Abstract
In canine leishmaniosis (CanL), kidneys are affected in virtually all dogs. Treatment of CanL is limited in Europe to meglumine antimoniate and miltefosine. This study evaluated the pharmacological, toxicological, and pathological effects of both drugs in healthy beagle dogs. Four male and four female dogs were divided into two groups. The animals in Group 1 were administered an oral solution of 2% of miltefosine at 2 mg/kg b.w. once a day, for twenty-eight days. The animals in Group 2 were administered a preparation of meglumine antimoniate at 100 mg/kg b.w. subcutaneously once a day for twenty-eight days. After treatment, all dogs were followed-up for a further twenty-eight days. Dogs were observed daily and clinically examined ten times throughout the study. On days -1 and 55 a renal biopsy was performed on all dogs and analyzed by light microscopy, immunofluorescence, and electron microscopy. All the examinations failed to demonstrate any lesions in the miltefosine-treated dogs. Conversely, all the meglumine antimoniate–treated dogs demonstrated severe tubular damage, characterized by tubular cell necrosis and apoptosis. In conclusion, although no clinical signs of renal disease were evident, the use of meglumine antimoniate in the pharmacological treatment approach of CanL-affected dogs should be carefully considered.
Introduction
Canine leishmaniosis (CanL) is a multisystemic disease with variable clinical signs (Baneth 2006; Ciaramella et al. 1997). The majority of affected dogs present with poor body condition, generalized muscular atrophy, lymphadenomegaly, and excessive skin scaling. Kidneys are affected in virtually all dogs with CanL, and renal disease might be the only apparent abnormality in infected dogs (Baneth and Shaw 2002; Costa et al. 2003).
Until recently, the most commonly used drug for the treatment of CanL has been the pentavalent antimony meglumine antimoniate (N-methylglucamine antimoniate, Glucantime, Mérial), alone or in combination with allopurinol (Miró, Cardoso, et al. 2008; Nieto et al. 1992, Noli and Auxilia 2005).
In 2007, the first oral therapy for CanL based on miltefosine became available in some EU countries. Miltefosine (1-O-hexadecylphosphocholine, Milteforan, Virbac), an alkylphosphocholine and a membrane-active synthetic ether–lipid analogue originally developed for the treatment of cutaneous metastasis from mammary carcinomas (Hilgard et al. 1993), has proved to be an effective treatment for both human cutaneous and visceral leishmaniosis (Jha et al. 1999; Pearson 2003; Sundar et al. 1998; Sundar et al. 1999; Verma and Dey 2004) and later on for canine visceral leishmaniosis (Mateo et al. 2009; Miró, Oliva, et al. 2009).
Although the treatment of CanL is currently limited in some EU countries mainly to meglumine antimoniate and miltefosine, the information on the impact of these drugs on the clinical and kidney status of healthy dogs is limited.
Since the evaluation of the impact of these drugs on the kidneys in CanL affected dogs is not practical owing to the influence of underlying renal disease, the aim of this study was to evaluate the pharmacologic, toxicologic, and pathologic effects of a standard treatment with both drugs in healthy beagle dogs.
Materials and Methods
Animals
Four healthy male and four healthy female beagle dogs from Harlan (Netherlands) were used. The animals were seven to nine months of age and weighed from 7 to 9 kg at the start of the study. They were housed in individual cages in an environmentally controlled room with a twelve-hour light cycle, a temperature of 18°C ± 3°C and 55% ± 10% relative humidity. They were fed daily with dry pet food (Virbac Vet Complex) according to their individual weight. Drinking water was available ad libitum. All procedures and maintenance of animals were in accordance with internal procedure and the principles of humane treatment as elaborated by the National Institutes of Health. All dogs were identified with a code tattooed on the inside of the ear flap.
Study Design
The dogs were divided into two randomized groups (Table 1). The animals in Group 1 were administered an oral solution of 2% of miltefosine (Milteforan, Virbac) at a dose rate of 2 mg/kg b.w. once a day, for twenty-eight consecutive days (1 mL/10 kg/day). The oral solution was poured onto the dogs’ food at mealtime. The animals in Group 2 were administered an injectable preparation of meglumine antimoniate (Glucantime, Mérial) at a dose rate of 100 mg/kg b.w. subcutaneously once a day for twenty-eight consecutive days (3.3 mL/10 kg/day).
