ON THE BIOLOGICAL ASSAY OF TOXICITY OF PERTUSSIS VACCINE USING MICE III. CONSTRUCTION OF CHARTS OF CRITICAL REGIONS

Abstract

The definition of an absolute toxicity in terms of animal reactions is unsatisfactory, because such an“animal unit”is inevitably variable due to the variation in the mean response of stocks of animals in different laboratories and at different times.
In contrast to the failure to make reproducible measurements in terms of animal units, a much greater success was expected by the introduction of a comparative method, in which estimation of a toxicity of an unknown preparation was made by comparison with a stable and defined reference preparation and the toxicity was expressed as a relative value to the reference. The comparative method of assay is constructed on the basis of some fundamental assumptions (Jerne and Wood, 1949) .
In our studies on toxicities of pertussis vaccine, the mice-body weight decreasing toxicity (“toxicity ”) was first taken because it is adopted in the present regulation on the toxicity test of pertussis vaccine in Japan (Minimum Requirements, 1962) . And a possibility to adopt biological methods for reproducible as well as quantitative assay for the“toxicity”has been studied, considering the fundamental conditions for a valid comparative assay, and a method of estimation of the“toxicity”of a pertussis vaccine preparation was established (Kurokawaet al., 1962 a, 1962 b) .
Based on the informations about possible values of some parameters accumulated in the experiments, a form of assay for a routine work has been proposed (Kurokawaet al., 1962 b) as follows. The 3 point assay method shall be preferable, one dose for the test preparation and two doses for the reference, and a t least 15 mice are assigned for each dose. The preferable interval of the two doses for the reference shall be 4 times. The choice of the measure of the body weight change is the percentage of the body weight on the third day of vaccine inoculation to that on the day of inoculation, because it is easy to measure and calculate, and convenient to compare the variance between species, strains and ages, irrespective of difference in the body weight. The result shall be analysed statistically using the equations given for calculation of relative potency and its variance (Finney 1952) .
It was demonstrated that there existed a certain relationship between the relative“toxicity”of a vaccine preparation thus estimated and its outcome of the assay performed according to the current regulation, the relative“toxicity”of the vaccines rejected by the current regulation being larger than 0.33, and that of the vaccines passing it smaller than 0.5. As to the vaccines, of which the estimated relative“toxicity ”fell between 0.33 and 0.5, some of them had been rejected and others passed.
Our next step was to reform the assay procedure for toxicity of pertussis vaccine described in the present regulation, applying this method. In general, an assay procedure should involve a criterion for making decision on whether the test preparation should be accepted or rejected. In this criterion our proposed method requires somewhat tiresome computations of confidence limits to the relative“toxicity”from the result of every assay trial. In such a situation, we are interested in constructing a chart of critical regions, which allows to make the decision graphically.
In the present paper, several charts of critical regions for sequential and single test procedures of routine assay are presented.