Báo cáo Nghiên cứu khoa học Development of an improved capability in support of national bio-security for the surveillance and control of foot and mouth disease in cattle and pigs

1 Ministry of Agriculture & Rural Development CARD Project Technical Report Development of an Improved Capability in support of National Bio-security for the Surveillance and Control of Foot & Mouth Disease in Cattle and Pigs Milestone 6 Epidemiological and sero-surveillance programs operational By Debbie Eagles & Chris Morrissy 1 Table of Contents 1. Institute Information ___________________________________________________ 1 2. Project Abstract_________________________________________________________ 3 3. Executive Summary ____________________________________________________ 3 4. Introduction & Background _____________________________________________ 4 5. Epidemiological and Sero-surveillance programs ____________________________ 4 5.1 Implementation Highlights ________________________________________________ 4 5.2 Capacity Building _______________________________________________________ 36 5.3 Publicity _______________________________________________________________ 36 6. Implementation & Sustainability Issues ___________________________________ 37 6.1 Issues and Constraints ___________________________________________________ 37 6.2 Options________________________________________________________________ 37 6.3 Sustainability___________________________________________________________ 37 7. Next Critical Steps ____________________________________________________ 38 8. Conclusion __________________________________________________________ 39 1 1. Institute Information Project Name Vietnamese Institution Regional Animal Health Centre, Ho Chi Minh City (RAHO - 6 ), South Vietnam. Vietnamese Project Team Leader Dr. Dong Manh Hoa Australian Organisation Australian Animal Health Laboratory (AAHL), PMB 24, Geelong, 3213, Australia Australian Personnel Mr Chris Morrissy Date commenced 01/06/2005 Completion date (original) 01/06/2008 Completion date (revised) Reporting period Contact Officer(s) In Australia: Team Leader Name: Mr Chris Morrissy Telephone: +61 3 5227 5000 Position: Diagnostic Virologist Supervisor Mammalian Virology Fax: +61 3 5227 5555 Organisation Australian Animal Health Laboratory (AAHL), PMB 24, Geelong, 3213, Australia Email: chris.morrissy@csiro.au In Australia: Administrative contact Name: Mr Chris Morrissy Telephone: +61 3 5227 5000 Position: Patents Contracts Officer Fax: +61 3 5227 5555 Organisation Australian Animal Health Laboratory (AAHL), PMB 24, Geelong, 3213, Australia Email: christopher.morrissy@csiro.au In Vietnam Name: Dr. Dong Manh Hoa Telephone: + 84 8 8568220 Position: Director Fax: + 84 8 8569050 Organisation Regional Animal Health Centre, Ho Chi Minh City (RAHO - 6 ), South Vietnam. Email: rahchcmc@hcm.vnn.vn 2 2. Project Abstract The project’s purpose was twofold - to develop capacity for FMD (and other disease) surveillance and diagnosis at both a laboratory and field level, and to investigate the serotypes of FMDV circulating in Vietnam and the reason for vaccine failures. Regional laboratories were set up with the reagents and methods to allow a diagnostic capability for FMDV diagnosis and serology. Control strategies for understanding of FMD epidemiology have been implemented through veterinary and laboratory training workshops. The project has highlighted the importance of having a laboratory network to identify what is happening in the field and how to prevent and control disease outbreaks. The pilot zones were established in provinces near the borders of Vietnam to study serotypes circulating in Vietnam and to determine their origin. The number and quality of samples increased with each round of the project giving more data on the FMD situation in Vietnam. Virus isolation and molecular studies can now be carried out on FMD samples from the field and molecular epidemiological studies of the FMDV isolates in these provinces has provided insights into the effectiveness of border control and origin of circulating FMDV. Improved diagnostic capacity for FMD allows for the early detection and identification of disease enabling better control of disease and helps reduce loss of livestock and therefore increases productivity. 3. Executive Summary The CARD FMD project was ambitious in that it had 2 very broad and diverse aims. The first objective was capacity building – at the laboratory, epidemiological and field levels. The second major aim was to investigate possible causes of vaccination failure by evaluating isolates of circulating strains and sero-surveillance data. The project was very successful at achieving its objective in relation to capacity building. As documented in the final report, the four collaborating Vietnamese laboratories improved their FMD diagnostic capacity and have been able to apply their new skills to disease investigations and surveillance projects. In addition, there have been important and measurable improvements in both epidemiological and field areas. When this project began there was no epidemiology department at any of the laboratories. There is now a fully functional epidemiology department, with 3 full-time staff, at RAHO – 6. This group has been instrumental in supporting this and other international projects and has provided advice and training to field and provincial veterinarians. They have also been crucial to investigation of disease outbreaks such as HPAI and PRRS, particularly in southern Vietnam. Through implementation of this project one of the lessons learnt has been the importance of complete and accurate field information. Whereas the laboratories were already accustomed to recording results, the recording of information such as vaccination and infection data in the field was not commonplace. Data collection and management has improved dramatically throughout the project. The only downside of this is that data quantity and quality for the initial rounds is difficult to analyze for the purpose of investigation of vaccine failure. In addition to this the accuracy of some of the data is questionable, as is discussed throughout this report. 3 With the laboratory and epidemiological capacity now available in the collaborating laboratories, particularly HCMC, there is now the potential for a smaller, more focused study on vaccination failure. This would be best limited to a smaller number of provinces in southern Vietnam, with a study protocol aimed specifically at investigating vaccination effectiveness. 4. Introduction & Background Serum samples and information were initially to be collected from 10 provinces – An Giang, Binh Phuoc, Dong Thap, Kien Giang, Kom Tum, Lang Son, Long An, Quang Nam, Quang Ninh and Tay Ninh. No samples were ever collected from Lang Son. Samples were also only collected intermittently from the other northern province in the project, Quang Ninh. The central provinces of Kom Tum and Quang Nam provided samples for all but the 3rd round. In the southern provinces samples were not available from Long An in the final round and An Giang did not provide pig samples in round 1, 5 or 6. The epidemiological support and interest at RAHO - 6 is almost certainly a contributing factor to the better provision of samples from the southern provinces. A combination of fewer specialized staff in the northern provinces and laboratories and the required allocation of resources to outbreak response is likely to have reduced their ability to collect samples. NCVD did not have a epidemiology section. In 2005, the field data collection and the use of forms was not well developed. As a result, for most provinces the only information collected was species, sampling location (district/commune/village) and in some instances vaccination information, sampling date and animal age. In 2006 the data collection form was further developed and standardised with the following fields – district, commune, village, species, age, sex, sampling date, vaccination date, vaccine name/manufacturer/serotypes, last date of infection, last serotype of infection and field sample number. This form improved data collection dramatically although not all fields were completed in every round for each province. 