This study was performed primarily at the Cornell University, College of Veterinary Medicine. PCR and MAT were also performed at IDEXX Laboratories in Sacramento, CA and North Grafton, MA, respectively. The study was partially supported by a grant from IDEXX Laboratories, and was presented in poster form at the 2010 ACVIM Forum, Anaheim, CA.
Corresponding Author: R.E. Goldstein, Department of Clinical Sciences, CVM, Cornell University, Ithaca, NY 14853; e-mail: Rg225@cornell.edu.
Bacterin-based canine Leptospira vaccines could present a challenge for the use of whole blood real-time polymerase chain reaction (PCR) as a diagnostic tool. Recent vaccination could induce positive results if the targeted DNA fragment is present within the vaccine and in the blood of the recently vaccinated dog.
The objective of this study was to assess whether 2 available 4-serovar vaccines induce a positive real-time PCR reaction in the blood of healthy recently vaccinated dogs.
Twenty healthy dogs.
This was a prospective study. Dogs were assigned to 1 of 2 vaccine groups. Both vaccines were culture-based and include Leptospira interrogans serovars Pomona, Canicola, and Icterohaemorrhagiae and Leptospira kirschneri serovar Grippotyphosa. Whole blood for real-time PCR and serum for the microscopic agglutination test (MAT) were collected prior to and 3 and 7 days after vaccination and weekly thereafter for 8 weeks. Two real-time PCR tests targeting 2 different genes were performed independently in a blinded fashion.
Both Leptospira vaccines produced positive real-time PCR reactions when assayed undiluted or diluted 1 : 100 in canine blood. However, blood samples drawn from all dogs at all time points after vaccination were negative on PCR. All dogs developed MAT titers.
Conclusions and Clinical Importance
Recent vaccination with 2 commercially available vaccines does not interfere with the use of real-time PCR for the identification of acute Leptospira infection in dogs.
There are currently four 4-serovar vaccines commercially available in the United States. All are bacterin based and include Leptospira interrogans serovars Icterohaemorrhagiae, Canicola, and Pomona as well as Leptospira kirschneri serovar Grippotyphosa. At the time this study was performed there were two 4-serovar vaccines approved for dogs in the United States. Those included Vanguard L41 and Leptovax 42; the Vanguard L4 is an ultrafiltrated whole cell bacterin, whereas the Leptovax 4 is labeled as a subunit vaccine. Although Leptospira vaccines are not considered core vaccines in dogs according to the guidelines of the American Animal Hospital Association, they are recommended for many dogs throughout the United States.
A potential limitation of these Leptospira vaccines is interference with the MAT, the most commonly used diagnostic test for acute leptospirosis.[2, 3] This test is performed by multiple commercial laboratories in the United States, and involves incubation of patient sera with antigen derived from cultured Leptospira serovars. Despite the high frequency with which the MAT is used in the diagnosis of leptospirosis in dogs, it lacks the degree specificity and sensitivity desired of a “gold standard” test.[5, 6] The lack of sensitivity includes many false negatives early in the disease process prior to the development of antibodies. The lack of specificity mainly includes lack of differentiation between vaccine-induced antibodies and infection-induced antibodies. The MAT has also recently been shown to lack repeatability between laboratories and even in the same laboratory when samples were taken from the same dog on different days, also exemplifying the inaccuracy of the highest MAT titer in a given sample to correctly identify the serogroup. The inability of traditional serologic testing such as the MAT to provide reliable data in acute cases of leptospirosis in humans and dogs has led to the development of novel testing modalities including real-time PCR. This modality, performed on whole blood, has been used successfully in the diagnosis of human disease[9, 10] and is currently offered commercially by a number of veterinary laboratories. A recent study performed on experimentally infected dogs also showed good sensitivity and specificity when blood and urine were used for real-time PCR.3 We have obtained similar results in our lab from real-time PCR of renal tissue. A potential drawback of the use of real-time PCR is the possibility of positive results in recently vaccinated dogs from vaccinal DNA. Although we are not aware that other veterinary killed bacterial vaccines have been studied, killed viral vaccines have been shown to potentially interfere with real-time PCR.[11, 12] If vaccinating a dog with the culture-based vaccines induces a positive PCR result then this test would not be useful in the scenario of a clinically ill dog that is suspected of Leptospira infection and has been recently vaccinated. The aim of this study was to determine if positive results occur frequently in healthy dogs recently vaccinated for canine leptospirosis with 2 commonly used 4-serovar vaccines, utilizing 2 separate real-time PCR assays.
