Scotland ' s Rural College Newcastle disease vaccine virus I-2 fails to acquire virulence during repeated passage in vivo

Background This study determined whether the naturally attenuated, thermotolerant Newcastle disease vaccine virus I-2 could acquire virulence after five in vivo passages through SPF chickens. Methods Study design was to international requirements including European Pharmacopoeia, Ph. Eur., v9.0 04/2013:0450, 2013. I-2 Working Seed (WS) was compared with five-times-passaged I-2 WS (5XP WS) in intracerebral pathogenicity index (ICPI), Fo cleavage site sequencing and Safety tests. Results The first passage series used a 50% brain: 50% tracheal tissue challenge homogenate and was unsuccessful as I-2 was not detected after the fourth passage. A second passage series used 10% brain: 90% tracheal tissue homogenates. I-2 was isolated from tracheal tissue in each passage. However harvested titres were below the minimum challenge level (107 EID50) specified for the ICPI and Safety tests, possibly reflecting I-2’s inherently low pathogenicity (interestingly caecal tonsils yielded significant titres). Given this the WS and 5XP WS comparisons proceeded. ICPI values were 0.104 and 0.073 for the WS group and the 5XP WS group respectively confirming that I-2, whether passaged or not, expressed low pathogenicity. F0 amino-acid sequences for both WS and 5XP WS were identified as 112 R-K-Q-G-R-↓-L-I-G119 and so compatible with those of avirulent ND viruses. In safety, no abnormal clinical signs were observed in both groups except for two chicks in the 5XP WS group, where one bird was Open Peer Review


