Materials and Methods

Relevant institutional and government animal ethics committees approved all procedures. Dogs were fed and housed according to the National Research Council and National Health and Medical Research Council of Australia Guide for the Care and Use of Laboratory Animals. All dogs had ivermectin concentrations of 0 ng/mL at the start of each study, and the veterinarians and dog attendants ensured that no products containing ivermectin were used at any time.

Implants—The implants^a used in this study were silicone CRs^10,11 with an inner core containing the active ingredient ivermectin^b (a commercially available formulation consisting of a mixture of not less than 90% ivermectin B1a and not more than 5% ivermectin B1b) and the additives sodium desoxycholate^c and sucrose^d in a silicone^e matrix and an outer wall made of silicone only. The first step in the preparation of the implant is to mix specific amounts of the materials in a preparation mixer to make part A (silicone polymers and catalyst) and part B components (silicone polymers and cross-linker) either with (core) or without (wall) ivermectin. The part A and part B components are mixed to form the core and wall and are coextruded in a clean room by use of an extruder.

When the part A component is mixed with the part B component, the hydride group of the cross-linker reacts with the vinyl group of the silicone polymers in the presence of the platinum catalyst. The result is an ethyl bridge between the polymers with fast curing time, no volatile by-products, and formation of cross-linked silicone-oxide polymer chains. After packaging, the implants are sterilized by use of an electron beam.

Ivermectin analysis—Ivermectin (B1a and B1b) in serum or plasma samples stored at −20°C was analyzed by use of an HPLC method with fluorescence detection.^f Serum was collected for preclinical studies (studies 1, 2, and 3) and plasma for clinical studies (studies 4, 5, and 6). For plasma samples, lithium heparin was used as an anticoagulant. Ivermectin is very stable when stored frozen, even with repeated freeze thawing. Ivermectin was extracted with acetonitrile. The acetonitrile extract was evaporated to dryness under vacuum, and the residue was dissolved in a hexanedichloromethane mixture. The organic extract was concentrated by use of a silica column,^g and ivermectin was eluted with diethylamine ethylacetate. The extracted ivermectin was derivatized with acetic anhydride/1-methyl imidazole/dimethyl formamide, and the derivatization mixture was purified by use of a C18 column.^g The methanol eluate containing ivermectin was concentrated to dryness and made up to volume with the mobile phase prior to quantitation by use of reverse-phase HPLC with fluorescence detection. The limit of quantitation for the assay was 0.1 ng/mL.^f The method required 1 mL of serum or plasma, and linear calibration graphs (r² = 0.997) over the range of concentrations studied were used to validate the method. Recovery was > 75%, coefficient of variation was 4.10%, and relative residual SDs for accuracy and precision were < 8% and < 6%, respectively. The amounts of ivermectin B1a and ivermectin B1b were measured, with results expressed as total ivermectin (B1a + B1b).

Measurement of drug content in CR implants—After measurement of the weight of the formulation, the formulation was axially cut into 4 sections. Methanol was added at 2 mL/10 mg of formulation. The formulation was then incubated for 3 days at 37°C to elute the ivermectin. The ivermectin concentration in the eluate was measured by use of reverse-phase HPLC to determine drug content in each formulation.^h Analysis by use of HPLC was performed with a C18 symmetry columnⁱ (0.46-cm internal diameter × 5 cm; mean particle size, 3.5 mm), mobile phase A was water, and mobile phase B was methanol. Flow rate was set at 1.5 mL/min. Mobile phase B was increased from 70% to 80% in 4 minutes, and then 80% was retained for 8 minutes. Column temperature was set at 50°C. Detection was performed by measuring absorption at 245 nm.

Pharmacokinetics—At various intervals after implantation of the CR, serum of each treated dog was obtained. All dogs received the implant SC through the loose skin between the scapulae. Serum samples were stored frozen (–20°C) prior to analysis. Three studies (1, 2, and 3) were undertaken as part of the preclinical development of the CR implant, and Cmax and AUC were determined. The AUC was calculated by use of the linear trapezoidal rule.¹²