5. Epidemiological and Sero-surveillance programs 5.1 Implementation Highlights Analysis by Province The analysis for the information in each province is divided into cattle and pigs. For each species there are tabular results and a graph followed by a description of the results for each round. The graph only displays information for rounds and serotypes for which animals have been vaccinated ≤ 6 months prior to sampling. 4 An Giang Cattle Year (Round) Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval August 2005 (pigs, O) 2008 (6) O,A 2mths 36 89 100 51 August 2005 (pigs, O) 2008 (5) O,A 5mths 10 10 22 9 August 2005 (pigs, O) 2007 (4) O,A 1mth 21 61 58 0 August 2005 (pigs, O) 2007 (3) O,A 6-7mth 37 66 47 31 August 2005 (pigs, O) Higher proportion Asia 1 +ves amongst 3ABC+ group 2006 (2) O,A 6mth 38 19 56 34 August 2005 (pigs, O) Proportions across the 3ABC- /3ABC+ similar 2005 (1) O,A Unknown 76 63 70 56 August 2005 (pigs, O) Unknown history of infection 0 10 20 30 40 50 60 70 80 90 100 Percentage 1 2 3 4 5 6 Round Percent cattle O/A positive when vaccinated for that serotype O A 2005 (Round 1) Without vaccination or infection information available no judgment can be made on vaccination response. 75% cattle were NSP ELISA positive, and more than half of these are positive for all 3 serotypes. This suggests that these cattle have been vaccinated with either a 5 bivalent or trivalent vaccine, given that infection with more than one serotype simultaneously is rare. Further testing of the sera is necessary to determine the serotypes present, ie titration of the sera to a endpoint against each sera type. There was a reported outbreak of serotype O infection in pigs in 2005. 2006 (Round 2) All cattle were vaccinated for serotypes O/A, 6months prior to sampling. The greatest serological response was to serotype A at 50%. It is likely that there was a problem with the sensitivity of the O ELISA for this batch of samples, as it would be unlikely for animals to have been vaccinated for serotype A and not serotype O. It is almost certain that some animals were also vaccinated for Asia 1, given the serological response to this serotype, absence of outbreak history and the negative 3ABC result in the majority of those that were Asia 1 positive. 2007 (Round 3) Vaccination was 6 months prior to sampling with O/A vaccine. Despite lack of infection history, 37% were 3ABC ELISA positive. In general the serological response to serotype O was better than in the previous round which may be more indicative of changes to the assay as opposed to differences in vaccine response. Close to a third of cattle were also positive on the Asia 1 ELISA and of these 2/3 were also 3ABC positive. This is suggestive of both unrecorded vaccination and/or unreported infection (which may be related to animal movement). 2007 (Round 4) Vaccination occurred one month prior to sampling with an approximately 60% response rate to both serotype O and A (bivalent vaccine administered). A smaller % were positive on the 3ABC ELISA in this round than in round 3. 2008 (Round 5) The serological response to the O/A vaccine is very poor, regardless of the 5 month interval between vaccination and sampling. Following the protocol listed below in Appendix 1 (Investigating Vaccination Failure Checklist) may assist in determining the reasons for vaccine failure. Records suggest that the same vaccine was used in each round. 2008 (Round 6) There was an excellent serological response to vaccine in this round, in which the vaccination-sampling interval was 1 month. Over 50% of cattle also seroconverted to Asia 1 despite no history of recent vaccination for this serotype. Half of these were also 3ABC positive, despite no infection history in the sampled animals or the province (see below). 6 Conclusions The % of cattle 3ABC was > 1/3 in 4 of the 6 rounds. In the absence of outbreak history or isolates from this province since 2005 this suggests that: ™ There has been movement (transboundary or between provinces) of infected (diseased or carrier) animals ™ There have been unreported or undetected (due to mild clinical signs) infections in the surveyed communes. ™ There is a large number of animals previously exposed to FMD or carriers that remain – at least intermittently – NSP ELISA positive. Some variation in the results between years may also be due to the inclusion of 4 communes (An Phu, Tinh Bien, An Nong and Nhon Hung) which were variably sampled in the different rounds. Vaccination response was only – in the final around - above the required herd protected level of 80%. Pigs Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) No samples 2008 (5) No samples 2007 (4) O 1mth 0 5 0 0 August 2005 (pigs, O) 2007 (3) O 6mths 0 5 0 0 August 2005 (pigs, O) 2006 (2) O 6 mths 0 0 0 0 August 2005 (pigs, O) 2005 (1) No samples 7 0 10 20 30 40 50 60 70 80 90 100 Percent 2 3 4 Round Percent pigs O/A positive when vaccinated for that serotype O Samples were collected from vaccinated pigs in 2006 and both rounds of 2007. The highest serological response for serotype O was 5% despite the fact that on one occasion vaccination was administered just one month prior to sampling. In rounds 2 and 3 all sampled piglets were >6 months. In the remaining rounds some piglets were as young as 2.5 months, so could not have been vaccinated on the date recorded. Binh Phuoc Cattle Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) O,A,Asia1 1mth 24 94 93 93 2006 (Cattle, pig; O) Last infection Aug 05 2008 (5) O,A,Asia1 5 mths 28 34 51 18 2006 (Cattle, pig; O) ½ that have been infected are Asia + 2007 (4) O, A 1 mth 14 89 88 38 2006 (Cattle, pig; O) 2007 (3) O,A 5-6mths 9 41 81 55 2006 (Cattle, pig; O) 2006 (2) O,A,Asia1 9 mths 11 50 35 35 2006 (Cattle, pig; O) Previous infection (O/A?) 2005 (1) O, A 4 mths 10 20 5 24 April, Aug, Oct 2005 (Cattle, O/A; pigs, O) 8 0 10 20 30 40 50 60 70 80 90 100 Percentage 1 3 4 5 6 Round Percent of cattle O/A/Asia1 positive when vaccinated for that serotype O A Asia1 2005 (Round 1) There is little field information available for this year, as the forms for information collection had not been developed. Cattle were vaccinated 6 months prior to the sampling date with O/A vaccine, which is just at the extent of the expected vaccination protective period. The vaccination response rate in this year was very poor (20% positive for serotype O and 5% positive for serotype A), however it is difficult to pass judgment on first year results from either field or laboratory perspective. 2006 (Round 2) In this year the recorded information shows that vaccination was with a serotype A vaccine called Trivale. However discussions at sub-DAH confirm that, as the vaccine name would suggest, this is more likely a trivalent vaccine. The manufacturer is unknown. The vaccination date was 9 months prior to sampling date, so it is not surprising that the seropositives were relatively low at 50%, 35% and 35% for serotypes O, A and Asia 1 respectively. 2007 (Round 3) In this year records suggest that all cattle were vaccinated with serotypes O/A vaccine. Sub- DAH staff again suggested that cattle may have been vaccinated for Asia1. This would fit the ELISA results, with 41%, 81% and 55% positive for O, A and Asia1 serotypes respectively. The vaccine was administered 5 months prior to sampling, which may contribute to the variation in % positive for each serotype. 9 2007 (Round 4) In this year cattle were vaccinated with serotype O/A vaccine one month prior to sampling. The seropositivity for both these serotypes was around 90% suggesting excellent vaccination response. 38% of cattle were also positive to serotype Asia1 (see round 3). A relatively small % of animals were 3ABC positive. 