COVID19 Vaccines From Moderna And Pfizer Are Almost Here But MIT’s Machine Learning System Has Uncovered Efficacy Gaps
Numerous pharmaceutical giants such as Pfizer Inc. (NYSE: PFE), Moderna Inc (NASDAQ: MRNA), and others are preparing to roll out their COVID-19 vaccines, but there is a chance that those vaccines might not work for some people.
Massachusetts Institute of Technology researchers recently published a study that tapped into machine learning to detect vaccine efficacy gaps based on genetics. The study revealed that the vaccines currently in phase 3 clinical trials might be less effective on people with Asian or black genes compared to the vaccine performance in white people.
David K. Gifford, a Computer Science and Artificial Intelligence Laboratory (CSAIL) technician at MIT, pointed out that initial results indicate that the vaccines might be less effective on blacks or Asians. The findings are based on DNA data, and researchers continue to explore numerous other factors that may influence vaccine effectiveness. Understanding how new vaccines work is the best way to understand why the reasons for the existence of potential efficiency gaps.
How are mRNA vaccines designed?
Messenger RNA is the system that the body uses to deliver instructions. mRNA vaccines contain instructions for the body to manufacture a small part of the SARS-CoV-2 spike protein, which is then used as an immune system trigger. Once the immune system recognizes the spike protein as a foreign body, it manufactures antibodies to combat the foreign components. The approach allows the body to be prepared if it comes into contact with the actual SARS-CoV-2 virus.
Pfizer and Moderna have been using the mRNA to develop their pipeline vaccines for the coronavirus. The vaccines are supposed to be up to 95% effective, and the mRNA vaccine approach is relatively new, thus raising concerns about its effectiveness and how it will affect users.
The MIT report suggests that the mRNA vaccines do not have adequate viral particle data sets to promote similar immune responses in diverse populations based on genetic composition. This lack of consistency across all data sets suggests that the vaccines might not yield favorable results in Asians and Blacks.
Will the machine learning findings reflect vaccine performance once launched?
The MIT study was conducted using in silico computer designs, and it leveraged machine learning to make immune system predictions based on protein models and patient data. It means that the study was not conducted based on the actual vaccines, and the vaccine results might not necessarily reflect the study findings.
The report from the study findings has also not been peer-reviewed, which means that the findings are not verified, so they should not be considered final. However, they do provide an insight into the potential inadequacies of the upcoming mRNA-based coronavirus vaccines.
AstraZeneca, Pfizer, and Moderna recently announced preliminary findings of their phase 3 clinical trials of the COVID-19 vaccine. The pharmaceuticals observed that the mRNA vaccines are 94% to 95% more efficient at preventing COVID-19 than the placebo, which means that the vaccines could help secure a win in the fight against the coronavirus.
Sources:
https://edition.cnn.com/2020/12/02/health/coronavirus-vaccine-volunteer-side-effects/index.html
