key: cord-0263385-5cjczygh
authors: Freuling, Conrad M.; Busch, Frank; Vos, Ad; Ortmann, Steffen; Lohr, Frederic; Hedimbi, Nehemia; Peter, Josephat; Nelson, Herman Adimba; Shoombe, Kenneth; Shilongo, Albertina; Gorejena, Brighton; Kaholongo, Lukas; Khaiseb, Siegfried; van der Westhuizen, Jolandie; Dietze, Klaas; Geurtse, Goi; Müller, Thomas
title: Oral rabies vaccination of dogs – experiences from a field trial in Namibia
date: 2022-04-22
journal: bioRxiv
DOI: 10.1101/2022.04.21.488865
sha: dbee71638a95407b5b6f9008a42a15dea2f56bd1
doc_id: 263385
cord_uid: 5cjczygh

Dog-mediated rabies is responsible for tens of thousands of human deaths annually, and in resource-constrained settings, vaccinating dogs to control the disease at source remains challenging. Currently, rabies elimination efforts rely on mass dog vaccination by the parenteral route. To increase the herd immunity, free-roaming and stray dogs need to be specifically addressed in the vaccination campaigns, with oral rabies vaccination (ORV) of dogs being a possible solution. Using a third-generation vaccine and a standardized egg-flavoured bait, bait uptake and vaccination was assessed under field conditions in Namibia. During this trial, both veterinary staff as well as dog owners expressed their appreciation to this approach of vaccination. Of 1,115 dogs offered a bait, 90% (n=1,006, 95%CI:91-94) consumed the bait and 72.9% (n=813, 95%CI:70.2-75.4) of dogs were assessed as being vaccinated, while for (11.7%, n=130, 95%CI:9.9-17.7) the status was recorded as “unkown” and 15.4% (n=172, 95%CI: 13.4-17.7) were considered as being not vaccinated. Smaller dogs and dogs offered a bait with multiple other dogs had significantly higher vaccination rates, while other factors, e.g. sex, confinement status and time had no influence. The favorable results of this first large-scale field trial further support the strategic integration of ORV into dog rabies control programmes. Given the acceptance of the egg-flavored bait under various settings worldwide, ORV of dogs could become a game-changer in countries, where control strategies using parenteral vaccination alone failed to reach sufficient vaccination coverage in the dog population.

The Tripartite (WHO, OIE and FAO) considers rabies control a priority but also an entry 64 131 

Oral rabies vaccinations were conducted using 3 rd 

Vaccine baits were shipped according to IATA guidelines on dry ice (UN 1845) directly 156 from the manufacturer to the Central Veterinary Laboratory (CVL), Windhoek, using a 157 commercial courier service. After temporary storage at CVL the vaccine baits were further 158 transported to the Ondangwa branch of CVL located in the Oshana region of the NCAs. Upon 159 arrival in Windhoek and at the field study areas the vaccine baits were stored in standard style 160 freezers at -18 --20 °C until further transportation or use in the field, respectively. Maintenance 161 of the cold chain was checked and documented using temperature data logger and integrated 162 electronic measuring. Prior to shipment and the prior to start of the field trial, the quality of the 163 baits and the vaccine titre was checked independently by the national and OIE reference 164 

Vaccination campaigns were announced via local radio the evening before and the 179 morning the campaigns took place. Both door-to-door as well as central-point vaccinations 180

were conducted. Vaccine baits were distributed to the targeted dog population using the hand-181 out and retrieve model [58] . Immediately prior to the field trial, a two day staff introduction 182 session and workshop was conducted during which staff was trained on the objectives of the 183 field trial, oral rabies vaccination, vaccine bait handling, safety issues, techniques for 184 approaching free-roaming dogs, best practice on offering vaccine baits to dogs, data collection 185 (bait handling by individual dogs -duration, consumption, perforation and/or swallowing of 186 sachet), and interpreting effectiveness of vaccination attempt. The importance of retrieving the 187 discarded vaccine sachet after bait consumption as described [50] was highlighted followed by 188 a door-to-door vaccination training in the field. The field trial was carried out at the end of the 189 dry season during the second half of October 2021. During this time, vaccinations were 190 performed over eight full working days (including two half days). 191

Vaccine baits were transferred to portable cool boxes the evening before field use, 192 allowing them to thaw before they were offered to the dogs. Baits unused at the end of the 193 vaccination day were kept at refrigerator temperatures (4-8 °C) and offered to dogs the next and remotely loaded to the handsets using 3G. Data were entered offline and stored locally on 217 the handset where it could be reviewed on a map the same day. The app was also used to 218 assigned working zones for each vaccination team (different colours -gold, red, green and 219 blue) on the App backend platform the day before with demarcated boundaries for each zone 220 automatically synchronized to the App on each teams' handset via internet connection. 221