2008 (Round 5) In this year cattle were vaccinated with a trivalent vaccine. The seropositives were 34%, 51% and 18% again for O, A, and Asia 1 serotypes respectively. The vaccine was administered 5 months prior to sampling, which may partly account for the lower seropositivity than seen in rounds 4 and 6. However, this time interval is still within the expected protective period of the vaccine. Staff at Sub-DAH suggest that of the cattle presented for vaccination approximately 70% are re-presented for sampling, the remaining 30% may be different animals. Although all cattle in the district should have been vaccinated at the same time, some of the animals presented for sampling may be new to the province and possibly unvaccinated. One quarter of cattle sampled were positive for 3ABC ELISA suggesting previous infection. Most of these were positive for all 3 serotypes, and as vaccination in this year was also for all 3 serotypes it is not possible to determine the serotype of infection. Titration of positive samples may have assisted, in there was clearly a 4-fold difference in titrations between one serotype and the remaining 2 serotypes. 2008 (Round 6) There was an excellent response to the O/A/Asia 1 vaccine (Aftopor) used in this round with 90% or more seropositivity for all serotypes. As per the first round in 2008, one quarter of the cattle were also 3 ABC positive, indicating previous infection. Conclusions (cattle): Vaccination response rates for cattle in Binh Phouc were generally very good and, with the exception of round 5, improved throughout the project. As expected, the proportion of vaccinated animals seropositive was much higher in those sampled 1 month after vaccination, as opposed to those vaccinated 5 months previously. However, this is a concern as vaccination is meant to be protective for 6 months. Trans-boundary movement and movement of animal between provinces may also have contributed to some of the variation in results, particularly if these animals have previously been infected, and not recently vaccinated. Interestingly, there was a spike in % positive on 3ABC ELISA in the 5th round despite no history of vaccination since 2006. This is the same round in which there was poor vaccination response, suggesting a possible influx of unvaccinated, previously infected animals to the district. 10 Pigs Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) O 1mth 0 0 0 0 2006 (Cattle, pig; O) Most piglets 2-3 mths 2008 (5) None None 14 0 0 0 2006 (Cattle, pig; O) 2007 (4) O 1mth 0 0 0 0 2006 (Cattle, pig; O) Age unknown 2007 (3) O 1mth 0 0 0 0 2006 (Cattle, pig; O) Age unknown 2006 (2) None None 0 17.5 0 0 2006 (Cattle, pig; O) 2005 (1) None None 0 0 0 0 April, Aug, Oct 2005 (Cattle, O/A; pigs, O) 2005 (Round 1) There is no vaccination record for pigs in this year and no positives on any of the ELISAs. 2006 (Round 2) There is no history of vaccination in pigs in 2006, although a small % were serotype O positive but 3ABC negative suggesting previous vaccination, not infection. 2007 (Round 3) The records for the pigs suggest that all were vaccinated with a serotype O vaccine. However, there are no pigs positive on the O ELISA. Discussions with Sub-DAH have revealed that is very uncommon for the same pig to be presented for both vaccination and sampling – that commonly sows and boars are vaccinated, but piglets were presented for sampling. This would explain the non-existent vaccine response. Ages for the sampled piglets were not recorded in this year to confirm this theory. The O ELISA is also known to have lower sensitivity in pigs than in cattle. 11 2007 (Round 4) As for round 3, the results in pigs do not match the vaccination history with no seropositive pigs despite serotype O vaccination 1 month prior to sampling. There may be a number of reasons for this, including specifics of the ELISA assay, sampling after only one vaccine (rather than the initial two required for complete vaccine response) or the vaccinating and sampling of 2 different groups of pigs. Again age records have not been kept for these pigs. 2008 (Round 5) Pigs were not vaccinated for this round as there was not sufficient vaccine available. 2008 (Round 6) As with other rounds the results for pigs do not correlate with vaccination history. Further investigation and vaccine trials would be required to determine if the apparent vaccination failure. This also highlights the need to be absolutely certain that those piglets presented for sampling were definitely vaccinated in this round also. As piglets presented for sampling in this round were only 2-3 months old they are likely to have only had one vaccine, if any. Conclusions (pigs) In pigs field and laboratory information did not concur. Further investigations and trials would be necessary to determine the issues but it is likely a major contributing factor is the presentation of different pigs for vaccination and sampling, or sampling after just one vaccination. 12 Dong Thap Cattle Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) O,A O, A, Asia 1 1-3mth 4mth 24 20 61 60 70 90 5 10 September 2008 (pig, O) 2008 (5) Unknown None 1mth None 10 26 80 68 90 57 60 25 2007 (4) O,A,Asia1 None 1-3mth 25 39 44 52 51 63 29 34 January 2007 (pig, O) 2007 (3) O,A,Asia 1 O,A,Asia1 O,A 6mth 2-3mth 3mth 59 67 80 48 87 85 31 88 95 17 67 75 4 of the O/A have history of infection 2004 (O) 2006 (2) O O O O,A,Asia1 O,A,Asia1 None None <6mths + I (O) <6mths >6mths 1mth + I (O) 1mth + I 100 0 17 0 0 4 38 100 25 30 44 75 46 62 100 38 67 44 25 42 55 57 25 67 44 75 40 38 2006 (pig, O) 2005 (1) None 40 43 23 24 June-August 2005 (pig, O) January 2004 (pig, O) 0 10 20 30 40 50 60 70 80 Percentage 1 2 3 4 5 6 Round Percent cattle positive when vaccinated for serotype O A Asia1 13 2005 (Round 1) There was no vaccination history for this round. 40% show indication of previous infection. Outbreaks of FMD in pigs were reported in 2004 and 2005 but no infection in cattle is recorded. 2006 (Round 2) The large variation in vaccination history and the resultant small group size for each category make this information difficult to analyse. In general, the vaccine response appears to be reasonable. 2007 (Round 3) In this round cattle were vaccinated with trivalent vaccine 2, 3 or 6 months prior to sampling or O/A vaccine 3 months prior to sampling. The graph below compares the % cattle with antibodies to all 3 serotypes (and 3ABC) based on vaccination-sampling interval for those vaccinated with trivalent vaccines. This clearly shows the drop over time of protective antibodies. A large proportion of cattle reportedly only vaccinated with a bivalent vaccine were also positive on the Asia 1 assay. 0 20 40 60 80 100 % 3ABC O A Asia 1 Elisa type % Cattle positive by O/A/Asia 1 Elisa in relation to vaccination-sampling interval 6mths 2-3mths 2007 (Round 4) About ½ the cattle in this round were vaccinated with a trivalent vaccine, and there is no record of vaccination for the remaining cattle. Interestingly the % positive for each of the 3 LP ELISAs was greater in the unvaccinated group as opposed to the vaccinated. Overall the vaccine response in the vaccinated group was poor, given the short vaccination-sampling interval (1-3 months). 14 2008 (Round 5) In this round the vaccine type was unknown but likely to be either bivalent or trivalent, given the high seropositivity on all 3 ELISAs (80%, 90% and 60% for O, A and Asia 1 respectively). 