Dogs were observed daily and clinically examined ten times throughout the study. At the initial pre-inclusion visit, day -4, dogs were clinically examined and randomized. On the same day, the animals were weighed, and this procedure was repeated weekly until the end of the study (to adjust the drug dose rates according to body weight). Dogs underwent weekly complete clinical examinations on days -1, 6, 13, 21, 27, 34, 41, 48, and 55. On days -1, 27, and 55, a blood sample was taken from the jugular vein. Furthermore, a urine sample was collected from the bladder, by cystocentesis, prior to the biopsy procedure. The analyses performed included routine hematology, biochemistry, and protein electrophoresis on the blood and complete urinalysis. All the samples were analyzed by the same laboratory (Vébiotel Laboratoire de Biologie Vétérinaire, Arcueil, France). On days -1 and 55, an ultrasound-guided renal biopsy was performed under anaesthesia on all dogs (Table 2). After the clinical evaluation, an intravenous catheter (22G) was placed in the left or right cephalic vein and connected to an infusion line, for fluid therapy (normal saline solution, 0.9% NaCl at the rate of 10 mL/kg/hour) and anaesthesia induction. Anaesthesia was performed using an association of tiletamine chloride and zolazepam chloride (Zoletil 50, Virbac), at the dose of 0.15 mL/kg injected intravenously. The dog was positioned in right lateral recumbency for the first biopsy that was performed on the left kidney on day -1, and vice versa on the right kidney, with the dog recumbent on the left side on day 55. This approach was chosen to avoid that the second biopsy would be performed on scarred tissue following the previous procedure. For both procedures, the haircoat was clipped and the skin area prepared aseptically.
The biopsies were ultrasound guided (GE, mod Logic E, with microconvex 5–9 Mhz probe) and performed using an 18G True-cut disposable biopsy needle driven by a gun, with a spring trigger system (Magnum Bard). Based on an immediate evaluation by the pathologist, using a stereomicroscope, one or two biopsies were taken from the renal caudal pole of each dog.
To avoid clot formation in the renal pelvis, fluid therapy was maintained for about thirty minutes after each biopsy until the dogs woke up, whereas the femoral pulse and the capillary refill time (CRT) were continuously monitored. Before discharge, each dog was checked by ultrasound for signs of hemorrhage at the biopsy site.
Once the dogs were completely awake and clinically normal, they were taken to their individual cages. For the handling of the biopsy specimens, a standard protocol was followed. Renal biopsies were divided into three parts for light microscopy, immunofluorescence, and electron microscopy purposes. Renal biopsies were evaluated at the Department of Public Health, Comparative Pathology and Veterinary Hygiene–Anatomical Pathology Section of Padua University.
Light Microscopy
For light microscopy, biopsies were fixed in formalin, embedded in paraffin, and serially sectioned at 4-μm intervals. Alternate sections were stained with: hematoxylin-eosin (HE), periodic acid-Schiff (PAS), acid-fuchsin orange G (AFOG), Masson’s Trichrome, Periodic acid-Schiff methanamine (PASM), and Miller’s elastin stain. The evaluations were performed blinded by three independent pathologists (LA, LC, MC). Discordant data were reevaluated by the investigators, and a consensus evaluation was used for the definitive analysis.
Immunofluorescence
For direct immunofluorescence examination, fresh unfixed renal biopsies were embedded in optimal cutting temperature (OCT) compound, snap-frozen in liquid nitrogen, and stored at −80°C. Subsequently, 5-μm-thick sections were fixed with acetone for fifteen minutes. After washing with phosphate buffered saline (PBS) (two passages), slides were incubated with FITC-labelled anti-goat IgA, IgG, IgM and complement C3 antibodies (Bethyl Laboratories, Inc., Montgomery, TX, USA). Primary antibodies were omitted as negative controls and substituted with PBS.
