Dirofilariasis mouse models for heartworm preclinical research.

Introduction: Dirofilariasis, including heartworm disease, is a major emergent veterinary parasitic infection and a human zoonosis. Currently, experimental infections of cats and dogs are used in veterinary heartworm preclinical drug research. Methods: As a refined alternative in vivo heartworm preventative drug screen, we assessed lymphopenic mouse strains with ablation of the interleukin-2/7 common gamma chain (γc) as susceptible to the larval development phase of Dirofilaria immitis. Results: Non-obese diabetic (NOD) severe combined immunodeficiency (SCID)γc-/- (NSG and NXG) and recombination-activating gene (RAG)2-/-γc-/- mouse strains yielded viable D. immitis larvae at 2-4 weeks post-infection, including the use of different batches of D. immitis infectious larvae, different D. immitis isolates, and at different laboratories. Mice did not display any clinical signs associated with infection for up to 4 weeks. Developing larvae were found in subcutaneous and muscle fascia tissues, which is the natural site of this stage of heartworm in dogs. Compared with in vitro-propagated larvae at day 14, in vivo-derived larvae had completed the L4 molt, were significantly larger, and contained expanded Wolbachia endobacteria titres. We established an ex vivo L4 paralytic screening system whereby assays with moxidectin or levamisole highlighted discrepancies in relative drug sensitivities in comparison with in vitro-reared L4 D. immitis. We demonstrated effective depletion of Wolbachia by 70%-90% in D. immitis L4 following 2- to 7-day oral in vivo exposures of NSG- or NXG-infected mice with doxycycline or the rapid-acting investigational drug, AWZ1066S. We validated NSG and NXG D. immitis mouse models as a filaricide screen by in vivo treatments with single injections of moxidectin, which mediated a 60%-88% reduction in L4 larvae at 14-28 days. Discussion: Future adoption of these mouse models will benefit end-user laboratories conducting research and development of novel heartworm preventatives via increased access, rapid turnaround, and reduced costs and may simultaneously decrease the need for experimental cat or dog use.

Recording drug responses from adult Dirofilaria immitis pharyngeal and somatic muscle cells.

Despite being considered one of the most pathogenic helminth infections of companion animals, members of macrocyclic lactone class are the only drugs available for the prevention of heartworm disease caused by Dirofilaria immitis. Alarmingly, heartworm prevention is at risk; several studies confirm the existence of macrocyclic lactone resistance in D. immitis populations across the United States. To safeguard the long term prevention and control of this disease, the identification and development of novel anthelmintics is urgently needed. To identify novel, resistance-breaking drugs, it is highly desirable to: Unfortunately, none of the three above statements can be answered sufficiently for D. immitis and most of our hypotheses derive from surrogate species and/or in vitro studies. Therefore, the present study aims to improve our fundamental understanding of the neuromuscular system of the canine heartworm by establishing new methods allowing the investigation of body wall and pharyngeal muscle responses and their modulation by anthelmintics. We found that the pharynx of adult D. immitis responds to both ivermectin and moxidectin with EC50s in the low micromolar range. We also demonstrate that the somatic muscle cells have robust responses to 30 µM acetylcholine, levamisole, pyrantel and nicotine. This is important preliminary data, demonstrating the feasibility of electrophysiological studies in this important parasite.

Heartworm and Wolbachia: therapeutic implications.