If applicable, brain material of all dogs with laboratory confirmed rabies from the field trial 222 area (Oshana, Omusati) collected in a period of three months after the end of the trial were 223

shipped to the FLI, Germany, for virus characterization (OIE, 2018). For distinguishing field 224 from potentially vaccine virus induced rabies cases a discriminatory realtime PCR using SPBN 225 Table 4) . 226 227

A dog was considered 'interested' if the animal had any direct contact (smelling, licking) 229 with the bait offered, irrespective of subsequent handling. Animals were regarded as 230 successfully 'vaccinated' if the bait chewing and intensity (thoroughness) was detectable 231 and/or perforation of the sachet clearly visible. Any dog that swallowed the bait immediately, 232 or walked away with it and could not be observed, or chewed inappropriately on the bait without 233 visible perforation of the sachet was assigned an 'unknown' vaccination status. The status 234 'non-vaccinated' was assigned if the dog was not interested or the bait was only shortly taken 235 up and immediately dropped with the bait casing and sachet still intact. The latter also applied 236 to dogs that showed interest (and accepted the bait) but were interrupted by external factors 237 (other dogs, humans, cars, etc) and discontinued bait handling. initial review and analysis. Spatial information was analyzed and displayed using QGIS 241

Geographic Information System (QGIS.org, 2022.http://www.qgis.org). 242

Fisher's exact test) and followed by a multiple logistic regression (MLR). The dependent 244 variable was "vaccination success" (yes/no), and datasets for dogs with an "unknown" status 

The majority of dogs (78%) encountered and offered a bait during the study were owned and 285 free-roaming. The proportion of ownerless free-roaming dogs was 3%, while the remaining 286 dogs were assessed as confined during the vaccination. With 63%, there was a gender bias 287 towards male dogs. Larger dogs (>30 kg) were rare (8%), whereas medium (57%) and small 288 (<10kg; 33%) were dominating in the dog population. Dogs were offered baits both at central 289 places (crush-pens, village centres) or at the individual homestead in the respective areas 290

( Figure 1 ). The mean distance between individual baitings per team was 533m, with the lowest 291 mean distance (226m) at the last day of the study when semi-urban areas were included. The 292 longest distance between two baitings was 10km (Figure 2, a) . Overall, under field study 293 conditions, the average number of dogs vaccinated per hour was 7, with a maximum of 28 294 dogs vaccinated per hour for one team (Figure 2, b) . were assessed as being vaccinated, for 11.7% (n=130, 95%CI:9.9-17.7) the status was 298 recorded as "unkown" and 15.4% (n=172, 95%CI: 13.4-17.7) were considered as being not 299 vaccinated. In 54.9% (n=552) of dogs observed, the vaccine blister was swallowed, while 300 43.4% (n=437) of dogs that consumed a vaccine bait discarded the blister. For the remaining 301 dogs, the status of the bait could not be verified, as e.g. the dog ran away with the bait and 302

could not be observed anymore. Only 9.8 % (N=43) of all blisters retrieved were not perforated. 303

For the statistical analysis, 985 entries with a vaccination assessment (yes/no) were available. 304

A statistically lower vaccination rate (p=0.0048, Chi-square test) was observed on the last 305 (69.8%) and first day (76.7%) of the campaign (Figure 3, a) . Differences in vaccination rates 306 during time of the day (Figure 3, b) and the different teams were not significant (Figure 3, a) . 

While there was no statistical difference in vaccination status in regard to the confinement 312 status (Fig 4b, Supplementary Table 1 ) or the sex of the dog (Fig 4b) , smaller dogs (p=0.0166, 313

Chi-square test) and dogs offered a bait with multiple other dogs present (p=0.0494, Fisher's 314 exact test) had significantly higher vaccination rates (Fig 4c-d) . All variables with a p<0.20 315 identified , i.e. date, size and social situation of the dog, in the univariate analyses were 316 included in a multivariate logistic regression model, but only size and social situation had a 317 significant impact (Supplementary Table 2 ). Vaccination success was higher in small dogs and 318 when more than one dog were together and were offered a bait. 