2008 (Round 6) Most of the cattle were vaccinated with bivalent (O/A) vaccine in this round, 1-3 months prior to sampling. There were also 9 cows that reportedly received trivalent vaccine, yet there was a very low proportion of cattle seropositive on the Asia1 ELISA from this group, suggesting that they perhaps received a bivalent vaccine also. It is possible that the O ELISA has a lower sensitivity than the A ELISA, as evidenced by the lower % positives when vaccinated against both serotypes. Conclusions Dong Thap provided excellent vaccination history – it appears that vaccination history was recorded separately for each animal sampled as opposed to a generic vaccination history for all animals. Despite this, vaccine responses were still variable but best in the final round. Dong Thap experienced a number of outbreaks of FMD (both O & A serotype) during the project period and this, along with their commitment to data collection, would make this province ideal for any further studies. Pigs Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) O 1mth 2 8 7 5 September 2008 (pig, O) 2008 (5) None NA 0 0 4 0 2007 (4) O None 1-2 mths NA 0 0 42 66 0 0 0 0 January 2007 (pig, O) 2007 (3) O 5mths 0 12.5 0 0 2006 (2) O O None <6mths >6mths NA 0 0 0 0 0 0 0 0 0 0 0 0 2006 (pig, O) 2005 (1) None None 0 0 0 0 June-August 2005 (pig, O) January 2004 (pig, O) 15 0 20 40 60 80 100 Percent 2 3 4 5 6 Round Percent pigs elisa positive when vaccinated for serotype O The highest proportion of pigs positive to serotype O in this province was in round 4. In this round three pigs, aged just 1-1.5months, had not been vaccinated but 2 out of the 3 were positive on the O ELISA, presumably due to maternal antibodies. The remaining piglets were aged from 2-5 months. As opposed to most other provinces, the dates of vaccination appear to correlate with the age of the piglets ie. older piglets have an earlier vaccination date that younger ones. It is possible that the better records are a result of an outbreak of FMD in pigs in the province in that year, which would increase willingness of farmers to present piglets for vaccination and sampling and enhanced awareness and enthusiasm on the part of veterinarians. 16 Kien Giang Cattle Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) O,A 1mth 24 51 96 2 2006 (Cattle, O) 2008 (5) O,A 5mth 29 50 90 13 2006 (Cattle, O) 2007 (4) O,A 1mth 31 82 64 43 2006 (Cattle, O) 2007 (3) O,A 5mths 30 27 25 19 2006 (Cattle, O) 2006 (2) O,A Unknown 25 53 51 40 2006 (Cattle, O) 2005 (1) O,A Unknown 57 60 11 8 0 10 20 30 40 50 60 70 80 90 100 Percentage 3 4 5 6 Round Percent of cattle Elisa positive when vaccinated for serotype O A 17 2005 (Round 1) In this round the vaccination and infection history is unknown. There are no known recorded outbreaks in this province from this or preceding years but there is a very high proportion of seropositives for the 3ABC ELISA. Of those cattle that are 3ABC ELISA positive 81% are also O ELISA positive. (Of the 3ABC negative group only 33% are O ELISA positive), suggesting a possible previous (recent) infection with that serotype. 2006 (Round 2) Although vaccination is known to have been with O/A vaccine in this round the date of vaccination is unknown. If vaccination occurred within the 6 month protective period the % positives are very low, sitting just above 50% for both vaccinated serotypes. 2007 (Round 3) Vaccination occurred 5 months prior to sampling, which is within the protective period of the vaccination, the response rate on ELISAs is very poor. All cattle were vaccinated with O/A vaccine yet there was only around ¼ seropositives on each of these ELISAs. 2007 (Round 4) In this round the % seropositives is relatively good, as would be expected for a vaccination- sampling period of 1 month. However it is still of some concern that if all cattle were definitely vaccinated with O/A vaccine that 1 month post vaccination only 64% were positive for serotype A. There are two main possibilities for this – that there is an antigen “mismatch” between the vaccine and the ELISA, or that some animals have been vaccinated with a serotype O vaccine only. 2008 (Round 5) All cattle were vaccinated 5 months prior to sampling with O/A vaccine. In both this and round 6 there is a significantly lower proportion seropositive on the O ELISA than the A ELISA. It would be extremely unlikely for cattle to be vaccinated for serotype A only, so this suggests a problem with the sensitivity of the O ELISA assay. In addition to this the proportion of ELISA A+ is greater amongst the 3ABC +ve group than the 3ABC –ve group, which suggest there may also be some animals previously infected with this serotype. 2008 (Round 6) In this round the vaccine is the same as for round 5 but with a vaccination-sampling interval of 1 month. The proportions positive to the O and A serotypes are not dissimilar to round 5. Proportions of O and A ELISA positives are both roughly the same between infected and uninfected groups. Conclusions Vaccine response to serotype A improved each round. In contrast serotype O vaccine response declined from round 4 to rounds 5 and 6. As vaccination with serotype A alone is not common, this is suggestive of an issue with the sensitivity of the O ELISA. 18 Pigs Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) (Aftopor) 1mth 1 35 11 1 2006 (Cattle, O) 2008 (5) O 5mths 28 31 13 0 2006 (Cattle, O) 2007 (4) O Various 0 13 0 0 2006 (Cattle, O) 2007 (3) O 5mths 0 0 0 0 2006 (Cattle, O) 2006 (2) O Unknown 0 12 0 0 2006 (Cattle, O) 2005 (1) O Unknown 0 0 0 0 0 5 10 15 20 25 30 35 Percentage 3 4 5 6 Round Percent pigs elisa positive when vaccinated for serotype O 19 In Kien Giang pigs were vaccinated in each round, with the sampling-vaccination interval of either 1 or 5 months, or in some cases unknown. The percentage seropositives for the O ELISA varied from 0% (when vaccination date 5 months previous, or unknown) to 35% when the vaccine was administered 1 month prior to sampling. In all rounds were ages were recorded pigs were greater than 4 months of age. Kom Tum Cattle Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) O,A 2mth 14 15 10 0 None 2008 (5) O,A 5mth 5 8 36 0 None 2007 (4) O,A,Asia1 2mth 37 43 61 40 None 2007 (3) None No results. Samples unsuitable. 2006 (2) O,A,Asia1 7mth 0 60 10 23 None 2005 (1) O,A,Asia1 Unknown 0 31 4 12 None 0 10 20 30 40 50 60 70 Percentage 4 5 6 Round Percent of cattle Elisa positive when vaccinated for serotype O A Asia1 20 2005 (Round 1) Again, vaccination date and infection history are unknown for this round. A large vaccination-sampling interval may explain the variation in proportion seropositive against each vaccinated serotype. 2006 (Round 2) The 7 month vaccination-sampling interval may partly explain the variation in seropositives between O, A and Asia1 serotypes for which these animals were vaccinated. However, it is also possible that some animals were vaccinated with O vaccine only (due to the significantly higher % of positives on this ELISA). No animals had history of infection or reacted on the 3ABC ELISA. 2007 (Round 3) No samples collected or tested in this round. 2007 (Round 4) Given the short period between vaccination and sampling the % seropositives for each of the 3 vaccinated serotypes is very low. In this round, as compared to others, there was a very high proportion of animals positive on the 3ABC ELISA. These animals did not have a history of infection and there were no recorded outbreaks in the province. As such, the high number of animals NSP ELISA positive may be due to any one or combination of the following: ™ Movement of infected animals (from other provinces or countries) ™ Unreported/undetected (mild clinical signs) infection ™ Large number of carriers that remain (at least intermittently) NSP ELISA positive Interestingly the same communes and villages were used in rounds 2 (no samples collected in round 3) and round 5. Most of the 3ABC positive animals were from one commune (Dak Nong), only one animal was 3ABC positive from the other commune (Bo Y). Samples from this round were taken late (ie. January 2008) and the samples in round 5 were taken in April 2008. With just 3 months between sampling, it would be expected that if the same animals were tested the proportion of those positive to 3ABC ELISA would be higher than the 5% seen in round 5. Titrating the positive samples for 3ABC and for each of the LP ELISAs in this round may have helped determine the serotype responsible for infection. This may give some indication of the source of infection (if there were reported outbreaks in nearby provinces/countries) however questioning of farmers may also be required. 