A safer, more effective adulticidal treatment and a safe method for reducing microfilaremia and breaking transmission of heartworm disease early in the treatment are needed. The present study evaluated efficacy of ivermectin (IVM) and doxycycline (DOXY) alone or together (with or without melarsomine [MEL]) in dogs with induced adult heartworm infection and assessed the ability of microfilariae from DOXY-treated dogs to develop to L3 in Aedes aegypti mosquitoes and subsequently to become reproductive adults in dogs. Thirty beagles were each infected with 16 adult heartworms by intravenous transplantation. Six weeks later, dogs were ranked by microfilarial count and randomly allocated to 6 groups of 5 dogs each. Beginning on Day 0, Group 1 received IVM (6 mcg/kg) weekly for 36 weeks. Group 2 received DOXY (10 mcg/(kgday)) orally Weeks 1-6, 10-11, 16-17, 22-25, and 28-33. Groups 3 and 5 received IVM and DOXY according to doses and schedules used for Groups 1 and 2. At Week 24, Groups 3 and 4 received an intramuscular injection of MEL (2.5 mg/kg), followed 1 month later by two injections 24h apart. Group 6 was not treated. Blood samples were collected for periodic microfilaria counts and antigen (Ag) testing (and later immunologic evaluation and molecular biology procedures). Radiographic and physical examinations, hematology/clinical chemistry testing, and urinalysis were done before infection, before Day 0, and periodically during the treatment period. At 36 weeks, the dogs were euthanized and necropsied for worm recovery, collection of lung, liver, kidney, and spleen samples for examination by immunohistochemistry and conventional histological methods. All dogs treated with IVM + DOXY (with or without MEL) were amicrofilaremic after Week 9. Microfilarial counts gradually decreased in dogs treated with IVM or DOXY, but most had a few microfilariae at necropsy. Microfilarial counts for dogs treated only with MEL were similar to those for controls. Antigen test scores gradually decreased with IVM + DOXY (with or without MEL) and after MEL. Antigen scores for IVM or DOXY alone were similar to controls throughout the study. Reduction of adult worms was 20.3% for IVM, 8.7% for DOXY, 92.8% for IVM + DOXY + MEL, 100% for MEL, and 78.3% for IVM + DOXY. Mosquitoes that fed on blood from DOXY-treated dogs had L3 normal in appearance but were not infective for dogs. Preliminary observations suggest that administration of DOXY+IVM for several months prior to (or without) MEL will eliminate adult HW with less potential for severe thromboembolism than MEL alone.

Ability of an oral formulation of afoxolaner to protect dogs from Borrelia burgdorferi infection transmitted by wild Ixodes scapularis ticks.

A randomized, blinded, negative controlled study was conducted to determine whether treatment with afoxolaner (NexGard®, Merial, Inc.) would prevent the transmission of Borrelia burgdorferi to dogs by wild caught Ixodes scapularis ticks. Twenty healthy dogs were randomly assigned to two groups of ten dogs each. Ten dogs were treated orally on Day 0 at a dose near the minimum recommended dose of afoxolaner of 2.5mg/kg (actual doses 2.5-3.1mg/kg) and ten control dogs were not treated. On Day 28, each dog was infested with approximately 50 adult unfed wild caught I. scapularis that had a 67% B. burgdorferi infection rate (determined by polymerase chain reaction). On Day 33, live ticks were counted and removed. No ticks were found on treated dogs while control dogs had an average of 21.4 ticks. To detect infection, the B. burgdorferi-specific C6 antibody SNAP® 4Dx® test (IDEXX) was performed on serum collected before infestation (all dogs seronegative on Days -6 and 27) and on Days 48, 63, 77 and 92. The ten treated dogs remained seronegative through the end of the study (Day 92), while nine out of the ten control dogs were infected, as demonstrated by their seroconversion to being positive for the presence of the B. burgdorferi-specific C6 antibody starting on Day 48. In this study, all dogs treated with NexGard® 28days prior to challenge with wild caught I. scapularis ticks were protected from B. burgdorferi infection, while nine out of the ten untreated control dogs were infected.

Evaluation of the efficacy of emodepside plus praziquantel topical solution against ascarid infections (Toxocara cati or Toxascaris leonina) in cats.