Electron Microscopy
Renal tissue specimens were fixed in Karnovsky fixative, stored in 0.1 M sodium cacodylate (pH 7.2), and postfixed in 2% 0s04, buffered with 0.1 M sodium cacodylate (pH 7.2). After rinsing twice for thirty minutes each time with distilled water, tissue specimens were stained with 2% uranyl acetate twice for two hours each time at room temperature, dehydrated in a graded series of acetone, and embedded in Durcopan ACM (Fluka Ag., Buchs, Switzerland). Ultrathin sections (50 nm) were stained with lead citrate and examined with a LEO electron microscope (Carl Zeiss electron microscopes, NY).
Results
Baseline Characteristics
At the pre-inclusion visit (day -4) and at the day -1 visit, all dogs were clinically healthy. The blood analyses performed on all dogs on day -1 did not show any significant abnormalities. The urinalysis performed on day -1 indicated an abnormal urine protein/urine creatinine (Up/Uc) ratio in only one dog from Group 2 (Up/Uc ratio = 1.74).
Body weight
The dogs in Group 1 had a weight gain of 0.5 to 0.9 kg between day -4 and day 55. Three of the four dogs in Group 2 had a weight gain of 0.15 to 0.7 kg during the same period. One dog from Group 2 (H7E 0886) had a weight loss of 0.15 kg between Day -4 and Day 55.
Clinical Results
Globally, all the animals remained clinically healthy throughout the study period.
Blood Results
The blood examinations performed on day 27 did not show any relevant abnormality. The hematological and biochemical results were within the normal range for all dogs. At day 55, one dog in Group 2 (H7E 0886) showed an increased urea value (0.8g/L; cutoff level: 0.6 g/L).
Urine Results
The urinalysis performed on day 27 indicated increased urine protein levels in four dogs (two dogs in each group), with all other parameters remaining within normal reference ranges. At the examination performed on day 55, three out of the four dogs still showed increased urine protein levels, without any other detected abnormalities.
Light Microscopy
Sampling for light microscopy was adequate in all biopsies, with a mean glomerular count of 17.4 (range, ten to thirty-six glomeruli). Light microscopy evaluation of the renal biopsies showed no alterations in the eight dogs at baseline (day -1). After fifty-five days, the two groups of dogs treated with the two compounds presented different histology features. The four dogs treated with miltefosine showed normal glomeruli and minimal vacuolization of the proximal tubular epithelial cells with a scattered distribution. The vascular compartment was within normal limits. The four dogs treated with meglumine antimoniate showed a normal glomerular pattern. The predominant findings were restricted to the tubular compartment and were similar in all dogs treated with this drug. Diffusely, tubular cells showed marked swelling with pale cytoplasm, and in multifocal areas, coagulative necrosis features were evident. Sloughed tubular cells were present in tubular lumina, and individual tubular cell necrosis with picnotic nuclei was characteristic in these tubules (Figure 1). Some tubular cells underwent apoptosis with phagocytosis of apoptotic bodies by neighboring epithelial cells (Figures 2 and 3). No tubular casts or crystalline deposits were evident. The luminal borders were markedly irregular, with simplified cells alternating with enlarged, hypereosinophilic cells. Focal interstitial edema with minimum mononuclear inflammatory infiltrates without tubulitis was evident. Vessels were normal.
Immunofluorescence
Immunofluorescent reactions for IgG, IgM, IgA and C3 were negative in the renal biopsies at baseline (day -1) and at the control time (day 55) in both groups.
Electron Microscopy
Electron microscopy confirmed the severe tubular damage seen in the meglumine antimoniate–treated dogs. Proximal tubules exhibited epithelial simplification with reduced organellar content, loss or attenuation of brush border, cellular detachment from the tubular basement membrane, apical blebbing, widened intercellular spaces, individual cell necrosis, and focal shedding of cytoplasmic debris into the tubular lumen. No lesions were detected at the ultrastructural level in the miltefosine-treated dogs.
Discussion
The kidneys are affected in virtually all dogs with CanL, and renal disease might be the only apparent abnormality in infected dogs. Renal disease can progress from asymptomatic proteinuria to the nephrotic syndrome or chronic renal failure with glomerulonephritis, tubulointerstitial nephritis, and amyloidosis (Koutinas et al. 1999). Glomerular lesions developing in dogs during infection with Leishmania organisms are associated with the presence of immune complexes (Nieto et al. 1992) and have been classified histologically as mesangial glomerulonephritis, membranous glomerulonephritis, membrano-proliferative glomerulonephritis, and focal segmental glomerulosclerosis (Aresu, Pregel, et al. 2008). Tubulointerstitial nephritis has been reported with different incidences by different authors in CanL-affected dogs, but it has never been observed as a solitary lesion (Aresu, Rastaldi, et al. 2008). The occurrence of tubulointerstitial lesions is generally considered as a consequence of the progression of immune-mediated glomerular lesions, which are classically primary lesions (Zatelli et al. 2003).