Overall, the results from this first ORV field trial in Namibia demonstrate a high acceptance for 336 this method both by the veterinary/technical staff as well as the dog owners. In the field, the 337 apparent efficiency in vaccinating dogs, particularly those that cannot be easily handled, was 338 well acknowledged both by the veterinary staff involved as well as by the owners of dogs. For 339 many dogs, this was the first time they had ever been vaccinated. Only very few individuals 340 did not give their consent to vaccinate their dog using a novel vaccination approach and a 341 vaccine that is not yet licensed. This is surprising and very promising for future vaccination 342 campaigns in dogs, as for human diseases there seems to be an increasing hesitancy for 343 vaccination, e.g. for COVID-19 [63]. Public announcement prior to the campaign by radio, and 344 the direct interaction with the dog owner by DVS likely played an important role in the 345 acceptance of this approach. 346

The egg-flavoured baits were highly attractive to the dogs and with 90% bait acceptance the 347 results were higher than with the same bait in other countries, e.g. Navajo Nations, US (77.4%) 348

[38], Goa State, India (77.5%) [33], and Thailand (78.8%) [39] and Bangladesh (84%) [59]. The 349 true percentage of dogs vaccinated by ORV in this field trial was at least 72.9% but likely 350 higher, because a number of dogs disappeared with the bait and were considered "unknown". 351

While about half of the vaccine blisters were swallowed, when blisters were retrieved, more 352 than 90% were perforated, suggesting that if the bait was consumed, a large proportion of dogs 353 have likely had contact with the vaccine and can be regarded as vaccinated. In any case, the 354 observed vaccination rate was higher than with the same bait in Thailand with 64% [39] . The 355 reasons for these differences are not clear. Ownership practices and the integration of bait 356 offering in feeding routine by the owner might contribute to the higher acceptance of the bait 357 compared to previous studies. 358

Although the assessment of vaccination was based on individual observation, the small 359 differences between teams suggest that the overall bias was not affecting the outcome of the 360 analysis ( Figure 2c) . Also, the fluctuation of vaccination rates along the study days are likely 361 due to learning (day 1) where the success was lower than average, and the final day, where a 362 certain working fatigue may have occurred. Also, during the last day, two teams were assigned 363 areas in semi-urban settings where dogs that were approached got up much faster and more 364 often walked away when the bait was offered. 365

The fact that smaller dogs had a higher interest, better consumption of the bait and a higher 366 vaccination rate as opposed to mid-sized and large dogs is interesting. Partly, these small dogs 367 comprised of younger puppies that were very interested in the bait and readily consumed it 368 ( Figure 5) . Also, in situations when more dogs were around, smaller dogs tended to be more competitive towards consuming the bait, even though several baits were offered to avoid 370 hierarchic feeding behavior. A similar observation was made in Thailand, where small and 371 young dogs had higher bait acceptance rates [39] . Dogs that chewed very long (>60sec) and 372 dogs that discarded the sachet were more likely not vaccinated. Dogs chewing very long rarely 373 swallowed the sachet (12.4%), meanwhile most dogs that chewed very short swallowed the 374 sachet (74.2%). 375

In the frame of this field trial with more than 1,100 baits handled, vaccine exposure to humans 376 that would require intervention did not occur. This adds to the high safety profile of this live 377 vaccine when using the hand-out-and retrieve model [64] . Spillage of vaccine is not considered 378 a source of contamination for potential contact to humans since the enveloped virus has a 379 reduced viability in the environment. In the study area, the sandy floor, the high temperatures 380 and the constant sunlight are further factors that decrease virus' persistence. 381

There are some limitations to this study. For statistical reasons, datasets with vaccination 382 status "unknown" had to be removed thus leading to higher proportions of dogs being 383 interested, consuming the bait and being assessed as vaccinated than if they were included. App also prevented from vaccinating dogs in a shorter time interval when several dogs were 391 presented for vaccination. One aspect that was identified to limit the potential of ORV in the 392 field was the requirement of owners' consent prior to vaccination as was laid down in the study 393 plan. Future campaigns should address this by indicating a general consent when the dog is 394 free roaming at the time of vaccination. Another practical issue that emerged during the 395 campaign was the provision of a vaccination certificate. Principally, the ORV method aims at 396 the herd immunity and not the immune response in any individual dog, but specific ORV 397 certificates may be issued during campaigns when ORV is included. In this field trial, both 398 central-point vaccination as well as a door-to-door was used. As for the latter, with a highly 399 dispersed human and dog population, partly absent dog owners, and distances between one 400 and ten kilometers between individually vaccinated dogs (Figure 2c) if not even higher in other 401 areas, this approach would be very inefficient and against the background of increasing costs 402 for fuel, inappropriate under many settings. Rather, dog owners should be instructed to bring 403 their dogs to a central point where parenteral and oral vaccination is conducted with a higher 404 efficiency than parenteral alone. While dogs may be stressed due to the unfamiliar territory, other dogs and the transportation by leash as experienced before [44] , in our study, we did not 406 see a reduced bait uptake or vaccination rate when more dogs were present. However, to 407 prevent negative influence dog owners could be instructed to keep their dogs at a certain 408

In any case, a central point approach would again disregard those dogs that cannot be handled 410 and brought to a vaccination point. To overcome this dilemma, the oral rabies baits could be 411 handed over to the dog owners and vaccination would occur at their own premises, as has 412 been demonstrated with non-vaccine baits in Tunisia [65] . A similar approach was also 413 

The Road to Dog Rabies

Control and Elimination-What Keeps Us from Moving Faster? Front Public Health

Rabies in terrestrial animals

Rabies -Scientific Basis of the Disease and its Management

Can rabies 483 be eradicated?