Eleven controlled studies were conducted in the United States and Europe to evaluate the efficacy of a topical solution of emodepside (3 mg/kg)+praziquantel (12 mg/kg) (Profender, Bayer AG, Leverkusen, Germany) against infection with various stages of the ascarid nematodes Toxocara cati and Toxascaris leonina. Infections were induced by administration of larvated ascarid eggs, and stage-specific efficacy was evaluated by treating cats at scheduled intervals post-inoculation. All studies featured random allocation to treatment groups, placebo-treated control animals and assessment of outcome measures by masked personnel. The product (emodepside+praziquantel topical solution) was 100% effective against mature adults and immature adult T. cati. In addition, it was 96.8% effective against third stage larvae and at least 99.4% effective against fourth stage larvae of T. cati, respectively. Efficacy against mature, immature adult and L4 stages of T. leonina exceeded 93.4%, but regulatory "adequacy of infection" criteria were not met in some studies. No adverse reactions to treatment were noted in cats treated with the emodepside+praziquantel topical solution.

Antifilarial agents. 1,2-Cyclobutanediamines as analogues of diethylcarbamazine. Status of structure-activity relationships among diethylcarbamazine analogues.

cis-and trans-1,2-cyclobutanediamines bearing appropriate N-methyl and N-acyl substituents were prepared as analogues of diethylcarbamazine (DEC). None displayed activity against Litomosoides carinii in the gerbil despite substantial structural and sterochemical similarities to the parent drug. The inactivity of these drugs is rationalized in terms of eclipsed pharmacophore configurations and the increased population of unfavorable rotational conformations made possible by the exocyclic position of both pharmacophores. To provide perspective for these conclusions, the literature on DEC analogues is briefly summarized and structure-activity data are discussed in terms of critical structural factors associated with microfilaricidal activity. Generalizations on structural principles governing activity are advanced which encompass test results for the large majority of DEC analogues.

Synthesis and biological activity of certain alkyl 5-(alkoxycarbonyl)-1H-benzimidazole-2-carbamates and related derivatives: a new class of potential antineoplastic and antifilarial agents.

A series of methyl and ethyl 5-(alkoxycarbonyl)-1H-benzimidazole-2-carbamates (7-19) and methyl 5-carbamoyl-1H-benzimidazole-2-carbamates (24-34) have been synthesized via the reaction of an appropriate alcohol or amine with the acid chloride derivatives 6a or 6b at room temperature. Reaction of an alcohol with acid chloride 6a at reflux temperature afforded transesterified products 20-23 in good yield. Treatment of methyl 5-amino-1H-benzimidazole-2-carbamate with substituted benzoyl chlorides furnished the methyl 5-benzamido-1H-benzimidazole-2-carbamates (36-38). Compounds 9, 16, 20, and 22 demonstrated significant growth inhibition in L1210 cells with IC50's less than 1 microM. Growth inhibition by this series of compounds appears to be associated with mitotic spindle poisoning. All the compounds tested, 9, 10, 19, 20, 22, and 23, caused significant accumulation of L1210 cells in mitosis. Compounds 7, 9, 19, 25, 26, 27, and 36 showed significant in vivo antifilarial activity against adult worms of Brugia pahangi, Litomosoides carinii, and Acanthocheilonema viteae in experimentally infected jirds.

Synthesis and antifilarial activity of N-[4-[[4-alkoxy-3-[(dialkylamino)methyl]phenyl]amino]- 2-pyrimidinyl]-N'-phenylguanidines.

A series of N-[4-[[4-alkoxy-3-[(dialkylamino)methyl]phenyl]amino]- 2-pyrimidinyl]-N'-phenylguanidines have been synthesized for antifilarial evaluation. Reaction of the appropriate benzenamines with N-cyanoguanidine, followed by condensation of the resultant N-phenylimidodicarbonimidic diamides (V) with ethyl 4,4,4-trifluoro-3-oxobutanoate provided the intermediate N-(4-hydroxy-2-pyrimidinyl)-N'-phenylguanidines VIa. Alternatively, compounds VIa were synthesized by reaction of the requisite beta-keto esters (VII) with N-cyanoguanidine to give the (4-hydroxy-2-pyrimidinyl)cyanamides (VIII), followed by treatment with the desired benzenamines. Chlorination with POCl3 and condensation with the appropriate benzenamines (IX) generated the desired guanidines (X). Antifilarial activity was confined to adult Litomosoides carinii infections, and a structure-activity relationship for this activity is discussed. Lack of activity against L. carinii microfilaria and adult Brugia pahangi infections preclude further work in this area pending evaluation in additional experimental models.

