 2 = moderate (2–6 times per week), and 3 = severe (every day).

^fAscites and edema score — 0 = none, 1 = positive for ascites and edema, which decreases after imatinib administration, and 2 = positive for ascites and edema, which does not change or is exacerbated after imatinib administration.

^gmean ± standard deviation (n = 6).

^*P < 0.05, significantly different from the Pre value.

Hematological and blood biochemical examinations

Blood was collected from the jugular vein of each dog at the pre- and post-investigations. A 0.5-mL volume was mixed with ethylene diamine tetraacetic acid (EDTA) for blood cell counts, and a 1.0 mL volume was mixed with heparin for plasma biochemical measurements. In addition, a 3.0-mL volume was mixed with EDTA containing aprotinin for atrial natriuretic peptide (ANP) measurement and 1.5 mL was transferred to a tube for serum collection. After centrifugation, the plasma or serum was separated and stored at −40°C for analysis. The blood cell counts were determined using an automatic blood cell analyzer. Blood urea nitrogen, creatinine, total bilirubin, total cholesterol, triglyceride, and inorganic phosphorus concentrations, as well as, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, and creatine phosphokinase activities were measured using an automatic biochemistry analyzer (Fuji Dri Chem 3500V; Fuji Film Medical, Tokyo, Japan). Plasma sodium, potassium, and chloride levels were measured using a clinical electrolyte analyzer (Fuji Dri Chem 800V; Fuji Film Medical). Serum N-terminal pro-brain natriuretic peptide (NT-proBNP) concentrations were measured using an enzyme-linked immunosorbent assay (ELISA) at the reference laboratory (IDEXX Laboratories, Tokyo, Japan). Plasma ANP levels were measured using a chemiluminescence enzyme immunoassay at another reference laboratory (Fukuyama Medical Laboratory, Hiroshima, Japan), which was the method validated for canine plasma (21). Each blood sample was analyzed immediately or sent to the reference laboratory within 1 d.

Echocardiography and circulation parameter measurements

Systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean blood pressure (MAP) were measured by the oscillometric method using a noninvasive blood pressure monitor (Dinamap 8300; Critikon, Florida, USA) attached to the tail ridge in the prone position. Transthoracic 2-dimensional; M-mode; and pulsed, continuous wave and tissue Doppler echocardiography were performed with the dogs in right or left lateral recumbency using a digital ultrasonography system (Prosound α7; Hitachi Aloka Medical, Tokyo, Japan) with a 5-MHz probe. The HR was calculated from the preceding R-to-R interval on the electrocardiogram recorded simultaneously.

Using the M-mode method, the left atrium/aorta ratio (LA/Ao) was measured from the left ventricular outflow tract view in the long-axis plane under right lateral recumbency. Left ventricular fractional shortening (FS) (22), left ventricular ejection fraction (EF) (22), end-diastolic interventricular septum wall-thickness (IVSd), end-diastolic left ventricular inner dimension (LVIDd), end-diastolic left ventricular posterior wall-thickness (LVPWd), end-systolic interventricular septum wall-thickness (IVSs), end-systolic left ventricular inner dimension (LVIDs), and end-systolic left ventricular posterior wall-thickness (LVPWs) were measured in the left ventricular short-axis view.

Using pulsed Doppler echocardiography in left lateral recumbency, the early diastolic transmitral flow (E) wave, late diastolic transmitral flow (A) wave, ratio of peak velocity of E to peak velocity of A (E/A), and deceleration time of the E wave (DecT) were recorded in the left apical 4-chamber view. In the apical 5-chamber view, a pulsed-wave sample volume was placed just under the aortic valve and the cross-sectional area (CSA) of the left ventricular outflow tract, aortic ejection flow velocity (AEV) and time velocity integral (TVI) were measured, and stroke volume (SV) (23) and cardiac output (CO) were calculated. The CO was calculated as SV × HR. Furthermore, time (a) from the end of the left ventricular inflow to the initiation of re-inflow was measured from the left apical 4-chamber view. Time (b) from the onset to the end of the left ventricular ejection flow was measured from the apical 5-chamber view. The left ventricular Tei index was calculated as (a − b)/b (24). Likewise, the right ventricular Tei index was determined from time (a) from the end of the right ventricular tricuspid inflow to the initiation of re-inflow in the apical 4-chamber view, and time (b) from the onset to the end of the right ventricular ejection flow in the apical short-axis view (24). Using continuous wave Doppler echocardiography, the maximum systolic mitral regurgitation velocity (MRmax) was measured in the left apical 4-chamber view. The maximum systolic tricuspid regurgitation velocity (TRmax) was measured in the left and right apical 4-chamber views and the left aortic short-axis view, and the highest value was used. The end-diastolic pulmonary regurgitation maximum velocity (PRmax) was measured in the apical short-axis view and observed in 2 dogs only (Cases 5 and 6). The sPA was calculated by adding the estimated right atrial pressure to the systolic right ventricle-to-right atrial pressure gradient, which was calculated using the modified Bernoulli equation:

