Cross-reactions of sera from dogs infected with Angiostrongylus vasorum in commercially available Dirofilaria immitis test kits

Background Dirofilaria immitis and Angiostrongylus vasorum are both important potentially fatal canine nematodes with overlapping endemic areas, especially in Europe. The preadult and adult stages of both species are living in the Arteria pulmonalis and the right heart, and diagnostically detectable circulating parasite antigens have been demonstrated for both species. For the detection of D. immitis infections, a variety of commercial tests have been developed, however, they have not been evaluated for cross-reactions against circulating antigens of A. vasorum. Methods In this study, potential cross-reactions of sera from 16 dogs, which were experimentally infected with A. vasorum and which had circulating antigens as confirmed by a species-specific ELISA, were evaluated for the detection of A. vasorum antigen in six commercially available D. immitis test kits. Results In three fast tests (Witness® Dirofilaria, SensPERT® Canine Heartworm, SNAP® 4Dx® Plus), all sera were negative. One fast membrane ELISA (SNAP® HTWM RT Test) was positive with four sera (25%), and one serum delivered a non-valid result twice. In the PetChek® HTWM PF Test, depending on the interpretation protocol, 5 or 8 dogs (31.2 – 50%) were positive. With the DiroCHEK®-ELISA, a single A. vasorum-infected dog (6.2%) tested positive. Conclusions Due to potential cross-reactions with A. vasorum in commercially available test kits for the detection of D. immitis antigen, the simultaneous use of highly specific diagnostic methods for the differentiation of these two canine heart worms is recommended.


Background
The adult stages of Dirofilaria immitis, a filarial nematode, and Angiostrongylus vasorum, a metastrongylid nematode, are both localized in the Arteria pulmonalis and the right heart of their definitive hosts. Dogs, foxes and some other carnivores are the definitive hosts of both parasites, while Culicidae and Gastropoda are the intermediate hosts of D. immitis and A. vasorum, respectively.
In Europe, D. immitis is present in coastal Mediterranean areas with expansion northwards, while in North America the parasite has expanded from the southeastern coastal areas northwards and westwards [1] up to Canada [2]. A. vasorum was diagnosed for the first time in France in 1913 [3], but it is only recently that this parasite has regained attention within the veterinary community [4,5]. Its presence has been increasingly reported from several new areas in and outside Europe (reviewed in [6]). Reports of an increasing number of cases of canine angiostrongylosis, as well as the development of new diagnostic tools such as ELISAs [7][8][9][10] or biomolecular techniques [11,12] may have contributed and also incentivised epidemiological studies, confirming the presence of this parasite in dogs, foxes and snails throughout Europe. The Atlantic provinces of Newfoundland and Labrador are the only regions actually affected by A. vasorum in North America, with a potential for expansion to further regions [4,13,14]. Overlapping areas in large parts of southern Europe with the presence of both A. vasorum and D. immitis have therefore to be accounted for. Furthermore, in non-endemic areas of D. immitis, this agent has to be considered based on anamnestic information (travelling with pet dogs or imports) and differentiated from A. vasorum infections.
Fatal clinical consequences of D. immitis infection are usually prevented by the monthly use of macrocyclic lactones in known endemic areas [15][16][17], and treatment of dirofilariosis is based on the intramuscular application of the arsenic derivate melarsomine [18] and/or, alternatively, by eliminating the endosymbiont Wolbachia with doxycycline supported by administration of macrocyclic lactones [19]. Dogs infected with A. vasorum, instead, are treated using macrocyclic lactones such as moxidectin [20] or milbemycin-oxime [13], or applying fenbendazole [21]. Prophylactic treatment (with macrocyclic lactones) against potentially fatal canine angiostrongylosis is, as for dirofilariosis, recommended in highly endemic and well known areas [20].
The currently used diagnostic laboratory methods for the detection of these parasites are divergent. The diagnosis of D. immitis is based on the detection of microfilariae or circulating antigens released by mature adult female worms into the blood circulation, both being detectable starting from 6 months after infection [18]. A variety of tests have been developed for the detection of circulating antigens, employing lateral flow immunochromatographic techniques, membrane ELISAs or conventional ELISAs [22][23][24]. Test evaluations showed that the sensitivity of heartworm antigen tests depends on the worm burden, and the sex and age of the parasites [24][25][26][27][28], while the specificity of the kits is regularly indicated to be very high, between 95% and 100% [23,25,26,29,30]. However, only occasionally were potential cross-reactions evaluated in animals with natural or experimental infections with other helminths, mainly against other filarial nematodes such as Dipetalonema reconditum [22,31] or Dirofilaria repens, indicating that modern test kits may overcome cross-reactions detected in previously developed test kits for these parasites [32,33], and, rarely, against intestinal parasites such as Ancylostoma caninum and Trichuris spp. [22,34]. The most current diagnostic method for detection of A. vasorum infections in dogs is the isolation of first stage larvae (L1) from faecal samples, which are produced by the parasites approximately 6-7 weeks after infection. Larval migration techniques such as the Baermann-Wetzel method [35] are commonly adopted. Furthermore, ELISAs for the detection of antibodies against A. vasorum have been described [7,9], and, recently, tests for the detection of circulating antigen of A. vasorum have been developed. These latter ones have been evaluated for cross-reactions against Crenosoma vulpis [8,10] and also against intestinal parasites (Toxocara canis, Ancylostoma caninum) and, importantly, against D. immitis [8], showing a high specificity (94-100%).
Due to their common localization within the definitive hosts, their considerable size and particularly the well documented production of circulating antigens [31,36], it was argued that antigens of A. vasorum and D. immitis may share epitopes responsible for potential crossreactions in antigen detection tests. This hypothesis has been confirmed during the development of the ELISA for the detection of circulating antigen for A. vasorum [8].
The aim of the study was to evaluate potential crossreactions of sera from dogs experimentally infected with A. vasorum in six different commercially available tests for the detection of D. immitis antigen.