Preclinical study 1—Dogs were part of a commercial dog breeding operation located in South Eastern Victoria.^j The percentage of ivermectin and excipients varied with each CR implant. Thirty-six dogs consisting of Beagles and Cocker Spaniels of mixed sex, weighing approximately 20 kg (range, 10 to 27 kg) and between 2 and 7 years of age, were selected. Dogs were housed in fenced yards with shelter. The dogs were fed a mixture of fresh meats consisting of chicken and beef with a high proportion of bone and water ad libitum. Six CR implant formulations were tested. Six dogs were allocated to each of the 6 implant formulations (Table 1); serum was prepared from blood collected at 0, 1, 2, 3, 4, 6, 8, 10, 12, and 16 weeks after implantation; and the amount of ivermectin in the serum was determined by use of HPLC. Dogs were randomly allocated to treatment groups. Dogs in each group were treated with the same length and type of CR implant. Each dog received 10 implants, each measuring 1.6 × 12 mm.

Preclinical study 2—For further testing as a potential 12-month product, implant formulations E and F were selected on the basis of results of the first study. For these 2 formulations, implants of 4 lengths each were studied to determine the most appropriate dose for use in dogs. All implants had an external diameter of 1.6 mm. Implants were cut to lengths of 12, 18, 24, or 30 mm. Thirty-two Beagles and Cocker Spaniels, of mixed sex and weighing approximately 20 kg (range, 10 to 30 kg) were selected for 8 groups of 4 dogs each (4 implant lengths and 2 formulations). This study was undertaken in the same location as study 1, and dogs were housed and fed as described for study 1. Dogs received a single implant only. Serum was obtained from blood collected for measurement of ivermectin concentration at 0, 4, 10, 16, 20, 24, 30, 34, 42, 46, 52, and 57 weeks after implantation. Dogs received an ivermectin dose from 4 to 10 mg, depending on the composition of the implant and its length (Table 2).

Preclinical study 3—Study 3 was undertaken to determine whether a CR silicone implant (implant E) containing 7.3 mg of ivermectin and measuring 1.6 × 24 mm was sufficient to maintain a putative protective ivermectin concentration¹³ against heartworm infection for 52 weeks. This study was undertaken in South Eastern Victoria, per the previous 2 preclinical studies. Dogs in this study were housed and fed as described for study 1. In this study, 14 Beagles (4 females and 10 young adult males; body weight range, 12 to 23 kg) received a single CR silicone implant, and serum samples obtained from blood collected at 0, 1, 2, 4, 8, 10, 12, 21, 33, 37, 52, and 56 weeks were used for measurement of ivermectin concentrations.

Challenge studies 1 and 2—Challenge studies 1^k and 2^l were randomized controlled trials and had identical formats. Dogs used were hounds, Beagles, and mongrels (body weight range, 18 to 22 kg).^m,n For each study, 12 dogs (clinical study 1, 12 females; clinical study 2, 6 males and 6 females) received a single 1.6 × 24-mm CR implant containing 7.3 mg of ivermectin° on day 0, SC, beneath the loose skin between the scapulae. A further 12 dogs were enrolled in each study as placebo-treated controls that received an implant containing sucrose instead of ivermectin. All dogs were tested for heartworm by use of a commercial heartworm antigen test^p and modified Knott test. All dogs had negative results for heartworm antigen and microfilaria at the start of the trials. Dogs were monitored for signs of ivermectin toxicosis before implantation and 1, 2, 3, 4, 6, 12, and 24 hours after implantation. After 24 hours, dogs were monitored twice daily for signs of ivermectin toxicosis (10 min/dog). Dogs were observed once daily for the first week for any signs of implant site reactions and again at 30, 90, 180, and 365 days after implantation. On day 365, all dogs received 50 D immitis infective L₃ that were cultured and administered according to published methods.¹⁴ The infections were allowed to establish between days 365 and 510. On day 510, all dogs were euthanatized and the numbers of adult heartworms in the heart and pulmonary vessels of each dog were counted.¹⁵ Percentage effectiveness of treatment was calculated by comparing the geometric mean total adult heartworm counts from control and treated dogs according to the following formula:

View Expanded

Field study—A field study was conducted in Northern Australia^q with 15 client-owned dogs of mixed breed, age (range, 2 to 9 years), and sex (7 males and 8 females) that weighed between 5.3 and 36.5 kg and were recruited by means of a client database through 2 veterinary practices located in Cairns, Queensland. Dogs were included on the basis of being in good health and having negative results of heartworm antigen tests, body weight from 5 to 40 kg, and a > 12-month history of prophylactic heartworm medication with good compliance history as determined from veterinary practice records. Informed written consent was obtained, and preimplantation blood samples were collected at −4 and 0 weeks for measurement of plasma ivermectin concentrations and heartworm antigen. Heartworm antigen was measured by use of a canine heartworm antigen test kit according to the manufacturer's instructions.^r Covered-rod implants measured 1.6 × 24 mm and contained 7.3 mg of ivermectin; dogs received implants at a rate of 1 implant/20 kg of body weight (ie, 5- to 20-kg dogs received 1 implant and 21- to 40-kg dogs received 2 implants). Physical examinations were conducted, and blood samples were collected at 0, 1, 18, 26, and 52 weeks after administration of the implant for measurement of plasma concentrations of ivermectin and circulating canine heartworm antigen. A dog with positive results of the heartworm test at ≤ 18 weeks would be excluded from the study because any infection detected at this time would have been caused by pretrial infection.

Statistical analysis—Area under the concentration-time curve was calculated from 0 to 4 and 0 to 16 weeks in study 1 and 0 to 57 weeks in study 2 and was determined by use of the linear trapezoidal rule.^s Comparisons of multiple implants (study 1) were performed via 1-way ANOVA.^t The AUCs for 4-week and 16-week samples from study 1 were calculated by use of mean serum concentrations for each sampling point. Both 4-week and 16-week determinations were used. Area under the concentration-time curve comparisons were made by use of the 4-week AUC data. Comparisons of AUC and Cmax between implant lengths (study 2) were performed by use of 1-way ANOVA. The AUC and Cmax comparisons between groups that received implants E or F were assessed by use of a Student t test for unpaired data. Results with values of P < 0.05 were considered significant. Where a significant difference in pharmacokinetic parameters caused by effects of implant size was observed, the data were further analyzed by the use of Fisher protected least squares differences post hoc tests.

Results

Preclinical study 1—Peak mean serum concentrations (19.2 to 28.0 ng/mL) were reached 1 to 2 weeks after implantation. The highest mean concentration (28.0 ng/mL) was observed in the group of dogs given implant F. After reaching peak serum concentration, there was a steady decrease in serum ivermectin concentration for implants A to E, whereas a second peak occurred 4 weeks after implantation for implant F. For dogs that received implants E and F, mean serum ivermectin concentrations at 16 weeks were 8.1 and 6.8 ng/mL, respectively. No significant main effect of implant formulation on AUC (P = 0.32) or Cmax (P = 0.14) at 4 weeks was detected (Table 3).

Preclinical study 2—Mean serum ivermectin profiles (Figures 1 and 2) obtained from dogs that received E and F implants and pharmacokinetic values were determined (Table 4). A significant (P = 0.03) main effect of implant length on AUC but not Cmax (P = 0.58) was detected. Fisher protected least squares difference post hoc tests revealed a significantly greater difference in AUC between implants of 30 mm and 12 mm (P = 0.005) and between implants of 30 mm and 18 mm (P = 0.02). Mean AUC for weeks 0 to 57 ranged from 60.5 ng•wk/mL to 115.2 ng•wk/mL.

View Full Size

Mean serum ivermectin concentrations in dogs (n = 4/group) that were administered CR silicone implants (implant E [various lengths]) on day 0.

Citation: American Journal of Veterinary Research 67, 9; 10.2460/ajvr.67.9.1564

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Mean serum ivermectin concentrations in dogs (n = 4/group) that were administered CR silicone implants (implant F [various lengths]) on day 0.

Citation: American Journal of Veterinary Research 67, 9; 10.2460/ajvr.67.9.1564

• Download Figure
• Download figure as PowerPoint slide

Mean Cmax for implants of different lengths ranged from 2.6 to 3.9 ng/mL. No significant differences were detected in serum ivermectin concentrations between implants E and F when administered at the same size (P = 0.93, 0.35, 0.10, and 0.14, for 12-, 18-, 24-, and 30-mm lengths, respectively).