$$[\text{pressure\hspace{0.17em}difference\hspace{0.17em}}(\mathrm{\Delta }\text{P})=4\times {\text{TRmax}}^2].$$( 25)

The estimated right atrial pressure was 15 mmHg in 3 cases (Cases 1, 5, and 6) and 10 mmHg in the other cases (Cases 2, 3, and 4). The end-diastolic pulmonary arterial pressure was calculated by adding the estimated right atrial pressure to the pressure gradient, which was calculated using the modified Bernoulli equation in 2 dogs (25).

Using tissue Doppler imaging, the mitral annulus motion velocity wave was recorded from the left apical 4-chamber. The early diastolic mitral annulus motion velocity (Em) and atrial systolic mitral annulus motion velocity (Am) were measured, and the ratio of Em to Am (Em/Am) was calculated. The ratio of E wave to Em (E/Em) was also calculated. The endomyocardial (Vend) and the epimyocardial velocities (Vepi) were measured at the posterior wall of the left ventricle in the short-axis view, and the myocardial velocity gradient (MVG) was calculated by dividing the difference (Vend-Vepi) by a distance between 2 points (26). The MVG was measured at mid systole (MVGs), early diastole (MVGe), and atrial systole (MVGa).

The measurement of each echocardiographic parameter was performed at least 3 times, and the average value was adopted as data. All measures and follow-up on each dog were performed by the same investigator.

Statistical analysis

Statistical analyses were performed with commercially available software (StatMate3; ATMS, Tokyo, Japan). A paired t-test was used for comparison between pre- and post-data for blood biochemical, echocardiographic, and circulation variables. The LSD test was used for multiple comparisons among the before-pre, pre-, and post-data for sPA. For comparison between pre- and post-clinical score data, Wilcoxon’s signed rank test was used. The level of significance in all tests was P < 0.05.

Results

Clinical manifestation score

Clinical manifestation data analyses are summarized in Table 2. The cough score was significantly lower in the post-scores compared with pre-scores. Exercise intolerance, syncope, and ascites, and edema scores tended to be lower in the post than in pre, but not significantly. The total score was significantly lower at post-evaluation compared with pre-evaluation.

Echocardiographic and circulation variables

The changes in sPA values in the 6 dogs are shown in Table 3. During treatment with general therapeutic drugs, sPA increased significantly from 14 to 40 d to immediately before imatinib administration. The elevated sPA decreased significantly and markedly after imatinib therapy. Moreover, post-sPA decreased significantly compared with the before-pre value.

Table 3.

Changes in the systolic pulmonary arterial pressure (mmHg) before and after imatinib administration in 6 dogs

Case number	Before-Prea	Preb	Postc
1	64.0d	96.7	60.1
2	46.0e	42.5	26.7
3	74.0f	83.2	39.3
4	20.2g	41.6	21.2
5	32.6h	40.3	24.5
6	54.4i	75.7	44.5
	48.5 ± 19.9j	63.3 ± 24.9*	36.1 ± 14.8*†

^aBefore-Pre — 14–40 days before imatinib administration.

^bPre — immediately before imatinib administration.

^cPost — 30 days after imatinib administration.

^d30 days before imatinib administration.

^e35 days before imatinib administration.

^f20 days before imatinib administration.

^g40 days before imatinib administration.

^h14 days before imatinib administration.

ⁱ14 days before imatinib administration.

^jmean ± standard deviation (n = 6).

^*P < 0.05, significantly different from the Before-Pre value.

^†P < 0.01, significantly different from the Pre value.

