All enrolled dogs were handled in compliance with Merck Animal Health Institutional Animal Care and Use Committee (IACUC) guidelines and approval was obtained prior to study initiation (protocols E-2020-64 and E-2021-45).

For vaccine preparation, the CPI strain was grown in Vero cells at 37 °C ± 2 °C for 48–144 h, and the B. bronchiseptica strain was grown in a fermenter, harvested, and concentrated. The CPI and B. bronchiseptica antigens were combined with stabilizer (hydrolyzed gelatin, N-Z Amine, sorbitol, and sodium phosphate, dibasic), HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) and sterile water, lyophilized, and then reconstituted with 1.0 mL of sterile water prior to use. In the CPI DOI study, the CPI fraction was at minimum protective dose (MPD) (> 6.0 log₁₀HAID₅₀/dose) (Hemadsorption Antibody Infectious Dose), and the B. bronchiseptica fraction was at field dose (> 1.0 × 10⁸ cfu/dose). In the B. bronchiseptica DOI study, the B. bronchiseptica fraction was at MPD (> 1.0 × 10⁷ cfu/dose), and the CPI fraction was at field dose (≥10-fold higher than the MPD). The placebo vaccines consisted of all components of the test vaccine except the antigen being evaluated. The test vaccine is commercially available as Nobivac® Intra-Trac® Oral BbPi.

The heterologous CPI strain was grown on primary dog kidney cells to a titer of > 5.0 log₁₀FAID₅₀/mL and frozen at < −50 °C. On the day of challenge, the frozen challenge virus was thawed, and 1.4 mL was administered to the upper respiratory tract of each dog. The challenge dose was optimized in preliminary studies to induce CPI viral shedding in at least 50% of the animals.

The heterologous B. bronchiseptica strain was grown on tryptic soy with 5% sheep blood agar plates. The plates were incubated for approximately 18 h at 36 ± 2 °C, and the pure bacterial growth was washed off the agar plates with tryptose phosphate broth and centrifuged. The pellet was adjusted to > 1.0 × 10⁸ cfu/mL, and 1.0 mL was administered by the intranasal route. The challenge dose was optimized in preliminary studies to generate clinical signs of disease in at least 50% of the animals.

Six- to eight-week-old, purpose-bred beagles that had not been exposed to or vaccinated for B. bronchiseptica or CPI were enrolled in the two studies. Dogs were housed in an isolation facility in barrier rooms and randomized into two treatment groups using housing units and litter as randomization factors for the CPI study or using pre-vaccination antibody titers and litter as randomization factors for the B. bronchiseptica study. The dogs were separated by treatment group prior to vaccination and remained as such until challenge when they were comingled to ensure an equal number of vaccinated dogs and placebo-vaccinated control dogs were housed in the same room. All dogs were vaccinated orally once (study day 0) with a 1 mL dose of the B. bronchiseptica + CPI vaccine (n = 23) or placebo vaccine (n = 23) using the Immuno-Mist-R™ applicator attached to a 3-mL Luer-Lok™ syringe for each study. Prior to vaccination, whole blood for serum and nasal swabs were collected. Following vaccination, whole blood for serum was collected monthly until the dogs were challenged, and nasal swabs were collected just prior to challenge.

In the CPI study, 23 test-vaccinated and 23 placebo-vaccinated control dogs were challenged with CPI via the upper respiratory tract 12 months following vaccination. During the 14-day post-challenge observation period, clinical signs including nasal and ocular discharge, coughing, sneezing, depression, fever (defined as ≥103.5° F), and dyspnea were monitored daily, and nasal swabs were collected daily for 10 days.

In the B. bronchiseptica study, 22 test-vaccinated dogs and 23 placebo-vaccinated control dogs were challenged by the intranasal route 13 months following vaccination with B. bronchiseptica. One dog from the test vaccine treatment group was euthanized prior to challenge due to aggressive behavior that raised animal welfare concerns as well as safety concerns for animal care personnel. During the 28-day post-challenge observation period, clinical signs including coughing, nasal discharge, depression, and dyspnea were monitored for 30 min each day, twice daily, and nasal swabs were collected twice a week.