Although there were no significant differences between formulations E and F with respect to pharmacokinetic variables, there were differences in maintenance of serum ivermectin concentration ≥ 0.2 ng/mL for all dogs in each group. All dogs given E implants, at either 24 mm or 30 mm, had serum ivermectin concentrations ≥ 0.2ng /mL at 52 weeks, with mean serum ivermectin concentrations of 0.77 and 1.09 ng/mL respectively. Mean ivermectin concentration was > 0.2 ng/mL for all dogs given E implants (any length) and F implants at 24 mm and 30 mm. Ivermectin was not detectable in approximately half the dogs receiving implant F at 24 or 30 mm and in dogs receiving implant E at 12 or 18 mm.

Preclinical study 3—Serum ivermectin concentrations of 14 dogs from 0 to 56 weeks after implantation with a single CR implant were measured. Mean serum ivermectin concentrations at 52 and 56 weeks were 0.5 and 0.7 ng/mL, respectively; the range at 52 weeks was 0.23 to 1.1 ng/mL and at 56 weeks was 0.44 to 1.8 ng/mL.

Clinical studies 1 and 2—A mean of 29.8 adult heartworms/dog were recovered from the 24 dogs in the 2 control groups (Table 5). No adult heartworms were recovered from the 24 treated dogs. Mean plasma ivermectin concentration of treated dogs at challenge (ie, 365 days after implantation) was 0.4 ng/mL (range, 0.15 to 0.64 ng/mL). A single silicone implant containing 7.3 mg of ivermectin was 100% effective in preventing infection with 50 infective L₃D immitis inoculated 12 months after treatment.

Field study—Thirteen of the 15 dogs enrolled in the study completed the full 12 months of the study. One dog was hit by a car and killed and was consequently lost from the study. A second dog had a positive heartworm antigen test result at 18 weeks and was excluded from the study; a subsequent canine heart worm antigen test from a repeat plasma sample yielded negative results. Of the 13 dogs that completed the field study, all had plasma concentrations of ivermectin > 0.2 ng/mL 52 weeks after implantation. Mean plasma ivermectin concentrations at 1, 18, and 26 weeks after implantation were 5.6 ± 2.93 ng/mL, 1.07 ± 0.64 ng/mL, and 1.04 ± 0.39 ng/mL, respectively. Mean plasma ivermectin concentration at 52 weeks was 0.89 ± 0.61 ng/mL (range, 0.37 to 2.8 ng/mL). None of the 13 dogs had detectable circulating heartworm antigen at 78 weeks after implantation.

Discussion

Members of the macrocyclic lactone family are highly effective in preventing heartworm infection when used consistently, irrespective of their mode of application.¹⁶ The products are also extremely safe because they are used at very low concentrations. Nevertheless, their effectiveness is often compromised by lack of owner compliance, which exposes dogs to infection at critical times of the year.¹⁷ Additionally, shifting weather patterns may alter the distribution and concentration of vectors responsible for transmission of D immitis, making seasonal administration of these compounds a less reliable means of preventing infection. Year-round administration is thought to avoid the difficulties associated with correct timing of initiation and cessation of preventative administration. Development of an injectable form of moxidectin that provides protection for 6 months was intended to remove lack of owner compliance from prevention programs; however, not all regions have a transmission period as short as 6 months. Development of a product providing 12 months of protection that is both safe to use and efficacious would be of considerable benefit to dog owners and veterinarians.

Dirofilaria immitis, especially the fourth-stage larvae, are susceptible to ivermectin. On the basis of results of laboratory and field studies, it has been determined that ivermectin administered at 6 μg/kg PO, once per month, provides protection against canine heartworm disease.³ Even when administered every 2 to 3 months, ivermectin remains highly efficacious in preventing canine heartworm infection, and doses of 1 to 2 μg/kg have good efficacy in field studies.³ In a study conducted by Daurio et al,¹³ ivermectin administered PO at 6 μg/kg reached a peak plasma concentration of approximately 3.0 ng/mL in 2 to 4 hours and this concentration decayed exponentially after this. Within 72 hours of its administration, plasma concentration was approximately 0.5 ng/mL and by 14 days had decreased to < 0.2 ng/mL. On the basis of these considerations, it was anticipated that an implant delivering a serum concentration of ≥ 0.2 ng/mL would provide 100% protection against canine heartworm infection. It is not known whether lower concentrations of ivermectin would be required when administered in a continuous fashion (as with CR implants), as opposed to the pulsatile release from monthly oral administration.