The other results for echocardiographic and circulation data analyses are shown in Table 4. The HR decreased significantly after imatinib therapy. The DBP was significantly higher after imatinib intervention. The TRmax was markedly elevated prior to intervention, but decreased significantly after intervention. Post-SV, CO, FS, EF, and TVI increased significantly compared with pre-values. Both right and left ventricular Tei indexes showed markedly high levels prior to intervention and decreased significantly after imatinib administration. LA/Ao decreased significantly after imatinib administration. Post-E wave, A wave, E/A, and Am did not show a significant difference from pre-data. Both DecT and Em/Am increased significantly after intervention; in contrast, E/Em decreased significantly after imatinib administration. Post AEV and post-MRmax were not significantly different from the pre-values. The PRmax in 2 dogs (Cases 5 and 6) decreased after intervention. The end-diastolic pulmonary arterial pressure decreased from 30.3 mmHg at pre to 24.6 mmHg at post in 1 case and from 32.8 mmHg at pre to 21.3 mmHg at post in the other case. Mid-systole Vend-Vepi increased significantly after intervention. There was no significant difference between the pre and post-IVSd, LVIDd, LVPWd, IVSs, LVIDs, and LVPWs values. Myocardial velocity gradient showed a tendency to increase after therapy, but not significantly.

Table 4.

Echocardiographic and circulation variables before and after imatinib administration in 6 dogs

Variable	Prea	Postb
Systolic blood pressure (mmHg)	138 ± 9h	138 ± 6
Diastolic blood pressure (mmHg)	82 ± 14	93 ± 13**
Mean blood pressure (mmHg)	102 ± 8	105 ± 8
Heart rate (beats/min)	150 ± 27	132 ± 22**
Left atrium/aorta ratio	1.61 ± 0.22	1.53 ± 0.18*
Left ventricular fractional shortening (%)	43.9 ± 4.5	49.9 ± 3.9**
Left ventricular ejection fraction (%)	82.0 ± 4.1	87.2 ± 3.0**
End-diastolic interventricular septum wall-thickness (mm)	6.4 ± 1.4	6.0 ± 1.2
End-diastolic left ventricular inner dimension (mm)	19.5 ± 3.9	20.8 ± 3.4
End-diastolic left ventricular posterior wall-thickness (mm)	6.4 ± 1.3	6.2 ± 1.3
End-systolic interventricular septum wall-thickness (mm)	10.2 ± 1.8	9.7 ± 1.1
End-systolic left ventricular inner dimension (mm)	10.9 ± 2.5	10.5 ± 2.0
End-systolic left ventricular posterior wall-thickness (mm)	9.3 ± 1.9	9.6 ± 1.3
Early diastolic transmitral flow wave (cm/s)	71.4 ± 23.2	74.7 ± 26.1
Late diastolic transmitral flow wave (cm/s)	86.9 ± 20.1	84.3 ± 25.0
E/Ac	0.82 ± 0.16	0.94 ± 0.31
DecTd (ms)	85 ± 21	110 ± 11**
Left ventricular outflow tract cross-sectional area (cm2)	0.49 ± 0.18	0.51 ± 0.19
Aortic ejection flow velocity (cm/s)	75.7 ± 6.9	82.1 ± 9.2
Time velocity integral (cm)	7.1 ± 1.0	8.7 ± 0.9***
Stroke volume (mL)	3.5 ± 1.5	4.5 ± 1.9**
Cardiac output (L/min)	0.50 ± 0.27	0.60 ± 0.32**
Left ventricular Tei index	0.39 ± 0.10	0.30 ± 0.07*
Right ventricular Tei index	0.74 ± 0.35	0.42 ± 0.27**
Maximum systolic mitral regurgitation velocity (cm/s)	408 ± 176	360 ± 246
Maximum tricuspid regurgitation velocity (cm/s)	348 ± 86	234 ± 71***
End-diastolic pulmonary regurgitation maximum velocity (cm/s)	203 ± 11i	140 ± 20i
End-diastolic pulmonary arterial pressure (mmHg)	31.6 ± 1.8i	23.0 ± 2.3i
Early diastolic mitral annulus motion velocity (cm/s)	6.9 ± 1.0	8.4 ± 1.7*
Atrial systolic mitral annulus motion velocity (cm/s)	8.5 ± 1.8	7.9 ± 2.5
Em/Ame	0.83 ± 0.14	1.11 ± 0.22*
E/Emf	10.4 ± 3.5	9.0 ± 3.2**
Mid systole Vend-Vepig (cm/s)	1.59 ± 0.52	2.27 ± 0.86*
Early diastole Vend-Vepi (cm/s)	−2.06 ± 1.05	−2.05 ± 0.52
Atrial systole Vend-Vepi (cm/s)	−1.43 ± 0.4	−1.51 ± 0.43
Mid systole myocardial velocity gradient (s)	2.16 ± 0.90	3.24 ± 1.60
Early diastole myocardial velocity gradient (s)	−2.05 ± 1.01	−2.18 ± 0.53
Atrial systole myocardial velocity gradient (s)	−2.04 ± 0.45	−2.17 ± 0.58