Materials and Methods
Twenty staff and student owned dogs were included in the study, between September 2009 and March 2010, after acquiring owner consent. The dogs could not have been vaccinated with any vaccine containing Leptospira antigens within the 3 months prior to the study and were assessed as healthy by their owners at the time and by a physical examination. Each dog was vaccinated with one of two 4-serovar Leptospira vaccines: Vanguard Plus 5 L41 or Duramune Max 5/4L2 based on owner preference or random assignment until 10 dogs were included in each vaccine group. Both vaccines include attenuated strains of Canine Distemper virus, Canine Adenovirus 2, Canine Parainfluenza virus, Canine Parvovirus as well as culture-based L. interrogans serovars Canicola, Icterohaemorrhagiae, and Pomona and L. kirschneri serovar Grippotyphosa. Dogs that lacked evidence of prior vaccination within the previous 12–16 months were given 2 vaccines of the same type 2–4 weeks apart and the presample was acquired prior to the 1st vaccine but the additional samples were acquired subsequent to the second dose. All the vaccinations were given subcutaneously in a standard fashion using a 21-gauge needle. The dogs were monitored for 30 minutes postvaccination for acute reactions. The Cornell University IACUC approved the study.
Sample Collection and Handling
Blood was drawn for real-time PCR (whole blood in EDTA) and MAT titers (serum in in clot tubes) at approximately the following time points: days 0 (prevaccination), 3, 7, 14, 21, 28, 35, 42, 49, and 56. As a result of occasional lack of owner compliance, not all dogs were available for sample collection at every time point. The blood samples were drawn and refrigerated immediately. The whole blood in EDTA was frozen at −20°C within 48 hours of collection until DNA was purified within 60 days. Serum was separated and frozen at −20°C within 48 hours and stored until shipped for MAT evaluation within 60 days. The DNA was purified from whole blood in a routine fashion using QIAamp DNA Blood Mini kit.4
Real-Time PCR at Cornell University
Real-time PCR was performed in triplicates using a previously validated 16S rDNA primer and probe set that is used routinely in this laboratory after further validation in canine blood. The samples were run in 96-well plates read by an AbiPrism 7000 Sequence Analyzer. A standard 2-step 40-cycle QPCR reaction was used. Probes had a FAM5 fluorescent tag on the 5′ end and a black hole quencher on the 3′ end. Reaction volumes of 30 μL containing 96 ng of DNA were used with a primer/probe set designed to target a Leptospira 16S rDNA sequence. Every plate contained standard positive (canine blood spiked with cultures of Leptospira organisms) and negative (blank, water, master mix, and unspiked canine blood) controls as well as 6 positive standards logarithmically scaled between 10 whole bacteria genome copies and 1 × 106 whole bacteria genome copies per reaction. Leptospira DNA standards were obtained from live culture and bacterial genome concentration was estimated by photometric analysis of extracted DNA. The software bundled with the sequence analyzer determined CT values for the QPCR reactions automatically. These values were fitted to the standard CT curve to give genome counts.
In addition to the samples from the study dogs, a real-time PCR assay was performed at Cornell University on DNA that was purified from the reconstituted vaccines that were used to give to the dogs as an undiluted sample and again when diluted 1 : 100 in canine Leptospira negative control blood. Other vaccines were tested in an identical fashion, including a bacterin based Lyme vaccine (LymeVax2), the Duramune and Vanguard vaccine without the Leptospira component.