Despite the high prevalence of kidney damage, increase of serum creatinine and urea resulting from primary kidney failure is evident only when the majority of nephrons are lost, which happens rather late during disease progression (Baneth et al. 2008).
Treatment with antimonials has sometimes been pointed out as being responsible for the deterioration of renal conditions of already affected kidneys in leishmaniotic dogs. Nevertheless, the information available on the toxicology, pharmacokinetics, and pharmacodynamics of this drug in the dog is scarce. A specific study with meglumine antimoniate has shown that the pharmacokinetic behavior of this drug in healthy dogs differs considerably from that described in man (Tassi et al. 1994). This and similar studies (Belloli et al. 1995) have shown that the subcutaneous route of administration is the most suitable to maintain the serum level of antimony (Sb) over time. After subcutaneous administration, the availability of the drug was close to 100%, and the maximum concentration of antimony was achieved within three to five hours, with an immediate decline in the drug concentration, which reached values close to the detection limit only eighteen hours after injection. The half-life was of 121 ± 6 minutes (Tassi et al. 1994). Meglumine antimoniate has been reported to be rapidly eliminated by the kidneys in healthy dogs, mainly by glomerular filtration. After subcutaneous injection, more than 80% of the drug is eliminated through the kidneys (Belloli et al. 1995; Tassi et al. 1994).
Monitoring of Sb urinary excretion indicates that the kidneys are the almost exclusive route of elimination of meglumine antimoniate (Hantson et al. 2000); however, no specific guidelines exist for dosage adjustment in renal failure. Even if nephrotoxicity has rarely been related to this treatment in humans, some cases of acute renal failure as a result of acute tubular necrosis, followed by death after receiving Glucantime, have been reported in human patients (Rodrigues et al. 1999).
In animals, renal function was assessed in rats treated with the pentavalent antimonials, Glucantime (meglumine antimoniate, Rhodia) or Pentostam (sodium stibogluconate, Well-come). When administered at a dose rate of 30 mg of Sb (Glucantime or Pentostam) per 100 g of body weight per day for thirty days, renal functional changes were observed consisting of disturbances in urine concentrating capacity, suggesting an interference of the drugs in the action of antidiuretic hormone on the distal tubules and collecting ducts. The disturbance in urine concentration was reversible after a seven-day period without the drugs being administered. No significant histopathological alterations were observed in the kidneys of the rats treated with the drugs. Only the rats treated with a high dose of Pentostam (200 mg/100 g body weight/day) demonstrated functional and histopathological alterations of acute tubular necrosis (Veiga et al. 1990)
Adverse effects of antimonial therapy in humans include lethargy, anorexia, pneumonia, nausea, urticaria, fever, vomiting, abscess formation, cardiotoxicity, hepatotoxicity, and nephrotoxicity (Davidson 1998; Ikeda-Garcia et al. 2007).
In dogs, the most commonly observed adverse effects associated with meglumine antimoniate are apathy, anorexia, vomiting, diarrhea, and pain at the site of injection (Baneth and Shaw 2002); however, the frequency and severity of these adverse effects are unknown (Bravo et al. 1993; Noli 1999). Miltefosine has only recently become available for the treatment of CanL, therefore very little is known of its effect on the kidney of leishmaniotic dogs in the field. Several pharmacokinetic studies have been performed with miltefosine, in laboratory animals and in the target species; miltefosine was also assayed in dog plasma, urine, and feces by HPLC-MSMS analyses (Virbac, data on file Study Nos. F-107.010000-60008; F-107.010000-60010; F-107.010000-60027).