^aPre — immediately before imatinib administration.

^bPost — 30 days after imatinib administration.

^cE/A — ratio of peak velocity of early diastolic transmitral flow to peak velocity of late diastolic transmitral flow.

^dDecT — deceleration time of early diastolic transmitral flow wave.

^eEm/Am — ratio of early diastolic mitral annulus motion velocity to atrial systolic mitral annulus motion velocity.

^fE/Em — ratio of peak velocity of early diastolic transmitral flow to early diastolic mitral annulus motion velocity.

^gVend-Vepi — endomyocardial velocity minus epimyocardial velocity.

^hmean ± standard deviation (n = 6).

ⁱn = 2.

^*P < 0.05,

^**P < 0.01, and

^***P < 0.001, significantly different from the Pre value.

Hematological and blood biochemical variables

Pre ANP concentrations were markedly elevated (mean ±SD, n = 6; 96.6 ± 56.3 pg/mL), but decreased significantly (P < 0.001) after intervention (77.8 ± 53.4 pg/mL). Similarly, NT-proBNP and blood urea nitrogen levels were elevated prior to intervention (3121 ± 3645 pmol/L and 30.0 ± 13.8 mg/dL, respectively), but tended to decrease thereafter (2777 ± 3230 pmol/L and 23.7 ± 8.2 mg/dL, respectively), although not significantly. Packed cell volume, white blood cell and platelet counts; blood creatinine, total bilirubin, total cholesterol, triglyceride and inorganic phosphorus concentrations; aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, and creatine phosphokinase activities; and plasma sodium, potassium, and chloride values were not affected by therapy.

Discussion

In humans, a dose of 100 to 200 mg/day imatinib is typically used for PAH cases (18,27), which is approximately 1/6 to 1/3 of the antineoplastic dosage. A dose of 10 mg/kg BW/day of imatinib has been used as an antineoplastic drug in dogs (14). It has also been reported that a low dose of imatinib reduced pulmonary fibrosis more effectively than a high dose in mouse models (28). Furthermore, it has been reported that imatinib administration decreased the plasma PDGF concentration after 4 wk in PAH patients (27). Based on these reports, the dosage of imatinib in the present study was selected to be 3 mg/kg BW per day, approximately 1/3 of the antineoplastic dosage for dogs and the duration of administration was 30 d. The present study was the first to demonstrate that administration of low-dose imatinib for 30 d led to significant and marked improvement of PAH in dogs with chronic heart failure, suggesting that one possibility for improvement with imatinib could be reduction in pulmonary artery remodeling.

In this study, all the dogs had MR as well as tricuspid regurgitation (TR). We deduced that 4 dogs showed pulmonary venous PAH mainly due to MR, and 2 showed pulmonary arterial PAH mainly due to chronic Dirofilaria immitis infection. Therapeutic drugs, such as angiotensin-converting enzyme inhibitors, calcium-channel antagonists, prostacyclin pharmaceuticals, and phosphodiesterase inhibitors, have been used for PAH in dogs (20,29). However, in some cases, worsening of symptoms is not prevented by this polypharmacy approach. For such patients, it has been reported that imatinib therapy improved pulmonary vascular resistance and successfully relieved symptoms in humans (18,27,30). In our study, cough and the cumulative clinical manifestation score in dogs with PAH were improved by treatment with imatinib.