Introduction
Newcastle disease (ND) is one of the most important viral diseases of poultry and occurs in both commercial flocks and also in scavenging rural (village, backyard, sector 4,) chickens (Alders & Spradbrow, 2001;Cattoli et al., 2011;FAO, 2007;Spradbrow, 2000). The latter birds contribute significantly to the economies of poor households by providing eggs and meat for consumption and sale or bartering. These birds also are sources of readily available cash and gifts and may also have ceremonial or ritual value (Moreki et al., 2011;Perry et al., 2002;Peters et al., 2012a andPeters et al., 2012b). They require the lowest capital investment of any livestock species, and have a short production cycle (Copland & Alders, 2004). However their behaviour makes them prone to or exposes them to the spread of ND and other similar diseases, as despite being from multiple households, chickens will often congregate when scavenging so in effect forming one large village flock of all-ages and with unknown genetics. This enhances the transmission of infectious agents which has major implications for vaccination strategies (Msoffe et al., 2010).
Control of ND by vaccination is widely practised in commercial poultry flocks. However the vaccines used in the commercial sector are less suited for use in the village chickens of low-middle income countries (Aini et al., 1990;GALVmed-PANVAC-IIAM, 2009). Typically, commercial vaccines are produced in large dose vials, are often insufficiently thermotolerant so require a dependable cold-chain and generally are too expensive for use in village flocks. These issues led to the characterisation of thermotolerant ND virus vaccine strains that could be produced inexpensively in smaller batches by local laboratories for use in all-age backyard flocks (Campbell et al., 2019;Domingue et al., 2017;Spradbrow, 1993Spradbrow, /1994). An example of such a vaccine is the naturally attenuated, thermotolerant I 2 ND virus, now commonly known as I-2, which has long been known to be a suitable vaccine for use in developing countries. This is due partly to its high titre yield in embryonated eggs, its lack of virulence, a low pathogenicity coupled to a high immunogenicity which confers substantial protection, its thermotolerance (at least 12 weeks when stored at 22 o C in 1% gelatin), its straightforward delivery routes including by eye drop, its contact spread between birds and its safety and efficacy in very young (8 day) African local ecotype chicks (ACIAR, 2005;Bensink & Spradbrow, 1999;Copland & Alders, 2004;Dias et al., 2001;Domingue et al., 2017;Henning et al., 2009;Kattenbelt et al., 2006;Tu et al., 1998;Wambura et al., 2000;Wambura et al., 2006;Wambura et al., 2007). While the ND I-2 vaccine is thermotolerant, it eventually loses its potency if exposed to excessive sunlight or temperatures for long periods, i.e. it is not thermostable (Copland & Alders, 2004). This was partly addressed for thermotolerant ND vaccines in general by Domingue et al., 2017 who demonstrated that a preparation of 10X field dose of Clone 30 could offset viability loss due to high temperature (24 h, 32.3°C) while retaining safety and efficacy.
I-2 Master Seed (MS) is maintained by the University of Queensland, Australia and owned by the Australian Centre for International Agricultural Research (ACIAR) who make it available at no cost to low-middle income countries wishing to establish local ND vaccine production. (Alders & Spradbrow, 2001). ND I-2 vaccine use has also been allowed in these low-middle income countries because local registration requirements are relatively relaxed, but the vaccine has not been registered in those countries where registration requirements are more demanding.
The Global Alliance for Livestock Veterinary Medicines (GALVmed) is a not-for-profit organisation that helps develop and register veterinary medicines for livestock in those markets that are not attractive to the global commercial animal health industry. GALVmed has prioritised livestock diseases in lowmiddle income countries depending on perceived unmet need Amendments from Version 1 Shahn P.R. Bisschop et al. Some of the reviewers acknowledged the importance or excellence of the study and its appropriate design. Also none of our interpretation of the regulatory demands and the results was found to be incorrect. No corrections of Figures and Tables were necessary. There were no updates required to the Author list. Here we state our response to the comments from the 3 reviewers. We have responded by making 14 amendments and rebuttals which are in this revised version and also we refer readers to our already submitted pointby-point responses which include some rebuttals.
Throughout: in response to Prof Alders we have corrected "Newcastle Disease "to "Newcastle disease" so that it is in line with OIE standards.
Abstract: in response to Prof Alders the last sentence now reads as "From an international regulatory perspective, the study provides further definitive data demonstrating that Newcastle disease vaccine virus I-2 is safe for use." Introduction: " Unfortunately the behaviour of village birds make them prone to disease spread" has been replaced with "However their behaviour makes them prone to or exposes them to the spread of ND and other similar diseases" in response to Prof Alders and Dr Msoffe.
"and with unknown genetics" has been deleted in response to Dr Msoffe's comment.
We have deleted "Currently only one serotype of ND virus is" in response to Dr Morrow and "especially to help AU-PANVAC in its pro-poor aims" in response to Dr Msoffe.
We deleted "This has delayed or prevented registration in a number of low-middle income countries" and inserted "of an international regulatory standard" in response to Prof Alders and Dr Msoffe's comments.
Methods: "SPF" has been inserted throughout as per Dr Morrow's comment.
Discusssion: "Dr Spradbrow" has been changed to "Professor Spradbrow" in response to Prof Alders.
Any further responses from the reviewers can be found at the end of the article irrespective of species and target diseases, and so has included ND in poultry for development funding.
Despite I-2's long known suitability, there is a need to develop a globally acceptable I-2 registration dossieras agreed with the Bill and Melinda Gates Foundation. Indeed one of the obstacles to the wider use of the I-2 vaccine has been that no comprehensive dossier of an international regulatory standard has been compiled on the vaccine. Data confirming that the vaccine virus does not increase in virulence after serial passage through chickens is an essential component of such a dossier.
Therefore there was a universal requirement to re-investigate the naturally attenuated I-2 ND vaccine strain but under appropriate regulations to determine whether it would acquire virulence after five serial in vivo passages in SPF chickens. Accordingly, GALVmed sponsored a trial to good laboratory principles (GLP) (OECD, 1998) , 1981 to meet the requirements of a large number of countries so that the vaccine might be widely accepted by international registration authorities.

I-2 Newcastle disease (ND) virus
Master seed (MS) was a gift from ACIAR (Australian Centre for International Agricultural Research). Preparation of MS and working seed (WS) was as described (ACIAR, 2005). Briefly MS was reconstituted, titrated, diluted and inoculated into embryonated eggs. After 4 days incubation the allantoic fluids were pooled and the WS harvest was titrated, adjusted to 4.0 × 10 9 EID 50 ml -1 / 8.4 × 10 10 GC, genomic copies, ml -1 and stored at -70 o C.