Methods
A total of 16 sera from dogs experimentally inoculated with 200 third stage larvae (L3) of A. vasorum were obtained during previously performed studies [20,37]. Infection of dogs was confirmed by positive Baermann-Wetzel analyses [35], by the detection of circulating A. vasorum antigen [8] and by the presence of adult worms at necropsy adopting an established method of reverse lung perfusion [20]. The day of sample collection (between 55 and 356, mean 101) after inoculation (dpi) and the number of detected parasites at necropsy are shown in Table 1. Worm burdens varied between 10 and 170 adult parasites, with a mean of 66 worms per dog. Dirofilaria immitis infection was excluded based on the fact that the dogs were living in a non-endemic area under controlled experimental conditions, and at necropsy.
All tests were performed blinded with an identity code from 1 to 16, by veterinarians (fast tests) or by experienced laboratory technicians from the IPZ (ELISAs).
All sera were non-haemolytic, stored at −20°C and tested within 11-40 months after collection. The following test kits were used, adopting the manufacturer's instruction and within the indicated expiry dates:

Results
Single test results are shown in Table 1. All tests, with one exception (see Tab. 1, dog Av 8) fulfilled the criteria for test validity based on the results of positive and/or negative controls.
The ELISA for the detection of circulating A. vasorum antigen was highly positive for all experimentally infected dogs, with absorbance values (optical density read at 405 nm, OD) varying between 0.263 and 2.007 (cut-off value: 0.159, as previously described [8]), with a mean of 1.096.
In three fast tests (Witness W , SensPERT W , SNAP W 4Dx W Plus) all sera resulted negative, while in one fast membrane ELISA (SNAP W HTWM RT) four A. vasorum infected dogs were positive for D. immitis antigen, and one serum delivered a non-valid result twice. In the PetChek W -ELISA two methods for interpretation were adopted: following the instruction for veterinary practitioners based on eye detection, 5 dogs resulted positive for D. immitis infection, while following the instructions under laboratory conditions with OD measurements, a total of 8 dogs were seropositive. With the DiroCHEK W -ELISA, a single A. vasorum infected dog was D. immitis seropositive. With one exception (76 dpi), all crossreactions were observed in dogs infected with A. vasorum for more than 90 days, with worm burdens varying from 10-170.

Discussion
This study provides evidence of false positive reactions in D. immitis antigen detection kits with sera of dogs infected with A. vasorum. The two ELISAs for D. immitis detection (PetChek W and DiroCHEK W ) and the membrane ELISA SNAP W HTWM RT showed single crossreactions against A. vasorum, which had not been considered so far. In contrast, the adopted ELISA for the detection of circulating A. vasorum antigen has been developed evaluating different monoclonal antibodies, which had been selectively chosen based on their absence of cross-reactivity against D. immitis circulating antigens, resulting in an overall high specificity [8].  Generally, D. immitis antigen tests are considered to be more sensitive than microfilariae concentration methods or other procedures [38]. In particular, the ELISA technology has been shown to be more sensitive than lateral flow immunochromatography [26] for the diagnosis of heartworm infected dogs. Reasons for the occurrence of false negative results with sera of D. immitispositive dogs have been discussed in previously performed studies evaluating different D. immitis test kits. Low worm burden and low number of female worms have been shown to reduce sensitivity of the tests [25,26,39]. However, increased sensitivity may be coupled with lower specificity and, importantly, with potential cross-reactions against A. vasorum. An unknown number of dogs with travel anamnesis and testing positive for circulating heartworm antigen may have falsely been diagnosed positive due to A. vasorum cross-reactions, and erroneously treated with melarsomine and/or macrocyclic lactones. Therefore, serological results for D. immitis should be confirmed or excluded by additional diagnostic tests (Knott's test for microfilariae of D. immitis, or serology or Baermann migration test for L1 of A. vasorum) or diagnostic imaging frequently delivering pathognomonic findings for heart dirofilariosis [40,41] or angiostrongylosis [42,43].

Conclusions
In this study we confirmed that sera of dogs infected with A. vasorum cross-react in commercially available test kits for the detection of circulating D. immitis antigen. The simultaneous use of highly specific diagnostic tools is recommended for epidemiological studies where both heart worm species occur or for individual dogs with a suspected heart worm infection.