Real-Time PCR IDEXX Laboratories
DNA for real-time PCR and serum for MAT were batched and shipped for analysis by IDEXX Laboratories5 in a blinded fashion, including all study samples as well as positive and negative controls. These samples were then analyzed using the hap-1 based Leptospira real-time PCR.3 An 18S rRNA real-time PCR test was used to assess DNA quantity and integrity. The assay is based on IDEXX's proprietary real-time PCR oligonucleotides as previously described.3 Briefly, hap-1 gene sequences were aligned and a region was selected for primer and hydrolysis probe design using PrimerExpress.6 Real-time PCR was run with standard primer and probe concentrations (ABI) using the Roche LightCycler 480 Probes Master mastermix.6 Real-time PCR was performed using default cycling conditions on a Roche LC480 instrument7 in the 384-well plate configuration.
MAT was performed at by a large commercial veterinary reference laboratory.5 The laboratory was blinded to the identity of the submitted samples. Titers are reported as negative if 50% agglutination was not apparent at a serum dilution of 1 : 100. The highest dilution that maintained at least 50% agglutination was reported as the titer in positive samples.
Of the 20 dogs included in the study, 13 were purebred representing 11 breeds as well as 7 mix breed dogs. The median age of the dogs was 4 years with a range of 3 months to 11 years at the initiation of the study. The median weight of the dogs was 26.8 kg with a range of 7.0–61.0 kg. Seven dogs had been previously vaccinated against Leptospira, the most recent was 3 months prior to the beginning of the study. Those 7 dogs received 1 dose of vaccine at the initiation of the study and the remaining 13 dogs received 2 doses of the vaccine 2–4 weeks apart. Dogs that received the Vanguard vaccine received all components of the reconstituted vaccine, whereas 9/10 dogs receiving the Duramune vaccine only received the Leptospira component (found in the diluent fraction) and 1/10 dog received all components of the reconstituted vaccine. As a result of a lack of owner compliance, not all dogs were sampled at every time point. Out of 10 time points for every dog (200 total time points) sample were not obtained at 24 time points. No single dog missed more than 4 time points.
Nine of the 10 dogs that received the Vanguard vaccine were MAT negative for all serogroups on the pre-sample; 1 previously vaccinated dog was positive for Leptospira serogroups Grippotyphosa and Autumnalis. Eight of the 10 dogs that received the Duramune vaccine were MAT negative for all serogroups on the presample; 1 previously vaccinated dog was positive for Leptospira serogroups Canicola and Autumnalis and 1 dog was not sampled prior to vaccination. Eighteen of 20 dogs mounted a positive MAT response by day 7 to multiple serovars, 2 dogs never mounted a positive MAT response; 1 of these 2 dogs was diagnosed with Lymphoma at the conclusion of the study. MAT titers reached a value of ≥ 1,600 for at least 1 serogroup at at-least 1 time point in 5/10 dogs of each vaccine group. One of the 5 in each group was the dog that had positive titers on a sample collected before vaccination. Figure 1 depicts the median MAT titers for all serovars for all dogs by vaccine group.
Real-Time PCR Results
Both 4-way vaccines used in the study were positive on 16S rDNA (Cornell University) real-time PCR; this includes those samples when DNA was purified from the reconstituted vaccine as an undiluted sample and when diluted 1 : 100 in canine Leptospira negative control blood. All other vaccines, including a bacterin based Lyme vaccine (LymeVax2), and the Duramune and Vanguard vaccine without the Leptospira component were negative. Both the 16S rDNA and Hap1 (IDEXX Laboratories) real-time PCR tests were run on all samples from all 20 dogs. The results for both tests were identical on all samples. Nineteen of the 20 dogs sampled were negative on the presample. One dog's 1st sample was drawn on day 3 following the vaccine and was negative. All 20 dogs were negative at all time points for Leptospira DNA.