Study practice and design
Design complied with international regulatory requirements (Ph. Eur. v9.0 04/2013Eur. v9.0 04/ :0450, 2013VICH GL 41, 2007;EU Council Directive 81/852/EEC, 1981) to ensure global acceptance. Study practice including data collection and storage was to Good Laboratory Practice (GLP) -like principles (OECD, 1998). All Study practice and design activities were performed under the umbrella of the GALVmed Quality Control system (VICH GL 41, 2007 standard). The study flow designs are shown in Figure 1a and 1b.
Briefly, the prescribed test for "increase in virulence" (European Pharmacopeia, Ph. Eur. v9.0 04/2013:0450, 2013) requires that native I-2 ND vaccine virus be administered to birds at "the least attenuated passage level present between the MS lot and a batch of vaccine". Since only one vial of ND I-2 MS was available, the least attenuated passage level was confirmed as Working Seed (WS; Ph. Eur. Helpdesk, personal communication).
The test for increased virulence, i.e. acquisition of virulence in the case of the naturally attenuated I-2, starts with the 5 times in vivo passage with at least 5 chicks per passage, of I-2 WS (i.e. native or non-passaged) to produce 5XP WS (i.e. 5X passaged WS).
The prescribed test then continues as the detection of any virulence changes in ICPI (Intracerebral pathogenicity index), F o amino acid sequence and Safety assessments (see below and Figures 1a and 1b).

Five-times in vivo passage
Ph. Eur. v9.0 04/2013:0450, 2013 requires the administration by the intra-ocular (IO) route, of a quantity of I-2 virus in homogenates containing both brain and tracheal tissues, that will allow recovery of I-2 virus through the 5 passages.
There were two passage series and each used seven SPF 1-day-old chicks per passage. In both series, to attempt the required I-2 recovery, the initial IO challenge, i.e. the challenge into the first passage birds, was 2.0 × 10 8 EID 50 / 4.2 × 10 9 GC via 2 x 25 µl drops of native I-2 WS at 4 × 10 9 EID 50 ml -1 / 8.4 × 10 10 GC ml -1 . Next, for both series, four sequential IO challenges i.e. numbers 2 to 5 (2 x 25µl eye drops of pooled organ homogenate from the previous passage) and passages followed. Caecal tonsils were also collected in both series and investigated for viral titres, as although not demanded by the Ph. Eur. v9.0 04/2013:0450, 2013, this was normal procedure at the study site; N.B. caecal tonsils were never incorporated into any homogenates.
Critically, the two series differed in their homogenate compositions. The first passage series was for "range-finding" and used an initial IO challenge homogenate made up of 50% brain: 50% tracheal tissue. During each passage, birds were observed twice daily for 4 days, subsequently euthanised (cervical dislocation) and a pooled suspension of homogenised brain and tracheal tissue was prepared for the next passage. The presence of I-2 virus in the pooled suspension of homogenised brain and tracheal tissue from each bird in each passage was detected by egg passage and subsequent EID 50 in HA and by real-time reverse transcription-polymerase chain reaction (rRT-PCR) (see below). Unfortunately, this first passage series was incomplete as I-2 virus could not be detected after the fourth passage and throughout I-2 virus yields from brain tissue were very low or negative (see Results).
In accordance with the Ph. Eur. v9.0 04/2013:0450, 2013 and implementing advice from the Ph. Eur. Helpdesk, the second passage series was performed. An "augmented" homogenate consisting of 10% brain: 90% tracheal tissue to minimise the dilution effect of the brain material, was used as the challenge inoculum for passages 2 to 5. This approach was successful in that I-2 was isolated to the end of the fifth passage, albeit in very low titres. Although these were below the minimum challenge level of >2.0 × 10 8 EID 50 ml -1 (i.e. 50 µl containing10 7 EID 50 ) specified by the Ph. Eur. v9.0 04/2013:0450, 2013 for the challenges for the subsequent ICPI and Safety tests, an exploratory approach was taken and the five times passaged I-2 was assessed against native WS I-2 in the ICPI, amino acid sequencing of the F 0 cleavage site and Safety tests as described below.
A. ICPI: a daily scoring of signs for 8 days after intracerebral (IC) administration (0 = normal, 1 = clinical signs of disease, 2 = dead). For a virus challenge of not less than 10 8 EID 50 or for a virus challenge less than 10 8 EID 50 but not less than 10 7 EID 50 , a virus would comply with the test if the ICPI induced was not greater than 0.5 or 0.4, respectively.
B. Determination of the encoded amino acids at the I-2 F 0 cleavage site that imparts pathogenicity in ND viruses.
C. Safety, an absence of clinical signs after ocular challenge, IO, the recommended vaccination route for I-2.