Current 485 status of rabies and prospects for elimination

WHO expert consultation on rabies, third report. WHO Tech 488 Rep Ser 1012

Evidence-based 490 control of canine rabies. a critical review of population density reduction

The history of rabies in the Western Hemisphere

Rabies control in Europe: an overview of past, current and future 496 strategies

Rabies in the Americas

Progress 501 towards eliminating canine rabies

Caribbean. policies and perspectives from Latin America and the Caribbean

Mexico is free from human rabies transmitted by dogs

World Rabies Day: Latin America and the Caribbean on the cusp of eliminating 508 rabies deaths

Estimating the 513 global burden of endemic canine rabies

The feasibility of 516 canine rabies elimination in Africa. dispelling doubts with data

Scaling-up the delivery of dog vaccination campaigns against rabies in Tanzania

Role 522 of Oral Rabies Vaccines in the Elimination of Dog-Mediated Human Rabies Deaths

Oral vaccination of 525 dogs:. a well-studied and undervalued tool for achieving human and dog rabies elimination

Field Studies Evaluating Bait Acceptance and Handling by Free-Roaming Dogs in Thailand

Assay of

Oral Vaccination of Dogs Against Rabies in Tunisia with the Vaccinal Strain Sad(Bern)

Tunisian dogs with the lyophilised SAG2 oral rabies vaccine incorporated into the DBL2 dog bait

The feasibility of oral 593 vaccination of dogs in Turkey -a European Union supported project

The safety 596 and efficacy of the oral rabies vaccine SAG2 in Indian stray dogs

Immunogenicity of 599 the Oral Rabies Vaccine Strain SPBN GASGAS in Dogs Under Field Settings in Namibia

Humoral Immune Response of Thai Dogs after Oral Vaccination against 603 Rabies with the SPBN GASGAS Vaccine Strain

Feasibility of canine oral 606 rabies vaccination in Sri Lanka--a preliminary report

Field trial with oral vaccination of dogs against 608 rabies in the Philippines

Evaluation of 610 immune responses in dogs to oral rabies vaccine under field conditions

First trials of oral 613 vaccination with rabies SAG2 dog baits in Morocco

Feasibility and Effectiveness Studies with Oral Vaccination of 617

Global rabies 619 control: the role of international organisations and the Global Strategic Plan to eliminate dog-620 mediated human rabies

Rabies (infection with rabies virus and other lyssaviruses) (NB: 622 Version

Mediated Rabies in Namibia-Implementation of a Rabies Elimination Program in the Northern 626 Communal Areas

Application of the 628 GARC Data Logger-a custom-developed data collection device-to capture and monitor mass 629 dog vaccination campaigns in Namibia

Overexpression of the rabies virus glycoprotein results in enhancement of apoptosis and 633 antiviral immune response

Single Amino 635

Acid Change in Rabies Virus Glycoprotein Increases Virus Spread and Enhances Virus 636

Safety studies 638 with the oral rabies virus vaccine strain SPBN GASGAS in the small Indian mongoose (Herpestes 639 auropunctatus)

Oral bait handout as 641 a method to access roaming dogs for rabies vaccination in Goa

Oral bait 644 preferences and feasibility of oral rabies vaccination in Bangladeshi dogs

One million 648 dog vaccinations recorded on mHealth innovation used to direct teams in numerous rabies control 649 campaigns

Ecology 651 and epidemiology of rabies in humans, domestic animals and wildlife in Namibia

Environmental 654 distribution of certain modified live-virus vaccines with a high safety profile presents a low-risk, 655 high-reward to control zoonotic diseases

COVID-19 657 vaccine acceptance and hesitancy in low-and middle-income countries

Review of Oral Rabies 660

Vaccination of Dogs and Its Application in India

Field 662 evaluation of a dog owner, participation-based, bait delivery system for the oral immunization of 663 dogs against rabies in Tunisia

Oral vaccination of backyard 665 pigs against classical swine fever

Integrating the landscape epidemiology and 669 genetics of RNA viruses. rabies in domestic dogs as a model

Predictors of free-roaming domestic dogs' contact network centrality and their relevance for rabies 673 control

West and Central Africa: A review