African swine fever (ASF), caused by the ASFV virus, has reached pandemic proportions, severely impacting the global pig industry and wild pig populations. The lack of effective commercial vaccines necessitates ongoing research into safe and efficacious alternatives. Live attenuated vaccines (LAVs) show the most promise, with ASFV-G-AMGF (AMGF) demonstrating a strong efficacy profile. This study evaluates AMGF's safety using an in vivo reversion to virulence study following VICH guidelines, involving five serial passages in domestic pigs to assess genetic stability and potential virulence reversion.

The introduction section reviews the devastating impact of ASF, the urgent need for a vaccine, and the challenges associated with developing safe and effective live attenuated vaccines. It cites various publications highlighting the global spread of ASF, the limitations of current control measures, and the potential of LAVs as a promising approach. The review emphasizes the critical need to balance efficacy with safety concerns, particularly regarding potential genetic mutations and virulence reversion in live vaccines.

The study employed a stringent experimental design adhering to VICH guideline 41. Five groups of ten 6–10-week-old weaner pigs were inoculated intramuscularly with the AMGF master seed virus (MSV). Organ samples were collected at day 7 (except passage 5, where it was 21 days), with the highest viral load tissues selected for subsequent passages. Clinical observations, including body temperature and clinical scores, were recorded daily. Pathological examinations were conducted post-mortem. Quantitative PCR (qPCR) assessed ASFV genome loads in various tissues and swabs. Whole-genome sequencing (WGS) identified genetic changes in the virus. A tailored qPCR was developed to distinguish between the original MSV and any emerging variants. In vitro growth kinetics compared the replication of the original and variant viruses in primary macrophages.

No clinical abnormalities were observed in the first passage. However, passages 2–5 showed transient fever (up to 42.1 °C) in multiple animals, with mild clinical signs like reduced appetite and apathy. All animals recovered fully by the end of the study. qPCR revealed increasing ASFV genome loads from passage 3 onwards, with detectable viral genome in swabs. WGS of passage 4 samples identified a virus variant (AMGFnV) with a large deletion at the 5′-end and a duplication at the 3′-end, resulting in the loss and duplication of multiple genes. The AMGFnV was initially detected in one animal of passage 1 and subsequently spread through the remaining passages. Despite the emergence of AMGFnV and its association with increased replication and mild clinical signs, no significant virulence reversion was observed. Comparative in vitro growth kinetics showed no significant difference in replication between MSV and AMGFnV in primary macrophages. Retrospective analysis of samples from previous efficacy tests showed no evidence of AMGFnV.

The study demonstrates that while the AMGF vaccine candidate showed high efficacy in previous studies, the emergence of the AMGFnV variant necessitates a thorough safety assessment before potential field application. The observed mild increase in virulence, manifested primarily as transient fever, raises concerns about the long-term consequences and transmission potential of the variant. The authors acknowledge the artificial nature of the serial passaging method used in the study, which might not fully reflect natural transmission dynamics. The discussion considers the trade-off between the vaccine's efficacy and the emergence of a less attenuated variant, emphasizing the need for a comprehensive benefit-risk assessment. The unique genetic changes observed in AMGFnV suggest a potential common mechanism of ASFV adaptation under selective pressure. The findings underscore the importance of comprehensive safety testing and suggest the need for genetic tools to differentiate between infected and vaccinated animals (DIVA).

The study provides valuable insights into the safety profile of the ASFV-G-AMGF vaccine candidate. While the emergence of AMGFnV indicates a need for further investigation of its long-term effects and transmission characteristics, the lack of significant virulence reversion suggests potential for future development and controlled field application after a thorough benefit-risk assessment. Further research should focus on the long-term effects, transmission characteristics of AMGFnV, and the development of DIVA strategies.

The study's artificial passaging method might not fully reflect natural transmission dynamics. The in vitro growth kinetics showed no significant difference between MSV and AMGFnV, indicating limitations in the model's ability to capture complex in vivo interactions. The study's duration might not capture long-term effects of the vaccine or the variant virus.