Bis-cationic heteroaromatics as macrofilaricides: synthesis of bis-amidine and bis-guanylhydrazone derivatives of substituted imidazo[1,2-a]pyridines.

A series of guanylhydrazone, amidine, and hydrazone derivatives of 2-phenylimidazo[1,2-a]pyridine have been prepared and evaluated for macrofilarial activity against Acanthocheilonema viteae and Brugia pahangi in jirds. Compounds with 4',6-bis-substitution by cyclic guanylhydrazone groups show activity. 4',6-Bis-amidines show some activity but are more toxic; 4'- or 6-monosubstituted compounds are inactive. 2,6-Bis-substituted compounds lacking the phenyl ring are inactive. 4',6-Bis-substituted compounds having additional double bonds inserted between the heterocyclic ring and the phenyl ring or between the substituent and the ring system show reduced activity.

2-acetylpyridine thiosemicarbazones. 13. Derivatives with antifilarial activity.

Several members of a series of 2-acetylpyridine thiosemicarbazones possess in vivo and in vitro macrofilaricidal properties. The most promising of the group tested is N4-(2-aminophenyl)-2-[1-(2-pyridinyl)ethylidene]-hydrazinecarbothioam ide (4), which suppressed 100% of the macrofilariae of Brugia pahangi and 94% of those of Acanthocheilonema viteae in the jird at a dose of 25 mg/kg per day x 5. Compounds 4 and 14 were also shown to inactivate or kill Onchocerca gutturosa and Onchocerca volvulus adult worms as measured by the loss of their motility or the inhibition of the conversion by the worms of the dye MTT to formazan.

Synthesis of potential antifilarial agents. 1-(5-Benzoylbenzimidazol-2-yl)-3-alkyl- and -arylureas.

A series of 1-(5-benzoylbenzimidazol-2-yl)-3-substituted ureas have been synthesized by reacting an appropriate isocyanate with 2-amino-5-benzoylbenzimidazole or by reacting methyl (5-benzoylbenzimidazol-2-yl)carbamate with various amines. Several of the compounds have demonstrated antifilarial activity against Brugia pahangi and Litomosoides carinii.

Antigenic role of the endosymbionts of filarial nematodes: IgG response against the Wolbachia surface protein in cats infected with Dirofilaria immitis.

Filarial nematodes harbour intracellular endosymbiotic bacteria, which have been assigned to the genus Wolbachia. These bacteria appear to play an important role in the pathogenesis of filarial diseases through their lipopolysaccharides. In view of the presence of Wolbachia endosymbionts in the body of filarial nematodes, one might also expect that proteins from these bacteria play an antigenic role in humans and animals affected by filariases. To test this hypothesis, we produced in recombinant form the surface protein WSP and a portion of the cell-cycle protein FTSZ from the Wolbachia of Dirofilaria immitis. Western immunoblot assays were then performed using cat sera to test the immunogenicity of these proteins. Sera were collected from owners' cats, which were either sero-negative or sero-positive for D. immitis and from cats before and after experimental infection with D. immitis. FTSZ was recognized in Western blots by sera from both positive and negative cats and from both uninfected and experimentally infected cats. WSP was recognized only by sera from positive cats and from cats experimentally infected with D. immitis; this protein was not recognized by sera from negative cats and from cats before experimental infection with D. immitis. The results of Western blot assays on WSP thus support the hypothesis that infection with filarial nematodes induces the production of antibodies against Wolbachia proteins.