This study demonstrated that recent vaccination did not influence real-time PCR in the diagnosis of leptospirosis in dogs. This included 2 commercially available 4-way vaccines in the United States, and 2 separate and distinct real-time PCR methods, one at a research university and another at a commercial laboratory. The observation that the diluted and nondiluted vaccines themselves gave a positive real-time PCR result on the 16s real-time PCR prompted the initiation of this clinical study. The goal of the study was to assess whether a positive real-time PCR result in a recently vaccinated dog should be considered artifactual, in which case this testing modality could not be used reliably in these dogs. The secondary goal was to assess how long this period of interference lasted if it did exist so that clinical recommendations could be made. The fact that no vaccinated dog was positive at any time point on either of the 2 real-time PCR methods indicates that there is no clinically relevant interference of recent vaccinations with real-time PCR for canine leptospirosis and that this modality can be recommended even in these dogs. The obvious rise in MAT titers is supportive of the dogs actually receiving the vaccines. All dogs but 2 mounted positive titers with approximately half the dogs mounting titers of or above 1 : 1,600 to at some time point during the study. These high titers are consistent with those often seen from multiserovar vaccines,[2, 3] reiterating the limited diagnostic value of a single MAT titer in a recently vaccinated dog.
The use of real-time PCR is expected to become more and more common in the future in the diagnosis of this disease as more commercial laboratories offer it and it is validated on more field samples. One of the driving forces for the development of new diagnostics for leptospirosis in dogs is the obvious limitations of the most commonly used testing modality, the MAT.[2, 8] The limitations include major inter- and intralaboratory inconsistencies, negative result early on in infection, and false positive results in vaccinated dogs. These limitations most often necessitate the use of 2 separate MAT 7–10 days apart to attempt to achieve a diagnosis with reasonable confidence. The preliminary data from experimental infection studies in dogs,3 suggest that real-time PCR may be both sensitive and specific when blood and urine samples are both assayed. Today, PCR is already considered the best method for identification of urinary shedding in dogs.[2, 14]
Despite its potential advantages, there are many possible limitations to the use of PCR. A single negative real-time PCR result can never rule out infection for any bacterial disease, because it requires a sufficient quantity of bacterial DNA in the PCR sample. Despite the positive data regarding the sensitivity and specificity from experimental infections, data regarding clinical cases and real-time PCR in natural canine infections are lacking. It is unclear when a real-time PCR result will be positive within the time frame from the appearance of clinical signs through end organ damage and the development of chronic disease. The pathogenesis of canine leptospirosis is consistent with the assumption that a whole blood sample could yield a positive result within the first few days after the development of clinical signs and later on the urine would turn positive and the blood would revert to negative. Hence the solicitation of clinical samples that include blood and urine from every case by commercial laboratories, however, the exact time frame as well as the possibility that a window exist that neither sample would be positive requires additional study. It is also possible that antibiotic use prior to obtaining these samples will induce a false negative result.[2, 6]
Although this study provides adequate answer for the clinical question posed in its outset and is strengthened by the use of 2 different real-time PCR techniques at 2 different institutions in a blinded fashion, it has some limitations as well. A small number of dogs were used for the studies and blood was not collected on all dogs at all time points, including 1 dog lacking a prevaccinal sample. A small number of dogs were MAT positive prior to initiation of the study as a result of previous vaccination. Although there is no reason to believe that the results of the study would be different if different vaccines were utilized they do only pertain definitively to the 2 vaccines that were tested in the study. Despite these limitations, we believe that this study shows that recent vaccination does not interfere with the use of real-time PCR in the diagnosis of canine leptospirosis and this modality can be utilized in these dogs.
Pfizer Animal Health, New York, NY
Fort Dodge Animal Health, Overland Park, KS
Leutenegger CM, Palaniappan R, Elsemore D, et al. Analytical sensitivity and specificity of a real-time PCR assay detecting pathogenic Leptospira in dogs based on the Hap-1 gene. JVIM 2009; 23:772 (abstract).