2008 (Both rounds 5 and 6) In both these rounds cattle were vaccinated for serotypes O and A. In round 5 this was 5 months prior to sampling and in round 6 it was 2 months prior to sampling. The vaccine response rate is extremely poor for all serotypes in both rounds. This is suggestive of major vaccine failure or, contrary to the recorded information, no vaccination. 21 Pigs Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) O 2mths 0 3 0 0 None 2008 (5) None NA 8 0 0 0 None 2007 (4) O 2 mths 0 48 0 0 None 2007 (3) None No samples 2006 (2) None NA 0 0 0 0 None 2005 (1) O Unknown 0 0 0 0 None 0 20 40 60 80 100 Percent 4 6 Round Percent pigs elisa positive when vaccinated for serotype O In Kom Tum pigs were vaccinated in rounds 1, 4 and 6. The only year in which there was a serological response to the vaccine was 2007, in which pigs were vaccinated 2 months prior to the sampling date. There are no ages available for these pigs, but 48% were positive on the O ELISA. 22 Long An Cattle Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) No samples. 2008 (5) O,A,Asia1 3mths 7 48 83 61 2007 (4) O,A 1mth 0 0 0 0 2007 (3) Unknown 28 45 55 70 2007 (pig, O) 2006 (2) O,A,Asia1 O 1mth 2mths 25 23 58 42 50 32 48 23 2006 (pig, O) 2005 (1) Unknown 48 50 28 22 2005 (pig, cattle; O) 0 10 20 30 40 50 60 70 80 90 Percentage 2 4 5 Round Percent of cattle Elisa vaccinated when vaccinated for serotype O A Asia1 23 2005 (Round 1) No vaccination or infection history for these cattle is available, but there was an outbreak of serotype O in cattle and pigs in 2005 in Long An. This is consistent with the 3ABC ELISA and O ELISA results suggesting that close to 50% of cattle had previously been infected. A higher proportion of 3ABC positive cattle were O ELISA positive than those that had not previously been infected. 2006 (Round 2) In this round, approximately ½ the cattle were administered a trivalent vaccine and ½ the monovalent (O) vaccine. Given the short period between vaccination and sampling the vaccine response rates were very poor ie. in almost all cases less than 50%. The slightly higher number of seropositives to the O ELISA in both groups is likely due to previous infection (with around ¼ cattle 3ABC positive and of these all are serotype O positive). It is likely that some of the cattle recorded as having just O vaccine have been administered trivalent vaccine due to the seropositive percentages of 32 and 22% respectively for A and Asia 1. Of these none were 3ABC positive. 2007 (Round 3) Both vaccine and outbreak information is missing in this round, making interpretation of the results difficult. It can be surmised from the results that at least some animals have been recently vaccinated with a trivalent vaccine. 2007 (Round 4) Vaccination was with O/A vaccine 1 month prior to sampling but no animals were positive on any ELISA. This should have been further investigated immediately at both the laboratory and field level. Given the range of P.I. values for each of the ELISAs, and the fact that it was all assays (thus unlikely to be a specific antigen problem), then in the absence of operator error the field information is likely to be incorrect or there may have been a problem with the batch of vaccine (including poor storage or transport). 2008 (Round 5) Vaccine response rates are highly variable between serotypes. This is difficult to explain when a trivalent vaccine was administered to all cattle 3 months prior to sampling. In addition, it is the A ELISA which has the higher proportion of positives – and this is less likely than O to be used on its own. The proportion of 3ABC positives is also very low. 2008 (Round 6) No samples. Conclusions The overall vaccine response in cattle in this province was poor. Long An would potentially prove to be a good province for further investigations of vaccine failure. 24 Pigs Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) No samples. 2008 (5) (aftopor) 3mths 3 0 0 0 2007 (4) O 2-3 mths 0 31 0 0 2007 (3) None NA 0 0 0 0 2007 (pig, O) 2006 (2) O 2-4mths 0 32 0 0 2006 (pig, O) 2005 (1) None NA 0 0 0 0 2005 (pig, cattle; O) 0 20 40 60 80 100 Percentage 2 4 5 Round Percent of pigs elisa positive when vaccinated for serotype O In 2 of the rounds in which pigs were vaccinated in this province there was a seroconversion of around 30%. In one of these round ages are not available, in the other (round 2), the pigs ranged from 3 months - >4 years. 25 Quang Nam Cattle Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) O,A 2mths 29 64 21 59 2008 (5) Unknown 0 77 0 0 2007 (4) O,A,Asia1 O,A,Asia1 None 7mths 2mths NA 0 0 0 70 80 78 75 80 87 80 80 90 2007 (3) O,A,Asia1 O,A,Asia1 None 10mths 6mths NA 6 0 8 47 42 51 59 60 45 18 28 38 2007 (Pig, O) 2006 (2) 2006 (Buffalo, cattle; O) February 2006 (goats; A) No samples 2005 (1) Unknown 5 0 0 15 June/July 2005 (Buffalo, pig, cattle, O) 2004 (goats/sheep, A) 0 10 20 30 40 50 60 70 80 Percentage 3 4 6 Round Percent of cattle Elisa positive when vaccinated for serotype O A Asia1 26 2005 (Round 1): There is no vaccination or infection history provided for this year, although there were outbreaks in the province in 2004 and 2005. There were no positives for either O or A ELISA assays and only very low for both 3ABC and Asia 1. 2006 (Round 2): No samples. 2007 (Round 3): Vaccination was with trivalent vaccine at either 6 months or 10 months prior to sampling or no vaccination at all. In the latter group the results were still suggestive of previous vaccination. Given the time post-vaccination it isn’t surprising that antibody levels for O and A are around the 40-60% mark. However, it does seem unusual that in both groups the Asia 1 positives are significantly lower, ie. around 20-30%. 2007 (Round 4): All animals, irrespective of vaccination status, had excellent response to all 3 serotypes (>70% but commonly 80%) on ELISA. Records suggest, despite this, that some animals were not vaccinated and others were vaccinated 7 months previously. This result is unexpected given the results from other rounds (both this and other provinces) in regards to the vaccination- sampling interval. It is possible that the vaccination date has been recorded incorrectly. 2008 (Round 5): Vaccination history is unknown, but the only test in which there were any positives was the O ELISA. In this assay 77% were positive. In the absence of infection it is certain that a large number of cattle were vaccinated and this information not recorded. 2008 (Round 6): Again, these results are unusual in that the history suggests vaccination with O/A serotypes just 2 months prior to the sampling date, yet the results for the A ELISA show relatively few positives on this assay (this could be vaccine-ELISA antigen mismatch) yet high percentage of positives (similar to those for the O ELISA) for the Asia 1 ELISA. There is also a spike in 3ABC ELISA positives in this round. Looking at the proportions of O, A and Asia 1 positives between the 3ABC positive and negative groups, the A positives are slightly overrepresented in the infected group. Proportions of O and Asia 1 positives are similar between the 3ABC positive and negative groups. A different commune is used to that used in other rounds (where recorded) so infection may have occurred in this area only. However there is no record of outbreaks in this province in cattle. An outbreak of serotype O occurred in pigs in 2007. It is possible that there was a concurrent unrecorded outbreak in cattle. Conclusions It appears that some of the recorded vaccination details (dates and presence/absence of vaccination) may have been incorrect for Quang Nam. This makes interpretation of the results difficult. Vaccination response seems good in round 4 but poor in round 6. Information is insufficient to make judgment for round 1, 2 and 5 and vaccination in round 3 is at the length/beyond the vaccine protective period. 