Effects of tetracycline on the filarial worms Brugia pahangi and Dirofilaria immitis and their bacterial endosymbionts Wolbachia.

Wolbachia endosymbiotic bacteria have been shown to be widespread among filarial worms and could thus play some role in the biology of these nematodes. Indeed, tetracycline has been shown to inhibit both the development of adult worms from third-stage larvae and the development of the microfilaraemia in jirds infected with Brugia pahangi. The possibility that these effects are related to the bacteriostatic activity of tetracycline on Wolbachia symbionts should be considered. Here we show that tetracycline treatment is very effective in blocking embryo development in two filarial nematodes, B. pahangi and Dirofilaria immitis. Embryo degeneration was documented by TEM, while the inhibition of the transovarial transmission of Wolbachia was documented by PCR. Phylogenetic analysis on the ssrDNA sequence of the Wolbachia of B. pahangi confirms that the phylogeny of the bacterial endosymbionts is consistent with that of the host worms. The possibility that tetracycline inhibition of embryo development in B. pahangi and D. immitis is determined by cytoplasmic incompatibility is discussed.

Tetracycline treatment and sex-ratio distortion: a role for Wolbachia in the moulting of filarial nematodes?

Filarial nematodes harbour intracellular bacteria of the genus Wolbachia. These bacteria are thought to be beneficial to the host nematode. Indeed, tetracycline treatments reduce the population of Wolbachia in filarial worms and have detrimental effects on the nematode. Even though various antibiotic-curing experiments have been performed on filariae, the actual role of Wolbachia in the biology of these nematodes is not yet clear. To address this issue, we designed a first experiment on a model filaria (Brugia pahangi), maintained in the gerbil (Meriones unguiculatus). In this experiment, timing of tetracycline treatment was set on the basis of the larval stage of the nematode. This first experiment showed that 2 weeks of treatment started after the L(4)-L(5) moult of males, but before the moult of females, led to significant sex-ratio distortion of the nematodes. We thus hypothesised that tetracycline interferes with the moult in B. pahangi. To test this hypothesis, we designed a second experiment in which antibiotic treatments were started (1). before the moult of both sexes, (2). after the moult of males but before the moult of females, or (3). after the moult of both sexes. Treatment 1 determined a reduction of worm recovery with no sex bias. Treatment 2 led to a male-biased sex-ratio. Treatment 3 had no effect on either worm recovery or sex-ratio. These results thus support the hypothesis that tetracycline treatment interferes with the L(4)-L(5) moult of B. pahangi. The nematodes recovered from the treated and control animals were examined for the presence of Wolbachia using both immunohistochemistry and real-time PCR. In general, nematodes from treated animals showed a dramatic reduction in Wolbachia content. In one group, Wolbachia depletion, as observed at the end of the treatment, was followed by a rebound to 'normal' values 160 days later. Prospects for antifilarial therapy using Wolbachia-targeted tetracycline treatments should thus take into account the possibility of Wolbachia rebound.

Pharmacokinetics of UMF-078, a candidate antifilarial drug, in infected dogs.

The pharmacokinetics of the filaricidal benzimidazole compounds UMF-078 and UMF-289 were evaluated in beagle dogs experimentally infected with Brugia pahangi. Twenty-four infected microfilaremic beagles were selected and randomly allocated into 4 treatment groups of 6 dogs each: oral (PO) UMF-078, PO UMF-289 (the HCl salt form of UMF-078), intramuscular (IM) UMF-078, and untreated controls. Equivalent doses of 50 mg/kg of the free base were given twice a day for 3 days to the 3 groups of treated dogs. Oral absorption is rapid compared with IM dosing; the absorption half-life (K01-HL) for the IM treatment is approximately 14 hr compared with 1 and 2 hr for the PO regimen of salt and free base forms, respectively. The elimination half-lives (K10-HL) for the PO regimens are 13 and 15 hr for the salt and free base forms, respectively. Because of sustained absorption following IM dosing, the K10-HL is prolonged. In contrast to oral administration, IM dosing of UMF-078 provides sustained, relatively low plasma drug levels, with good tolerance and efficacy.

