Modeling for influenza vaccines and adjuvants profile for safety prediction system using gene expression profiling and statistical tools

Abstract

Historically, vaccine safety assessments have been conducted by animal testing (e.g., quality control tests and adjuvant development). However, classical evaluation methods do not provide sufficient information to make treatment decisions. We previously identified biomarker genes as novel safety markers. Here, we developed a practical safety assessment system used to evaluate the intramuscular, intraperitoneal, and nasal inoculation routes to provide robust and comprehensive safety data. Influenza vaccines were used as model vaccines. A toxicity reference vaccine (RE) and poly I:C-adjuvanted hemagglutinin split vaccine were used as toxicity controls, while a non-adjuvanted hemagglutinin split vaccine and AddaVax (squalene-based oil-in-water nano-emulsion with a formulation similar to MF59)-adjuvanted hemagglutinin split vaccine were used as safety controls. Body weight changes, number of white blood cells, and lung biomarker gene expression profiles were determined in mice. In addition, vaccines were inoculated into mice by three different administration routes. Logistic regression analyses were carried out to determine the expression changes of each biomarker. The results showed that the regression equations clearly classified each vaccine according to its toxic potential and inoculation amount by biomarker expression levels. Interestingly, lung biomarker expression was nearly equivalent for the various inoculation routes. The results of the present safety evaluation were confirmed by the approximation rate for the toxicity control. This method may contribute to toxicity evaluation such as quality control tests and adjuvant development.

Genomic Approaches Enable Evaluation of the Safety and Quality of Influenza Vaccines and Adjuvants

“…In other words, if the factor of effectiveness becomes excessive, toxic effects may appear. We are currently working on establishing a safety assessment system based on the WPV as a toxicity indicator [60]. The WPV is an effective and excellent vaccine, but the frequency of adverse reactions is high, and currently, it is only rarely used, especially as a prepandemic vaccine.…”

“…Therefore, we believe that the WPV can be a safety indicator. In other words, we think that there is a high probability that adjuvants and vaccines with innate-immunityinducing activities that exceed the activity (toxicity) of WPVs will cause adverse reactions [60].…”

Immunogenicity and Toxicity of Different Adjuvants Can Be Characterized by Profiling Lung Biomarker Genes After Nasal Immunization

“…We have previously reported that it is possible to profile the immunogenicity of inactivated influenza vaccines by analyzing the expression patterns of lung genes within 16 h of priming using genomic analyses (38)(39)(40)(41)(42). In addition, we constructed a novel safety evaluation system for an adjuvanted inactivated influenza vaccine using WPV as a toxicity control via mathematical analyses (43). It has also been reported that these evaluation methods can be applied in mouse (40) and rat (38,39) models, as well as for the evaluation of nasal influenza vaccine (41).…”

“…Therefore, the 18 clustered biomarker genes expressed in the lungs were designated as immunogenicity indicators for influenza vaccines ( Table 1) (38). The biomarker gene expression levels were altered by several adjuvants contained in SV (41)(42)(43). As the present study focuses on influenza vaccine, we used the biomarker gene set similar to those used in our previous studies (39)(40)(41)(42)(43).…”

“…The biomarker gene expression levels were altered by several adjuvants contained in SV (41)(42)(43). As the present study focuses on influenza vaccine, we used the biomarker gene set similar to those used in our previous studies (39)(40)(41)(42)(43). We previously performed a logistic regression analysis to estimate the similarity between the immunogenicity rate of a test vaccine to that of WPV using biomarker gene expression profile data (43).…”