Whole blood collected into serum separation tubes (BD; Franklin Lakes, NJ) was allowed to clot for a minimum of 2 h at 15–30 °C and then centrifuged at 1000 × g to separate the serum. Serum samples were tested in a serum neutralization assay to evaluate the serological response post-vaccination. Briefly, each serum sample was serially diluted and incubated with CPI for 30–60 min before inoculated onto a confluent monolayer of dog kidney cells planted on 96-well tissue culture plates (Falcon; Corning, NY). Plates were incubated at 36 ± 2 °C with 4–6% CO₂ for 7 days. Monolayers were then fixed and stained with fluorescein-conjugated CPI antiserum (VMRD; Pullman, WA), and titers were calculated using the Spearman-Karber method.

Polyester fiber-tipped swabs (Fisherbrand; Pittsburgh, PA) were wetted in transport media [Dulbecco’s Modified Eagles Medium (Corning; Corning, NY), gentamicin (Gibco; Grand Island, NY), and amphotericin-B (Corning; Corning, NY)] and inserted into each nostril of the dog. In the laboratory, swabs were removed from the transport media, and the liquid contents were centrifuged at 500 × g. The supernatant was used to inoculate a confluent monolayer of dog kidney cells that were planted on 96-well tissue culture plates (Falcon; Corning, NY). Plates were incubated at 36 ± 2 °C for 7 days, fixed with 80% acetone (Fisher Scientific, Pittsburg, PA), and stained with fluorescein-conjugated CPI antiserum (VMRD; Pullman, WA). Virus titers were calculated using the Spearman-Karber method.

Whole blood collected into serum separation tubes (BD; Franklin Lakes, NJ) was allowed to clot for a minimum of 2 h at 15–30 °C and then was centrifuged at 1000 × g to separate the serum. Serum samples were tested in an agglutination assay to evaluate the serological response post-vaccination. Briefly, each serum sample (along with a positive and negative control sample) was serially diluted 2-fold in a U-bottom microtiter plate (Nunc; Rochester, NY) using normal saline containing 0.1% gelatin as the diluent. B. bronchiseptica antigen was added to each well and mixed for 15–30 s on a microtiter plate shaker (Lab-Line Instruments, Inc.; Melrose Park, IL; Model 4,625). The plates were incubated at 36 ± 2 °C for 1–3 h and then incubated at 2–7 °C for 36–72 h until they were visually assessed for agglutination. The titers were expressed as the reciprocal of the highest dilution showing complete agglutination.

Two calcium alginate swabs (McKesson; Irving, TX) were used to collect specimens from each nostril using one swab per nostril and placed in transport media [Tryptose phosphate broth (BD; Franklin Lakes, NJ) with 10% glycerol (Fisher; Pittsburgh, PA)]. In the laboratory, swabs were removed from the transport media, and 10-fold serial dilutions of the nasal swab material were performed in tryptose phosphate broth. Each dilution that was expected to yield 30–300 colonies (0.1 mL/plate) was spread onto three replicate MacConkey agar plates (Thermo Scientific; Waltham, MA). The plates were incubated at 36 ± 2 °C for 48–96 h, colonies counted, and results recorded (cfu/mL).

Statistical analysis for duration of CPI shedding was performed using R software version 3.5.0. Statistical significance for serologic geometric means and viral load were performed using a two-tailed Student’s T-test assuming equal variances in Microsoft® Excel® version Microsoft 365® and significance was declared for p-values < 0.05.

An affected dog in the B. bronchiseptica study was defined as having spontaneous coughing or spontaneous coughing with retching on 2 or more consecutive days during the post-challenge observation period. The proportion of affected dogs in the vaccinated and placebo-vaccinated control group was calculated and compared using Fisher’s Exact Test. Statistical significance for serologic geometric means and bacterial load were calculated and compared using a two-tailed Welch’s T-test in Microsoft® Excel® version Microsoft 365®. Significance was declared for p-values < 0.05.