Filariasis testing in a jird model: new drug leads from some old standbys.

A total of 65 compounds, most of which were from chemical classes having members known to be active against one or more parasitic organisms, were evaluated against Brugia pahangi and Acanthocheilonema viteae for macrofilaricidal activity in male Mongolian jirds (Meriones unguiculatus). Sixteen of the 65 compounds tested suppressed the number of parasites. Of these 16, three were suppressive for B. pahangi, 10 for A. viteae, and three for both parasites. The antibiotic nigericin and the antihistaminic isothipendyl were found to be most active.

The ferret (Mustela putorius furo) as an experimental host for Brugia malayi and Brugia pahangi.

Ferrets inoculated subcutaneously with 150--200 infective larvae of Brugia malayi (subperiodic strain) usually developed patent infection during the 3rd month post inoculation. Microfilaremia was transient, and most animals became amicrofilaremic after the 6th month of infection. Ferrets developed a persistent eosinophilia at the time of patency. At necropsy, 5--8 months post infection, adult worms were recovered principally from lymphatic vessels and recovery ranged from 0.5--13% of the inoculated larvae. The inflammatory response of ferrets to microfilariae was characterized by nodules 1--5 mm in diameter in the liver, lungs, spleen, and submucosa of the gastrointestinal tract. The center of these lesions contained a degenerated microfilaria or the cast of a microfilaria embedded in Splendore-Hoeppli substance. The Splendore-Hoeppli substance was surrounded by eosinophils and/or foreign body giant cells. Identical lesions were observed in ferrets experimentally infected with Brugia pahangi. Sera from ferrets infected with B. malayi demonstrated a 3- to 5-fold increase in IgG by the 4th month of infection and these sera produced 2--3 precipitin bands in double gel diffusion assays with an extract of B. malayi microfilarial antigen. Skin tests with B. malayi microfilarial antigen showed that the majority of the infected ferrets had immediate hypersensitivity responses, but none had Arthus or delayed hypersensitivity responses.

Physicochemical studies of lidocaine-menthol binary systems for enhanced membrane transport.

The melting properties of lidocaine and l-menthol binary systems were studied using differential scanning calorimetry (DSC). A eutectic mixture was obtained for the lidocaine:menthol ratio of 30:70 (w:w) with a eutectic point of 26 degrees C. The binary melt systems formed within a range of 30:70-50:50 (w:w) remained as homogeneous oils at ambient temperature. The solubilities of pure lidocaine and lidocaine from the binary melt systems were determined with and without propylene glycol in pH 8.0 phosphate buffer. Lidocaine from the melt systems was less soluble in the buffers due to the partition of lidocaine between the oil and aqueous phases. The addition of propylene glycol to the buffer significantly increased both the solubility and heat of solubilization of lidocaine. The permeation rates of lidocaine from the binary melt systems across shed snake-skin were concentration dependent and significantly higher than those from the reference solutions.

HPLC assay of Lidocaine in plasma with solid phase extraction and UV detection.

