四参数模型和平行线分析法在病毒疫苗抗原生物检定统计中的应用

Application of 4-parameter model and parallel line assay in biological test statistics for antigens of viral vaccines

摘要 目的 比较四参数模型与平行线分析法在Sabin株脊髓灰质炎灭活疫苗（Sabin inactivated poliovirus vaccine，sIPV）D抗原和肠道病毒71型（enterovirus 71，EV71）抗原生物检定统计中的应用。方法 对sIPV-D抗原和EV71抗原检测数据分别进行四参数模型和平行线分析法的方法验证，并将两种方法应用于不同含量样品检测结果的分析计算。结果 sIPV-D抗原检测数据四参数曲线拟合度好（决定系数为0.998，残差值≤0.059），平行性分析可靠（线性回归：参比品F=727.49，P〈0.01；样品F=1 169.44，P〈0.01。平行性方差分析F=0.24，P〉0.05）。EV71抗原四参数曲线拟合度好（决定系数为1.000，残差值≤0.025），平行性分析可靠（线性回归：参比品F=126.63，P〈0.01；样品F=192.74，P〈0.01。平行性方差分析F=4.00，P〉0.05）。对于sIPV和EV71不同抗原含量的生物检定统计，两种方法计算结果的差异无统计学意义（t=0.248～1.023，P值均〉0.05）。 结论 四参数模型和平行线分析法均能较好地应用于sIPV-D抗原和EV71抗原的ELISA检测体系。

Objective To compare 4-parameter model with parallel line assay method in the application in biological test statistics for antigens of Sabin inactivated poliovirus vaccine （sIPV） and enterovirus 71 （EV71）. Methods The 4-parameter model and parallel line assay were methodologically validated using the detection data of sIPV-D antigen and EV71 antigen. Then, both methods were used to calculate the detection result of samples with different antigen content. Results The 4-parameter model was fit for sIPV-D antigen detection data （coefficient of determination 0.998, residual value ≤0.059）, and parallel line analysis was reliable （regression： reference F=727.49, P〈0.01; sample F=1 169.44, P〈0.01. parallelism： F=0.24, P〉0.05）. For EV71 antigen, 4-parameter model was appropriate （coefficient of determination 1.000, residual value≤0.025）, and parallel line analysis was reliable （regression： reference F=126.63, P〈0.01, sample F=192.74, P〈0.01; parallelism： F=4.00, P〉0.05）. There was no significant difference between the calculation results of two methods for different antigen content of sIPV and EV71 （t=0.248-1.023, all P〉0.05）. Conclusion Both 4-parameter model and parallel line assay can be applied in the ELISA detection system of sIPV-D and EV71 antigens.