Administration of ivermectin results in a proportional relationship between the dose administered and the serum concentration attained, when the dose is within the standard therapeutic range. The serum concentration of ivermectin measured in dogs given 10 implants (preclinical study 1) was 10 times the serum concentration achieved after administration of commercial products that contain ivermectin for heartworm control in dogs.¹³ The dose of ivermectin delivered from each CR implant was greater than the putative therapeutic value, suggesting that each CR implant resulted in a peak serum concentration of 2.5 to 3.0 ng/mL, similar to that observed after administration of ivermectin PO at 6 μg/kg.¹³ This was supported by the finding in the second preclinical study that dogs given a single CR implant (1.6 × 12 mm) identical in composition to implants used in the first study had a Cmax from 2.9 to 3.0 ng/mL.

Changes in the dissolution additives can be used to increase the serum concentration of ivermectin. However, care must be taken to ensure that the rate of release, although resulting in therapeutic concentrations of ivermectin, is not too rapid. Comparison of the single CR implants E and F in preclinical study 2 revealed that there was little difference in the Cmax irrespective of the amount of sodium desoxycholate in the 2 formulations. Administration of both formulations resulted in serum ivermectin concentrations greater than the theoretical minimal therapeutic concentration. There were, however, significant differences between the 2 formulations in the duration that the therapeutic concentrations were maintained. In preclinical study 2, differences in the length (ie, 24 or 30 mm vs 12 or 18 mm) of implant were associated with differences in sustained duration of therapeutic concentrations of ivermectin without any differences in Cmax. This observation was similar to results of Maeda et al,¹⁰ who reported that for hydrophobic ivermectin, the ivermectin is only released from the ends of the CR implant because the outer silicone layer provides an impermeable barrier to drug dissolution and release. For CR implants that contain ivermectin, the cross-sectional area restricts release. Implants with the same formulation and same cross-sectional area but of different lengths yield the same rate of release, despite the greater gross amount of active drug in longer implants, because the percentage of ivermectin remains unaltered, along with the rate of dissolution. The extended release profile of longer implants is attributable to the greater amount of ivermectin contained in the implant and the increased time required for water to penetrate the entire length of the implant and dissolve the drug.

Results of clinical studies 1 and 2 indicated that the implants were highly efficacious in preventing infection with D immitis in dogs challenged on day 365 after administration of the implants with high numbers of larvae, which are not likely to be encountered via natural challenge. Plasma concentrations of ivermectin at the time of challenge (ie, 365 days after administration) ranged from 0.15 to 0.64 ng/mL. Hence, a plasma concentration of approximately 0.2 ng/mL can be considered to provide 100% protection against heartworm infection in dogs. The lowest protective concentration of ivermectin cannot be concluded from these studies. Plasma concentrations < 0.2 ng/mL may be equally effective, and in support of this, a single treated dog with a plasma ivermectin concentration of 0.15 ng/mL was found to be free of infection at necropsy.

Effectiveness of the implants was evaluated in a field study undertaken in Northern Australia, an area with historically high prevalence of canine heartworm infection. Heartworm infection was not detected in any of the dogs. Plasma concentrations of ivermectin at 52 weeks were > 0.2ng/mL.

The higher plasma concentrations of ivermectin in dogs in the field study were most likely attributable to most dogs receiving 2 implants at day 0 because of body weights > 20 kg. In this situation, the concentrations derived from each implant were additive. In 1 dog in the field study, a 52-week plasma ivermectin concentration of 2.8 ng/mL was observed, reflecting the much smaller size of this dog (5 kg).

None of the dogs (n = 120) treated with CR implants had any adverse systemic reactions, even when 10 implants were administered. No swellings were detected at the site of implantation. Veterinarians and dog owners participating in the field trial did not report any concerns or signs of discomfort in their dogs.

Some breeds of dogs, in particular Collies,¹⁸ have increased susceptibility to the toxic effects of ivermectin, which is caused by a defect in the multi–drugresistance gene (MDR1) that encodes for P-glycoprotein, an integral component of the blood-brain barrier,¹⁹ resulting in high concentrations of ivermectin in the brain. Toxicosis is observed when Collies are given ivermectin at concentrations 15 times the therapeutic dose. The CR implants used in the study reported here resulted in a peak plasma concentration from 3 to 7 ng/mL; this concentration of ivermectin is not likely to result in toxicosis in Collies.