Bird numbers
To allow for mortality and to ensure that no treatment group included fewer birds than required in the Ph. Eur. monograph above, the treatment group sizes were increased slightly above monograph requirements per group: Passage -7 birds per passage for the first four passages and 10 birds for the fifth passage ICPI -12 birds

Safety -22 birds
Groups of 6 birds each were present as negative controls for each test that included virus passages and safety tests on the WS and 5XP WS. Further groups of 6 chicks served as controls for the ICPI techniques.

Location
The study was conducted at the National Department of Agriculture-approved BSL3 laboratory isolation unit at the University of Pretoria, South Africa.

Ethics
Approval for the study was obtained in advance from the University of Pretoria Animal Use and Care Committee as well as the Research Committee of the Veterinary Faculty. Approval for the study was also obtained in advance from the National Department of Agriculture. These approvals and the GALVmed Quality Control System ensured that the study followed the principles of the Animal Research: Reporting of In Vivo Experiments (the ARRIVE full set 2.0) guidelines (Kilkenny et al., 2010). Any trial amendments were also approved by the above committees of the University of Pretoria.

Chickens
Mixed sex, specific pathogen free (SPF) Leghorn chickens were obtained either as eggs on the point of hatching or as day-old  • Chicks were clinically normal and in good health.

Exclusion criteria
• Unhealthy chicks • In the event that more healthy chicks hatched than were required for the study, the study director excluded the smallest birds from the study.

Withdrawal criteria
• Any abnormal signs considered to compromise the welfare of the chicken.
• Any chicken requiring concurrent medication that could compromise its suitability for the study.
Randomisation and bias reduction. Randomisation plans were prepared by an independent biometrician. Chicks were individually identified using food colouring and/or coloured leg bands, so they could be assigned numbers and randomly allocated to treatment groups. Chicks were randomised to treatment and randomised between isolation cabinets. For some passages, no randomisation was performed as each treatment group was placed sequentially and on different days. For all five passages, chicks in each isolator were individually marked using coloured leg bands. This was solely for the purpose of monitoring, so that investigators were able to determine if an individual chick had been in protracted recumbency or had consistently exhibited particular clinical signs. Postchallenge observation and clinical scoring was conducted by two observers and by different observers throughout the different phases of the study in order to reduce observer bias.
Five-time in vivo passage (5XP) of I-2 WS and comparisons of 5XP I-2 with native I-2 WS Tissue homogenisation (for preparation of passage challenges). Post-mortem, organs were placed into a solution at a dilution of 1 part tissue: 4 parts PBS (phosphate buffered saline, pH7.4) containing antibiotics (penicillin 1,000 units ml -1 and streptomycin 10,000 µg ml -1 ) and an anti-mycotic (amphotericin B 25 µg ml -1 ) and blended.
I-2 quantification. All challenges for the in vivo passaging, ICPI and safety tests were quantified at time of use in the HA and rRT-PCR assays below as were all test yields.

Quantification by haemagglutination inhibition (HAI) test.
After each in vivo passage in chickens, organ homogenate was inoculated into the allantoic cavity of a series of 11-day embryonated SPF eggs and the EID 50 calculated according to the method of Reed & Muench (1938) using the macroscopic haemagglutination technique (Anon, 1989). Cycling parameters were as follows: 55°C for 10 minutes -(reverse transcription), 95°C for 1 minute -(initial denaturation); followed by 40 cycles of amplification followed by 95°C for 5 seconds (denaturation), 55°C for 5 seconds (primer binding), 60°C for 20 seconds (hydrolysis of probe and acquisition of fluorescence); cooling to 40°C for 10 seconds.

Quantification by rRT-
In each run, in order to quantify virus, at least three standard solutions containing known amounts of Newcastle disease virus I-2 GC were also analysed -the standards were prepared after purification of ND I-2 on a sucrose gradient, extraction of RNA from purified virus, and quantification of RNA (8.1 ng of NDV RNA represents 10 9 NDV GC. From a stock containing 10 9 GC µl -1 , 10-fold serial dilutions were made from 10 9 to 1.0 GC µl -1 . Quantification of RNA copies per ml of homogenised sample were calculated by multiplication of the genomic copies per PCR reaction by the dilution factor used during RNA purification. Assessment of possible increase in virulence ICPI test. Each bird in the WS group (n=12) was given an intracerebral (IC) challenge of 50 µl from pooled allantoic stock containing 4 × 10 9 EID 50 ml -1 / 8.4 × 10 10 GC ml -1 of I-2 ND virus (i.e. 2 × 10 8 EID 50 / 4.2 × 10 9 GC total).
Two control groups (n=6 each), one for each test group above, received a 50 µl IC challenge of PBS intracerebrally. Test and control birds were observed (clinical administration post vaccine administration, CEPVA) at 1 h and 4 h posttreatment and then at least twice daily for 8 days.
An ICPI was calculated for the chickens according to the Ph. Eur. v9.0 04/2013:0450, 2013. The ICPI is the mean of the scores per bird per CEPVA, performed twice a day, over an 8-day period: 0 = clinically normal; 1 = clinical signs of disease; 2 = dead. I-2 would comply with the test if the ICPI induced was not greater than 0.5 or 0.4 respectively.