A sensitive and reliable method based on solid-phase extraction and reversed-phase liquid chromatography was developed and validated for the quantitation of Lidocaine (Lid) in dog plasma. Phenacemide was used as an internal standard (IS) in the extraction which employed C18 solid-phase extraction cartridges. The washing and eluting solutions were 2 ml acetonitrile-pH 9.0 phosphate buffer (10:90 v/v) and 0.5 ml acetonitrile-pH 4.0 phosphate buffer (40:60 v/v). respectively. The eluent obtained from the cartridge was directly analyzed on a reversed-phase ODS column with UV detection at 210 nm. A clean chromatogram and high sensitivity were achieved at this wavelength. The mobile phase was acetonitrile and pH 5.9 phosphate buffer (20:80 v/v). The retention times were 6.4 and 7.2 min for Lid and IS, respectively, at a flow rate of 1.0 ml min(-1). The mean absolute recovery was 96.6% (n = 9) with a CV of 3.8% for Lid and 81.7% with CV of 2.5% (n = 3) for IS. The limit of quantitation was 20 ng ml(-1), with the intra- and inter-day precisions (n = 5) of 4.4 and 3.4%, respectively, and the intra- and inter-day accuracies (n = 5) of -4.3 and -5.0%, respectively. For the analyses of Lid in spiked plasma samples at 20, 100 and 200 ng ml(-1), the overall mean intra- and inter-day precisions (n = 15) were 3.9 and 4.9%, respectively, and the overall mean intra- and inter-day accuracies (n = 15) were -3.7 and -4.6%, respectively. The correlation coefficients for calibration plots in the range 20-1000 ng ml(-1) in plasma were typically higher than 0.998. The suitability of the method was demonstrated by the study in a beagle dog receiving a low intravenous dose of Lid.

Efficacy of long-term monthly administration of ivermectin on the progress of naturally acquired heartworm infections in dogs.

This study was designed to evaluate the efficacy of prolonged monthly ivermectin treatment against Dirofilaria immitis in client-owned dogs with naturally acquired infections and to clinically monitor the animal's response to the slow killing of heartworms, with death of the worms distributed over a period of up to 2 years. A total of 17 male and female dogs of different breeds and ages were used. Prior to treatment, all of the dogs tested positive for heartworm antigen (Ag) and all but two had microfilariae (mf). The dogs were randomly allocated to one group of seven dogs which received a commercial formulation of ivermectin (minimum, 6 mcg IVM/kg) plus pyrantel (minimum, 5 mg PP/kg) (Heartgard Plus Chewables, Merial, Ltd.), another group of seven dogs which received a commercial formulation of IVM (min, 6 mcg/kg) (Heartgard Chewables, Merial Ltd.), and a group of three dogs which served as an untreated controls. All dogs were evaluated prior to initiation of treatment and thereafter at 3- to 5-month-intervals for mf, Ag, and radiographic and echocardiographic findings. All of the 17 dogs, with the exception of two dogs in the IVM group, had circulating mf of D. immitis prior to the 1st monthly dose, and a few also had mf of Dirofilaria repens. After 4 monthly doses, only one dog in the IVM/PP group and two dogs in the IVM group had a patent heartworm infection, and no heartworm mf were seen in the 14 treated dogs thereafter. After 10 monthly doses, the number of Ag-positive dogs in both of the treated groups decreased gradually. Efficacy, based on the reduction in number of Ag-positive dogs, was similar for the IVM/PP and IVM groups, with overall efficacy scores for the 14 dogs of 21, 21, 43, and 71% after 10, 14, 19, and 24 monthly doses, respectively. Two of the seven dogs treated with IVM/PP, one of the seven treated with IVM, and two of the three untreated controls showed echocardiographic evidence of a parasitic burden prior to treatment, and all of these scores had decreased by the end of the study. Only one dog (IVM/PP group) had a cardiovascular pattern of heartworm disease by echocardiography prior to treatment, but this dog's score increased to two and the scores of two additional dogs increased from zero to two (IVM group) or three (IVM/PP group) by the end of the study. Only 1 (IVM/PP group) of the 17 dogs showed a pulmonary pattern of heartworm disease by radiography prior to treatment, but this dog's score increased to three by the end of the study. The radiographic scores of two additional dogs in the treated groups increased from zero to three (IVM/PP) or two (IVM) by the end of the study. Thus, monthly administration of IVM to dogs with clinical, radiographic or echocardiographic evidence of heartworm disease is ill-advised and such treatment of even the asymptomatic dog should be done only with much caution and frequent monitoring by the veterinarian.