Imatinib therapy herein reduced TRmax and sPA and increased FS, EF, and SV and CO that were calculated. The reduction in the right ventricular Tei index after imatinib therapy may in part be caused by a decrease in right ventricular afterload, because of a decrease in pulmonary arterial pressure. Furthermore, the results of this study suggest that imatinib therapy reduces right ventricular diastolic pressure, resulting in a decrease in right atrial and peripheral venous pressures. In the left heart system, pulmonary venous pressure could be also reduced by dilation of pulmonary veins due to imatinib treatment (19), resulting in a decrease in left atrial pressure and an increase in left ventricular blood flow. In addition, the decrease in left atrial pressure might reduce left mainstem bronchial compression, resulting in a decrease in cough frequency. LA/Ao and E/Em decreased significantly after imatinib therapy which would be expected with a decrease in left atrial pressure. The decrease of the right ventricular afterload by treatment with imatinib may induce an elevation of the left ventricular preload and then increase CO via the Frank-Starling mechanism. In this respect, it was expected that the dogs with MR may develop pulmonary congestion due to an increase of the left ventricular volume load after imatinib therapy, but this was not the case. Therefore, these effects may be mainly due to the decrease in pulmonary vascular resistance by imatinib-induced PDGF receptor antagonism (16) or the potent pulmonary vasodilatory activity of imatinib (19), as well as the anti-inflammatory and antifibrotic actions of imatinib (31).

In this study, E/Em was reduced after imatinib treatment of the dogs with PAH, indicating a decrease in left atrial pressure. It has been reported that sPA correlates positively with the left ventricular regurgitant fraction in canine MR (32). Therefore, the reduction of left atrial pressure after imatinib treatment may have been caused by a decrease in sPA. The improvement of circulation parameters, including sPA, SV, and CO after imatinib treatment observed in this study was in agreement with the results of a previous report conducted in rats (33).

In the dogs with refractory MR, increase in LVIDd and elevation of both HR and LA/Ao have been reported to be amongst prognostic factors for death (34,35). In the present study, both HR and LA/Ao were reduced significantly after imatinib treatment. In addition, both the left and right ventricular Tei indexes decreased significantly after imatinib treatment, suggesting that imatinib is effective in improving cardiac failure in dogs with PAH. On the other hand, ANP or NT-proBNP levels have also been shown to be prognostic indicators in dogs with heart failure (34,36,37). In the present study, plasma ANP levels decreased significantly, and serum NT-proBNP levels tended to decrease after imatinib treatment. These results might be due to the reduction of the right ventricular preload by imatinib treatment, although the drug may increase the left ventricular preload by increasing flow across the pulmonary vascular bed. A positive correlation between pulmonary arterial pressure and cardiac sympathetic nerve activity has been shown in human patients with chronic cardiac failure (38). The present results revealed that both HR and sPA were reduced by imatinib treatment, suggesting suppression of cardiac sympathetic nerve activity. Therefore, imatinib therapy for PAH may improve prognosis through cardioprotection, including inhibition of the hemodynamic loading on the heart and inhibition of the activated sympathetic system.

Regarding the safety of imatinib, when 5 persons with PAH were treated with low-dose imatinib (100 mg/day), renal dysfunction was observed in 3 cases during 24 wk of imatinib therapy; no other adverse events, including digestive symptoms, were observed (27). In the present study, no hematological and plasma biochemical abnormalities suggestive of renal and liver pathology were observed during imatinib therapy for 30 d. This indicates that low-dose imatinib could be used safely without any apparent adverse effect in dogs with PAH.

The present results have to be interpreted with caution based on limitations in the study. The primary limitations were the small sample size, which reduces the power of the study; the lack of a placebo controlled group; the owners and investigators not having been blinded to avoid treatment bias; and no assurance of repeatability of the results.

In conclusion, low-dose imatinib therapy in dogs with PAH herein led to reduction of pulmonary arterial pressure without decreasing systemic blood pressure or CO and improved cardiac function and hemodynamics. Furthermore, treatment with imatinib was effective for PAH caused by MR or chronic filariasis in dogs. To the best of our knowledge, this is the first report demonstrating the usefulness of this therapy for treating dogs with PAH. However, this is a pilot study and further work with larger, placebo-controlled, randomized, and blinded studies will be necessary to determine the efficacy of imatinib in reducing PAH due to various etiologies and relieving symptoms due to PAH.