27 Pigs Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) O 2mths 0 8 0 0 2008 (5) No samples 2007 (4) None O O NA 1wk 2mths 0 0 0 3 18 14 0 0 0 0 0 0 2007 (3) None NA 0 0 0 0 2007 (Pig, O) 2006 (20 2006 (Buffalo, cattle; O) February 2006 (goats; A) No samples 2005 (1) June/July 2005 (Buffalo, pig, cattle, O) 2004 (goats/sheep, A) No samples Samples were collected for pigs in round 3, 4 and 6 with pigs vaccinated in the latter 2 of these rounds. Vaccine response for serotype O ranged from 8-18% in pigs vaccinated up to 2 months prior to sampling. Quang Ninh Cattle Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) June/July 2008 (Buffaloes; O) No samples 2008 (5) Unknown Unknown 0 77 0 0 2007 (4) (O,A,Asia1) 1mth 1 79 98 96 2007 (3) February 2007 (unknown, O) No samples 2006 (2) March 2006 (Goats; A) No samples 2005 (1) Unknown 7 0 0 0 January 2000 (unknown; O), May 2001 (goats;O) 28 2005 (Round 1) As with other provinces there was little history provided with these samples and thus the results are relatively meaningless. There is indication that a small proportion of the population has been previously infected. 2006 (Round 2) There were no samples for this round. 2007 (Round 3) No samples. 2007 (Round 4) In this round cattle were vaccinated one month prior to sampling with a trivalent vaccine and the ELISA result suggests excellent response to the vaccine. 2008 (Round 5) There is no history available for this round, and it appears that samples were only tested by 3ABC and O ELISA. It is almost certain that the cattle have recently been vaccinated against at least serotype O, due to the 77% of positives on this assay. 2008 (Round 6) No samples. Conclusions There was only one round from Quang Ninh for which there were sufficient samples and information about which to make a judgment of vaccine response. The vaccine response was excellent in that round. However both sample and data collection need to improve so that Quang Ninh can actively participate in the National Control Program. 29 Pigs Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Previous Infection date in province (species, serotype) Comment Type Vaccination- sampling interval 2008 (6) June/July 2008 (Buffaloes; O) No samples 2008 (5) No samples 2007 (4) Unknown 1mth 0 0 0 0 2007 (3) February 2007 (unknown, O) No samples 2006 (2) March 2006 (Goats; A) No samples 2005 (1) January 2000 (unknown; O), May 2001 (goats;O) Samples were only collected from pigs in round 4 of the project, and all pigs were negative on all 4 ELISAs. 30 Tay Ninh Cattle Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Comment Type Vaccination- sampling interval 2008 (6) O,A 1 mth 20 35 90 0 2008 (5) none none 42 27 9 3 No history of vacc or infection 2007 (4) O,A 1mth 3 69 69 9 2007 (3) O,A 5 mth 5 0 15 0 2006 (2) O,A 1 mth 48 96 54 12 2006 (pig, buffalo; O) No record of infection 2005 (1) Unknown Unknown 27 35 5 0 May 2005 (pig; O) No infection or vacc information 0 10 20 30 40 50 60 70 80 90 100 Percent 2 3 4 6 Round Pecent cattle Elisa positive when vaccinated for serotype O A 2005 (Round 1) There are no vaccination records for cattle or pigs for the first round. About ¼ of the cattle were 3ABC positive, suggesting previous infection. Of these, the majority were also O 31 ELISA positive, with a small number A ELISA positive. Around 15% of the cattle were O ELISA positive but not previously infected (3ABC negative) indicating vaccination. 2006 (Round 2) In this year 50% of cattle showed signs of previous infection. Outbreaks recorded for previous years were only in pigs and buffalo. It is possible that there was either ™ Movement of infected animals (from other provinces or countries) ™ Unreported/undetected (mild clinical signs) infection in cattle ™ Large number of carriers that remain (at least intermittently) NSP ELISA positive It is difficult to determine the serotype to which these cattle had been exposed, as the relative proportions of O ELISA and A ELISA positives between infected and non-infected groups is approximately the same. Titration of positive samples would also be required. There is some discrepancy between the O and A ELISA results, with almost 100% positive to the O ELISA and 50% positive on the A ELISA after vaccination one month previously with bivalent vaccine. This could be due to either a vaccine-ELISA antigen mismatch, or vaccination of some cattle with monovalent (O) vaccine only. 2007 (Round 3) Cattle were administered O/A vaccine 5 months prior to sampling. The vaccination response rate is extremely poor – so low that it is unlikely to be just waning antibodies post- vaccination. It is more likely that different animals were presented for vaccination and sampling or inaccurate recording of vaccination. There was no indication of previous infection in cattle. The village used in this round was also one of the villages sampled in round 2, in which there was a very high proportion of 3ABC positive cattle. The dramatic decrease in positives on this ELISA suggests few were still carrying the virus, or that animals were moved out of the village. 2007 (Round 4) In the second round of 2007 cattle were vaccinated with O/A vaccine and sampled 1 month later. The vaccination response was reasonable with 70% cattle positive on both the O and A ELISA. About 10% were also positive for antibodies to Asia 1 without any positive to the 3 ABC ELISA, suggesting previous vaccination with Asia 1 also. 2008 (Round 5) Vaccination occurred later than normal in this year as vaccine was not available for the normal March vaccination period. As such, the cattle were not vaccinated prior to sampling and the ELISA results are consistent with this. 32 Again, there is a relatively high (40%) 3ABC positive ELISA response and most of the O ELISA positives were also 3ABC positive suggesting that this was the serotype of infection in these animals (however this would need to be confirmed by titration of the ELISA positive samples). These animals came from 2 different communes. Without recent outbreaks of serotype O in cattle in the province this is suggestive of: ™ Movement of infected animals (from other provinces or countries) ™ Unreported/undetected (mild clinical signs) infection in cattle ™ Large number of carriers that remain (at least intermittently) NSP ELISA positive 2008 (Round 6) In this round cattle were vaccinated 1 month prior to sampling with O/A vaccine. There was excellent response to the A component of the vaccine (100% positive on the ELISA) but only 25% on the O ELISA. This is most likely to be a problem with the sensitivity of the O ELISA, given the response to the A component of the vaccine and assurances that O/A vaccine was administered to all cattle. Conclusions Vaccination responses in cattle were very variable in Tay Ninh. In addition, 3ABC ELISA results varied and were not entirely consistent with outbreak history. Combined these results suggest significant movement of cattle into or through this province. 33 Tay Ninh - Pigs Year Vaccination % 3ABC + % O ELISA + % A ELISA + % Asia 1 + Comment Type Vaccination- sampling interval 2008 (6) O,A 1-2mth 8 21 41 0 2008 (5) None NA 0 0 0 0 2007 (4) O 1mth 0 0 0 0 Almost all <4mths 2007 (3) None NA 0 0 0 0 2006 (2) None NA 0 8 0 0 2006 (pig, buffalo; O) 2005 (1) None NA 0 0 0 0 May 2005 (pig; O) There was neither vaccination history nor seropositive results for pigs in rounds 1, 3 or 5. In round two 8% of pigs were seropositive on the O ELISA, but there is no vaccination history for this group. There was an outbreak of serotype O FMD in pigs in Tay Ninh in 2006. Although there was history of vaccination of pigs in the villages in round 4 (with serotype O vaccine) it is possible that younger, previously unvaccinated piglets were presented for sampling, which would explain the zero antibody response to vaccine. The last vaccine had been administered 8 months prior to sampling, and most of the piglets sampled in this round were under that age. Records for pigs in round 6 show that they were vaccinated with O vaccine, but only 20% had a serological response to the vaccine. However 40% were positive on A ELISA and this does not match the vaccination history. If these are larger farms that have been sampled it is possible that they are also vaccinating for serotype A with their own, self-purchased vaccine but not informing sub-DAH. It is also possible that pigs have moved to the area from other provinces and/or countries. 