In rats and dogs, miltefosine showed an absolute bioavailability of 82% and 94%, respectively, with the time for maximum concentration values (Tmax) ranging from four to forty-eight hours. In dogs, after repeated oral administration of miltefosine in food for twenty-eight days, plasma clearance was determined to be 3.40 ± 0.447 mL/kg/hr, corresponding to an overall body extraction ratio of about 0.06% for a 10-kg dog. This finding suggests that, in dogs, the metabolic efficiency to transform miltefosine into its different metabolites is low and that there is no hepatic first-pass metabolism.
The terminal elimination half-life in rats was approximately eighty hours (3.3 days), and 153 hours (153 ± 13.7 hr) in dogs, equivalent to 6.3 days. This long terminal half-life can be explained by the low plasma clearance of miltefosine. Considering this terminal half-life in dogs, a steady state can be expected after about three to four weeks of repeated daily administration of miltefosine.
In dogs, repeated administration of 2 mg/kg/day of miltefosine for twenty-eight days led to an accumulation index of 7.65 ± 1.99. The accumulation index is the ratio between the amount of miltefosine (AUC0–24hr) reached at a steady state and that obtained after the first administration of miltefosine. After the last administration, the Cmax was 32582 ± 4030 ng/mL with a mean Tmax of 5.0 ± 2.0 hours.
These data indicate an increase of plasma miltefosine concentration within the first two weeks of treatment, reaching a “steady state” until the end of treatment (twenty-eight days). At the end of treatment, there is a slow and linear decrease of plasma miltefosine with complete elimination in a further four weeks. Miltefosine is widely distributed in body organs, undergoes a slow metabolic breakdown, and is metabolized in the liver into choline and choline-containing metabolites (Berman 2005).
Miltefosine is slowly excreted in the feces. The mean fecal clearance observed was low (0.32 ± 0.13 mL/kg/hr), which is very similar to the bile flow rate in dogs, suggesting that miltefosine fecal clearance is actually a biliary clearance. The contribution of fecal clearance to total body clearance was 10% ± 4.86%, meaning that the elimination of miltefosine is likely to undergo an extensive, but slow, metabolism.
As fecal clearance represents about 10% of total clearance, it can be concluded that only about 10% of the administered dose is eliminated as its parent drug in the feces (i.e., about 200 μg/kg/day).
The miltefosine urinary concentration was rather low and below the limit of quantification (20 ng/mL of miltefosine) after three days. The contribution of renal clearance to total body clearance was considered as negligible (about 0.03%). These results suggest that urine is therefore only a minor route of miltefosine parent drug elimination.
As in human beings, side effects induced by miltefosine in dogs are mostly gastrointestinal, such as occasional vomiting and diarrhea. They are dose-dependent and at the recommended dose rate (2 mg/kg/day) are generally mild and transient, disappearing without therapy after the end of treatment period.
Our results confirm that the impact of miltefosine on kidney functions and on the clinical status of dogs is limited. The only abnormal clinical value was limited to an abnormal Up/Uc ratio in one dog from the meglumine antimoniate–treated group (Up/Uc ratio = 1.74) at the pre-inclusion visit, which can be explained by blood contamination of the urine samples during cystocentesis. In the miltefosine-treated group, light microscopy examinations demonstrated normal glomeruli and minimal vacuolization of the proximal tubular epithelial cells with scattered distribution. Immunofluorescent examinations failed to detect any immune deposits, and electron microscopy examinations did not demonstrate any lesions. Although there is a lack of clinical evidence of kidney involvement, all the meglumine antimoniate–treated dogs demonstrated morphological changes consistent with severe tubular damage. In light microscopy examinations, diffuse marked swelling with pale cytoplasm of tubular cells and multifocal areas of coagulative necrosis were seen. Sloughed tubular cells and individual tubular cell necrosis with picnotic nuclei were also evident in these tubules. Ultrastructural examination confirmed the damage seen at the light microscopy level. This severe damage was evident four weeks after the end of the therapy without signs of regeneration of the damaged tissues.
In conclusion, although in both groups of treated animals no clinical signs of renal disease were evident, in light of our morphological results on renal biopsies, the pharmacological treatment approach of CanL-affected dogs should be carefully evaluated.
Footnotes
Figures and Tables
Financial Disclosure: All expenses required for trial (availablity of dogs, materials for procedures and laboratory analyses) were financed by Virbac SA.