**2. Current vaccine safety tests**

The microarray technology was recognized in the toxicology research community after its introduction in the 1990s [3–5]. Subsequently, toxicological studies using the microarray technology have given rise to a new field termed toxicogenomics [6]. By integrating the genomic technology and bioinformatics, toxicogenomics has garnered a great deal of attention as an alternative means of addressing drug safety by studying the fundamental molecular mechanism of toxicity, which was difficult to detect with conventional toxicological methods [6]. In fact, microarray-based toxicogenomics remains a major breakthrough. Using microarrays, we can monitor the expression levels of tens of thousands of genes at the same time and evaluate gene expression profiles altered by various compounds or changes in gene expression profiles associated with different physiological conditions. Moreover, testing a large number of genes together gives the opportunity to identify genetic patterns and signatures that provide unique insights into drug toxicity, which are difficult to obtain by conventional animal-based techniques [7]. Thus, toxicogenomics is expected to revolutionize the traditional approaches to toxicity assessment and has been considered a paradigm shift in toxicology. To date, many studies have revealed the value of toxicogenomics [8–15]. For example, it has been suggested that toxicogenomic biomarkers can identify drug candidates that are more likely to cause toxicity in susceptible patient populations despite the lack of conventional toxicity indicators, such as hematological parameters, body weight changes, blood biochemical data, and histopathological data, which are examined in preclinical studies [16, 17]. Similarly, more sensitive biomarkers for the detection of early toxicity can be analyzed at "subtoxic doses" of a candidate therapeutic agent, where the injury is at the genetic level, but does not occur at the

phenotypic level or cannot be detected by clinical-chemistry measurements [18].

carried out, when compared to the testing of chemical and synthetic drugs.

technology.

114 Influenza - Therapeutics and Challenges

Just as chemical and synthetic drugs, biological therapeutics are evaluated for their toxicity by safety tests involving animal experiments, as part of preclinical studies. In addition, to guarantee the quality and homogeneity of the preparation, a portion of the biological preparation is subjected to toxicity tests for each lot [19, 20]. These toxicity tests are based on the aforementioned conventional assays, and phenotypic alterations such as body weight changes, hematological changes, pathological changes, and similar parameters are the evaluation criteria [19, 21]. Tests of the safety and quality control of vaccines include the abnormal toxicity test (ATT, also known as a general toxicity test) [21], and the leukopenic toxicity test (LTT) [19]. In all preclinical trials, in addition to these tests, a pathological examination is carried out. Although these tests have historically been practiced for a long time, it is expected that genomics techniques will be incorporated into these tests to improve their sensitivity and to obtain information on toxicity. For biologicals, however, toxicity studies using the genomics technology have not yet been actively

Therefore, we have been using the genomics technology to search for vaccine safety assessment markers since the late 2000s. In particular, we have been conducting research on the use of genomics technology for studying pertussis vaccine [22, 23], Japanese encephalitis vaccine [24], and influenza vaccine [25]. This chapter provides an introduction to the genomics technology in the safety assessment and quality control evaluation of influenza vaccines and describes a new evaluation method involving the biomarkers obtained by the genomics This section describes the lot release safety and quality control testing methods implemented for the influenza vaccine. The ATT has been conventionally conducted as an animal-based test to evaluate the contamination by phenol, which is used in the process of inactivating endotoxins, viruses, and bacteria [26, 27]. The method of ATT is simple: 5 mL of a sample is injected into the abdominal cavity of a guinea pig, and its survival and 7-day body weight changes are measured [19]. It has been suggested that these 7-day body weight changes reflect the biological activity of the vaccine. Indeed, if the animals were inoculated with a different type of influenza vaccine, their body weight changes would show different profiles [21]. The whole-particle inactivated influenza vaccine (WPV) that has high reactogenicity [28]. Nevertheless, it causes highly frequent adverse reactions, such as pain, swelling, and fever [29]; on the other hand, the hemagglutinin-split influenza vaccine (HAV), whose effectiveness is inferior to that of the WPV [28], has been reported to cause almost no adverse reactions [29]. The current seasonal influenza vaccines are based on the HAV, and the WPV is manufactured only as a pandemic vaccine. This approach also includes avoiding adverse reactions caused by WPVs. Therefore, the WPV also serves as a toxicity index in the quality control testing of the HAV by the LTT, which is described later.

The LTT is a safety test that assesses the leukopenic toxicity induced by the WPV as a toxicity index [19]. In this method, mice are inoculated with 0.5 mL of a WPV as a toxicity reference vaccine, and the leukopenic activity rate induced by the WPV at that time point is set to 100% leukopenic activity. At that time point, the test sample confirms whether the leukopenic activity rate is within 20% or not. The test criterion is as follows: the leukopenic activity rate should be less than 20% of the toxicity control. On the other hand, the ATT is an assay that evaluates the body weight loss of guinea pigs, and the transition during their recovery. When the same experiment was carried out in rats, WPV-injected rats showed a severe body weight loss, unlike HAVinjected rats [21]. A vaccine showing a statistically significant body weight loss in a population, when compared with a homogeneous preparation, would be rejected in terms of its quality.

Thus, safety and quality of the HAV are mainly ensured by two tests. Safety assessments of the HAV have been conducted using WPVs, which is one of the safety indices.