34 Results from Vaccine Trials and HCMC trials HCMC Trials In this trial 61 pigs were vaccinated with serotype O vaccine. Their ages ranged from 2.5 months to >12 months, and vaccination-sampling interval ranged from 20 – 122 days (<1mth-4mths). 72% of pigs were seropositive on the O ELISA vaccine. Pigs at the lower and upper ends of both the age group and the vaccination-sampling interval were positive on this assay. Interestingly 16% and 20% of pigs were positive to the A and Asia 1 ELISA respectively. All of the A+ pigs were >12months and only ½ were also O ELISA positive, while most of those that were Asia 1+ were aged 3-4 months and predominantly O ELISA positive. No pigs had either a history of infection or a 3ABC positive result. The A ELISA positives are almost certainly due to previous vaccination for this serotype, particularly due to the age of the pigs. It is possible some of the younger pigs also received different vaccines. 0 20 40 60 80 % O A Asia1 Elisa type % Pigs Elisa positive after O vaccine Forty cows ranging in age from 12months – 7 years were vaccinated with O/A/Asia1 with a vaccination-sampling interval of 43-154 days (1.5-5 months). The seropositives by ELISA type were 70%, 80% and 100% for O, A and Asia1 respectively. 38% of cattle were also 3ABC positive, despite no infection history. These cattle ranged in age from 1.5-6years. Those at the upper end of the age range may have had multiple vaccinations. It is known that NSP ELISAs can be positive in animals that have received multiple vaccinations (or unpurified vaccines). The representation of O/A/Asia1 positives between the 3ABC ELISA positive and negative groups was similar. Although 19 buffalo were also vaccinated as part of this trial, the vaccine dates and serotypes are unknown, thus limiting the evaluation of the data. 35 Vaccine Trials In this trial, 20 cows were vaccinated with O/A vaccine and 76 with O/A/Asia 1. Samples were taken 3 weeks after the 2nd vaccination. All were Merial vaccines. The overall vaccine response was >88% (up to 100%) for each serotype where administered. No cattle were 3ABC positive. Ages are not available. It would also have been useful to test these animals at 5-6 months post-vaccination, given that in the field there appears to be a reduction in herd immunity over the vaccination coverage period. Trial Conclusions Vaccination and sampling in the vaccine trial group was obviously performed under ideal conditions. In these conditions vaccine response was excellent, and would be sufficient for herd immunity. In HCMC trials the vaccine and sampling were conducted under conditions similar to those seen in the field (for example, a range of animal ages and vaccination-sampling intervals). However, in this environment the vaccination response rate was significantly higher in both the pig group than seen in the provincial results. This suggests that properly administered vaccination is effective in pigs and indicates that pigs in the field may not be correctly receiving vaccine or that there are different animals vaccinated and sampled, as appears to be the case on a number of occasions. Vaccination response rates in cattle appeared to be better than in most rounds in the provinces. Conclusions • In cattle the vaccines do not appear to be lasting the full 6 months, with antibody levels regularly good at 1-3 months post vaccination but poor 4-6 months after vaccination. • In pigs the vaccination response is generally very poor. The best response to vaccination with serotype O was just under 50%. However in at least 13 rounds where serotype O vaccine was administered <6 months prior to sampling the proportion positive was <10%. This is in contrast to the HCMC trial results in which over 70% of pigs were seropositive on the O ELISA with a vaccination-sampling interval of up to 4 months and a variation in piglet/pig ages. It is highly likely that some of this ambiguity is due to the presentation of piglets that have not yet been vaccinated, or those have only received one vaccination (and therefore may not have full immune response). Some of the remaining ambiguity may be due to incorrect history (of vaccination). The results from the HCMC trials indicate that the sensitivity of the O ELISA in pigs is good. In a large number of rounds pigs were not vaccinated. 36 • One of the major difficulties in the collection of interpretable information is the lack of animal identification, which would assist in confirming that the same animals are both vaccinated and sampled, and also allow for records of previous vaccination and infection for each animal to be accessed. • It appears that in many cases, based on the available laboratory data, animals are vaccinated with different vaccines to those that they are recorded as receiving. This assumption can be made where animals have no infection history and are 3ABC negative yet have a LP ELISA positive result for a serotype for which they have not been vaccinated. This interpretation is best made on a group rather than individual basis. • Problems did occur with variation in the ELISA results from round to round, particularly with the O LP ELISA. The production of an local antigen will help solve this problem and produce more consistent results. • The use of serology to determine the circulating FMDV is possible but expensive and labour intensive and is only carried out in focused studies where further investigation of FMD outbreaks is warranted. 5.2 Capacity Building The project has provided training and technology transfer of FMD diagnostics to each laboratory involved in the project. Reagents and standard methods were supplied to each laboratory giving them the diagnostic capability for FMDV diagnosis. Serology and serotyping (detection of antigen) by ELISA is now being practiced at all of the laboratories and RAHO – 6 and NCDV have also established virus isolation, cell culture, virus neutralisation for serology, molecular and sequencing techniques. Training and education of field veterinarians in sample and data collection showed an impact with an increase in quality and numbers of samples collected and submitted to the laboratory. These skills will be vital in implementation of the National FMD Control Program. 5.3 Publicity The CARD AusAID project received publicity in Vietnam, Australia and internationally through the training programs and also through the achievements in understanding FMD in Vietnam. FMD is a disease of importance in Vietnam and the region and this put this project into the lime light. The project has been publicised through a press releases in Australia and articles in newsletters including the SEAFMD newsletter and on the internet. The results from the project have been presented at: o OIE/SEAFMD meetings o EU FMD 2008 o WAVLD 2007, 2009 o Lower and Upper Mekong Working Group meetings in the region. Note: copies of presentations stead with milestone 3 report. 37 6. Implementation & Sustainability Issues 6.1 Issues and Constraints Throughout the project period DAH and field staff were regularly required to spend time controlling AI & PRRS outbreaks in Vietnam and this reduced the time available for the project. Although the sample and data collection improved throughout the project the initial difficulties in this area highlighted the need for further training of field veterinarians throughout Vietnam in disease investigation techniques with a focus on sample and data collection. There is a need for large scale training of field veterinarians to ensure better knowledge and control of disease in Vietnam. This could be achieved though better collaboration of aid agencies in Vietnam especially those supplying training for AI. Training in data analysis by DAH staff is also an area that needs further input. 