F o amino acid sequencing.
Pooled, homogenised 10% brain: 90% trachea samples were analysed. The encoded amino acids at the F 0 cleavage site in RNA preparations from I-2 WS and 5 XP ND I-2 were determined via the RNA sequencing service of the Molecular Epidemiological and Diagnostic Programme, Agricultural Research Council -Onderstepoort Veterinary Institute, Pretoria, South Africa. Sequencing reactions were prepared using a BigDye Terminator v3.1 sequencing kit (Applied Biosystems, USA) and a 3500/3500xL Genetic Analyzer (Applied Biosystems). Multiple sequence alignments were performed using the BioEdit v 7.1 sequence alignment (Hall, 1999). The derived amino acid sequences were obtained using the ExPASytranslation tool (v 2003;Gasteiger et al., 2003).
Negative control birds (n=6) for each safety test received PBS pH 7.4 in two 25 µl eye drops. All birds were observed at 1h and 4h post-treatment and then at least twice daily for 21 days.
Control birds. All controls were examined for I-2 virus as above to detect possible accidental cross-inoculation except that for the rRT-PCR, choanal cleft swabs were pooled and analysed. Also "sentinel" chickens were placed to monitor biosecurity throughout the study.

Control birds
Negative, untreated and placebo control groups were present in the same study rooms during the I-2 in vivo passages, the safety tests and the ICPI tests. No abnormal clinical signs were observed in any negative control chickens. ND I-2 was never detected in any tissue samples. Again I-2 was never isolated from "sentinel" chickens indicating that biosecurity was satisfactory throughout the study.

In vivo passages.
Only one bird was withdrawn from the study. This bird subsequently died from a cloacal prolapse which was judged not to be linked to the challenge. No birds died in any of the in vivo passage groups, or in the control groups.
No signs of clinical disease in chickens were observed as a result of challenge with ND I-2 WS, whether native or five-times-passaged material (organ homogenate), in any of the passages. The first attempt at passaging was not successful in that virus could not be detected after the fourth passage but it was noted that viral loads in the tracheal tissues were 3-5 times greater than in the brain tissues (data not shown).
In the second series of passages, an augmented homogenate consisting of 10% brain: 90% tracheal tissue was used for the IO challenges for each group and this proved more successful as 5XP I-2 WS was obtained. The HA and rRT-PCR analyses of the tissues recovered from each passage are shown in Table 1. Significantly, I-2 was not detected in brain tissues through all passages and overall, virus titres in all harvested tissues were below the minimum challenge level of >2.0 × 10 8 EID 50 ml -1 (i.e. 10 7 EID 50 in 50 µl per chick) specified in the Ph. Eur. v9.0 04/2013:0450, 2013 for the challenge for the ICPI and Safety tests: the highest tracheal yields for I-2 were 5.1 × 10 6 EID 50 ml -1 / 4.1 × 10 7 GC ml -1 (passage 5) while those for caecal tonsils were 1.8 × 10 7 EID 50 ml -1 / 6.0 × 10 8 GC ml -1 (passage 2). Despite the low HA titres, it was decided that, given the exploratory nature of this work, to continue with the ICPI and Safety tests with a challenge homogenate composed of both brain and tracheal tissue to ensure compliance but with the former at only at 10%, to minimise the dilution effect of the brain material.  ill-health and an unwillingness to drink associated with the effects of the ND challenge as observed post-mortem. Two other birds were also seriously ill by the end of the study. There were two deaths due to dehydration shortly before the end of the study in this group -one bird was sick on day 7 and was found dead on the morning of day 8. One bird was noted as sick on the morning of day 8 and died on the final morning, day 9. Additionally, one bird was recorded as being sick on days 8 and 9 while a final bird was found sick on the morning of day 9. The ICPI value for the native ND I-2 WS was 0.104 for 96 possible observations i.e. 10/ 96 = 0.104. The native WS I-2 therefore complied with the test with an ICPI not greater than the cut out value of 0.4.