All dogs were seronegative (<2) for antibodies against CPI prior to vaccination, indicating no prior exposure. Vaccination with the placebo vaccine did not induce antibodies specific to CPI. In contrast, vaccination with the test vaccine induced CPI serum neutralization titers of ≥ 23 in 21 (91%) of 23 test-vaccinated dogs, with peak titers ranging from <2 to 2,435 (GMT = 150) 5 weeks after vaccination (Figure 1). Titers decreased thereafter and remained constant throughout the 12-month vaccination period. CPI serum neutralization titers were significantly higher in the test-vaccinated dogs compared to the placebo-vaccinated control dogs at all time points tested (study days 35, 70, 98, 126, 161, 189, 217, 252, 280, 315, 343, and 371; p-values < 0.01).

CPI was not isolated from nasal swabs prior to challenge, but 2 days following challenge, dogs from both treatment groups started shedding CPI (Figure 2). In total, 22 (96%) of the 23 placebo-vaccinated control dogs shed CPI, compared to only 13 (57%) of the 23 test-vaccinated dogs (data not shown). Test-vaccinated dogs averaged 0.2 log₁₀FAID₅₀/mL of CPI shed during the post-challenge phase (p-value = 0.0008) compared to 1.1 log₁₀FAID₅₀/mL of CPI shed from the placebo-vaccinated control dogs. Specifically, the vaccinated treatment group demonstrated significant reduction in viral load on study days 2–9 (Figure 2, p-values ≤ 0.04).

Duration of virus shedding post-challenge was the primary variable to evaluate vaccine efficacy and was defined as the first through the last day that CPI was isolated from nasal swabs. The median duration of shedding for the placebo-vaccinated control dogs was 6 days, which was significantly more than the 1 day for the vaccinated dogs (Figure 3, p-value = 0.001).

Consistent with previous laboratory CPI challenge studies (14, 15), there were very few clinical signs of disease during the post-challenge period. None of the placebo-vaccinated control dogs showed clinical signs of a CPI infection, and only 1 of the test-vaccinated dogs spontaneously coughed on two separate days.

Prior to vaccination, all dogs in the B. bronchiseptica study had low IgG antibody titers (≤ 4) to B. bronchiseptica. Vaccination with the placebo vaccine did not induce high IgG antibody titers specific to B. bronchiseptica. In contrast, vaccination with the test vaccine induced titers ≥ 16 in all of the vaccinated dogs, with peak titers ranging from 16 to 128 (GMT = 45) 5 weeks after vaccination (Figure 4). Titers decreased thereafter and remained relatively constant throughout the 13-month vaccination period. Serum antibody titers were significantly higher in the test-vaccinated dogs compared to the placebo-vaccinated control dogs at all time points tested (study days 35, 70, 112, 133, 168, 196, 259, 287, 315, and 343; p-values < 0.05). On study day 168, both the placebo-vaccinated control group and test-vaccinate group had an increase in titers; however, this increase occurred at nearly the same magnitude across both treatment groups suggesting the increase was assay related. In addition, titers returned to expected levels on study day 196.

Coughing was the primary variable to evaluate efficacy in the B. bronchiseptica study. An affected dog was defined as having spontaneous coughing or spontaneous coughing with retching on two or more consecutive days during the post-challenge observation period. Following challenge, 17 (74%) of the 23 placebo-vaccinated control dogs were considered affected, compared to only 2 (9%) of the 22 test-vaccinated dogs (Figure 5; p-value < 0.0001). In addition, the test vaccine significantly reduced the number of dogs that coughed on study days 3, 8, 9, 10, 13, 15, 16, 17, and 27 (Figure 6; p-values ≤ 0.047). Other clinical signs of disease such as nasal discharge, dyspnea, or depression were not observed from any of the dogs throughout the post-challenge observation period.

B. bronchiseptica was not isolated from nasal swabs prior to challenge, but following challenge, 11 (48%) of the 23 placebo-vaccinated control dogs and 10 (45%) of the 22 test-vaccinated dogs shed B. bronchiseptica. The test-vaccinated dogs shed less bacteria, but the difference between the groups was not significant.