6.2 Options The government of Vietnam is looking at increasing the support to DAH and has increased funding for AI diagnosis and is looking to do the same for FMD diagnosis. The increased funding for AI has seen laboratories updated with new equipment some of which will improve all disease diagnosis which includes FMD. The money available to for AI diagnostics can improve all disease diagnosis if this money is used wisely. For example, veterinarians should be trained in general disease investigation skills rather than focussing solely on HPAI. Quality assurance is also something that can be applied across all testing procedures at the laboratory. With the laboratory and epidemiological capacity now available in the collaborating laboratories, particularly HCMC, there is now the potential for a smaller, more focused study on vaccination failure. This would be best limited to a smaller number of provinces in southern Vietnam, with a study protocol aimed specifically at investigating vaccination effectiveness. 6.3 Sustainability As the collaborating laboratories are now implementing a range of FMD diagnostic techniques it is considered that these abilities will be sustained and transferred to other laboratories. Continuing training of field & provincial veterinarians by epidemiology staff at RAHO - 6 will ensure that the techniques learnt for sample and data collection and outbreak investigation in the field will also be sustainable. 38 7. Next Critical Steps Recommendations into Continuing Investigations into Vaccine Failure Data Collection • Where outbreaks have occurred in a province the strain of serotype involved should be recorded. This is important for matching of antigen between the ELISA and the field strain (and with knowledge of the antigen used in the vaccine). • It appears that the serotypes contained in the vaccines may vary despite the name staying the same. This variation can occur between years, and between species. It is possible for different people to have different ideas about what serotypes are contained in which vaccine. For this reason accurate recording of not just the manufacturer and the name of the vaccine, but the serotypes contained in it is absolutely vital for determining the level and effectiveness of the vaccine response. This information must be recorded by those in the field at the time of vaccinating. However, verification could be made by the DAH epidemiologists by obtaining a vaccine product guide from the supplier. Project Management • It is recommended that each province be presented with a report of their sero- surveillance results immediately after testing. This should be completed as soon as possible after samples have arrived at the laboratory. Graphs can be used to help illustrate the results and it may be useful to anonymously compare the results with those of other provinces. • Where possible results should be delivered in person by an epidemiologist, particular where there appears to be a failure in vaccine response. Staff at sub-DAH should be advised of such visits and provided with the checklist in Appendix 1 so that where possible information can be gathered in advance of the visit. The laboratory checklist should be completed first before the sub-DAH is visited (as this may uncover a laboratory error). Laboratory • The antigen used for the serotype A ELISA needs to match the circulating field strain. When animals are vaccinated with a bivalent or trivalent vaccine and the response on the A ELISA is significantly lower than on either the O or Asia1 ELISA another strain may have been used in the vaccine. This can be checked with the vaccine manufacturer and/or by testing the same serum on an A ELISA using a different antigen. This is important as it may indicate poor protection against serotype A in the field. • Samples should be tested as soon as possible after arrival at the laboratory. Rapid turn- around of results helps to maintain field interest and also ensures that ambiguous results can be followed up immediately. 39 Field • A temperature logger can be purchased relatively cheaply and could be used by the sub-DAH to ensure that both vaccine and samples stay cold. ie. the logger could be sent with the vaccine from the supplier to the sub-DAH, then into the field when vaccination is performed. It could then be used to monitor temperature of collected blood samples back to sub-DAH and onto the testing laboratory. • Field visits – at least to the sub-DAH office – are important for gaining an understanding of the provincial situation and to discuss results, particularly where there appears to be vaccine failure. 8. Conclusion The project achieved its objectives by helping to improve the FMD and general diagnostic capacity of the network of veterinary laboratories. The RAHO – 6 and Hanoi laboratories have provided and are continuing to provide support and training to other laboratories in the network. The project has highlighted the need for training of field veterinarians in the collection of data and how to ensure the correct information is obtained from the farmer. The staff at RAHO – 6 will continue to deal directly with the veterinarians in the field to ensure the correct information is collected in future surveys. The project has lead to better data collection not only in the project but in other projects due to the training and lesions learnt in this project. The collaborating laboratories and DAH now have standardised methods in place which are being applied in other areas and nationally. 40 Appendix 1 Investigating Vaccine Failure Checklist Field Laboratory Vaccinated for >1 serotype ≤ 6 months prior to sampling. Recognizably lower proportion of animals positive on one serotype than the other/s. 1. Were ALL animals vaccinated with the same vaccine? 2. Is it known if the same animals were both vaccinated and sampled? 3. Were all animals of this species in the village vaccinated at the same time? 1. Did the same technician perform all ELISA tests for this group of samples? 2. Were the samples retested? 3. Which antigen was used for this ELISA? a. Does this match the circulating field strain? b. Does this match the vaccine strain? Vaccinated ≤ 6 months prior to sampling with % positive for all serotypes <80% 1. What is the name of the vaccine ordered? 2. Who is the vaccine manufactured by? 3. What serotypes does the vaccine contain? 4. Was the same vaccine ordered and administered? eg. Was there leftover vaccine from last round that was used in this round – if so answer 1-3 above for this vaccine and do we know in which animals it was used? 5. Did the expiry date of the vaccine get checked prior to use? 6. Was the vaccine cold when it arrived from the supplier? 7. How many hours did it take to vaccinate all the animals? a. Did the vaccine stay cold during this time? b. How was it kept cold? 8. How many people vaccinated the animals? a. Has each of these people vaccinated animals previously? 9. Were all the animals (of this species) in the village vaccinated at the same time? a. From what age were animals vaccinated? 10. If only some animals in the village were vaccinated, how do we know that the same animals were presented for vaccination and sampling? 11. Is there substantial animal movement into this province from: 1. How long were the samples in transport? 2. Were the samples in suitable condition when they arrived? 3. Did they arrive as serum and/or whole blood? 41 a. other provinces? b. other countries? 12. Do we know if these animals have previously been vaccinated with FMD vaccine? a. If yes, with what and when? 13. How were the samples stored after collection? a. How long did it take to get them back to the laboratory? b. Were the samples sent as whole blood or serum? 14. Were all animals reportedly healthy at the time of vaccination? 42 Appendix 2 Interpreting the 3ABC ELISA Pathway captured in current (available) data 3ABC positive True positive False positive Vaccination Infection Local outbreak Unpurified Multiple vaccinations Reported Unreported “Imported” animals