Comparative tests
For 5XP WS, two of 12 birds (16.7%) showed depression and died shortly before the end of the study on day 8 of observation. On post-mortem examination, both birds showed signs of dehydration. Two (16.7%) other birds were also seriously ill by the end of the study and three birds showed decreased activity when compared with birds in the ICPI control group. The ICPI value for the 5XP I-2 WS was 0.073 for 96 possible observations i.e. 7/ 96 = 0.073. The 5XP I-2 virus seemingly complied with the test as its ICPI was less than 0.4 but this is qualified as the challenge used was less than that required by the Ph. Eur. v9.0 04/2013:0450, 2013.
For both test groups, the negative control chickens showed no abnormal clinical signs related to the IC PBS challenge.
Amino acid sequencing. Samples from WS and 5XP WS were identified by BLAST analysis of the cDNA sequence as ND I-2 vaccine strains. The translated amino acid sequences at the fusion protein cleavage site (F O ) for both WS and 5XP WS were identical, namely 112 R-K-Q-G-R-↓-L-I-G 119 and further, were consistent with those of avirulent ND viruses,

Safety.
Generally no clinical signs of abnormal health were seen in chickens from both native WS I-2 and 5XP WS I-2, except for two chickens in the 5XP WS group: one bird developed a vent prolapse and was consequently excluded from the study, while another bird showed neurological signs and lateral recumbency before death. The post-mortem for this bird was inconclusive but dehydration signs were observed. Both native WS I-2 and 5XP WS 1-2 (given the low challenge level) had complied with the test as the cut out of < 10% of birds dying was not exceeded. The control birds never showed abnormal clinical signs.

Discussion
The Ph. Eur. v9.0 04/2013:0450, 2013 requires that all tissue homogenate challenges for in vivo passage must contain brain tissue. The first five times passage series used a 50% brain: 50% trachea ratio, meaning that I-2 in tracheal tissue was diluted by brain tissue (I-2 not detected) rendering the first series invalid. The second passage series used a homogenate ratio of 10% brain: 90% trachea which was more effective but again I-2 was not detected in brain tissue. The titres in tracheal tissues were highest after the fifth passage at 5.1 × 10 6 EID 50 ml -1 but still below the minimum challenge level of >2.0 × 10 8 EID 50 ml -1 required for the ICPI and Safety tests (Ph. Eur. v9.0 04/2013:0450, 2013. This seemed to indicate that the birds were holding the infection at low levels throughout the passaging, which probably reflected the inherently low level of pathogenicity for ND I-2 (Dr Peter Spradbrow, personal communication

I confirm that I have read this submission and believe that I have an appropriate level of
However, there are a few items that need to be addressed to give the paper more strength and validity.
transmission. I am not sure that the EU helpdesk advice that in vitro amplification between passaging is completely excluded is correct. Usually if direct passage without in vitro progation has been attempted and is not successful twice (as described in this paper) applicants would make a case for using in vitro passage amplification and I would have thought most (valid) ICPI testing would have used allantoic fluid and it may not have been validated for the final passage material being assessed and therefore brain and tracheal homogenate would need extra controls in the ICPI test. At least in the final passage before the safety testing in vitro passage may be needed).
The statement that "Currently only one serotype of ND virus is recognised" is tautological as haemagglutinating viral isolates are identified as NDV by their serological characterization by sera raised against a NDV type strain raised antibody.
The stated aim of the project to generate local manufacture of inexpensive vaccine is now an anachronism with regulation creep imposing first world quality standards on vaccines in nearly all countries. The need for SPF eggs for vaccine manufacture and GMP standards are nearly universal and an expense. The need to protect chickens from viral and other contaminates is hard to argue against (having suffered through REV, EDS-76, ALV, CAV over the years). The materials and methods do not state whether SPF eggs were used for the initial propagation of the material studied.
It is not clear what SPF definition has been used for this work.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound? Partly

Are sufficient details of methods and analysis provided to allow replication by others? Partly
If applicable, is the statistical analysis and its interpretation appropriate?

Not applicable
Are all the source data underlying the results available to ensure full reproducibility? Partly Are the conclusions drawn adequately supported by the results? Yes significant reservations, as outlined above.